The Yale Law Journal

March 2015

Deviance, Aspiration, and the Stories We Tell: Reconciling Mass Atrocity and the Criminal Law

abstract. The historian Raul Hilberg once observed that we would all be happier if we believed the perpetrators of the Holocaust were crazy. But mass atrocity is never so simple. We may search in Germany, Bosnia, the Congo, or Rwanda for the madman or the deviant, but often we will find instead an ordinary person, one who commits a crime at the barrel of a gun or who succumbs to the awful indirect coercion that pervades entire communities in the throes of transformative violence. In the ashes of atrocity, criminal courts have been created, but many scholars have come to think that the basic structures of criminal law—built to address willful deviance from society’s norms—are inappropriate for dealing with the complex context of mass atrocity crimes.

This Article challenges this critique by making three contributions. First, it presents a novel descriptive account of how courts addressing mass atrocity crimes wrestle with the concept of deviance in criminal responsibility. Second, applying principles of domestic criminal law, the Article proposes a theory of “aspirational expressivism,” which envisions international criminal law as legitimately and positively setting forth aspirations for human behavior, rather than simply drawing a line between normalcy and deviance. Finally, the Article builds on the theory of aspirational expressivism to make the normative claim that courts can be more than forums for condemning the world’s horrors, as their role has been predominantly conceived. Instead, they can be—and should be—sites of storytelling, providing an opportunity for understanding how individuals choose to perpetrate unspeakable crimes, articulating how we hope people will behave in the most demanding of circumstances, and shaping our beliefs about the way we ought to behave under the unflattering light of the way we actually do.

author. Assistant Professor of Law, University of California, Berkeley, School of Law. This Article won the 2014 Association of American Law Schools Section on Criminal Justice Junior Scholars Paper Competition. For helpful comments and conversations, I am deeply grateful to Ty Alper, José Alvarez, Ken Bamberger, Nels Bangerter, Dan Bodansky, David Caron, Anupam Chander, Jen Daskal, Mark Drumbl, Avlana Eisenberg, Dan Filler, Laurel Fletcher, Stavros Gadinis, David Gartner, Andrew Guzman, Monica Hakimi, Angela Harris, Oona Hathaway, Kinch Hoekstra, Sandy Kadish, Chris Kutz, Máximo Langer, Katerina Linos, Larry May, Judge Gabrielle Kirk McDonald, Luis Moreno-Ocampo, Melissa Murray, Jens Ohlin, Darryl Robinson, Bertrall Ross, Andrea Roth, Peter Rush, Jonathan Simon, Gerry Simpson, David Sklansky, Avani Sood, James Stewart, Madhavi Sunder, Karen Tani, Jenia Turner, Kish Vina-yagamoorthy, Chuck Weisselberg, Andrew Woods, and participants at the ASU International Law Colloquium, Berkeley Law Faculty Retreat and Junior Working Ideas Group, Junior International Law Scholars Annual Meeting, Legal Scholarship 4.0 Conference, SMU Criminal Justice Roundtable, Vanderbilt Law School Faculty Workshop, UC Davis School of Law Faculty Colloquium, and the University of Melbourne Affective States in International Criminal Justice Symposium. Bita Assad, Musetta Durkee, Britt Harwood, Shayne Henry, and Maya Karwande provided outstanding research assistance. Ariela Anhalt and the staff of the Yale Law Journal provided excellent editorial assistance.


On July 16, 1995, twenty-three-year-old Dražen Erdemović shot and killed some seventy unarmed men and boys at the Branjevo farm in the Bosnian town of Srebrenica. Erdemović, a low-level soldier in the Bosnian Serb army, had been ordered to execute these individuals. When his commander, Brano Gojković, first instructed the members of the unit that they were to kill the civilians who would soon begin arriving at the farm, Erdemović refused. Gojković told him that he could either pick up his gun and start shooting, or line up with the victims and face death himself. Erdemović submitted to the order, drinking brandy as the hours went on to distract himself from the stench of bodies piling up in the hot sun.1

Twelve hundred men and boys were killed that day. The following year, Erdemović confessed his role in the slaughter to a French journalist. Soon after, he was charged with war crimes and crimes against humanity by the International Criminal Tribunal for the Former Yugoslavia (ICTY), the court created by the United Nations Security Council in 1993 to prosecute individuals responsible for atrocities committed during the Yugoslav civil wars.2 He had sought to assert a duress defense based on the threat against his life, but the court rejected his argument, holding that duress is not a complete defense for homicide and declaring him guilty of war crimes.3 Erdemović became the first defendant convicted by the ICTY.4

We typically think of the criminal law as punishing those who deviate from what society deems expected, normal, or good.5 And indeed, Erdemović did the unthinkable. He killed innocent people—some blindfolded, others watching their friends and neighbors slaughtered in line before them; some paralyzed by fear, others shouting defiantly until the moment they were silenced with a bullet. At the same time, Erdemović behaved in a way that, if we allow ourselves to imagine the darkest of moments, might be understandable. He had a gun to his head. He was scared and panicked. He feared what would happen to his wife and eight-month-old baby if he refused or ran. The judges who decided his fate recognized this, asserting that any person facing such a threat to his life would react the same way. They recognized the “human frailty” of us all and insisted that they would not “‘expect’ a person whose life is threatened to be [a] hero and to sacrifice his life by refusing to commit the criminal act demanded of him.”6 The court sentenced Erdemović to five years in prison, a conspicuously lenient punishment for the murder of seventy people; yet still, it branded him a criminal.7

This decision has become a touchstone for those who argue that criminal law as we know it cannot adequately address the ugly realities of mass atrocity. If Erdemović did what any person would have done—if, indeed, we can understand the choice he made—then what does it mean to convict him of a crime? Many critics of the decision, and of international criminal law more generally, describe it as hypocrisy, bristling at the uncomfortable juxtaposition of the criminal law’s sanction and the notion that any person, in the right circumstances, might commit an atrocious act.8 To understand all is to forgive all, the saying goes. But to conceptualize the law as a tool of either condemnation or understanding is to create a false dichotomy.

This Article posits an alternative interpretation, one that envisions a role for the criminal law in both denouncing mass atrocities as crimes and conceding that these crimes are often committed by ordinary people—not “alien others,” not madmen, not deviants.9 By understanding the criminal law as an instrument that legitimately sets forth aspirations for human behavior in trying situations, rather than one that simply reflects what is identified as normal or typical, this Article argues that international criminal courts need not go to the lengths they do to paint the defendants in front of them as deviants. And by recognizing the normalcy of those perpetrators, the courts can be more than forums for voicing outrage against the world’s horrors, as they have been principally conceived.10 Instead, these courts can serve as sites of storytelling, providing both an opportunity to derive insights into how individuals choose to perpetrate crimes and an occasion for articulating how we hope people will behave in the most hopeless of circumstances.

This Article intervenes in a longstanding debate over the appropriateness of domestic criminal law structures for offenses under international law. Many scholars and practitioners in the field have noted that although international criminal law shares traits with a national system of criminal law, it differs in significant ways from “ordinary” criminal law. International criminal law exists to address situations of exceptional violence, terror, and despair: organized massacres of hundreds of thousands, systematic rapes and mutilations motivated by hatred for the ethnicity or religion of the victims, forced pregnancy, enslavement, torture. International criminal law is seen as unique, as distinct from ordinary criminal law, because of the widespread savagery it must confront; instead of ordinary crime, it targets extraordinary crime.11

The average perpetrator, too, is said to be different in the international criminal law context. In contrast to the perpetrator of ordinary domestic crime, whose actions deviate from those of other members of society, the perpetrator in mass atrocity commits crimes alongside masses of other individuals who are doing the same. Whereas the perpetrator of ordinary crime acts in violation of widely accepted social norms, the perpetrator of mass atrocity participates in acts that, in his community, fall squarely within the parameters of widely accepted social norms.12 For individuals in conflict settings, participation in group crime may fortify a sense of community and belonging when society is otherwise disintegrating. In the time, place, and society in which they are committed, some crimes—even the most horrific of crimes—may be quite frighteningly normal.13

These unique features of mass atrocity have been the subject of a rich scholarly discussion. Mark Drumbl, Laurel Fletcher, Christopher Kutz, and Gerry Simpson, among others, have examined questions of deviance through the lenses of criminology, social psychology, philosophy, and sociology, inspiring a robust debate about whether the reality of mass atrocity is adequately addressed by the law that intends to respond to it.14 This Article steps into that conversation and seeks to transform it by offering three contributions to the existing scholarship: (1) a novel descriptive account of how international criminal tribunals conceive of the role of deviance in criminal responsibility; (2) a new theoretical understanding of expressivism in international criminal law; and (3) based on that theory, a normative critique of the courts’ approach to deviance and a path to a more productive role for the law in the wake of mass atrocity. Throughout its analysis, this Article takes the unconventional approach of drawing examples from domestic criminal law—from provocation to sexual assault to drunk driving—to disentangle some of the confusions that have beset international criminal law. This methodology is deliberate, built on a belief that the project of international criminal law is at its core an attempt to answer the same questions about choice and responsibility that arise in domestic criminal law. Accordingly, the ultimate aim of this Article—to illuminate in criminal law an opportunity to understand how individuals make decisions and to model behavior rather than just condemn it—finds focus here in the context of mass atrocity, but the Article has implications for situations of pressure and horror confronted by domestic criminal law as well.

This Article proceeds in four Parts. Part I briefly describes the critique of international criminal law that I refer to as the “deviance paradox”: the notion, put forward by many scholars, that international criminal law is ill suited to assign responsibility to perpetrators of mass atrocity crimes, many of whom are not deviant in the same way that perpetrators of ordinary crimes are. To explain why those who study mass atrocity maintain that these extraordinary crimes are understandable—even as they are horrific—this Part presents social science and historical research that suggests that social pressures can drive ordinary people with no propensity for violence to commit violent acts. Stanley Milgram’s shock experiment, Philip Zimbardo’s Stanford prison experiment, and Christopher Browning’s study of the ordinary men responsible for rounding up and killing over one million people in Nazi Germany provide powerful evidence to support the theory that in the right circumstances, evil can be quite ordinary.

