In the early 2000s, international monitors observed an increase in violence in the western Sudanese region of Darfur.1 The pattern of attacks defied prior Western understandings of the main geopolitical fault lines in Sudan. A second aspect of the violence that puzzled outsiders concerned the overlapping ethnic and religious identities of perpetrators and victims.2 Initial reports diverged, finally, on the question of whether the killings, rapes, and displacements in Darfur constituted war crimes, crimes against humanity, or acts of genocide.3
Against this backdrop, one humanitarian aid effort stood out for the simplicity of the relief it promised and the roster of support it received. This was a plan to distribute high-efficiency cook stoves to Darfuris living in refugee settlements in Sudan and neighboring Chad. Although the aims of this plan were multiple, a key goal was to reduce the vulnerability of women and girls to sexual assault during trips made outside the bounds of settlements in search of firewood.4 Better stoves, advocates argued, would decrease the quantity of wood needed for everyday tasks, thus reducing the frequency and duration of collection trips.5 Nongovernmental organizations working on the ground in Darfur eventually abandoned claims about the power of cook stoves to constrain wartime rape, but similar programs have been launched in subsequent conflicts.6 Through such “technological solutions,” critics of the contemporary international aid system contend, “sexual assault and abuse are depoliticized,” while “the perpetration of the violence and the social norms which legitimize and cause such violence … are made invisible.”7
My discussion in the previous chapter demonstrated the central place of social norms in explanations of mass atrocity. My aim in this chapter is to show that social norms are just as integral to efforts to prevent large-scale crimes. Compared to moral and legal norms, the preventive power of social norms has received scant scholarly attention.8 Existing studies of “bad norms”—those that undermine substantial interests of some or all members of particular groups, organizations, or societies—concentrate on social norms within unequal but relatively stable societies.9 Philosopher Cristina Bicchieri and political scientist Karisa Cloward have independently analyzed the social norms that support early marriage and female genital mutilation in communities where these practices are commonplace.10 Cloward holds that “persuasion” is the principal mechanism for eliminating these social norms. Bicchieri divides the process of transforming social norms into the discrete tasks of changing individuals’ empirical expectations, on the one hand, and changing their normative expectations, on the other. While these studies are instructive, it is not clear whether the specific strategies they recommend apply in the more turbid circumstances of mass atrocity.
In this chapter, I consider several ways in which attention to social norms can aid efforts to prevent large-scale crimes. I assess the value of eliminating certain existing social norms within particular populations, notably norms prescribing silence about past atrocities. And I explain how the intentional creation of new social norms, notably norms against incitement, can contribute to atrocity prevention. Along the way, I rebut one argument commonly encountered in the scholarly literature on rescue and rescuers, according to which these agents are distinguished from bystanders by the fact that they are unusually insensitive to social norms. This argument misrepresents the motives and conduct of most rescuers. More important, it overstates the significance of rescue for our understanding of atrocity prevention.
Throughout the chapter, I seek to distinguish different kinds of collectives that can be recruited for the task of constraining large-scale crimes. These collectives range from highly structured institutions like the United Nations to ephemeral coalitions of activists or citizens. Each kind of collective possesses different capacities for shared agency, but even collectives lacking expansive agential powers can sometimes help prevent mass atrocities by adopting or abandoning specific social norms.
The chapter proceeds as follows. In section 7.1, I survey how social norms typically feature in the literature on rescuer behavior during large-scale crimes and argue that rescue is of only limited significance for our understanding of atrocity prevention. Section 7.2 discusses strategies for eliminating social norms that prescribe silence about past or ongoing atrocities and explains how this can advance atrocity prevention efforts. Section 7.3 analyzes the opposite dynamic: the intentional creation of social norms designed to give loosely organized collectives greater capacities for constraining large-sale crimes. Here I propose the creation of social norms against incitement to atrocities. A brief conclusion follows in section 7.4.
It is difficult to frame a definition of rescue that applies to all cases of mass atrocity. In part, this reflects the political nature of past and present campaigns to identify and honor rescuers.11 More substantially, it reflects the complex character of rescue itself. That complexity becomes apparent when we consider a few of the distinctions scholars have drawn when discussing this broad category of action. One such distinction differentiates short-term and long-term acts of rescue.12 Another asserts a stark moral division between rescuers and “paid helpers.”13 A third distinction challenges our tendency to associate individuals with one particular category of action, forcing us to reflect that the same person may act as a rescuer at one moment, a collaborator at another, and a bystander for the balance of any specific episode of wrongdoing.14
For my purposes, it is sufficient to define rescuers as agents who, acting alone or in groups, offer aid to persons targeted for atrocities despite substantial costs or personal risks of harm.15 Social norms have been taken up in several ways by scholars seeking to explain rescuer behavior during large-scale crimes. One argument, often encountered in prosopographical studies, holds that individual rescuers hail disproportionality from groups occupying a marginal social position before or during mass atrocities. They may be foreign born, belong to a minority religion, or perform a socially stigmatized occupation such as sex work.16 This line of analysis receives some support from the philosophical theory of social norms, according to which individuals on the margins of any given population are less likely to internalize group norms and may be better positioned to avoid sanctions for violations.17 That said, acts of rescue are often legally, as well as socially, prohibited during mass atrocities. It is unclear whether a marginal social position tends to shield individuals and groups from legal penalties or instead renders them all the more vulnerable to punishment. Nor does this line of analysis distinguish differences in sensitivity to social sanctions, on the one hand, from differences in the perceived likelihood of actually being detected, and sanctioned, on the other.18
A second way that social norms inform accounts of rescue turns on the plurality of social norms. As Bicchieri and Cloward observe, there are usually multiple social norms that plausibly apply to agents’ deliberations and actions in any given situation. Individuals are often able to make strategic judgments about which particular norms to follow.19 Indeed, narratives of rescue often feature agents who find ways to call attention to, and stubbornly insist on, specific social prohibitions or prescriptions, in order to prevent the enforcement of other social or legal norms. This dynamic appears in Sara Brown’s interviews with Hutu women who rescued Tutsis during the Rwandan genocide, where Brown found that some of her subjects were able to exploit local norms prohibiting male nonrelatives from entering a family’s compound while the head of household was absent.20 It also describes the case of non-Jewish concierges in Budapest ghetto buildings during the Holocaust who, as historian Istvan Pal Adam has shown, sometimes chose to treat their Jewish tenants as employers deserving deference rather than as prisoners subject to the decrees of the occupying authorities.21
There is a third way that social norms have been incorporated into studies of rescue, an approach developed by Samuel Oliner and Pearl Oliner in their 1988 book, The Altruistic Personality.22 In that study, based on interviews with four hundred Holocaust-era rescuers, the authors identified three forms of rescuer motivation, which they call “empathic” motivation, “normocentric” motivation, and “principled” motivation.23 The smallest proportion of rescuers in their sample described principled motivations for rescue: deeply internalized beliefs about justice or due care that compelled the agents in question to offer aid to others.24 The second-largest proportion of rescuers described empathic reasons for helping: reasons rooted in personal concern for the well-being of specific individuals singled out for harm.25 But the plurality of rescuers, according to the Oliners, described normocentric reasons for their acts of rescue: reasons rooted in the norms circulating within one of the specific groups to which they belonged, such as their church, profession, or family.26 The authors conclude that the action-guiding influence of social norms need not undermine properly moral agency. Rather, “empathy and concern with social norms … represent alternative but equally profound ways of apprehending moral claims.”27
The Oliners’ study has shaped much subsequent research on rescuer behavior during large-scale crimes. Leading scholars such as Ervin Staub, Kristen Renwick Monroe, and Jacques Semelin draw heavily on their methodology and their general conceptualization of rescue.28 But there are three problems that should cause us to question whether any of the preceding accounts of the relationship between social norms and rescuer behavior can be applied directly to the study of mass atrocity prevention. The first problem is that none of the historical cases of rescue highlighted in these accounts actually succeeded in preventing large-scale crimes. The second problem is that coalitions formed for the purposes of rescue are usually either ephemeral or dedicated to peacetime missions far removed from atrocity prevention. The third problem concerns the hostility that individual rescuers often encounter within the fragmented communities left after atrocities end. I discuss each of these problems in turn.
In order to make the first point clear, we should distinguish between the moral significance of acts of rescue, on the one hand, and their preventive significance, on the other. Philosopher Lawrence Blum provides a perceptive account of the moral value of acts of rescue during large-scale crimes, focusing on the altruistic character of rescuers’ actions and identifying additional bases of moral value in the way such acts manifest resistance to racism or resistance to evil.29 Blum notes the paradoxical fact that when we learn about parents who have offered aid in historical cases of mass atrocity, thereby risking the lives of their own children, we are not on reflection inclined to blame them for this, but instead deem their conduct morally exemplary.30 He argues that such assessments reflect the existence of moral ideals that go beyond, and sometimes trump, even the strongest associative duties.
