Choosing Among Alternative Responses to Mass Atrocity: Between the Individual and the Collectivity

| September 18, 2015
Facebook Twitter Email
Print Friendly, PDF & Email
Rwanda Genocide Memorial. (Courtesy: DFID - UK Department for International Development/Flickr)

Rwanda genocide memorial. (Courtesy: DFID – UK Department for International Development/Flickr)

The most distinctive feature of mass atrocity,1 compared to other wrongs, is that the possibility of “collectivizing” our understanding of both wrongdoers and victims is always credibly within conceptual reach, and never beyond the bounds of serious consideration. To be sure, it is always possible, concerning virtually every kind of misconduct, to discern a collective or institutional aspect, and try to work that systemic or structural element into the legal foreground. Yet Western law finds its moral and methodological bearings in individualism, and therefore strongly disfavors this conceptual move in nearly all fields, where it is not simply foreclosed altogether.

With mass atrocity, though, we have conflicting impulses on the question, sometimes evenly balanced; these prompt us in opposing directions at the same time. These warring inclinations influence real-life events by inducing those engaged in decision-making concerning “transitional justice” to favor differing modes of response to similar events, creating the appearance, at least, of incoherence. Yet mass atrocity is a hydra-headed monster, and each approach to it—in how it alternately aggregates or isolates the characters—simply attends to its different aspects, often equally present in a given set of facts.

Each method of atrocity response offers a certain representation of events, and defines the characters as suits its specific purposes: factual truth-telling, causal explanation, retribution, deterrence, compensation, memorialization, or simply the expression of moral condemnation. Differences of opinion over which methods to adopt thus often turn on disagreements over which such purposes to prioritize, especially when these appear likely, in the case at hand, to run mutually afoul. In real life, however, none of these purposes can remain indifferent to, or extricate itself entirely from (at least some of) the rest, when pursued in the same political circumstances. Thus, for instance, the policy responses that focus on thwarting recurrence are more convincing, and hence more likely to win favor, when they capture something genuinely salient in demonstrable facts about who was chiefly responsible for past events, and chiefly suffered their consequences, in light of how and why mass atrocity actually occurred.

When someone opposes a given mode of accountability, she often argues that, if one fairly examines the relevant facts about who really did what to whom, one finds them inconsistent with how this particular response to atrocity insists on characterizing wrongdoer or victim; it treats one or both as individuals when they’re really collectivities,2 or the other way around. Thus, for instance, those who claim to find the chief sources of mass atrocity in the dysfunctional workings of some large institution, such as the armed forces or the inequities of a land tenure system, often express grave reservations about the frequent, post-atrocity focus on the criminal prosecution of individuals. Others retort that any effort to expand the analytic frame beyond discrete criminal acts itself distorts history by reducing the flesh-and-blood human beings, even the most obviously consequential, to mere playthings of vast impersonal forces.

Let us begin by very briefly observing how each of the leading forms of atrocity accountability adopts a distinctive stance on the question of whether to aggregate or individualize the wrongdoer and victim. People are often torn between these competing moves, in ways bearing powerfully on their choice of atrocity response. Why this is so becomes the focus of this article’s second half.


International Criminal Law

This body of law conceives the wrongdoers as individuals, and the victims chiefly as groups. With regard to wrongdoers, the International Law Commission and Rome Statute for the International Criminal Court reject the concepts of state criminality and of corporate criminal liability. By formal stipulation, then, collectivities are insusceptible to prosecution.

Peace Palace, International Court of Justice, The Hague. (Courtesy: United Nations Photo)

Peace Palace, International Court of Justice, The Hague. (Courtesy: United Nations Photo/Flickr)

Yet this decision to blame only individuals, though intelligible and appealing, encounters the problem, practical and conceptual, that even the most powerful leader cannot commit mass atrocities by himself. International criminal law therefore devises rules on so-called “modes of commission” to link the conduct and intentions of a particular, upper-echelon defendant to the acts of those directly performing the violence, often through a large number of intermediaries. Hence the novel doctrines of joint criminal enterprise, command responsibility, and control over a hierarchical administrative apparatus—all designed specifically to let us conceptualize the defendant as a true perpetrator, deserving the most serious sanction, not merely an accessory to the wrongs of others. With these legal rules, we discover a way to reinsert the collective dimension that is essential to understanding how the perpetrator managed to have so devastating an effect, that is, by means of his influence over or cooperation with others, in ways nearly unique to this type of episode. The crime of “aggression” similarly requires that the perpetrator exercise significant control over, or decision-making authority within, the war-making apparatus of a collective actor, the sovereign state. And by definition, the perpetrator of any crime against humanity acts not on his own but pursuant to “a state or organizational policy.”