Part II offers a look at how international courts have grappled with the role of deviance as they assign criminal responsibility. Based on a study of all final judgments and sentencing decisions and appeals therefrom in the ICTY and its sister court, the International Criminal Tribunal for Rwanda (ICTR),15 this Part shows that the emphasis on categories of deviance and normalcy that has dominated scholarship in this area also has infused the work of the courts. I establish this through two observations. First, I demonstrate how the courts, seeking to avoid aspirational standards and at the same time accepting the notion that ordinary people commit terrible crimes, create categories of individuals who, they assert, are not ordinary and thus should be expected to behave differently from the norm. The courts contend that even if the average person might succumb to coercion, the soldier should not; or that even if the average person might lose himself in a whirlwind of violence, the well-educated one should not. Second, I show that the courts, by augmenting punishments for leaders who have exploited an impressionable population’s trust, implicitly endorse the notion that ordinary people succumb to pressures to obey and conform. Put together, the acknowledgment of human frailty in atrocity and the accompanying expectation of better behavior from certain categories of people enable the courts to insist that even though perpetration of these crimes in these settings may be quite typical, for these defendants, it is a mark of deviance.

Part III disputes a central foundation of scholars’ notion of the deviance paradox and courts’ apparent understanding of responsibility in mass atrocity—that criminal law punishes behavior diverging from the ordinary—and introduces the theory of aspirational expressivism. I posit that criminal law is better understood as an instrument that often sets out aspirational goals rather than one that merely punishes deviation from ordinary behavior. The basic idea of aspirational expressivism will sound familiar to anyone immersed in Anglo-American criminal law, through concepts such as the reasonable person and casebook classics like Regina v. Dudley and Stephens.16 But this understanding of aspirational expressivism has been obscured in international criminal law, as scholars and practitioners in the field have envisioned the law’s capacity to punish legitimately as contingent on the offender’s deviation from the ordinary. Criminal law, however, operates not only in this negative space, identifying acts that individuals ought not undertake, but also in a positive space, proffering ideals of how individuals ought to behave—even if they typically do not.

Based on this understanding of criminal law, Part IV urges that international criminal courts interpret their work through the lens of aspirational expressivism, which would allow them legitimately to punish deviations from aspirational standards. The courts could then both condemn those who succumb to situational pressures and admit the normalcy, however horrific, of crime in certain situations—without having to devise arbitrary categories of individuals who should be expected to behave better than the average person. International criminal courts should acknowledge that even if the average person may behave just as a defendant did, the criminal law may still legitimately punish, as the law does more than just reflect average behavior: it can function as a voice of our moral imagination and move us to aspire beyond the ordinary.

This approach reveals a way out of the deviance paradox. By recognizing the criminal law as an instrument that can legitimately set out aspirational ideals of behavior—rather than fearing that an absence of deviance in the accused weakens the operation of the criminal law—we can see the normalcy of violence as a feature that makes that violence an even more appropriate target for the criminal law. To accept that criminal law regularly establishes aspirational standards for its subjects is to accept that criminal courts can acknowledge the ease with which individuals slip into violent conduct and in the same breath demand that they behave otherwise.

Recognizing the capacity for criminal law to punish deviations from aspirational standards not only provides a normative justification for prosecutions and convictions of seemingly ordinary individuals, but also creates a space for criminal judgments to assist in understanding how mass atrocity takes place. If international criminal courts admit that the defendants before them were drawn to violence by social forces rather than straining to insist that there was something deviant about the offenders, judges could use the trial process and the decisions they produce to explore how ordinary people came to commit horrific acts. After decades of “never again,” the community that has created international criminal courts needs to think about atrocity prevention in more nuanced ways. Examining the motivations of individual perpetrators has far more promise for contributing to an end to atrocities than do prison sentences meted out to a handful of offenders with no attempt to comprehend how they metamorphosed from ordinary people to extraordinary criminals.

Part IV concludes with a response to counterarguments. First, it considers the contention that punishment of an ordinary person making understandable choices is unfair. Criminal law often condemns such a person: when new laws punish conduct that remains common, such as texting while driving or even sexual intercourse in the absence of affirmative consent, the criminal law is punishing ordinary people who made understandable choices. We are thus already accustomed to criminal law extending its reach to such individuals in its aspirational norm-setting functions. Moreover, punishment of individuals who engaged in common or understandable behavior can be made more palatable through nuance and individualization in punishment.Beyond this, the cornerstone of criminal responsibility and punishment should not be typicality of the offender or the offense, but rather the exercise of choice and judgment by a person who engages in conduct prohibited by the criminal law. If we understand the criminal law as justified by its aspirational expressive purposes, then punishment of such individuals despite their ordinariness can serve to confirm that even in the commission of mass atrocity crimes, there are opportunities for choice, and there are opportunities to choose differently from the norm.

Second, Part IV addresses arguments that admitting the role of social and environmental forces in motivating perpetrators of mass atrocity would compromise the legitimacy of international courts. Critics argue that recognizing the larger contextual factors at work in driving perpetrators would dangerously draw attention to the role of foreign states and international organizations in creating the situation that led to the violence in the first place. This argument has some purchase. Consider the rotten social background defense of American criminal law, which would provide a partial defense to those who could show that their criminal acts were in part attributable to socioeconomic deprivation.17 One explanation for why the defense never got off the ground in the United States was its uncomfortable acknowledgment that the system that punishes an individual is the very same system that contributed to that individual’s wrongdoing.18 Nonetheless, I argue that conceding the role of the system in contributing to mass atrocity crimes can enhance the legitimacy of the courts rather than detract from it. I work through this contention by drawing lessons from domestic criminal law to propose that we should accept that institutions that contribute to crimes can also, and do also, contribute to their prevention. Recognizing the responsibility of bystander states and organizations in the perpetration of atrocities need not vitiate the responsibility of the individual perpetrators; the two can coexist. Moreover, recognizing the responsibility of those states or organizations may generate a more responsibly involved international community, one that contemplates not merely a duty to intervene to protect human rights, but also the longer-term and more complex consequences of intervention for the politics and stability of societies.

Ultimately, this Article is an examination of the stories the law puts forward about what it means to live through the world’s greatest horrors, and the way violence does and does not change us. It is an account of law’s most basic functions—not simply resolving disputes or determining guilt or innocence, but also shaping our beliefs about what is typical and deviant, about normal human weakness and extraordinary aspiration, about the way we ought to behave and the way we actually do.

I. the deviance paradox

One of the most ubiquitous critiques of the project of international criminal justice is the contention that the criminal law, built as it is around a foundational belief in responsibility for willful deviation from society’s norms, cannot properly account for the fact that the ordinary perpetrator of mass atrocity crimes is normal rather than deviant. I refer to this contention as the “deviance paradox,” deliberately choosing the word “paradox”—a seemingly irreconcilable situation that in fact has a solution—to indicate that there is indeed a way out of the conundrum. This Part describes the deviance paradox. After situating the paradox in international criminal law’s foundation in individual responsibility, this Part discusses social science and historical research upon which scholars rely to support their contention that in times of extraordinary violence, to commit a crime is to be normal.

A. Individual Responsibility

International criminal law is, in essence, the law of atrocity. Encompassing prohibitions against genocide, crimes against humanity, war crimes, and aggression,19 this body of law originated in the International Military Tribunal at Nuremberg, which tried twenty-two military and political leaders of Nazi Germany.20 Since that time, international tribunals have been established to prosecute individuals for crimes committed during the disintegration of Yugoslavia and the 100-day genocide in Rwanda, and now a permanent International Criminal Court has jurisdiction over atrocities committed around the world.21

International criminal law is distinct from any municipal criminal law in the narrowness of the crimes under its jurisdiction, but it resembles domestic criminal law in its general structure and fundamental principles, including its central principle of individual, as opposed to collective, responsibility. Whereas international law traditionally contemplates responsibility for only the state, rather than the individuals who run the state, international criminal law seeks accountability exclusively for individuals.22 Scattered precedents for this focus on individuals existed prior to World War II. The Treaty of Versailles, for example, charged Kaiser Wilhelm with the “supreme offence against international morality and the sanctity of treaties” and declared the Allies’ intention to prosecute him, while it also granted the Allied Nations the right to try additional individuals for “violations of the laws and customs of war.”23 Only in the aftermath of the Second World War, however, did leaders begin to consider the normative implications of individual responsibility as opposed to state responsibility. The declaration of the Nuremberg Tribunal that “[c]rimes against international law are committed by men, not by abstract entities” has become something of a slogan for the core tenet of international criminal law that blame should be placed on individuals, not on states or militaries or political organizations.24 For the architects of Nuremberg, this focus on individuals was instrumental: trials of individuals could pinpoint those directly responsible for the crimes of war, thus affirming that not all Germans were guilty.25

When international trials for mass atrocity were revived in the 1990s, this emphasis on individual responsibility persisted. The creators and supporters of the ICTY and ICTR pushed for individual responsibility in part because they believed that doing so could halt the cycles of collective blame that caused the atrocities in the first place. As stated in the ICTY’s first annual report, “If responsibility for the appalling crimes perpetrated in the former Yugoslavia is not attributed to individuals, then whole ethnic and religious groups will be held accountable . . . . [C]linging to feelings of ‘collective responsibility’ easily degenerates into resentment, hatred and frustration and inevitably leads to further violence and new crimes.”26 That is, if all Serbs are responsible for the massacres at Srebrenica, then peace can never exist between Serbs and Bosniaks; but if Milošević and Mladić and Karadžić and a handful of others are responsible, then perhaps the larger communities can coexist.27

B. Collective Perpetration

1. The Social Dynamics of Mass Atrocity

Although individual responsibility has been a source of comfort for those who fear the consequences of collective attributions of guilt, it has dismayed many scholars, too, because individual perpetration of mass atrocity is so often influenced by persons other than the individual. These criticisms often center on a perceived distinction between the perpetrator of “ordinary crime,” which “tends to be deviant in the times and places it is committed,”28 and the perpetrator of mass atrocity crime, which is “not so obviously deviant.”29 The distinction stems from two dynamics that are typical of mass atrocity. First, individuals who commit mass atrocity crimes often are doing the same thing as masses of others. In contrast to the paradigmatic perpetrator of municipal criminal law, whose criminal behavior diverges from most of the rest of society, the perpetrator of mass atrocity crimes is often acting no differently from those around him.30 By many accounts, hundreds of thousands of perpetrators participated in the genocide in Rwanda,31 and conservative estimates put the number of direct participants in killings in the Holocaust at over one hundred thousand.32 The individual killer looked no different from the masses; murder was consistent with descriptive social norms, at least in the narrowest vision of his community.33 One might argue that this fact has no salience when the harm caused is so grave, but to many critics, something intuitively unsettling attends the idea of labeling a person a criminal when everyone around him is doing the same thing. As Douglas Husak explains, “Each of us would be likely to feel somewhat indignant if blamed (in a moral context) or punished (in a criminal context) for conduct that everyone does.”34