Blum expressly rejects consequentialist evaluations of the moral significance of rescue. He argues that assessments of altruistic action that focus on outcomes do not illuminate altruism at all and suggests that the value of rescue as a gesture of resistance to evil or racism lies in its expressive qualities rather than its practical results.31 Even failed acts of rescue, or short-term acts of rescue that nevertheless did not suffice to save the lives of those assisted, realize profound moral value on this view.
If we accept Blum’s analysis, we should conclude that the moral value of rescue does not depend, even in part, on its preventive effects. But atrocity prevention is precisely the subject with which we are here concerned. It is true that some scholars of rescue and rescuers, including the Oliners and, more recently, Ervin Staub, suggest that we can learn from the study of rescue how to craft policies and institutional reforms aimed at atrocity prevention. These are, in large part, educational policies and reforms, exemplified by the Oliners’ call for schools to become “caring institutions” that help students “acquire an extensive orientation to others.”32 Whatever the general desirability of such initiatives, however, their contributions to atrocity prevention remain unproved. They cannot stand as evidence of a correlation between successful forms of rescuer intervention and successful strategies for atrocity prevention.33
A second reason for doubting whether studies of rescue meaningfully inform theories of atrocity prevention does not focus on the gap between the moral and the preventive significance of rescue, but rather on the long-term aims and capacities of the individuals and groups involved. Out of the many cases of rescue on which the Oliners based their study, a substantial number involved individuals acting together to offer aid to targets of large-scale crimes. Paralleling Christian Gerlach’s analysis of perpetrators, we may wish to call these “coalitions for rescue.” In some cases, these coalitions reflected preexisting social relationships, such as membership in the same church, school, or office. In other cases, individuals without preexisting social ties worked together to offer aid or ease escape. In either case, there is reason to question whether such groups will continue to exist and work together for shared goals once the immediate menace of atrocity ends. Churches and professional associations are just as likely go back to being focused centrally on the spiritual lives or the vocational skills of their members rather than directing their efforts toward the prevention of large-scale crimes. The “goal-oriented collectives” that engage in rescue commonly break up after direct threats dissipate, in some cases splintering because of differences in visions for the social and political future and in other cases because of urgent practical needs.34 Although an increasing number of permanent organizations are devoted to atrocity prevention, few of these feature in accounts of rescue. Nor does success at the relatively short-term task of rescue necessarily presage success at the longer-term challenge of mass atrocity prevention.
It might be objected that while groups or coalitions of rescuers often break up after atrocities end, individual rescuers may yet play a leading part in prevention efforts. After all, several prominent individual rescuers (as well as attempted rescuers) from specific episodes of large-scale crimes have gone on to become leading advocates of mass atrocity prevention. Roméo Dallaire, the commander of UN peacekeeping troops in Rwanda at the start of that country’s genocide, is one such example. The unnamed Holocaust-era rescuer interviewed by the Oliners who founded a “House of Reconciliation” at Versailles immediately after World War II is another.35 One way of viewing these cases is to say that rescuers, or attempted rescuers, enjoy the standing to serve as spokespeople for policies and institutions designed to prevent large-scale crimes. Is this not a clear connection between the work of rescue and atrocity prevention?
Against this objection, we should set the fact that across many different contexts of mass atrocity, rescuers are not welcomed as heroes or moral authorities in the immediate aftermath of conflict, but instead treated as social pariahs. This is true of the Polish rescuer Janka Polanska, interviewed by Nechama Tec, who, after the end of World War II, found “that to most Poles her protection of Jews was unacceptable,” and she soon “came to feel like an outsider in the country she loved.”36 It is also true of the Rwandan rescuer whom Sara Brown interviewed, who reported that after the genocide, her former neighbors “took us as the same as the Tutsis. They said we are traitors.”37 Another rescuer Brown spoke with never went public with her story, saying, “I always kept quiet. I didn’t want to be in the spotlight because I was afraid they would kill me.”38 The fragmented circumstances of transitional societies, so well described by Colleen Murphy, often encourage rescuers to play down their experiences after atrocities end, thus limiting the extent to which rescue can become a springboard for atrocity prevention.
Rather than transplanting findings about the influence of social norms on acts of rescue into proposals for atrocity prevention, I want to describe two independent pathways by which social norms can help constrain large-scale crimes. The first pathway focuses on the elimination of social norms that hinder atrocity prevention efforts by blocking or smothering information about past or present crimes—silencing not only rescuers, as in the case cited above, but also perpetrators, survivors, and bystanders to large-scale crimes. The second pathway focuses on creating new social norms within broader publics, focusing specifically on the advantages of a prospective social norm against incitement to atrocities.
Not every widespread pattern of behavior observed during large-scale crimes indicates the existence or influence of social norms. When perpetrators all wield similar weapons, this may simply reflect the tools available in their communities. When displaced persons require fuel for cooking, cleaning, or other domestic tasks, they may spend hours every day searching for firewood without forming any normative attitudes toward the foraging activities of their fellow refugees. To borrow Cristina Bicchieri’s language, these “action patterns” may be “created and sustained by the motivations of actors acting independently,” reflecting customs, perhaps, but not social norms.39
That said, some facts about the pattern of firewood collection in Darfur described at the start of this chapter do seem to warrant explanation via social norms. Responsibility for obtaining cooking fuel, here and in many others places around the globe, falls largely to women and young girls, though it is not the case in Darfur (as in many other developing countries) that women are generally confined to domestic work rather than wage-paying employment.40 At the same time, the risk of sexual assault that women and girls face during firewood collection trips appears closely tied to gender-based social norms: norms about sexual consent, norms about female “purity” or “honor,” and norms about the treatment due to women who report such crimes.41 Moving from theoretical posits to empirical proof of the power of social norms to help explain sexual violence in Darfur requires careful ethnographic research.42 But identifying, and altering, inequitable social norms clearly deserves a place within the project of atrocity prevention.
In what follows, I consider in more detail how social norms that escalate risks of large-scale crimes can be eliminated. My focus falls on several examples of social norms prescribing silence about ongoing or recently ended atrocities. Some of these social norms center on sexual or reproductive harms; others relate to diffuse forms of perpetration and victimization. I first explain the bad effects of social norms prescribing silence about atrocities as suggested by current theories of atrocity prevention. I then describe several pathways by which those norms might be eliminated.
The simplest cases of social norms prescribing silence about atrocities appear in groups directly implicated in large-scale crimes. Members of a military unit, for example, may share a norm according to which “ratting” on fellow members who violate the laws of war is prohibited. Norms of this kind are “bad” in the sense that they serve to protect the interests of one set of individuals (those who commit or permit atrocities) against the interests of victims of those crimes.43 Such norms are also bad from the perspective of more conscientious men and women in uniform, who have internalized the rules of engagement but may reasonably fear reprisals for reporting infractions by their peers.
Social norms circulating within the groups to which perpetrators of mass atrocities belong are not the only ones that can inhibit the disclosure of large-scale crimes. Social norms accepted and followed by victims of atrocities can also prompt silence about abuses. As I noted in the previous chapter, one explanation for the prominence of sexual violations in many twentieth-century cases of genocide and ethnic cleansing holds that perpetrators exploit preexisting social norms within target populations that subject women who have been assaulted to shame or exclusion. The same social norms, as Bina D’Costa has noted, may guide fathers or husbands whose daughters or wives have been raped to avoid disclosing those crimes due to the social sanctions they themselves might incur.44 Such patriarchal social norms also seem to fit the model of norms that are bad for some, good for others, though in the circumstances of mass atrocity, some prior beneficiaries of these norms may find their interests harmed by them instead.45
Finally, we should consider how social norms that prescribe silence about personal experiences of mass atrocity can arise from misperceptions about the normative beliefs and attitudes prevailing within particular populations. Holocaust scholars writing on shifts in the prevalence of acts of testimony by survivors over the last half-century often cite survivors’ early impressions of norms prescribing silence about their experiences within their countries of immigration or refuge. Annette Wieviorka, in her book The Era of the Witness, reports the advice one émigré survivor received from an aunt living in the United States: “If you want to have friends here in America, don’t keep talking about your experiences. Nobody’s interested, and if you tell them, they’re going to hear it once and then the next time they’ll be afraid to come and see you.”46 Lawrence Langer, in his Holocaust Testimonies, provides a comparable account of survivor silence in the initial postwar era:
The seeds of anguished memory are sown in the barren belief that the very story you try to tell drives off the audience you seek to capture. … Reluctance to speak has very little to do with the preference for silence.47
Philosophical studies of social norms suggest that these beliefs about the likely responses of hearers of testimony, even if unfounded, could in principle lead to the emergence of a social norm against sharing personal experiences of large-scale crimes. Such a norm, which would in fact be bad for everyone, could arise without anyone intending it, as the natural disposition of newcomers to err on the side of caution, “combined with an initial misestimate of the other’s intentions,” leads survivors to interpret the discomfort or avoidance their stories elicit from new neighbors as a sanction based on an existing social norm.48 Following Bicchieri, we may wish to say that the “reluctance to speak” that Langer described reflects a conditional preference for silence on the part of survivors, capable of being activated by a misjudgment of the preferences of their interlocutors.49
Some readers might object that the negative responses reported by survivors of large-scale crimes, such as physical discomfort or future avoidance by interlocutors, cannot reasonably be construed as sanctions and so cannot give rise to belief in the existence of a social norm prescribing silence. Such responses might be disheartening, but they are not punitive. They hardly suggest that a standard of conduct has been violated. However, as philosopher Linda Radzik has noted, “social avoidance” is in fact a common sanction for violations of shared norms.50 It is true that not all instances of social avoidance reflect practical commitments or normative attitudes; in some cases, social avoidance may follow from sheer physical unease on the part of the hearer.51 But this simply opens new opportunities for misperception and creates new chances for the emergence of a social norm prescribing silence.52
I have suggested that social norms prescribing silence about large-scale crimes should be of interest to scholars and practitioners of atrocity prevention. But how exactly do such social norms contribute to the etiology of mass atrocities? In the case of social norms against “ratting” in military units, the causal pathway is fairly clear: soldiers who are bad actors will not be eliminated from their units, and so will have further opportunities to violate legal or moral norms. In addition, those who fail to report them may thereby become susceptible to blackmail and face pressure to participate in future abuses. But what about silence among survivors of large-scale crimes? How can this sort of silence serve as a causal factor in future atrocities—such that eliminating social norms prescribing silence would aid prevention efforts?
Here we can distinguish several ways in which silence among survivors or witnesses to atrocities and the resulting lack of information about past assaults contribute to the etiology of large-scale crimes. In the first place, much like the military case, silence can leave perpetrators or planners of atrocities in positions of power, with barely reduced legitimacy, and so in a position to carry out further attacks. This pathway for large-scale crimes is particularly important for scholars who stress the rationality of mass killings or forced displacements, that is, the ways in which such acts advance the aims of calculating political leaders.
A second way in which norms of silence about past crimes render future abuses more likely is by impeding understanding of ongoing conflicts, and thereby hindering effective interventions. As noted in the discussion of Darfur at the start of this chapter, international observers are not always able to tell who the responsible parties are in given conflicts or what motives stand behind attacks. Where such understanding fails and no single group can be credibly charged with responsibility for mass atrocities, the political case for intervention becomes harder to make. As Alex Bellamy observes, states “tend to be risk averse and are unlikely to adopt strong positions when faced with incomplete, inconsistent, or uncertain information.”53
Finally, some scholars of atrocity prevention contend that intergroup dialogue is crucial to efforts to end cycles of violence in deeply divided societies. Where social norms obstruct such dialogue, they may hinder prevention. Political scientist Elazar Barkan provides a clear explanation of how lack of dialogue can hamper long-term prevention efforts. First, he argues, “Conflict between groups is often transgenerational, and the historical animosity [between those groups] remains if it is not addressed.” Second, he contends, both “fear and desire for revenge often linger … and are prone to be awakened by nationalists or fundamentalists as carriers of xenophobia,” thus establishing a reservoir of animus that can be tapped to recruit coalitions for violence. Historical dialogue, by contrast, “aims to defuse such hatred by engaging government and civil society in conversations about history and introducing empathy and even sympathy toward the other.”54
Barkan acknowledges that such long-term strategies for atrocity prevention remain undertheorized. He suggests that “for historical dialogues to be meaningful, new norms will have to be evolved.”55 This accords with Kerry Whigham’s observation that “ensuring free and open debate about the past is not always easy, especially when certain groups uphold a version of the past that rationalizes the violence that occurred.”56 But if the argument I have developed in this chapter is correct, it will not be enough, in confronting this challenge, to craft new norms for structuring intergroup dialogue. It will also be necessary to eliminate existing social norms prescribing silence about past atrocities.
How can these detrimental social norms be eliminated? Cristina Bicchieri provides the most detailed account of the multiple routes to social norm elimination. According to her model, the elimination of social norms requires the elimination of both shared empirical expectations that others will act in a certain way and shared normative expectations that others ought to act in a certain way.57 In addition, on Bicchieri’s account, the change in empirical expectations must generally precede the change in normative expectations when eliminating norms.58 Let us now consider several means by which changes in such expectations might be induced.
In some cases, the adoption of other kinds of norms, such as legal norms, within specific populations may help to erode social norms prescribing silence. The proposal by Mark Osiel, discussed in chapter 1, to make military officers civilly liable for atrocities committed by troops under their command is designed to erode the social norm against reporting on one’s fellow soldiers by vesting a form of legal accountability in figures responsible for shaping the culture of military units.59 The legal lustration programs adopted as part of many transitional justice mechanisms similarly seek to erode the norm-supported practice of blackmail against officials implicated in atrocities—in this case, through the rather drastic step of making sure that they do not hold office in the first place.60
In the case of social norms prohibiting disclosure of sexual assault, which are supported by sanctions such as social shaming, banishment, or other harsh treatment, two distinct strategies have been pursued, one of which is more akin to norm evasion, the other to norm elimination. In the former case, activists and criminal justice officials working with women who have suffered sexual assault in highly patriarchal societies have adopted practices of confidentiality, closed-door interviews, and other gender-sensitive approaches designed to avoid exposing victims who tell their stories to social stigma.61 In the other strategy, an increasing number of grassroots organizations, in a variety of postconflict societies, have come forward to advocate for public disclosure and legal accountability for sexual atrocities.62 Their actions reflect the finding that organized groups, as opposed to isolated individuals, may be better able to weather sanctions for violating existing social norms, and thus better able to effect changes in both the normative and empirical expectations underlying bad social norms.63
Finally, there are cases where social norms prescribing silence are disrupted by unexpected events or “crises” that serve as “catalysts” for norm erosion.64 The standard narrative about Holocaust survivors and their willingness to share testimony in their countries of immigration holds that shortly after the end of World War II, testimony collection efforts ended, so that it was only after the widely publicized trial of Adolf Eichmann that the “Era of the Witness” began.65 With that trial, in which more than a hundred Holocaust survivors gave public testimony transmitted to global audiences, it became much harder to sustain the empirical expectation that others would prefer not to speak or to hear about experiences of the Holocaust. Accordingly, the perceptions of a social norm prescribing silence that I already noted must have been seriously undermined by this trial. It is perhaps true, as Hannah Arendt famously argued just after these events, that a criminal trial is an inappropriate venue for writing history or recovering suppressed memories from victims of genocide. From a normative perspective, however, we may conclude that the trial produced effects extending far beyond what its architects intended.66 It is now by no means frowned upon for survivors of the Holocaust, or other large-scale crimes, to share their stories in most countries of settlement. Indeed, in many of these places, bearing witness is now socially prescribed.
The erosion of social norms enjoining silence about ongoing or past atrocities makes an important, but limited, contribution to mass atrocity prevention. Reliably constraining large-scale crimes depends on reliable information about such events; silence is inimical to such efforts, and so testimony matters.67 Reaching genuine understanding about the nature, causes, and responsibility for past atrocities may be the only way of ensuring that old distortions, fabrications, or scapegoating do not reemerge.68 It is clear, however, that these initiatives are limited in their effectiveness. They must be supplemented by legal developments and closer examination of existing moral norms. They may also be reinforced by efforts to create new social norms, to which I now turn.
I have argued that silence about past or ongoing mass atrocities generally impedes efforts at mass atrocity prevention. I have also sketched several ways in which social norms prescribing silence about such matters may be eliminated. But not all speech that occurs before, during, or after large-scale crimes serves to reduce intergroup tensions or ease reconciliation. Indeed, some speech, whether in the form of vocal pronouncements, published screeds, or other expressive acts, is explicitly intended to precipitate or prolong mass atrocities. In this section, I consider the significance of incitement to large-scale crimes from the perspective of atrocity prevention. I first briefly review the definition of incitement. I then examine key problems encountered by current legal efforts to define, prohibit, and punish incitement to genocide, mass killing, and other kinds of large-scale crimes. Finally, I defend an alternative preventive response to incitement, one based on the creation of new social norms.