Second, mass atrocity crimes are often prosocial in nature, intended to benefit others, and they are typically consistent with social norms rather than contravening them, at least at the community level. Even as killing, rape, or displacement violate both laws and norms at the national or international level, for the perpetrators these acts are moral insofar as they are undertaken in support of their community, often to protect the group against another population that is perceived as a threat.35 Moreover, the perpetrator of mass atrocity typically performs his actions in relation to some other actor; he may be responding to a direct order, feeling ignited by exhortations to slaughter, or succumbing to peer pressure in an overwhelming atmosphere of violence.36 Those criminal acts, then, are consistent not only with the community’s descriptive norms, but also with its injunctive norms.37

2. Authority and Conformity

These two dynamics of mass atrocity—the similarity of the average perpetrator to those around him and his conformity to local social norms—are particularly troubling to many scholars of international criminal law in light of observations made in the situationist tradition of social psychology. This theory of behavior ascribes individual choices not only to personality, but also to external circumstances: “[S]ocial behavior is, to a larger extent than people commonly realize, a response to people’s social context, not a function of individual personality.”38 In contrast to a belief that criminal behavior is rooted in bad character and that criminality is a fixed trait, situationism holds that other people can provoke behaviors in individuals. Situationist research demonstrates that individuals who would not otherwise commit certain acts can “be led, by the right set of subtle and not so subtle situational pressures and constraints, to commit similar transgressions or more generally to do things that they would condemn others for doing and that they believe themselves incapable of doing under any circumstances.”39 Critiques of international criminal law have seized on situationist research to argue that perpetrators of mass atrocity, who commit crimes in groups, do not exercise agency and free will to the extent that a legitimate exercise of criminal law would demand.40

Of particular interest to studies of mass atrocity is research on how individuals are affected by pressure from authority figures or peers, circumstances that often accompany widespread violence. Stanley Milgram’s obedience experiments powerfully demonstrated how an impulse to submit to authority influences human behavior. In this study, volunteers played the role of teacher, while another group played the role of learner; unbeknownst to the teachers, the learners were in on the experiment. The experimenters instructed the teacher to ask questions of the learner and, when the learner provided an incorrect answer, to administer an electric shock to that learner with increased intensity for each incorrect answer.41 The learners, of course, did not actually experience any shock, but the teachers believed they did. The learner would grunt at 75 volts, beg to be released at 150 volts, and scream in agony at 285 volts. At 330 volts, the learner would no longer answer questions; the teachers, meanwhile, were directed to treat silence as an incorrect answer.42

The study offered an opportunity to examine the effect of an authority figure—the experimenters, in this case—on the individuals in the role of the teacher. When the teachers expressed reluctance to proceed with the shocks, the authority figures told them with intensifying urgency to persist, from a request to “[p]lease continue,” to a statement that the “experiment require[d them] to continue,” to a directive that the teacher “must go on.”43 Despite the teachers’ discomfort with inflicting pain on the learners, when the authority figures told them to continue, most did. Ultimately, sixty-five percent of the participants in the study obeyed the orders they were given up through the maximum voltage.44 From the base experiment and a number of variations, Milgram concluded that the presence of an authority figure increased compliance by the teachers despite their discomfort with their actions.45 According to Milgram:

[O]rdinary people, simply doing their jobs, and without any particular hostility on their part, can become agents in a terrible destructive process. Moreover, even when the destructive effects of their work become patently clear, and they are asked to carry out actions incompatible with fundamental standards of morality, relatively few people have the resources needed to resist authority.46

Philip Zimbardo’s Stanford prison experiment, which also illuminated the impact of situation on human behavior, has been of interest to scholars of mass atrocity as well.47 Zimbardo set up a mock prison in the basement of the campus’s psychology building and assigned undergraduate volunteers, who had been screened for psychological abnormalities, to play prisoners or guards. Individuals in the role of guard quickly became abusive to the prisoners.48 The guards “repeatedly stripped their prisoners naked, hooded them, chained them, denied them food or bedding privileges, put them into solitary confinement, and made them clean toilet bowls with their bare hands.”49 Embracing their sheer power over the prisoners, the individuals playing the role of guard came to embody what they believed to be the way a guard behaves. From the participants’ quick recourse to cruelty, Zimbardo concluded that “[h]uman behavior is much more under the control of situational forces than most of us recognize or want to acknowledge.”50

While Milgram and Zimbardo offered observations on the role of authority, Solomon Asch’s experiments in the 1950s demonstrated the power of the desire to conform. Asch conducted studies in which he placed volunteers in a room with several “confederates”—actors who were cooperating with Asch in the experiment without the knowledge of the volunteers. Asch showed the groups two cards, one with one line and the other with three lines of varying lengths. Each person in the room was asked to identify out loud which line on the second card matched the length of the line on the first card. Prior to the experiment, all confederates had been instructed to give the same response; the volunteer always answered toward the end, after hearing most of the confederates’ answers. For the first two trials, the confederates gave the correct answer. On the third trial, they gave the incorrect answer. In more than 33% of the studies, volunteers also gave an incorrect answer to that third question, conforming to the confederates’ incorrect responses.51 Asch also set up a control group, in which no confederates were present. Individual volunteers responded to the questions on their own, with only the experimenter in the room, so that volunteers faced no pressure to conform to their peers’ answers. In the control group, volunteers gave the correct answer 92.6% of the time. The volunteers’ incorrect answers thus appeared to be given in response to the confederates’ incorrect answers, as opposed to a genuine mistake.52

Based on interviews with the participants, Asch concluded that the volunteers who conformed to the confederates’ incorrect responses did so for three types of reasons. The largest number experienced a “distortion of judgment”: they questioned their own judgment when they heard their peers’ responses. Smaller numbers of participants experienced a “distortion of perception”—they actually believed the lines were the same length and that the answers they gave were correct—or a “distortion of action”—they knew their answers were incorrect but decided to go along with the majority nonetheless.53

The observation that individuals change their behavior in response to situations—and even participate in acts that cause them feelings of moral discomfort—is replicated in historical research. In his groundbreaking work Ordinary Men,historian Christopher Browning chronicles German Reserve Police Battalion 101, which carried out the massacre and deportation of thousands of Jews in Poland during the Nazi occupation. Neither the rank and file nor the officers showed any characteristics that made them likely to engage in brutal violence.54 When the perpetrators committed these horrors, neither they nor their families faced any threats to their safety; indeed, commanders gave them the option not to participate.55 In Browning’s account, these were not motivated killers; they were, instead, ordinary people.56 Similarly, the work of journalists such as Philip Gourevitch and Jean Hatzfeld established that the participants in the Rwandan genocide were not experienced killers, but rather regular people—teachers, doctors, priests, children.57 As one survivor explained, “Killing became an ordinary activity, since our elders and everyone did it.”58

The Holocaust historian Raul Hilberg once asked, “Wouldn’t you be happier if I had been able to show you that all the perpetrators were crazy?”59 Of course, Hilberg was unable to do that, and the work of psychologists, sociologists, historians, and legal scholars suggests that, far from crazy, many of the individuals who commit the most horrifying acts are regular people who succumb to the pressure of situational coercion and the oppression of authority, people who had no prior intention to do anything wrong.

3. The Tension Between Individual Responsibility and Collective Perpetration

These insights from psychology and history are particularly troubling to scholars and observers of international criminal law because they suggest that the average perpetrator of mass atrocity60 is descriptively deviant neither in the smaller community in which he was operating, nor in the larger context of human nature. Indeed, he was doing what we might expect an average person to do in that situation.61 Because the perpetrator of mass atrocity crimes is normal, not deviant, some scholars argue, international criminal law and its foundation in individual responsibility cannot properly grapple with the reality of mass atrocity.62 Herein lies the deviance paradox: international criminal courts, scholars contend, deal in a system of law predicated upon the idea of a willfully deviant criminal, but the defendants they address are not willfully deviant within the context of their societies.63 To many scholars, a central problem of the international justice regime is that criminal law is utterly incompatible with the nature of wrongdoing in atrocity.64

To be sure, the creators of international criminal law have attempted, and scholars have encouraged, the creation of methods to address some of the dynamics of perpetration. One approach has been to turn an individualistic criminal law into one better equipped to address the activities of individuals within groups. This effort began at Nuremberg. Recognizing the ways in which political and military organizations facilitated the crimes of the German state, the drafters of the London Charter took account of the collective dynamics of atrocity and allowed the judges to declare the criminal guilt not only of the individual defendants, but also of particular organizations.65 Ultimately, the tribunal declared three entities—the Schutzstfaffel (SS), the Gestapo/SD, and the Leadership Corps of the Third Reich—criminal organizations.66 The legal concept of criminal organizations did not survive,67 but contemporary international criminal law features other doctrines built on an understanding of the importance of collectives in perpetration of mass atrocity. For example, joint criminal enterprise allows a person to be held responsible for a crime when he participates with others in a common plan and when commission of that crime is foreseeable;68 and co-perpetration assigns liability where two or more persons “each contribute to the commission of the crime” and one “‘could frustrate the commission of the crime by not carrying out his or her task.’”69 Nonetheless, although they account for the ways in which collectives operate in the context of mass atrocity,70 these doctrines are controversial, and they might fail to adequately consider the ways in which individual agency is affected in collective situations.71 Indeed, these doctrines enable courts to assign guilt based on an individual’s association with an act even in the absence of direct participation, whereas those who are concerned with the deviance paradox are often interested in ways to mitigate the blameworthiness of individuals who are enveloped by situations of violence.

Prosecutorial discretion, too, might be expected to play a role in alleviating some of the problems posed by the deviance paradox. To the extent that the individuals prosecuted are targeted on account of their willful deviation from social norms, singling them out as reprehensible would not alarm those who worry that prosecution of the non-deviant offender might undermine the foundations of international criminal law. Prosecutorial discretion, however, can only go so far in softening the blow of the deviance paradox. For one, international criminal law exists outside of the particular individuals who are prosecuted in international courts: even if the courts only touch a handful of people, the law, properly understood, still applies to anyone whose conduct is proscribed, prosecution or not. Moreover, the ripple effects of international accountability are felt in national courts, where criminal prosecutions have targeted low-level individuals in cases guided at least in part by international laws. As a result, even if the low-level soldier with a gun to his head, or the otherwise peace-loving teacher who got swept up in the violence around him, will not be haled before an international tribunal ever again, he might still be fair game for a national tribunal; and even if not, the notion that he could be subject to prosecution, in either a national or an international court, continues to trouble those who question the appropriateness of allowing criminal punishment to stretch its fingers into the realm of actions that many, perhaps most, could have done.