Incitement occurs when an agent uses words (or any other form of expression) to attempt to cause another agent to perform an illicit act. However suitable this may be as a general definition, in law the concept of incitement is subject to significant debate, both with respect to the state of mind (mens rea) required for legal charges of incitement and with respect to the requisite causal connection between one agent’s expression and another agent’s real or potential action.69 Incitement is generally construed as an inchoate crime—one that does not require that the particular action aimed at by the proscribed expression actually takes place.70 But the various rulings on incitement issued by the major ad hoc international criminal tribunals, as well as by the International Criminal Court, have not consistently interpreted the crime in this way.71 The confused state of jurisprudence surrounding incitement to genocide and other forms of mass atrocity has prompted legal scholars to advance alternative theories of incitement, grounded variously in speech act theory, communications research, or a more extensive analysis of different categories of “atrocity speech.”72
Legal scholar Gregory Gordon provides a particularly clear statement of the obstacles to atrocity prevention arising from current legal prohibitions on incitement to genocide or other large-scale crimes. On the one hand, because of inconsistencies in the way incitement has been interpreted and assessed at the various international criminal tribunals, there is a risk that “would-be defendants will not be put on notice as to what constitutes incitement.”73 This tends to diminish the deterrent effects of these legal prohibitions. On the other hand, unless the definition and limits of the crime of incitement are made clear, “repressive governments will continue to exploit incitement law for purposes of stifling legitimate speech.”74 Different countries afford different constitutional and statutory protections to the speech of their citizens, but the worry for all of them is that parties in power will use laws on incitement or more broadly on hate speech to silence or punish the legitimate speech of political rivals—thus deepening the social marginalization that makes atrocities more likely.75
It may be that Gordon and other legal scholars will eventually succeed in streamlining the jurisprudence of incitement and addressing the twin concerns about legality and political exploitation. This may occur by narrowing the range of speech acts that can be charged as incitement, as Richard Ashby Wilson recommends; alternatively, it may occur by adopting a “unified liability theory” for atrocity speech, as Gordon advocates.76 Still, some of the difficulties raised by legal prohibitions on incitement do not seem to be grounded in the failings of existing legal regimes. They are not, to use the language of jurists, problems of lex lata, or law as it exists, but rather problems stemming from the distinguishing features of all legal norms. Because legal norms are subject to standing procedural rules governing their creation and modification, there will always be centralized authorities (whether elected, appointed, or self-proclaimed) empowered to make decisions about the range of their applicability. But however thoroughly delegated this decision-making authority may be, it will always be susceptible to claims of bias against political rivals, thus undermining the legitimacy not only of laws against incitement but of law and legal institutions more broadly. Furthermore, the distinctive legal penalties available for incitement seem ill suited as specific deterrents for this crime, since being jailed does little to prevent an individual’s past speech, whether oral or written, from being reproduced or rebroadcast by others.
In light of this, it’s worth considering whether a different response to incitement, based not on legal but on social norms, could help constrain such expression while avoiding some of the drawbacks associated with legal prohibitions on speech. Such an approach would place certain kinds of expression outside the bounds of socially acceptable speech, including calls for the elimination or forced removal of whole groups of people from the political community. It would, if successful, work to arrest incitement without requiring the actual arrest of inciters.
To assess the promise of social norms in this domain, we must first consider how they would work: what behaviors they would prohibit and what sanctions might be used to enforce them. We can then consider how such norms might be established within particular populations. Finally, we may look at some lingering fears about social prohibitions on expression.
Existing accounts of the roles social norms perform in shaping political discourse typically focus not on the substance of discussion or debate but rather on the manner in which proposals concerning any given topic may be delivered.77 Philosophical studies of civility, for example, associate civil or uncivil speech with speech that does or does not conform to established social rules for how others should be addressed in public. These rules range from the use of titles, to tone of voice, to giving the other side a chance to speak. Following such rules, philosopher Cheshire Calhoun argues, is often morally significant because doing so signals respect for conversational partners.78 Critics of the enforcement of civility in political speech, including Iris Marion Young, also focus on formal requirements for speech and argue that these requirements are used to constrain opportunities for substantive argument, particularly among members of underrepresented groups, and are often tainted by biased judgments about what counts as respectful speech in specific contexts or with specific interlocutors.79
Whatever one thinks about these arguments for or against constraints on the manner of political debate, it is clear that social norms prohibiting particular styles of expression are neither necessary nor sufficient as a response to incitement. Although we commonly associate incitement with the demagogue’s screed or the fanatic’s shout, it is also possible to advance vicious proposals using perfectly polite phrases. Empirical investigation, of the kind advocated by Richard Ashby Wilson, may ultimately reveal whether particular forms of speech or writing are more or less likely to stir up hostile emotions and inspire violent action; the results of such inquiries will bear on our understanding of the contextual conditions for incitement.80 But what I am interested in investigating are social norms that discourage incitement by prohibiting certain contents of speech rather than by targeting specific modes of expression.
Unlike legal prohibitions on incitement, hate speech, or other forms of expression, social norms against incitement will not be governed by standing procedural rules concerning their emergence or modification; they will also lack institutions specifically charged with codifying, interpreting, or applying them. For this reason, the best way to think about how such norms would operate is to consider how they would affect particular agents’ practical deliberations. As we have seen, incitement occurs across a wide range of occasions and venues for expression, ranging from in-person political speeches to printed statements, to television, radio, or Internet streams. It does not typically extend to casual conversations between private persons—though some of the effects of a social norm against incitement might be felt in this context. Instead, incitement to atrocity generally originates with public figures, whether political officeholders, electoral candidates, military officials, or celebrity broadcasters. For this reason, social norms against incitement (as opposed to more generally applicable social norms against, say, hateful expression or demeaning or vulgar language) would need to be applied particularly to these figures.
Who might enforce a social norm against incitement? As already mentioned, social norms do not typically assign responsibility for enforcement to any particular persons. One reason that social norms may help pattern the behavior of unstructured groups (those without substantial capacities for shared agency) is that many different members of those groups can and do act in specific instances to enforce social norms. This wide distribution of sanctioning power supports John Stuart Mill’s claim that social sanctions, as compared to legal penalties, leave “fewer means of escape” and “penetrat[e] much more deeply into the details of life.”81 Of course, if the set of agents who can reasonably be accused of incitement to atrocities is limited to public figures, the specific ways in which sanctions can be applied may be affected by this fact, as such figures are often insulated against certain kinds of social sanctions, while being exceptionally exposed to others.
There are two further questions to be answered with respect to social norms against incitement. The first question is what kinds of sanctions can and should be applied against public figures who engage in incitement. The second question is how social norms countering such conduct can be spread across large populations. I will consider each of these questions in turn.
The sanctions used to enforce social norms range widely in their publicity, severity, and longevity, from mere “punitive criticism” all the way to deadly violence. Not all of these sanctions are morally permissible; not all are likely to succeed in sustaining particular social norms over time. Here I discuss a few of the most promising options.
Earlier, I cited Linda Radzik’s account of social avoidance as a form of social punishment. Radzik has in fact analyzed a broader range of social punishments, emphasizing in each case how they are distinguished from legal punishment by the lack of a centralized decision-making body that could determine proportionality and assign responsibility for enforcement. Among the various types of social punishment she has examined are moral rebukes, gossip, and boycotts.82 Moral rebukes, Radzik argues, are significant because they are communicative as well as punitive: they do not merely attempt to coercively change an agent’s behavior by setting back his or her interests, but also try to convey morally significant reasons for why that change should occur. Boycotts, by contrast, are more akin to social avoidance. By taking steps to withdraw from social interactions with agents deemed to have violated relevant norms and encouraging others to do the same, agents engage in a form of punishment that is open-ended and not directly conducive to moral reason giving. Both of these forms of social punishment, Radzik argues, have advantages and disadvantages. Both are more or less effective in specific circumstances.83
Can rebukes and boycotts be effective as sanctions for the kinds of public figures who are in a position to engage in incitement? Here, several considerations are relevant. First, public figures often enjoy certain forms of insulation from criticism, brought about by wealth, prestige, or the privileges of public office. Second, incitement is a form of action whose foreseeable (and intended) harmful effects are likely to follow very quickly, leaving only a small window for imposing preventive sanctions. Third, incitement itself often takes the form of a rebuke, in the sense of an articulation of highly moralized demands accompanied by a threat (more or less explicit) to back up those demands with coercive action if needed.84 In light of these features, we should consider how either rebukes or social avoidance might work as sanctions justified, at least in part, by their potential to deter specific acts of incitement.
First, consider rebukes for incitement. It is quite common for public figures to call out other public figures for their hateful or, more specifically, inciting, speech. In many cases, of course, those rebukes are offered in private, “between you and me,” in which case the analysis will closely follow the one provided by Radzik, where the communicative and punitive aspects of the rebukes are fairly evenly balanced. But the cases of rebukes that actually come to light are generally those where the relevant criticism is uttered publicly, either in person before a crowd or in a broadcast, public letter, or online forum. In these cases, it seems to me, the punitive aspect of the rebuke increases, for here what occurs is not only that a certain agent is criticized, but he or she is seen to be criticized by large numbers of third parties. Even where the addition of such punitive effects is not the intention behind making the rebuke public, as in cases where a former officeholder sees this as the only way to convey a moral message to his or her successor, this heightening of the punitive aspect will occur.
Whether such public shaming is actually effective in deterring incitement will depend on a variety of factors, but one of the most significant of those factors is how factionalized the public is that witnesses the rebuke. In cases of incitement, there will typically already be widespread factionalization on racial, ethnic, or other lines. This is one reason to think that rebukes may not be the most effective means of sanctioning incitement, though they may work to discourage offensive expressions less intimately tied to impending violence.
Turning to boycotts, we should note that in the case of public figures, a far larger proportion of individuals will have virtual ties to those persons than actual face-to-face ties, and so boycotting goods, speeches, or broadcasts may be a more widely available form of sanction than avoidance of physical contact. Where politicians are concerned, avoiding rallies or refusing requests for donations or other kinds of support may be the only way that most people can engage in social avoidance. Although it is sometimes said that elected officials are punished at the polls when voters cast ballots for their rivals, the plurality of reasons for the results of elections, and the many different spins that can be put on those results, make this a less communicative mode of punishment for incitement, even if it is in some ways effective as a means of preventing further wrongs insofar as it removes individuals from positions of authority.
One of Radzik’s main concerns in considering social punishment is with determining what forms (or extents) of punishment are proportional to particular violations of norms.85 Social norms, as we have seen, are not generally created, interpreted, or enforced by centralized bodies, and there is likely to be little “delegation” of authority for imposing sanctions in most cases. Where my analysis departs from Radzik’s is in noting that as far as public figures are concerned, often the only effective forms of rebuke or social avoidance require cumulative action by large numbers of persons. This is clearly true of boycotts and may also be true of effective rebukes, at least where factional feeling runs high. In such cases, it is not an appropriate test of proportionality to ask how many different persons participated in applying a sanction, for the effective sanction actually requires the parallel action of many persons. Instead, those concerned with proportionality might be asked to consider the appropriate duration of the boycott, especially once the incitement in question has ended or a public figure has lost his or her status.86 Should public figures who abuse their positions by engaging in incitement be shunned forever? Should they be blacklisted from all professional or highly remunerative employment, even outside politics or positions of cultural influence? And by what actions or expressions can such figures signal sincere changes of heart that might warrant an end to social sanctions? These are all questions worthy of further study. For now, however, I consider how a social norm against incitement might emerge within particular societies.
We saw previously in this chapter that there are many different pathways for the elimination of social norms. What was crucial in each of those cases was that a change in empirical beliefs about the likely conduct of others should proceed, or at least coincide with, a change in normative beliefs about what behaviors are prescribed, permitted, or prohibited. When considering the creation of social norms, the order of operations is reversed: first, new normative beliefs must emerge, followed by new empirical expectations about the conduct of others.87 With this in mind, we may examine how a social norm against incitement might intentionally be cultivated within modern political societies.
The pathways available for cultivating a social norm against incitement will differ depending on how many of the risk factors for mass atrocity exist within a given society. Where stable democratic institutions and free access to multiple modes of expression are present, campaigns for new social norms may be conducted freely in broadcast media, schools and classrooms, civil society settings, and so forth.88 Critiques of such initiatives will also receive wide airing. Where populations live in conditions of transitional justice, characterized by inequality, uncertainty, and widespread violations of purported moral and legal norms, the opportunities available for promoting new social norms of any kind will be more limited.89 Finally, in societies on the very brink of large-scale crimes, where incitement is not a distant prospect but a daily reality, the options for intentionally cultivating a social norm against incitement will be few and the risks to promoters of such a norm substantial.
One method for spreading a new social norm is to graft it onto an existing, widely accepted norm. Political scientist Richard Price defines grafting as a process of strategic exploitation of preexisting norms for the purpose of bringing emergent or previously unregulated modes of behavior under normative control.90 To say that grafting is strategic means that it does not involve an unconscious process of norm expansion, but rather proceeds via the conscious efforts of interested parties engaged in “active, manipulative persuasion.”91 Grafting seems to work best when the new norm being promoted shares substantial features with the preexisting norm (for example, it has the same modality or applies to the same reference group), and when the preexisting norm is seen as successful in regulating the behaviors it targets.92
Existing social norms against bullying are obvious candidates on which to graft a social norm against incitement. Like incitement, bullying is an action defined in part by its public nature; like bullying, incitement seems to be at the same time an expression of strength and of weakness. Furthermore, the modality of social norms against bullying and against incitement is the same. Both are prohibitive, and just as it makes no sense to saying bullying is to be avoided “except under the right conditions,” so too a prospective social norm against incitement would presumably apply without exceptions.
There is, however, a key difference between existing social norms against bullying and a prospective social norm against incitement. This concerns the size of the populations in which such social norms would circulate. Social norms against bullying are, in general, small-scale affairs, developed (where they are) in specific offices, schools, arenas, or other places where people regularly gather. This reflects the fact that bullying itself is generally carried on at a small scale against individuals or small groups of people by other individuals or small groups of people. There is, of course, a metaphorical sense in which one nation can bully another nation or one party within a state can bully another party, but these are not the core meanings of bullying, and it is less clear that well-developed social norms exist to constrain bullying in these cases. For that reason, it is not just a difference of scale that makes the possibility of grafting less clear here, but also the less obvious fact of success of a bullying norm at the level of whole populations.
Another pathway for creating new social norms, which does not always involve invocations of preexisting social norms, focuses instead on the persuasive action of norm promoters. Cristina Bicchieri has given substantial attention to the power of “trendsetters” to help create new social norms.93 Trendsetters, also sometimes referred to as “norm entrepreneurs,” are individuals who either visibly infringe existing norms or take unusual steps to establish new norms. They are typically individuals who are less sensitive to risks (particularly, social risks) than others and may also be less likely to perceive risks in specific populations. While these are figured by Bicchieri as differences in basic psychological profiles, there are also differences in social positions that can insulate trendsetters against risks while also making their actions more visible, and hence easier to emulate.94
How can trendsetters make a difference in campaigns against incitement? As noted, it is important to be aware of the different opportunities and costs of action in different kinds of societies, but certainly in cases of deeply divided societies, significant risks attach to public criticism of officials who engage in incitement. This is true both where the figures engaging in incitement control the security forces of a state and can use them to attack their critics, and in cases where state institutions are weakened to the point that private ruffians can be used to threaten or attack opponents. One of the most admirable aspects of public rebukes under such circumstances is that they may help disrupt the otherwise ubiquitous appearance that other members of society approve of the hateful messages of incitement that are being targeted at specific groups, thus potentially providing a basis for the development of new empirical, as well as normative, beliefs about the limits of legitimate political speech.
I have argued that scholars and policymakers should pay more attention to social norms in their search for effective strategies for preventing mass atrocities. My discussion, which has been largely prospective in focus, is likely to attract objections on a variety of points. In this section, I address three of these.
First, it might be objected that the particular social norms I have considered, focusing specifically on speech and silence, are not representative of social norms more broadly, and so they do little to demonstrate the potential of social norms to aid future efforts at atrocity prevention. Expressive acts differ qualitatively from other norm-guided actions that agents perform, insofar as they mainly aim to convey certain ideas, values, or commitments. The norms that most significantly affect whether societies experience mass atrocities, however, are not those that govern what can be said about such crimes, whether retrospectively or prospectively, but rather those that influence individuals’ decisions to participate or avoid participation in physical acts of violence.
This objection rightly directs our attention to the fact that the basis for discovering and interpreting particular social norms, especially from the perspective of an outsider, may be more or less difficult depending on whether those norms themselves concern expressive behavior or other kinds of conduct. Testimonial and documentary sources both provide many clues about norms concerning expression or silence, as when a witness to atrocity like Victor Klemperer notes that certain things could not be talked about publicly or when a document issued to military officers explicitly prescribes the use of particular euphemistic words and phrases. In the same way, it is perhaps easier to identify norms needing elimination or describe norms worth developing for preventive purposes when these norms target expression rather than other kinds of action. Nevertheless, I think this objection fails, insofar as social norms governing expression have the same two distinguishing features as other kinds of social norms. They are, in other words, sufficiently representative of social norms bearing on mass atrocity more broadly.