II. deviance in mass atrocity: three stories

The prominence of questions of deviance and normalcy in scholarly studies of international criminal law prompts consideration of whether courts, too, consider these kinds of dynamics. Accordingly, this Part examines how international criminal courts conceive of the role of deviance in assigning criminal responsibility. Based on a reading of the final judgments, sentencing judgments, and appeals therefrom in the ICTY and ICTR, I argue that the international criminal courts, like many of the scholars who study them, have seized on the categories of deviance and normalcy, and in their decision making they emphasize the ways in which the defendants before them differ from the average person and thus may be treated as deviant.

I offer three sketches of individuals who committed horrific crimes. I first return to Dražen Erdemović, whose story began this Article’s exploration of the deviance paradox. I then turn to individuals who, the courts say, should have known better because of their privileged education or upbringing. Finally, I look at individuals who were found to have taken advantage of an impressionable audience. In each section, I situate the acts of these individuals and their treatment by the tribunals in the context of similar cases. Challenging the dominant understanding that the tribunals fail to contemplate the deviance of offenders in mass atrocity situations,72 this Part devises an original descriptive account to explain how the tribunals approach the dynamics of individual responsibility in mass atrocity. Indeed, rather than ignoring dynamics of deviance and normalcy, the tribunals assess culpability through the lens of those categories.

Before turning to the cases, I offer a caveat. While the cases discussed reflect a broader trend in the tribunals, the analysis that follows is not meant to suggest that in every case the tribunals are guided by one consistent or coordinated approach to deviance. The trial chambers of the ICTY and ICTR operate separately, and decisions are made by many different judges.73 Moreover, the courts do not uniformly follow the approach I describe. In a few cases, for example, education or cultural background is used as a mitigating rather than an aggravating factor.74 Sentencing judgments are, moreover, scattered and vague; even when an aggravating factor is mentioned, there is no indication of whether it lengthens a sentence or is merely being noted.75 My aim is not to contend that one approach governs or to make a quantitative claim about, for example, the frequency with which certain features are treated as making a person more or less blameworthy. Instead, I intend in this Part to point out one approach that guides the tribunals in many situations and to draw from these observations a larger argument about our understanding of the criminal law and its vast possibility as an instrument of change.

A. The Reluctant Executioner

No individual defendant in international criminal law demonstrates the difficulty of the deviance paradox more sharply than Dražen Erdemović. Perhaps for that reason, arguably no decision in international criminal law is more notorious than the judgment in Erdemović’s case. He enlisted in the Bosnian Serb army in 1994, after brief stints in both the Bosnian and Croatian militaries soon after the eruption of the Bosnian civil war.76 He had never seen combat, and he was taken by surprise when, on a hot day in July 1995, his commanding officer instructed him and his fellow soldiers to shoot the Muslim men and boys who would begin arriving at the farm by bus that morning.77 Erdemović initially refused: “I said immediately that I did not want to take part in that and I said, ‘Are you normal? Do you know what you are doing?’”78 His commanding officer gave him a choice: either cooperate and kill, or, “if [he was] sorry for [the victims], line up with them” and be killed himself.79 Erdemović reluctantly agreed to participate.80 According to his own estimate, he killed some seventy unarmed men and boys that day.81

When the war ended a few months later, Erdemović left the army, tormented by what he had done at Branjevo farm, and he confessed his crimes to a journalist.82 Within days, he was arrested by Yugoslav authorities and charged by the ICTY with crimes against humanity and, in the alternative, war crimes.83 After he pleaded guilty and was sentenced to a ten-year prison term,84 Erdemović appealed on the grounds that his guilty plea had not been informed and unequivocal.85 Because Erdemović insisted when he was entering his plea that he had killed only because he was coerced, the Appeals Chamber of the ICTY had to examine the question of whether a claim of duress contradicted an admission of guilt.86 The Appeals Chamber thus had to grapple with the novel question of whether duress could be a complete defense to the killing of innocent persons or whether it could only offer mitigation at sentencing.87

Confronted with this issue of first impression, the Appeals Chamber undertook a detailed assessment of the meaning of the duress defense and its availability in this case. The Chamber ultimately split, with three of the five judges determining that duress could not be a complete defense to homicide.88 The majority’s conclusions were primarily explained in a separate opinion by Judge McDonald and Judge Vohrah, who analyzed the question on both doctrinal and policy grounds. In its evaluation of the doctrine, the opinion found that no unequivocal rule of customary international law existed as to the availability of duress as a defense to homicide.89 The opinion then turned to an examination of state practice and concluded that civil-law states generally allow duress to serve as a defense to homicide, whereas common-law jurisdictions sometimes limit the defense to non-homicide crimes.90 Based on this assessment, the majority concluded that it could not find a general principle of law recognized by civilized nations that would allow an assertion of duress as a complete defense to homicide crimes under international law.91

Having found no definitive rule of international law either prohibiting or allowing a duress defense, the opinion turned to policy considerations.92 The opinion of Judge McDonald and Judge Vohrah emphasized that the law “must serve broader normative purposes in light of its social, political and economic role.”93 Given that domestic jurisdictions concerned with mere “ordinary crimes” would preclude a defendant from asserting a duress defense, they argued, the law could not recognize such a defense for extraordinary crimes:

If national law denies recognition of duress as a defence in respect of the killing of innocent persons, international criminal law can do no less than match that policy since it deals with murders often of far greater magnitude. If national law denies duress as a defence even in a case in which a single innocent life is extinguished due to action under duress, international law, in our view, cannot admit duress in cases which involve the slaughter of innocent human beings on a large scale.94

Although it focused in this passage on the magnitude of violence at issue in international crimes, the opinion ultimately relied on a moral principle, rather than a question of quantity, to justify its decision to preclude a duress defense for Erdemović. The opinion declared that its ultimate goal was to enforce international humanitarian law, which is concerned with “the protection of humankind.”95 To allow a soldier to assert a duress defense against charges of murder would inevitably undermine that body of law by allowing him to prioritize his own safety above that of innocent civilians. Accordingly, the decision set out “an absolute moral postulate which is clear and unmistakable for the implementation of international humanitarian law”: no matter how dire the circumstances, and no matter how useless resistance to a threat may be, duress may not be a defense to the killing of innocents.96

Commentators both inside and outside the Tribunal have interpreted the opinion of Judge McDonald and Judge Vohrah as setting a standard of behavior that no person could live up to, as calling for heroism in circumstances under which only an extraordinary person could ever be a hero.97 In dissent, Judge Cassese lodged a scathing criticism of the majority’s judgment. Judge Cassese would have allowed a duress defense for two reasons. First, because no specific rule of customary international law prohibits the use of a duress defense against crimes against humanity or war crimes based on the killing of innocent persons, Judge Cassese argued that the general rule of allowing a duress defense for any crime should apply.98 Second, Judge Cassese differed from the majority in his approach to the purpose of the criminal law. In his view, “[l]aw is based on what society can reasonably expect of its members. It should not set intractable standards of behaviour which require mankind to perform acts of martyrdom, and brand as criminal any behaviour falling below those standards.”99 To Judge Cassese, Erdemović did what any person in his situation would have done. As a result, to describe that conduct as criminal would pervert the meaning of the criminal law.100

The fact that these victims would have died even without Erdemović’s participation was particularly meaningful to Judge Cassese101 and to Judge Stephen, who also wrote a dissent. Judge Stephen contended that there simply was no real choice available to Erdemović, as “the desire for self-preservation is not merely instinctive but rational.”102 Accordingly, he wrote, “a law which would require [that desire] to be contradicted is not consistent . . . with a rational system of law.”103 Outside of the tribunal, too, the Appeals Chamber’s refusal to allow Erdemović to assert a duress defense came to be understood as an unrealistic demand for heroism and a contortion of the basis for criminal law.104

This understanding of the majority opinion, however, misses an important feature of its reasoning. Instead of demanding one standard of behavior, the decision set out two: the standard of the ordinary person and the standard of the ordinary soldier. By constructing these alternative baselines of behavior, the opinion asserted that the criminal law should not demand heroism and insisted that the Appeals Chamber was merely criminalizing deviation from the ordinary, even as the court indeed demanded heroism from Erdemović.

The opinion directly took on the argument that “the law cannot demand more of a person than what is reasonable,”105 focusing first on the question of whether the court should allow a defense when the accused participates in the killing of victims who would be killed even without his participation.106 Rather than disputing that the ordinary person would yield to such a threat, the judges conceded that it is indeed unreasonable for a person to sacrifice his life when doing so would not save the lives of victims who would be killed despite that person’s resistance.107 Beyond this, the opinion admitted that it would be unreasonable for a person to yield to a threat of death even if the victims would not be killed anyway. Indeed, the decision stated that the person who, “when faced with a threat to his child’s life, . . . decid[es] to obey a command to shoot innocent persons in order to save the life of his child” is “act[ing] reasonably.”108 Thus, the decision did not expect or even imagine that a person would resist threats to his life or his family; instead, it accepted that any ordinary person in these circumstances would yield to the threat of death and take innocent lives.

Nonetheless, despite asserting that the ordinary person would yield to a threat of death even if that meant killing innocent people (and that an ordinary person would yield to a threat of death if those innocent people would be killed regardless), the opinion still reached the conclusion that Erdemović should have resisted—but not because the judges expected Erdemović to be a hero, they claimed.109 Instead, Judge McDonald and Judge Vohrah insisted that what they asked of Erdemović was not heroism, but ordinary behavior. The opinion conceded that a person faced with a threat of death would submit. But a soldier faced with death is another matter.

Early in their opinion, Judge McDonald and Judge Vohrah stated that it was confining its analysis to the question of whether duress provides a complete defense for a soldier who has been charged under international law with killing innocent persons.110 Notably, the prosecution had not restricted its arguments in this way; instead, it had examined the question of the availability of the duress defense to any defendant. It was the choice of Judges McDonald and Vohrah to inquire as to the availability of the defense to a soldier in particular.111 Restricting the analysis to the issue of a soldier faced with a threat of death, rather than evaluating any personunder the same circumstances, allowed the majority to transform Erdemović’s behavior from normal, understandable human behavior into deviant behavior that diverged from the conduct expected of him.112

The opinion suggested that Erdemović did what any person would have done; threatened with death, he killed people who would have been killed even if he had resisted.113 But if that is the case—if Erdemović did the same thing a reasonable person would have done in the same situation—then how can a criminal conviction and prison sentence be justified? Judges McDonald and Vohrah resolved this tension by turning Erdemović into a deviant, casting him not as an ordinary person, but as an ordinary soldier. Even though the ordinary person would yield to such a threat, Erdemović belonged to a category of person of whom we will—indeed, we must—ask more. Accordingly, the decision was able to maintain that it was not asking for heroism, as the dissent accused, but rather was simply punishing a deviation from ordinary behavior. To do so, however, the decision shifted its lens so that Erdemović’s actions diverged from the standards expected of him.