A second objection focuses not on my proposals for identifying or changing social norms, but rather on the one strategy I have rejected: that of modeling prevention on rescue. This objection holds that whereas I have argued that the study of rescue and rescuers provides an inadequate guide to effective prevention, in fact many of the pathways to prevention that I have discussed call for individuals to exhibit virtues very similar to those of rescuers. Accordingly, the objection continues, broad educational campaigns featuring rescuers might be just the sort of thing needed to give individuals who have suffered mass violence the courage to speak out or to inspire soldiers or civilians confronting incitement to make public their criticisms.
I have no problem with the general claim that the virtues manifested by rescuers overlap with those manifested by individuals and groups that attempt at some personal risk to prevent atrocities. But we should be wary of pushing this comparison too far. As Lawrence Blum points out, we may admire, and morally praise, the actions of rescuers even when these do not seem to match traditional conceptions of virtue.95 A traditional view of courage, for example, is that when carried to excess, it becomes the vice of rashness, but it seems that in cases of rescue (or attempted rescue) from atrocities, this outer limit to courage vanishes; otherwise, how could we praise those who put large numbers of lives at risk in order to save just a few? It is not obvious that the courage exhibited in preventive action has this structure, and this may be why some efforts at gathering testimony from assault victims in patriarchal societies take the form of evading, rather than publicly breaching, social norms (since not only the victim herself but her entire family may be at risk). Similarly, in philosophical discussions of virtue, it is common to draw a distinction between manifestations of virtue in short-term activities and in longer-term plans or projects, but here again, rescue and prevention stand largely on different sides. Accordingly, while persons seeking to prevent genocide and other forms of mass atrocity may learn from rescuers, rescue remains an inadequate model for atrocity prevention.
A final objection targets my proposed social norm against incitement and argues that it is morally unjustifiable or impracticable, or both. Those who believe legal prohibitions on incitement are at least potentially morally justifiable, if implemented in the right way, are likely to think the same about a social norm against incitement. But some liberals and libertarians, particularly those inspired by John Stuart Mill, may believe that even the ugliest forms of expression should be permitted, for epistemic reasons, in democratic societies. Hearing ugly speech or abhorrent proposals, on this account, is important even for those who detest them, because it provides access to the sincere views of some fellow citizens and an outlet for arguments that we must know about if we are to maintain our facility in defeating them.96 Mill himself might permit social sanctions on forms of expression that reliably lead to interpersonal harms more serious than mere suppression of speech, but given the distinction between incitement and instigation, as well as the general unreliability of claimed causal connections between expression and physical violence, it is hard to see whether, or in what cases, this exception applies.97 So, the objection goes, we should err on the side of protecting expression and reject social punishment even of vile speech.
One response to this objection, which I do not think is successful, is to say that the kinds of speech that count as incitement are not arguments at all but imperatives, and so the second of Mill’s reasons for allowing such expressions does not apply. It is true that many of the specific speech acts that have been prosecuted as incitement do not immediately appear to be arguments, with premises clearly lined up in order to prove a (practical) conclusion. However, what seems to me to be going on here is that those who engage in incitement, like many other political actors, are using enthymemes to advance their agendas. Enthymemes are arguments that, as expressed, contain one or more suppressed premises.98 An example in this context might be, “All Kulaks are enemies of the revolution,” (therefore) “All Kulaks must be eliminated,” where the suppressed premise is “enemies of the revolution must be eliminated.” My view is that we should not be too quick to deny that acts of incitement express arguments, even if those arguments are outrageous.
Rather than respond to the libertarian objection to prohibitions on incitement as above, I think it is more promising to point to the fact that social punishment of the kind I have recommended is much more likely to be proportional to the harms embodied in or aimed at by incitement than legal punishment, and so more likely to withstand Millian scrutiny. It is true that incitement need not actually lead to the perpetration of mass atrocities in order for it to be socially sanctioned, on my view, but it is a constitutive feature of incitement that the agent performing it intends that such a harm should occur and takes steps to bring it about. It would be disproportionate, in my view, to shoot such a person, or perhaps even to jail him or her for an extended period of time. But it is not disproportionate to ignore that person, to publicly rebuke him or her, to mount a recall election, or to boycott his or her speeches or business. Indeed, even philosophers like Jeff McMahan who strongly condemn boycotts of purportedly offensive speech and tactics like no-platforming make an exception for acts of incitement.99
This raises a final point about the relationship between legal and social norms against incitement. It is often argued that legal norms require for their legitimacy support from social norms within the particular populations in which they circulate. This is sometimes taken as a reason for creating or adopting laws against mass atrocity in postconflict societies, as in Sheri Rosenberg and Mark Drumbl’s accounts of the expressive functions of international criminal law.100 But a different lesson is that legal norms that are not closely paralleled by social norms in particular societies are unlikely to be capable of guiding conduct and may thus actually be invalid.101 If this reasoning is right, then readers who believe in the importance and justifiability of legal norms against incitement should also take up the project of establishing comparable social norms. Both types of norms may work together to arrest incitement and in this way help prevent mass atrocities. Indeed, this strategy encapsulates the overall argument of this book, which is that we must attend equally to moral, legal, and social norms if we are to explain and constrain large-scale crimes.
1. The history of Darfur, or the “Kingdom” (Dar) of the Fur people, goes back to at least the fourteenth century CE. Turkish and Egyptian colonization of the region began in the 1870s, with British colonial administration following in 1916. Since 1994, the region of Darfur has comprised three states: North Darfur, South Darfur, and West Darfur. For discussion of this region’s complex history, see Gerald Prunier, Darfur: The Ambiguous Genocide (Ithaca, NY: Cornell University Press, 2005).
2. Prunier, Darfur, 128–129.
3. Not everyone interested themselves in this question at first. As UN Secretary General Kofi Annan said in April 2004, “Whatever term it uses to describe the situation [in Darfur], the international community cannot stand idle.” Cf. “Darfur Destroyed,” Human Rights Watch 16, no. 6 (2004): 53. On March 4, 2009, however, the Pre-Trial Chamber of the International Criminal Court issued a warrant for the arrest of then-Sudanese president Omar Al Bashir on charges of war crimes and crimes against humanity. A charge of genocide was latter added to this warrant. See ICC Case Information Sheet, The Prosecutor v. Omar Hassan Ahmad Al Bashir (ICC-PIDS-CIS-SUD-02–006/18_Eng), April 2018.
4. Samer Abdelnour and Akbar Saeed, “Technologizing Humanitarian Space: Darfur Advocacy and the Rape-Stove Panacea,” International Political Sociology 8 (2014): 146.
5. Abdelnour and Saeed, “Technologizing Humanitarian Space.
6. Abdelnour and Saeed, “Technologizing Humanitarian Space.
7. Romy Listo, “Gender Myths in Energy Poverty Literature,” Energy Research and Social Science 38 (2018): 14.
8. Paul Morrow, “The Thesis of Norm Transformation in the Theory of Mass Atrocity,” Genocide Studies and Prevention 9, no. 1 (2015): 66–82; Heinrich Popitz, “Social Norms,” Genocide Studies and Prevention 11, no. 2 (2017): 11–12.
9. For two different ways in which norms can be called “bad,” see Geoffrey Brennan, Robert Goodin, Nicholas Southwood, and Lina Eriksson, Explaining Norms (New York: Oxford University Press, 2013), 176. For a critical account of “parochial” evaluations of norms as good or bad, see John Thrasher, “Evaluating Bad Norms,” Social Philosophy and Policy 35, no. 1 (2018): 196–216.
10. Cristina Bicchieri, Norms in the Wild (New York: Oxford University Press, 2017); Karisa Cloward, When Norms Collide (New York: Oxford University Press, 2016).
11. For discussion of the political dimensions of the designation “Righteous Among Nations,” accorded specifically to rescuers during the Holocaust, see Sarah Gensburger, “From the Memory of Rescue to the Institution of the Title of ‘Righteous,’” in Resisting Genocide: The Multiple Forms of Rescue, ed. Jacques Semelin, Claire Andrieu, and Sarah Gensburger (London: Hurst, 2011), 19–32.
12. Nechama Tec, When Light Pierced the Darkness: Christian Rescue of Jews in Nazi-Occupied Poland (New York: Oxford University Press, 1987), 70–84.
13. Tec, When Light Pierced the Darkness, 87–98.
14. Ernesto Verdeja, “Moral Bystanders and Mass Violence,” in New Directions in Genocide Research, ed. Adam Jones (New York: Routledge, 2012), 153–168.