B. The Lapsed Cosmopolitan

Erdemović is an especially hard case in international criminal law. The tribunals typically do not indict such low-level perpetrators,114 and Erdemović faced direct coercion in a way that other defendants generally have not.115 Nonetheless, the ICTY’s approach of carving out categories of actors who can be expected to behave differently from—that is, better than—the ordinary person represents a larger trend. Just as the ICTY asserted that soldiers should be expected to have considered the possibility that they would lose their lives, the courts have treated individuals who grew up with cosmopolitan or educated backgrounds as having the capacity to behave differently from the ordinary person. By failing to live up to the special standard of behavior the courts set for them based on their background, these defendants show themselves to be deviant and thus more deserving of punishment.116

Duško Tadić, for example, was the first individual tried by the ICTY. During the war, he was a member of the paramilitary forces that supported the Bosnian Serb army when it attacked the town of Kozarac.117 During the raid on Kozarac, which formed part of a larger effort to expel the entire non-Serb population from the area, Tadić participated in several killings and in the forced transfer of masses of civilians to detention camps.118 Tadić did not have a gun to his head; by most accounts, he was a thug, a man who seemed to enjoy the power he wielded so violently over others.119 There was no need to go beyond the acts he committed, and his attitude in committing them, in painting Tadić as criminally responsible. Yet at sentencing the Trial Chamber noted that, because Tadić was raised in an atmosphere of “ethnic and religious tolerance,” and because he was “capable of compassion towards and sensitivity for” others, his commission of crimes during the civil war required “an even greater evil will” for him than such crimes would require “for lesser men.”120 The court imagined, moreover, how these “lesser men” would behave. Citing the “virulent” and “endemic” propaganda that “contributed to the crimes in the conflict,” the Trial Chamber quoted the work of social psychologists Herbert Kelman and V. Lee Hamilton, who wrote that “‘moral restraints against killing or harming become less effective’” when the victims are dehumanized.121 Thus, the average person may be so affected by the circumstances of mass violence that he commits crimes he would not otherwise commit. By virtue of his upbringing, however, Tadić fell into a category of individuals of whom we should expect more.

This same type of reasoning emerges with respect to defendants who are well educated. For individuals like Paul Bisengimana, participation in the Rwandan genocide was particularly egregious in the view of the ICTR because he “was an educated person who could appreciate the dignity and value of human life and was aware of the need for and value of peaceful co-existence between communities.”122 Similarly, the ICTR concluded that because Augustine Bizimungu had an “exemplary education and military background,” he “was in a position . . . to halt the killings.”123 The ICC has pursued a similar approach in the one sentencing judgment it has issued to date, the sentencing of Congolese warlord Thomas Lubanga Dyilo. The ICC noted that Lubanga “is clearly an intelligent and well-educated individual, who would have understood the seriousness of the crimes,” and the court held that those factors were relevant in determining his sentence.124

These holdings recall the judgment by the U.S. military tribunal at Nuremberg in the Einsatzgruppen case, prosecuted against twenty-four defendants who were members of the SS death squads during the Second World War.125 The Einsatzgruppen leadership consisted of highly “educated and cultured” individuals,126 and the tribunal seemed particularly offended by the idea that men who were so “well-bred” did not know better than to carry out a campaign of extermination. The tribunal remarked:

The defendants are not untutored aborigines incapable of appreciation of the finer values of life and living. Each man at the bar has had the benefit of considerable schooling. Eight are lawyers, one a university professor, another a dental physician, still another an expert on art. One, as an opera singer, gave concerts throughout Germany before he began his tour of Russia with the Einsatzkommandos. . . . It was indeed one of the many remarkable aspects of this trial that the discussions of enormous atrocities w[ere] constantly interspersed with the academic titles of the persons mentioned as their perpetrators. If these men have failed in life, it cannot be said that it was lack of education which led them astray.127

In the treatment of both these men and the culturally and educationally privileged defendants before the ICTR and ICTY, courts contend that the average person might have committed the acts in question. At the same time, they also define categories of individuals who are different from—and purportedly superior in moral capabilities to—the average person, and who, by behaving as would the average person, show their deviance. The expectation that those who are educated or cultured should have known better than to get caught up in the violence around them indicates an approach to deviance and criminal responsibility that concedes that somethe “untutored aborigines,” in the words of the Nuremberg Tribunal,128 or the “ordinary soldier[] whose morals were merely loosened by the hardships of war,” in the words of the ICTY Trial Chamber129will be swayed by the powerful or will succumb to the pervasiveness of violence. At the same time, this approach also proposes that others, by virtue of their backgrounds, can be expected to do better, to know better, to withstand the whirlwind that surrounds them and resist the temptation to kill.

C. The Hateful Provocateur

The previous two sections address the ways in which international criminal tribunals create categories of people who are expected to behave differently from the average person. When the courts create these categories, they locate criminal blameworthiness in defendants’ failure to meet the particular standard of behavior set for the accused’s particular subgroup. Alongside the creation of these categories, in their decisions on leaders the tribunals also have offered a portrait of how they view the average person. In particular, the tribunals depict leaders’ culpability as deriving not only from their acts of masterminding, planning, or encouraging crimes, but also from their activities aiming to affect the choices of the individuals who carry out the crimes. In doing so, the tribunals construct an image of the ordinary person as participating in crimes because of natural susceptibility to pressure and coercion.

Hassan Ngeze provides one illustration.130 Ngeze, a Rwandan journalist, founded and served as editor for the newspaper Kangura, a publication that aimed to appeal to Hutu readers by branding itself as “the voice that seeks to awake and guide the majority people” in Rwanda.131 Soon after its creation, Kangura began to warn of an impending Tutsi uprising, printing lists of Tutsi leaders and Hutu “accomplices” who should be watched and publishing Ngeze’s “Hutu Ten Commandments,” which included such statements as “[e]very Hutu should know that all Tutsi are dishonest in their business dealings,” “[t]he Rwandan armed forces must be exclusively Hutu,” and “[t]he Hutu should stop having mercy on the Tutsi.”132

The ICTR charged Ngeze with multiple crimes, along with two co-defendants who were members of the steering committee that established the Radio Télévision Libre des Mille Collines, the radio station that broadcast messages urging Hutus to massacre Tutsis during the genocide. Ngeze was convicted of genocide, conspiracy to commit genocide, direct and public incitement to commit genocide, and persecution and extermination as crimes against humanity.133 In determining Ngeze’s sentence, the Trial Chamber considered not only his actions in disseminating hateful speech and calling the public to violent action, but also his position of power. Ngeze, said the Trial Chamber, “was in a position to . . . shape public opinion.”134 Instead of using that power to work toward peace, he “abused the trust of the public by using his newspaper to instigate genocide.”135 The Trial Chamber stated that Ngeze “poisoned the minds of his readers, and by words and deeds caused the death of thousands of innocent civilians.”136

In considering Ngeze’s “abuse[]” of the public’s trust as an aggravating circumstance, the ICTR recognized that individuals can be shaped by their environments. The judgment compared Ngeze and his co-defendants to Julius Streicher, the publisher and editor of the anti-Semitic Nazi weekly Der Stürmer who was convicted of crimes against humanity and executed in 1946.137 In its judgment, the Nuremberg Tribunal described Streicher’s work as “poison . . . injected into the minds of thousands of Germans which caused them to follow the National Socialist policy of Jewish persecution and extermination.”138 It is possible to understand the culpability of both Ngeze and Streicher as lying in their hatred and their desire to realize the destruction of entire populations. But both the ICTR and the Nuremberg Tribunal saw their crimes as social rather than individual: their culpability lay in their capacity to cause others to share in their hatred and to carry out their desires.

Ngeze is by no means an isolated case; this approach to culpability frequently appears in decisions of the ICTR and ICTY. Early in its work, the Trial Chamber of the ICTR held that “[a]buse of positions of authority or trust is generally considered an aggravating factor,”139 and the court has returned repeatedly to the idea that a person who abuses the trust of the community is particularly blameworthy. Simeon Nchamihigo, for example, was accused of genocide, complicity in genocide, and crimes against humanity based on his orders to kill Tutsi civilians. In its judgment, the Trial Chamber expanded upon its typical holding that abusing a position of authority will be considered an aggravating factor. It noted that because Nchamihigo, a deputy prosecutor in the Rwandan Ministry of Justice during the genocide, held a position in which he was “expected to uphold rule of law and principles of morality,” his actions were particularly powerful in influencing the public.140 This type of reasoning extends beyond leaders who had political power to those who had relationships of trust with their constituents. As Minister of the Interior, Calliexte Kalimanzira had little political authority, but the fact that “[h]e was loved and appreciated for his efforts at empowering his community by contributing to the agricultural development of his native region” aggravated his blameworthiness, according to the Trial Chamber of the ICTR.141 The respect that the community had for him “made it likely that others would follow his example.”142

In each of these cases, and in several more, the courts have conceded that for many individuals in mass atrocity settings, perpetration of crime stems not from any particular disposition toward violence, but rather from the circumstances—influential leaders encouraging perpetrators to act and reassuring them that there will be no consequences, or propaganda that dehumanizes the victims and characterizes violence as a social act.143 By drawing on narratives of leaders who abuse the trust of their constituencies, the courts offer a vision of ordinary people succumbing to social pressures. In the courts’ telling, the crimes of these leaders enable us to understand how ordinary people picked up machetes and garden tools and used them to slaughter their neighbors.

III. a more complete account of expressivism in international criminal law

The tribunals, it seems, seek to have it both ways: even as they have succumbed to the idea that criminal responsibility rests on deviance, they have found ways to neatly do away with the problem in some cases by devising reasons to expect more of the defendant than they ask of the average person. This approach anticipates and provides a defense against a challenge that would insist that the perpetrators’ crimes are understandable and thus inappropriate for criminal punishment. Nonetheless, we should be wary of embracing the tribunals’ artifice of identifying categories of people who should be expected to behave differently from the average individual. The following two Parts explain why the question of criminal responsibility in mass atrocity should be thought of as an opportunity for developing individual notions of responsibility, rather than a conundrum in which the concept of responsibility must be compromised in order to justify criminal punishment.

After pointing out the flaws in the international criminal courts’ deviance-focused approaches to decision making, this Part contends that these approaches rest on an unnecessarily narrow vision of criminal law, and it offers a more complete account that draws on familiar functions of domestic criminal law. Whereas believers in the deviance paradox warn that criminal law’s legitimacy may be compromised by the punishment of behavior that is not deviant, I argue that criminal law has two discrete and accepted functions. First, criminal law can legitimately target conduct that diverges from a standard of ordinary behavior. In so doing, the criminal law sends a message not only about what constitutes deviance, but also about what constitutes normalcy. Second, criminal law can legitimately target conduct that diverges from an aspirational standard of behavior. In so doing, the criminal law sends a message about behavior that might not be normal, but is desirable nonetheless. Understood through the lens of these dual functions, criminal prosecutions for mass atrocity crimes can serve a legitimate goal—even for perpetrators who are not deviant in the way that the paradigmatic perpetrator of criminal acts is—by giving voice to the better angels of our nature and setting out a model for behavior in the most demanding of times.