15. Compare this to the definition of “rescue” offered by Jacques Semelin, as “a set of actions, whether covert or not, that aim to legally or physically conceal the identity of wanted persons and/or organize their escape to a place where they will find safety.” Where the element of risk, or at least cost, is implicit in Semelin’s definition, it is explicit in my own. See Semelin et al., Resisting Genocide, 5–6.
16. Tec, Light Pierced, 153–154; Kristen Renwick Monroe, Ethics in an Age of Terror and Genocide (Princeton, NJ: Princeton University Press, 2012), 221.
17. Bicchieri, Norms in the Wild, 170–171; Cloward, Norms Collide, 79.
18. For the significance of this distinction, which Bicchieri characterizes as the distinction between “risk sensitivity” and “risk perception,” see Bicchieri, Norms in the Wild, 172–173.
19. Bicchieri, Norms in the Wild, 76–79. Bicchieri goes on to distinguish cases in which there are multiple norms from cases where there are multiple possible interpretations of the same norm; this latter dynamic also appears in some testimonial accounts of rescue.
20. Sara Brown, Gender and the Rwandan Genocide: Women as Rescuers and Perpetrators (London: Routledge, 2018), 73.
21. Istvan Pal Adam, Budapest Building Managers and the Holocaust in Hungary (Cham, Switzerland: Palgrave, 2016). In this case, since many of these building managers continued to receive tips from building residents, there is some controversy about whether they should be called rescuers or paid helpers.
22. Samuel Oliner and Pearl Oliner, The Altruistic Personality: Rescuers of Jews in Nazi Europe (New York: Free Press, 1988).
23. Oliner and Oliner, Altruistic Personality, 188.
24. Oliner and Oliner, Altruistic Personality, 209.
25. Oliner and Oliner, Altruistic Personality, 189.
26. Oliner and Oliner, Altruistic Personality, 221.
27. Oliner and Oliner, Altruistic Personality, 258. I understand the final quotation here to be making an epistemic point: social norms are often the means by which individuals come to know of properly moral norms. An alternative perspective on the relationship between social and moral norms, which may be relevant here, holds that many moral norms simply are social norms that have been “moralized,” that is, made the object of distinct kinds of practical commitments and normative attitudes. This view is defended in Chad van Schoelandt, “Moral Accountability and Social Norms,” Social Philosophy and Policy 35, no. 1 (2018): 217–236.
28. Ervin Staub, Overcoming Evil (New York: Oxford University Press, 2013); Monroe, Ethics in an Age; Semelin et al., Resisting Genocide.
29. Lawrence Blum, “Altruism and the Moral Value of Rescue,” in Moral Perception and Particularity (Cambridge: Cambridge University Press, 1994), 124–143.
30. Blum, “Altruism and the Moral Value,” 135.
31. Blum, “Altruism and the Moral Value,” 126, 135.
32. Oliner and Oliner, Altruistic Personality, 258. Erwin Staub also emphasizes “the possibility of education to reduce hate.” See Staub, Overcoming Evil, 363–364.
33. For an analysis of the shortcomings of current educational initiatives, along with proposals for tightening the connection between education about genocide and genocide prevention, see Deborah Mayersen, “So the World May Know All: The Importance of Education for Genocide Prevention,” in Last Lectures on the Prevention and Intervention of Genocide, ed. Samuel Totten (New York: Routledge, 2018), 215–221.
34. Philosopher Tracy Isaacs uses this term to describe perpetrators of some historical large-scale crimes; I believe it applies equally to many rescuer groups. Cf. Tracy Isaacs, Moral Responsibility in Collective Contexts (New York: Oxford, 2011), 27–29.
35. Oliner and Oliner, Altruistic Personality, 247.
36. Tec, Light Pierced, 59.
37. Brown, Gender and the Rwandan Genocide, 130.
38. Brown, Gender and the Rwandan Genocide, 131.
39. Bicchieri, Norms in the Wild, 15.
40. As Fahima Hashim notes, “Darfurian women are known for their significant economic contributions and their participation in market and agricultural economies.” By contrast, as Bina Agarwal observes, women in rural India who engage in firewood collection are largely excluded from market and agricultural work. Cf. Fahima Hashim, “Sudanese Civil Society Strategizing to End Sexual Violence against Women in Darfur,” in Darfur and the Crisis of Governance in Sudan, ed. Salah Hassan and Carina Ray (Ithaca, NY: Cornell University Press, 2009), 236; Bina Agarwal, Gender and Green Governance: The Political Economy of Women’s Presence within and beyond Community Forestry (New York: Oxford University Press, 2010), esp. chap. 2.
41. Listo, “Gender Myths.” For a discussion of the ways in which gender-based social norms have also influenced patterns of sexual violence against men in Darfur, see Gabrielle Ferrales, Hollie Brehm, and Suzy McElrath, “Gender-Based Violence against Men and Boys in Darfur: The Gender-Genocide Nexus,” Gender and Society 30, no. 4 (2016): 565–589.
42. For the place of “experience-near” ethnographic research in the study of genocide, see Alex Hinton, “The Dark Side of Modernity: Toward an Anthropology of Genocide,” in Annihilating Difference: The Anthropology of Genocide, ed. Alex Hinton (Berkeley: University of California Press, 2002), 33. For a more general philosophical discussion of ethnographic research as one of three distinct “levels” of norm analysis, see Thrasher, “Evaluating Bad Norms,” 207.
43. Brennan et al., Explaining Norms, 176.
44. Bina D’Costa, “Birangona: Rape Survivors Bearing Witness in War and Peace in Bangladesh,” in Women and Genocide, ed. Elissa Bemporad and Joyce Warren (Bloomington: Indiana University Press, 2018), 159–190. For an example of patriarchal social norms prescribing silence about domestic violence outside the context of mass atrocity, see Karima Manji, “Articulating the Role of Social Norms in Sustaining Intimate Partner Violence in Mwanza, Tanzania.” (PhD diss., London School of Hygiene and Tropical Medicine, 2018), 94–97.
45. D’Costa, “Birangona.” See also Elisa von Joeden-Forgey, “Gender and the Future of Genocide Studies and Prevention,” in Genocide Studies and Prevention 7, no. 1 (2012): 93–94.
46. Annette Wieviorka, The Era of the Witness (Ithaca, NY: Cornell University Press, 2006), 72.
47. Lawrence Langer, Holocaust Testimonies: The Ruins of Memory (New Haven, CT: Yale University Press, 1993), 61.
48. Brennan et al., Explaining Norms, 177.
49. Bicchieri, Grammar of Society, 20–22. Some historians dispute claims about a general hesitation among survivors to give testimony during the first decades after the Holocaust. Hasia Diner seeks to dispel the “myth of silence after the Holocaust” among American Jews. Laura Jockusch offers a Europe-centered corrective to claims about survivor silence, highlighting the extent of immediate postwar activism in European displaced persons camps. At the same time, Jockusch acknowledges that many Jews attempting to rebuild their lives after the war encountered hostility or indifference from their interlocutors and were led to “conceal their Jewishness in public, as well as to keep their memories of the recent past private.” In light of such controversy, I can only suggest that much more targeted historical inquiry will be required to confirm or disconfirm the hypothesized existence of social norms prescribing silence within specific communities of survivors during this period. Cf. Hasia Diner, We Remember with Reverence and Love: American Jews and the Myth of Silence after the Holocaust, 1945–1962 (New York: NYU Press, 2009); Laura Jockusch, Collect and Record: Jewish Holocaust Documentation in Early Postwar Europe (Oxford: Oxford University Press, 2012), 191.
50. Linda Radzik, “Moral Rebukes and Social Avoidance,” Journal of Value Inquiry 48 (2014): 648–650.
51. Radzik, “Moral Rebukes.” It should be noted that whereas Radzik is concerned with social avoidance as a response to actual violators of moral norms, I am interested in social avoidance merely as a response to real or perceived violations of social norms. I do not think victims of mass atrocities generally fear that they may violate moral norms by speaking about their experience; with social norms, however, such fears are intelligible.
52. Radzik suggests that social avoidance more generally risks misinterpretation because it typically takes the form of a withdrawal from communication; it is for this reason often much more ambiguous than explicit criticism or rebuke. See Radzik, “Moral Rebukes,” 650.
53. Alex Bellamy, Massacres and Morality: Mass Atrocities in an Age of Civilian Immunity (Oxford: Oxford University Press, 2012), 36. Note that this pathway assumes, as Bellamy’s study argues, that threats of international intervention are at least moderately effective deterrents of mass atrocity.
54. Elazar Barkan, “Historical Dialogue and the Prevention of Atrocity Crimes,” in Reconstructing Atrocity Prevention, ed. Sheri Rosenberg, Tibi Galis, and Alex Zucker (Cambridge: Cambridge University Press, 2016), 175–195.
55. Barkan, “Historical Dialogue,” 192.
56. Kerry Whigham, “Remembering to Prevent: The Preventive Capacity of Public Memory,” Genocide Studies and Prevention 11, no. 2 (2017): 65.
57. Bicchieri, Norms in the Wild. More specifically, Bicchieri refers to shared second-order normative and empirical expectations, meaning expectations about what other group members expect. Bicchieri’s use of the language of expectations differs from my use of beliefs, attitudes, and commitments, but that difference is driven mainly by her desire to give an operational account of norm transformation—one that can be tested directly in experimental settings.
58. Bicchieri, Norms in the Wild, 110. The opposite, as we shall see in the next section, is true of the creation of social norms: for new norms to be created, first new normative expectations must be inculcated and then new empirical expectations formed.
59. Mark Osiel, “The Collective Responsibility of Military Officers,” in Making Sense of Mass Atrocity (Cambridge: Cambridge University Press, 2011), 203–217.
60. For discussion of the relationship between blackmail and social, or “group,” norms, see Richard McAdams, “Group Norms, Gossip and Blackmail,” University of Pennsylvania Law Review 144 (1996): 2237–2292.
61. D’Costa, Birangona, 178. For a general discussion of problems and methods in collecting data about conflict-related sexual violence (CRSV), see Chein Reis, “Ethical, Safety, and Methodological Issues Related to the Collection and Use of Data on Sexual Violence in Conflict,” in Sexual Violence as an International Crime: Interdisciplinary Approaches, ed. Anne-Marie De Brouwer, Charlotte Ku, Renée G. Römkens, and L. J. Van Den Herik (Cambridge: Intersentia, 2013), 187–210.
62. Milli Lake, “Organizing Hypocrisy: Providing Legal Accountability for Human Rights Violations in Areas of Limited Statehood,” International Studies Quarterly 58, no. 3 (2014): 515–526.
63. Bicchieri, Norms in the Wild, 188–94.
64. For discussion of “catalysts” see Cloward, Norms Collide, 86–87. Phil Orchard uses the term crisis to refer to such events in his analysis of the evolution of legal protections for refugees. See Phil Orchard, A Right to Flee (Cambridge: Cambridge University Press, 2014), 33–34.
65. Wieviorka, Era of the Witness.
66. Hannah Arendt, Eichmann in Jerusalem: A Report on the Banality of Evil (New York: Penguin, 2006). For two different interpretations of the Eichmann trial and Arendt’s analysis of it, see Lawrence Douglas, The Memory of Judgment: Making Law and History in the Trials of the Holocaust (New Haven, CT: Yale University Press, 2006), esp. chap. 6; Christian Gerlach, “The Eichmann Interrogations in Holocaust Historiography,” Holocaust and Genocide Studies 15, no. 3 (2001): 428–452.
67. A few authors take a contrary view and argue that silence may in fact be what is most needed in post-atrocity societies in order to prevent future occurrences. Reporter and social critic David Rieff advances this view in his 2016 book, In Praise of Forgetting. However, as other scholars have pointed out, Rieff’s account evinces minimal awareness of the state of the art in mass atrocity forecasting or prevention and provides generally weak support for his polemical claims. See David Rieff, In Praise of Forgetting (New Haven, CT: Yale University Press, 2016); see also Whigham, “Remembering to Prevent,” 53–54.
68. For the importance of front-line journalism as a source of information about large-scale crimes that can inform, or indeed induce, political responses, see Roy Gutman, “Ending the Silence on War Crimes: A Journalist’s Perspective,” in Preventing Mass Atrocities: Policies and Practices, ed. Barbara Harff and Ted Robert Gurr (New York: Routledge, 2018), 73–92.
69. For discussions of the definitional difficulties surrounding incitement to genocide or other large-scale crimes in international law, see Richard Ashby Wilson, Incitement on Trial (Cambridge: Cambridge University Press, 2017), esp. chap. 1; Gregory Gordon, Atrocity Speech Law (New York: Oxford University Press, 2017), esp. chap. 8; and Susan Benesch, “Vile Crimes or Inalienable Right: Defining Incitement to Genocide,” Virginia Journal of International Law 48 (2008): 491–498.
70. Wilson Incitement on Trial, 72.
71. Wilson Incitement on Trial, 32.
72. Legal scholar Gregory Gordon uses the term atrocity speech to refer to incitement, instigation, persecution, and other legally proscribed forms of speech related to atrocities. See Gordon, Atrocity Speech.
73. Gordon, Atrocity Speech, 306.
74. Gordon, Atrocity Speech.
75. This worry is not confined to prohibitions on incitement, but extends to any legal regulation of speech. Jeremy Waldron offers a clear account of the basic concern, writing of the “massive power” modern governments enjoy “to suppress dissent, deflect criticism, and resist exposure of … malfeasances.” Jeremy Waldron, The Harm in Hate Speech (Cambridge, MA: Harvard University Press, 2014), 26.
76. Wilson, Incitement on Trial, 66–67; Gordon, Atrocity Speech, chap. 11.
77. The studies of social norms prescribing silence about sexual violations considered above may be an exception to this rule, though those norms typically are not specific to political speech.
78. Cheshire Calhoun, “The Virtue of Civility,” Philosophy and Public Affairs 29, no. (2000): 251–275.
79. Iris Marion Young, Inclusion and Democracy (Oxford: Oxford University Press, 2002), 47.
80. Wilson, Incitement on Trial, chap. 7.
81. John Stuart Mill, On Liberty, in Collected Works of John Stuart Mill, vol. 18, ed. J. M. Robson (Toronto: University of Toronto Press, 1977), 220, cited in Linda Radzik, “On Minding Your Own Business: Differentiating Accountability Relations within the Moral Community,” Social Theory and Practice 37, no. 4 (2011): 577.
82. See Radzik, “Moral Rebukes”; Linda Radzik, “Boycotts and the Social Enforcement of Justice,” Social Philosophy and Policy 34, no. 1 (2017): 102–122; Linda Radzik, “Gossip and Social Punishment,” Res Philosophica 93, no. 1 (2016): 185–204.
83. Radzik, “Moral Rebukes.”
84. Wilson, Incitement on Trial, chap. 7.
85. Wilson, Incitement on Trial, 658.
86. Radzik acknowledges the significance of duration as an element in determining the proportionality of social punishments, but does not provide more extended analysis of the appropriate duration of boycotts. See Radzik, “Boycotts,” 117, 119.
87. Bicchieri, Norms in the Wild, 110–111.
88. Such community-wide campaigns for transformation in social norms receive significant recognition and support from state authorities. For discussion of the normative and practical dimensions of state participation in the design of social norms, see Ryan Muldoon, “Understanding Social Norms and Changing Them,” Social Philosophy and Policy 35, no. (2018): 146.
89. Colleen Murphy, Conceptual Foundations of Transitional Justice (Cambridge: Cambridge University Press, 2017).
90. Richard Price, “Reversing the Gunsight: Transnational Civil Society Targets Land Mines,” International Organization 52, no. 3 (1998): 613–644.
91. Price, “Reversing the Gunsight,” 617.
92. Price, “Reversing the Gunsight.”
93. Bicchieri, Norms in the Wild, chap. 5.
94. Bicchieri, Norms in the Wild, 172–180.
95. Blum, “Moral Exemplars.”
96. John Stuart Mill, On Liberty (New York: Penguin Classics, 2007), chap. 2.
97. One reason to think that Mill would accept at least social sanctions against such harms is that they are not purely self-regarding, but rather (by definition) intended to harm others. See Radzik, “Minding Your Own Business,” 581.
98. This is at least one traditional way of defining an enthymeme. For discussion, see Edward Madden, “The Enthymeme: Crossroads of Logic, Rhetoric, and Metaphysics,” Philosophical Review 61, no. 3 (1952): 368–376.
99. Jeff McMahan, “I Was No-Platformed. Here’s Why It’s Counterproductive,” New Statesman, January 14, 2019, https://www.newstatesman.com/2019/01/i-was-no-platformed-here-s-why-it-s-counterproductive.
100. Rosenberg, Audacity of Hope; Mark Drumbl, Atrocity, Punishment, and International Law (Cambridge: Cambridge University Press, 2007).
101. Gerald Mackie, “Effective Rule of Law Requires Construction of a Social Norm of Legal Obedience,” in Cultural Agents Reloaded, edited by Carlo Tognato (Cambridge, MA: Harvard University Press, 2017), 313–334.