A. Unjustified Expectations

Is there any sense to expecting more of soldiers, or those who come from cosmopolitan backgrounds, or those who are well educated? Soldiers, of course, occupy a unique position in times of war. They are required to prioritize the lives of innocents over their own.144 For conscripts, this choice is not voluntary,145 and even those who voluntarily enlist, like Erdemović, might not truly anticipate losing their lives. Erdemović was never an eager soldier. He had been discharged from the Croatian forces when he tried to help a Serb cross the border, and when he enlisted in the Bosnian Serb army, he requested a non-combat position and was placed in a unit that carried out reconnaissance missions and manned border checkpoints.146 Erdemović, then, may well not have contemplated losing his life when he joined the army. But even if he had—even if a soldier voluntarily enlists and has full knowledge of the possibility that he will not survive the next week or month or year—facing a risk of death at the hands of an enemy differs dramatically from laying down one’s life in the face of one’s own threatening and criminal commander.147 Does it make sense to expect the average soldier to accept death in these circumstances? Unless we drastically redefine our understanding of the average battlefield, it does not.

Similarly, there is little reason to expect that educated persons or those who come from cosmopolitan backgrounds should be able to withstand the pressure of the violence around them any more than the average person would. In attributing to these categories of individuals a capacity to know better than the average person, the international criminal courts have created a doctrine akin to a converse of the rotten social background defense.148 Instead of affording a defendant the opportunity to introduce testimony on his psychological, cultural, educational, or economic background as a way of explaining his criminal behavior, and thus lessening his blameworthiness for that behavior,149 the tribunals use a defendant’s educational or social history against him, as a way of heightening the blameworthiness of the conduct under scrutiny. But whether we turn to history or literature or social science, we come up short when we try to find reasons to believe that a privileged educational or cultural background necessarily enables a person to resist the pull to violence any better than another.150 Perhaps this construction exposes the “black sheep effect” playing out in these courts: it is too horrifying for judges to accept that educated elites who look so similar to them can commit such horrific acts.151 When those judges insist that the defendants should have known better, they also insist that they themselves would have known better had they been in the same situation. But a person who has not been in such a situation does not know how he would behave and can only hope that he will never have to answer that question.152

Creating categories of people who are believed to differ from the average person—the soldier, the educated person, the cosmopolitan—is at its heart an exercise in othering. By separating people who are expected to behave differently, these decisions treat the defendant as abnormal, drawing a line between us and them. In so doing, the decisions declare that the criminal is distinctive in making the choice that he did.153 In Emile Durkheim’s terms, by identifying the deviants, the courts enable everyone else to unify around shared values, not simply around the rule of law or around a culture of accountability, but around a mutual belief that we never would do what those people in the dock did. If such line-drawing were successful, it could serve to legitimize the project of international criminal law by convincing onlookers of not only the moral responsibility of those who are tried and convicted, but also the moral authority of the courts to brand these individuals as deviant.

The line-drawing, however, is not only unsuccessful, but is also unnecessary. For one, it is apparent that at least some of the defendants were driven by their own sadism and cruelty, and pointing to their educational or cultural background seems simply beside the point in explaining their blameworthiness.154 But more generally, creating a class of deviants is necessary only if we narrowly envision criminal law as voicing condemnation for certain acts that diverge from the ordinary. That is, of course, an important basis of the criminal law, but it is not the only one.

B. Deviance and Positive Expressivism

First-year criminal law courses often begin by introducing the idea that criminal prohibitions declare society’s belief that certain conduct merits condemnation.155 Crimes are said to be those acts that diverge from social norms;156 ‘‘blame is reserved for the (statistically) deviant.’’157 Enforcing the law against individuals who commit those socially deviant acts can reinforce the norms themselves. Thus, by punishing murder, or robbery, or driving while intoxicated, the criminal law voices society’s assessment that those acts should not take place, that they are harmful, that they merit censure.

This interpretation of the purpose of punishment reflects a belief in an expressive function of criminal law. Rather than focusing exclusively on, for example, the retributive value or deterrent objectives of the law, expressive theories assert that law has an educative role and examine the “message” sent through trial or punishment.158 Expressivist theories may exist separately from other justifications for criminal law or may overlap with them. For example, guilty but mentally ill verdicts, which might be difficult to justify as a matter of deterrence, retribution, or even incapacitation, make more sense under a theory of expressivism as a way to express condemnation of the offender as blameworthy and brand him as a criminal.159 Hate crime laws, meanwhile, may reflect both expressive and retributive motivations, as they “signal that perpetrators of hate crimes are more culpable than those who commit parallel crimes.”160

Theorists of international criminal law have seized on the expressive functions that criminal law can serve.161 Indeed, theories of expressivism are not merely descriptive statements about what the law can do; they have become normative statements about what the law should do. In justifying the initiation of criminal trials for mass atrocities, scholars and practitioners point to the capacity for international criminal law and trials to communicate the outrage of the international community in the face of certain acts, and they further emphasize the need for the international community to demonstrate its intolerance of the criminal acts.162 As Antonio Cassese writes, “the international community’s purpose” in establishing a system of international criminal law was “not so much retribution as stigmatization” of certain conduct.163

As Cassese’s comment shows, expressivism in international criminal law has operated primarily in negative space: criminal law communicates to the public what conduct should not be undertaken.164 But even as stigmatization of particular acts forms an important component of what the criminal law does, and of what it should do, this is not its only task. Just as criminal law declares society’s belief that certain conduct is condemnable, it also declares that certain conduct is normal or acceptable.165 Examining the functions of domestic law, David Garland has emphasized this affirmative side of criminal law’s expressive function, noting that penalties communicate not only “how we should think about . . . evil, . . . pathological, . . . and illegitimate,” but also “how we should think about good . . . , normal . . . , [and] legitimate.”166 Garland’s analysis is largely rooted in the work of Emile Durkheim,167 whose thinking on the sociology of punishment has laid the groundwork for theorists of the expressive or educative functions of the criminal law. Durkheim describes punishment as a “sign indicating that the sentiments of the collectivity are still unchanged” despite the choice of the offender to diverge from those collective sentiments,168 and he argues that crime serves an integrative function by binding the community of law-abiders in their outrage against deviant criminals.169 Other scholars, too, have proposed that the criminal law serves to voice positive messages. Jean Hampton, for example, asserts that criminal punishment “‘takes back’ the demeaning message” communicated by the offender to the victim.170 In international criminal law, much has been made of the capacity for the trial to announce the triumph of the rule of law. Lawrence Douglas elegantly describes the Nuremberg Tribunal as “an act staged not simply to punish extreme crimes but to demonstrate visibly the power of the law to submit the most horrific outrages to its sober ministrations[,] . . . making visible . . . the sweeping neutral authority of the rule of law.”171

In calling attention to the positive side of expressivism, I highlight a phenomenon that is different from the rule of law rationales, one that operates, in the manner of Garland’s work, by means of affirmatively communicating to the public what is normal or acceptable alongside its declaration of what is abnormal and unacceptable. Of course, this declaration is not explicit. The criminal law only punishes; it does not offer gold stars to the most admirable citizens. But in its choices about which behavior is punished and which behavior is not, and in its decisions about how behavior is punished, the law reveals certain attitudes about what is normal and what is deviant.172

The law of provocation provides an instructive example. The partial defense of provocation provides that a person who kills another in the heat of passion, upon adequate provocation, is guilty of manslaughter rather than murder.173 When treated as an excuse,174 the defense arises out of the notion that the “heat of passion impairs a person’s agency,” as someone acting in hot blood “finds it more difficult to exercise self-control than a person in a cooler emotional state.”175 In early common law, finding one’s wife in the act of adultery was categorically considered “adequate provocation”; the defendant thus only had to establish that he was in a sudden heat of passion, without reasonable time to cool off, in order to meet the requirements of the defense. Witnessing the adulterous act was considered adequate provocation because lawmakers believed that any ordinary man would lose his self-control in that situation. As the Michigan Supreme Court explained, “In determining whether the provocation is sufficient or reasonable, ordinary human nature, or the average of men recognized as men of fair average mind and disposition, should be taken as the standard. . . .”176 Accordingly, because it is “ordinary human nature” to respond to adultery with hot blood, witnessing adultery is adequate provocation. As for the broader question of why losing one’s cool upon adequate provocation should be a defense at all, the Supreme Court of New Mexico explained the basis of the law as follows: “[T]he man who takes life under those circumstances is not to be punished; not because he has performed a meritorious deed; but because he has acted naturally and humanly.”177

Although the modern law of provocation has abandoned the early common-law approach of allowing a defense as long as the killer was responding to an act that fell into one of the discrete, defined categories of legal provocation,178 it still retains the notion that the person who can claim legally adequate provocation deserves mitigated punishment because that person has behaved naturally, as any normal person would.179 The drafters of the Model Penal Code, for example, characterize the provocation defense as a “concession to human weakness.”180 That is, it is “common experience” that anger compromises a person’s ability to behave appropriately.181 Thus the killer who merits a provocation defense is not an unusual person; instead, that person is acting in a way that, the law says, we can understand.182

In this sense, the criminal law of provocation serves not only a negative expressivist function—that is, voicing the law’s condemnation of the act of killing—but also a positive expressivist function: voicing the law’s assessment that the person who loses his cool in certain situations behaves in a way that is normal.183 This interpretation is not meant to suggest that the provoked killing is commendable; the killer is still punished for manslaughter and is still identified as a perpetrator of a serious crime. But by allowing the defense on the grounds that doing so recognizes common human frailty, the law announces that the person who cannot maintain his self-control in certain circumstances merits a reduced punishment because an ordinary person would have done the same.184

C. Normalcy and Aspirational Expressivism

If we imagine criminal law as legitimately punishing individuals only for deviating from the norm, or from behavior that can be expected of the ordinary person in a particular situation, then it is difficult to accept international criminal law as it is currently structured. Criminal law, however, does far more than this. In addition to defining ordinary behavior in contradistinction to deviance, criminal law—in certain cases—delineates behavior that may not be ordinary, but that should set the standard for appropriate conduct nonetheless. I describe this as “aspirational expressivism,” theorizing that the criminal law declares certain conduct to be behavior to which people should aspire.

The “reasonable person” standard that pervades Anglo-American law provides one example of aspirational expressivism. The reasonable person is sometimes conflated with the “ordinary person,” but the two are distinct and ought to be considered separately. While we might aspire to be reasonable at all times, the ordinary person—who might be defined as the average or typical person—may well fail to meet those aspirations. Accordingly, what the reasonable person does or thinks often differs from what the ordinary person does or thinks.185 Accordingly, if a defendant held an unreasonable belief that certain minorities are dangerous, that belief would not be relevant to determining the validity of a self-defense claim in a prosecution for the killing of a member of that minority, no matter how typical the belief.186

Outside of the reasonableness inquiry, too, criminal law sets forth standards that call on individuals to be better than the average person. State in the Interest of M.T.S. transformed rape law by interpreting New Jersey’s sexual assault statute to hold that, in the absence of “affirmative and freely-given permission” to the specific act of penetration, any act of sexual penetration constitutes the “physical force” necessary to establish sexual assault.187 The expectation that an individual will provide “affirmative and freely-given permission” may strike some as out of touch with ordinary patterns of sexual intimacy,188 but does that make it invalid as a standard enforced by the criminal law? There is no reason that it should.189 Even if the average person does not conform to these rules, the criminal law may encourage behavior to shift in that direction.190 So, too, with newer crimes like distracted driving. Philippa Curtis, a twenty-two-year-old woman in Oxford, England, was sentenced to a prison term of twenty-one months after she crashed into and killed Victoria McBryde while texting with friends. Many people sympathized with Curtis. Even a close friend of the victim said of Curtis, “‘[S]he seemed like such a normal girl. . . . Until Tory’s death I texted while driving, as have most people. I don’t think [Curtis] realized the danger she was causing.’”191 Curtis’s behavior may well have been normal and typical. Nonetheless, the criminal law still may legitimately punish that behavior—and may even seek to punish that behavior precisely because it is typical—in order to modify widespread perceptions of what is ordinary behavior.192

To be sure, all criminal law is aspirational in some contexts. If a law prohibiting murder is violated by one person in a community of thousands, then that law may be understood as representing an aspiration for the one, even as it is defined by deviance for the rest. I refer, however, to a type of law that sets out aspirational standards for a broader set of the community—a prohibition that identifies conduct that might not be ordinary or typical, but is desirable nonetheless—and, conversely, identifies conduct that might be ordinary and typical, but is undesirable nonetheless.193 As an aspirational tool, criminal law envisions a set of behaviors that might one day become the norm.

IV. a new vision for criminal courts in extraordinary times

In situations where masses of people victimize others, there is no denying that violence and cruelty are typical, ordinary, or normal. But despite this ordinariness, courts still have an opportunity to send a powerful message with their decisions. This Part builds upon the cases examined in Part II and the theoretical analysis provided in Part III to offer a way out of the seemingly paradoxical world of criminal responsibility for individuals in mass atrocity situations. Rather than creating categories of people who should be thought of as different from the norm so that defendants can be identified as diverging from their particular category’s higher standard of behavior, the courts should insist that even though the demands of the law may surpass what we might expect of the average person, they still provide fair standards around which to build criminal responsibility. By recognizing the criminal law as a statement of aspirational standards, we solve the paradox; instead of interpreting an absence of deviance and the application of the criminal law as self-contradictory, we can see the normalcy of violence as making it an even more appropriate target for the criminal law.

This Part goes beyond a defense of aspiration in criminal law, however, to argue for the promise of greater candor in decision making in international criminal law. Criminal law’s history of setting out aspirational norms—and the instances in which the law has succeeded in changing norms so that the aspirational standard became typical—would provide cover for decisions that punish even those who look like ordinary men,194 but straightforwardness in the courts’ reasoning is essential. Tom Tyler has written that people obey the law when they believe it is legitimate—when they believe it is the right thing to do, and when they believe that the authorities making decisions about the law are fair in doing so.195 The courts should thus abandon their efforts to craft stories of deviance about the perpetrators before them and instead more candidly identify the sources of criminal blameworthiness. These perpetrators are culpable not because in committing their crimes they failed to meet some standard of behavior for a particular class of people who should be expected to be better than ordinary, but rather because in doing so they failed to live up to the standard of behavior that, even if the average person would fail, we can all still hope to meet.

Before proceeding, I should note that this defense of the capacity of international criminal courts to achieve expressivist goals is not meant to suggest that the answer to the question of how to prevent atrocities or to enable communities to move forward is more criminalization and more prosecutions. To the extent that we rely on criminal prosecutions to achieve these goals, however, admitting that ordinary people undertake extraordinary crimes is a productive endeavor.

A. Judging and Understanding

1. Narratives of Perpetration

Studies of international criminal law abound with expectations that the courts can provide an officially sanctioned account of what happened in a time of mass violence.196 Narratives can serve as powerful antidotes for the collective amnesia that often accompanies mass atrocity. Criminal trials broadcast a story that challenges and ultimately triumphs over perpetrators’ insistence that crimes never took place or that they were not crimes at all.197 Narratives can also provide some insight into how these crimes came to take place. The stories presented in international criminal law are about years of conflict between communities; they tell of cycles of transgressions and years of accumulating anguish, campaigns of dehumanization, and systems of vulnerability. Alternatively, courts play out the stories of victims, giving voice to the many individuals who are silenced by atrocity. What is missing, however, is the story of the perpetrators.

“We tell ourselves stories in order to live,” wrote Joan Didion in The White Album. “We look for the sermon in the suicide, for the social or moral lesson in the murder of five. We interpret what we see, select the most workable of the multiple choices. We live entirely . . . by the imposition of a narrative line upon disparate images . . .”198 In the law, trials are the means by which judges or juries or the public at large try out different narratives—of a particular crime or of the larger world—and select one that will prevail. Law is “the open hearing in which one point of view, one construction of language and reality, is tested against another.”199 And ultimately in a court, one point of view prevails; one narrative emerges triumphant.200

To speak of crafting a story out of genocide might seem simply too ghastly, but it is the impulse of human nature, and it is a venerated goal of international criminal courts.201 The written opinions of these tribunals overflow with details of names and places and acts, but they provide little insight into motivations, feelings, or decision making by particular defendants at particular times. Why did Radislav Krstić, for example, go along with Mladić’s orders to eliminate all Muslims in Srebrenica? At trial, he made clear that he did not have a particular hatred for his victims. “We all went to school together, we socialised together, and we had a great respect for each other,” he said.202 Did he believe it was the only way to save his life? The only way to ensure the survival of his people? Was he too weak or too scared to resist? Did he simply not care, or did he grow to hate the victims? Did he actively choose to be a part of the massacre, as the ICTY contended?203

The impulse of international criminal tribunals to view perpetrators through categories of normalcy and deviance may offer greater moral clarity to the stories of mass atrocities; it is easier to accept that a person took another’s life or body or family because of the perpetrator’s distinctive monstrosity than it is to accept that many people may well have done the same. But blurring these categories, accepting that ordinariness and criminality can reside in the same person, and paying greater attention to how individuals decide to participate in violence may provide some answers to questions that are horrific to ask but nonetheless must be answered.204 Such information is valuable for victims, who may want to know how or why those responsible for the deaths, rapes, and torture of themselves and their loved ones came to commit such evil.205 It also is valuable for the world at large. The narrative function of international criminal law has been understood as a service to victims and to formerly warring communities.206 It is received wisdom that providing an accurate account of a crime allows individuals and peoples to move on and take solace in the official acknowledgment of what happened.207 Beyond that, however, more information about individual decision making can illuminate what allows a person to be drawn into mass violence. A greater understanding of these dynamics may be of use in creatively thinking about how to avoid future conflicts. International criminal law has made progress in holding accountable a handful of individuals who have committed atrocities, but it is not clear that the advent of prosecutions has affected individual choices. Massacres still take place, leaders and foot soldiers still commit crimes, and it seems doubtful that the unlikely prospect of trials will change this reality anytime soon. If the very existence of trials does not help to avert future atrocities, then perhaps at least we can learn something from those trials to move in this direction.

2. A Decision Grounded in Aspirational Expressivism

What would a judgment guided by aspirational expressivism look like? To return to Dražen Erdemović, the ultimate punishment—a term of five years’ imprisonment for killing seventy people—might look just the same. But the account of Erdemović’s culpability would look quite different. Instead of admitting that the ordinary person would have done the same as Erdemović but insisting that he should have behaved differently because of his status as a soldier, a judgment that accepted the aspirational role of criminal law would admit that Erdemović committed an act that many would do, and indeed, that many did. Such a judgment would acknowledge the normative role of the law not only in reinforcing clear prohibitions against killing,208 but also in voicing that the law operates in horrific situations—even in situations in which we might understand why the defendant did what he did—and in seeking to comprehend why Erdemović made the choice he did.209

One piece of this effort is recognizing that, no matter how horrific the circumstances, there are moments of choice. The Erdemović decision portrays agency at two extremes. The prosecution asserted, and Judges McDonald and Vohrah agreed, that the situation that Erdemović faced was characterized by a “lack of moral choice.”210 At the same time, the court’s quick leap to expecting more of soldiers resulted in portraying them as having no choice in the matter; this reasoning assumes that it should have been obvious to Erdemović that he should not have picked up his gun. Whereas critics of the international criminal legal regime argue that courts should pay more attention to the lack of agency exercised by individuals in these situations as a result of indirect and direct coercion,211 the courts here minimize the agency that actors exercise in the opposite direction: they judge these defendants as if they should have seen no alternative but to act lawfully.

Characterizing a person in these circumstances as having either no choice but to kill or no choice but not to kill fails to consider the complexity of the choice, and it misses the opportunity that resides in trying to understand that complexity. At trial, Erdemović testified that on other occasions he refused the orders of commanders to kill innocent people,212 a fact that, curiously, is never mentioned in the decision denying him a duress defense. What made him able to resist in those situations but not in this one? Why was there choice in those situations and not this one? Perhaps the lesson of Erdemović’s crime is that he was a model for critical thought and resistance—he tried to avoid combat; he refused orders he found unconscionable; he recognized the brutality of what he was being asked to do. The judges deciding his fate may have recognized these circumstances, and for that reason sentenced him to a prison term of only five years; but still, the decision on duress frames him as a deviant rather than a model. More attention to the real choices he made could transform what international criminal courts do. Scholars and practitioners who are troubled by the deviance paradox in international criminal law rally around the idea that agency is compromised in circumstances of mass violence213—but agency is not vitiated altogether. These judgments offer an opportunity to understand agency and choice as a spectrum rather than as extremes.

Understanding agency and choice in this way, moreover, may provide a response to those who would find unsettling and unjust the punishment of an “ordinary” person making “understandable” choices. Again, the criminal law does this all the time; punishing an individual for failing to adhere to the standard of the reasonable person is exactly that.214 More broadly, typicality may be a helpful proxy for what is right, and deviance may be a helpful proxy for what is wrong. But they are only proxies, and in the upside-down worlds in which mass atrocity takes place, that fact is critical. What is typical and what is deviant may be transformed by circumstances. The cornerstone of punishment and the assignment of criminal responsibility need not be deviance. Choice, judgment, the opportunity to do something else—these are better ways to identify the spaces in which there is opportunity for the criminal law to do some good.

Accordingly, emphasizing the choice that was available, and how that choice was made, can serve the goals of the criminal law. Drawing on Kant, Hannah Arendt criticized those who, like Eichmann, “refrain[ed] from critical judgment,” those who willingly surrendered their capacity—indeed, their responsibility—to choose.215 In expecting whole classes of people to behave differently on account of some generalized experience, however, international criminal courts attribute to defendants that same sin of “abnegation of the faculties of the mind.”216 But these are moments of choice, not choicelessness; and courts have an opportunity to affirm this, to grasp at some understanding of these moments, and to imagine that when a person next faces a similar point of decision, he may set off on a different path.

B. The Risks of Aspirational Expressivism

1. Nuance and Condemnation

Despite the opportunities presented by an aspirational expressivist vision of criminal law, there are risks as well. Whenever the law seeks to function as an agent of social change, there is a risk that overreach will delegitimize the law or the legal system.217 Especially in the context of the criminal law, and its harsh sanctions and lasting stigma, to punish those who are widely believed not to merit sanction may call into question the moral basis of the law. These concerns, however, are mitigated in situations in which a person causes a serious harm, such as death, and in situations in which punishment can be more nuanced than a stark choice between conviction and acquittal. The Queen v. Dudley and Stephens provides a helpful example.218 Dudley and Stephens were part of the crew of a shipwrecked yacht, The Mignonette.219 After weeks aboard a lifeboat with no remaining food and no fresh water, they killed and ate the body of the ship’s cabin boy, Richard Parker, who by that point was ill and weak.220 They were rescued four days later and, when they returned home, prosecuted for the murder of Parker.221 Although law students typically learn about the two sailors as an example of a court enforcing a sacred norm of ordinary behavior—the prohibition against killing222—the case is better understood as an example of a court imposing an aspirational norm—the prohibition against survival cannibalism on the high seas. A.W.B. Simpson’s fascinating account of the circumstances surrounding this decision, Cannibalism and the Common Law, explains that cannibalism at sea was not an uncommon occurrence during those days, and the British government was determined to put a stop to it.223 The prosecution of Dudley and Stephens represented an effort by the Crown to change the norms of the sea, to take what was a “normal” practice224 and recast it as deviant, as criminal.225

Dudley and Stephens ultimately were convicted and sentenced to death, despite the desperation of the circumstances on the lifeboat, the normalcy of their actions on the high seas at the time, and widespread public sympathy for their plight. The court may have been motivated to deliver a death sentence to prove to the public the inviolability of the norm that it sought to portray as already in existence, rather than conceding that it was merely an aspiration. Ultimately, however, the Queen commuted the sentence to six months’ imprisonment.226 The aspiration the law set out for Dudley and Stephens, and for all those who might find themselves in a similar situation, was accompanied by mercy. The case’s denouement may render it a cautionary tale for the fate of aspirational law in domestic systems in which decision makers have scant opportunity to temper punishment with understanding. But in the world of international criminal law, there is greater promise, as punishments are flexible, and judges have the freedom to write opinions explaining their decisions both as to guilt or innocence and as to the sentences they impose.227

2. Disavowal and Responsibility

Some might argue that the position put forward in this Article—that the international criminal courts should acknowledge that ordinary individuals, not monsters, commit crimes in atrocity situations because of the circumstances around them—is politically unfeasible. Those who hold this view would warn that a court’s admission that a defendant is an average person influenced by the circumstances around him—rather than one motivated by an evil disposition—would invite reflection on the degree to which the international community played a role in creating the very circumstances that drove the individual to wrongdoing. Such an admission, in turn, would invite a judgment that the courts themselves—representatives of that same international community—have no legitimate authority to stand in judgment.228 This understanding recalls the theory of the rotten social background defense. After proposing the defense in a D.C. Circuit decision, Judge David Bazelon was told that his idea “created more problems than it solved” because it called too much attention to society’s role in criminal wrongdoing.229 To this, Judge Bazelon responded that the problems had been there all along; his opinions proposing the defense “simply uncovered bullets that society has always refused to bite.”230

The United Nations Security Council created the ICTY and ICTR as efforts to hold accountable the perpetrators of horrible acts, certainly, but also as efforts to absolve the international community of its own sins of inaction. Atrocities took place as the world looked on; the efforts that the Council and individual states made to prevent these atrocities were anemic. The ICTY and ICTR were as much a way for the international community to pat itself on the back for a job well done as they were a way to ensure reconciliation and victim vindication.231

In elaborating on a rotten social background defense, Richard Delgado contends that society “does violence to an individual when it refuses to prevent the deprivation and suffering resulting from its social and economic order.”232 As Delgado notes, however, the state’s responsibility in such a situation will vary according to one’s conception of the state’s proper role.233 We can imagine two ideas of the state at work in the frequent refusal of the ICTY and ICTR to admit how context may mitigate a defendant’s blameworthiness and their silence on the role of the international community in creating situations of atrocity. The courts may be imagining an international community with no duties to prevent violent situations, or they may be imagining an international community whose only obligation is not to directly perpetuate such violence. In either case, this scaled-back vision conflicts with a rather grandiose one of the international community as creator and organizer of trials, as guarantor of justice, and as final decider of responsibility or impunity.234

Even if the international community expresses its outrage and intolerance for mass atrocity at the highest pitch, it should not be permitted to disavow all responsibility for mass atrocities, and international criminal law should dispel the myth that the crimes perpetrated in places like Rwanda, Bosnia, or Sudan are solely the making of particular individuals.235 As Kofi Annan stated at the ceremony marking the tenth anniversary of the Srebrenica massacre:

We can say—and it is true—that great nations failed to respond adequately. . . . We can say—and it is undeniable—that blame lies, first and foremost, with those who planned and carried out the massacre . . . . But we cannot evade our own share of responsibility.236

International criminal law has been so immersed in the project of identifying individual guilt, in order to eliminate the allegation of collective guilt, that it has failed to capture the reality of mass atrocity. These events did not materialize out of nowhere; they were preceded by small steps—from minor human rights abuses to campaigns of discrimination to systemic dehumanization.237 These events are distinct from isolated murders; in every instance, there was an opportunity for the escalation to stop, and there was an international community, a Security Council, that could have done more to try to stop it.238 Although international intervention is often thought of as existing separately from international criminal law, I urge that these two concepts be considered in tandem. An international criminal legal regime that places more emphasis on criminal behavior as a typical response to situational pressures could illuminate the role of the authority figures or peers who immediately influence the individual perpetrators, while it could also draw greater attention to the role of other political actors who could have played a part in stopping the atrocity.

At the same time, accepting that the international community played a role in the creation of the atrocious circumstances does not destroy this community’s capacity to play a role in resolving crises. In the American criminal justice system, despite the core belief in an individual’s responsibility for his own wrongdoing, we also acknowledge ways in which the broader society both contributes to criminal activity and can play a part in stopping it. Cities that fail their citizens with inadequate schools, job opportunities, or housing organize gun buybacks and tattoo removals for former gang members.239 Outside of government, too, institutions that may be seen as bearing some responsibility for failing to prevent crimes also seek to change individuals’ behavior. College campuses offer orientation programs in which students are alerted to the prevalence of acquaintance rape, urged to look out for one another, and exhorted not to take advantage of those under the influence.240 Bars offer free soft drinks to designated drivers.241

In these scenarios and others, we accept that institutions that contribute to crimes can also contribute to their prevention, and we still feel comfortable holding individuals accountable for their actions. So, too, can it be in the realm of international criminal law. Recognizing the responsibility of bystander states and organizations in the perpetration of atrocities need not vitiate the responsibility of the individual perpetrators. The two can coexist and, indeed, can point to a path forward. Recognizing the role of states or organizations in failing to intervene may convince these actors that ex post accountability mechanisms are insufficient and may motivate them not only to think more about intervention in impending atrocities, but also to consider more responsibly the longer-term consequences of intervention on the politics and stability of states and societies.242


Discussions of international criminal law often begin by cautioning that it vastly differs from any system’s domestic criminal law. Its goals are more ambitious, its audiences more varied, its crimes more horrendous, its perpetrators more ordinary.243 To be sure, some differences do exist, but the dichotomy between international and national may be overstated.244 The world of domestic criminal law is rife with situations of pervasive violence and coercion, situations in which the ostensibly deviant criminal behaves in an understandable or ordinary way. Rather than a mark of deviance in the community, for example, carjacking can be a rite of passage for young men entering the world of gang violence; from there, more serious crimes become the norm.245 For individuals living through the madness of domestic violence, the world turns upside down: in their view, authorities who are entrusted with enforcing the law offer little help, and killing the abuser can seem to provide the only way out.246 The powerful individuals who direct or enable their subordinates to sell toxic mortgages or to torture prisoners are blameworthy not only for their willful blindness or their direct orders, but also for their role in creating a community in which morals are inverted and in which actions once considered wrongful are now considered normal, appropriate, or productive.247 And, to be sure, the world of international criminal law is rife with situations of deliberate and uncomplicated wrongdoing. Sociopaths and sadists find new and expansive ways to hurt people and wreak havoc, to terrorize the vulnerable, and to use impressionable and scared publics as puppets to accomplish terrible ends.248

These situations are united by the fact that in these moments of wrongdoing, there is choice. And in these moments of choice, exercising agency is neither uncomplicated nor absolute. It is compromised, and it is complex. Criminal law can punish and blame in these moments, but to do only that would fail to fulfill the promise of criminal law. It can also seek to understand these moments of choice, to narrate and disseminate the stories of those moments, to concede that they are devastatingly ordinary. They are also devastatingly destructive, and for many observers that fact alone indicates that these acts deserve—indeed, demand—judgment, and their perpetrators deserve punishment. But the instrument of the criminal law need not be so blunt as to comprise only punishment and approval. In between these extremes is space for punishment accompanied by understanding, judgment accompanied by the woeful concession that ordinary people do terrible things. The criminal law can serve both needs by acknowledging the ordinariness of the acts it addresses, by recognizing that these acts merit punishment, and by using its educative and narrative voice to call for us all to do better.