As for the victims of international crimes, these were once conceived only as individuals, when the chief offenses were war crimes and piracy. Today, however, the offenses prioritized by prosecutors and in the scholarly commentary are genocide, persecution, gender crimes, extermination, enslavement, and apartheid. With all these crimes, we construe the true victim as a group. Crimes against humanity involve a “widespread or systematic attack” against a “civilian population.” The word “widespread” requires significant numbers of intended victims, while the term civilian “population” suggests that they possess some common feature, whether sociological or merely geographical.

Truth Commissions

Truth commissions conceive the wrongdoer chiefly as the sovereign state, or more precisely, certain of its agencies, whereas the country at large now becomes the victim. Only rarely are officers of the state individually named, though their names are always gathered, and sometimes confidentially referred to prosecutors for possible criminal investigation. Concerns about “due process” dissuade most truth commissions—agencies of the state, after all—from publicly naming an individual without granting him the opportunity to question his accusers.

The names of individual victims are often made public, with recommendations for their compensation. Yet the published report is nearly always written up to emphasize how the state as whole, through its security forces, harmed the country as a whole, in multiple ways eluding other forms of redress. Many believe that a truth commission should address questions about the origins and explanation of events, broadly conceived, in ways that neither criminal nor civil courts can or should do. From this standpoint, it is a mistake to emphasize the place of individuals, as either wrongdoers or victims, in officially recounting such large-scale events, often covering long periods and an expansive geography. That would only detract from this broader task of “getting history right” and suggesting far-reaching reforms in law or policy.

Vetting Personnel

This legal mechanism of accountability identifies the wrongdoers as individuals, but treats the victims as groups. When an entire organization, such as the secret police, is seen as the chief wrongdoer, it is simply abolished. Everyone within it is discharged, and often precluded, at the upper echelons at least, from any future public service involving policy responsibilities. The organization itself must be surgically excised from the body politic if its very design and constitution incorporate central elements, now repudiated, of the prior regime’s political ideology. More often, though, democratic reformers calculate that the government agency involved in atrocities, such as the police or military general staff, can be preserved and redirected toward different ends, simply by appointing key personnel to implement the new government’s policies.3

Still, it is necessary that the individuals at all echelons meaningfully engaged in the atrocities be removed from office if the agency is to be trusted to loyally serve the new government. Those below top ranks—provided they did not act “beyond the call of duty”—are then construed as indifferent paper-pushers and time-servers, who remain fit for service due to their lack of deep ideological commitment. Procedures for discharging specific employees vary greatly between countries in their degree of individuation, notably in how much reliable evidence must be presented against a particular civil servant to justify depriving her of tenured public employment. Personnel screening interprets the wrong to reside in the civil servant’s willingness to involve himself in the general process of implementing an evil policy, not in his discrete, intentional acts of injuring a specific person or persons. Here, the victim of repressive policies is conceived as a class or group of people—for example, all those whom a particular bureaucrat, through a single stroke of his pen, dispatched to the gulag.

Official Apology

It is necessarily an individual public servant who issues the government’s apology for mass atrocities, under his signature. Yet he does so always in his official capacity as an agent of the state, speaking in its name. It is the state as a collective actor, through its agencies and those therein employed, that constitutes the wrongdoer. The individual issuing the apology for the wrongs is never, in any event, among those responsible for causing them.

In fact, public apologies for a mass atrocity almost never name the individual leaders who initiated and oversaw it. Their followers sometimes remain too powerful to permit that measure of candor. But this reticence also seeks to capture the fact that such leaders could not and did not do this alone, that many others helped implement their criminal policies, within the state and at times beyond. Here, any attempt to individualize, by naming a few obvious names, diminishes the important sense in which the design and execution of these wrongful policies involved shared goals and truly collective, institutional dynamics.

The apology’s recipient—that is, the designated victims of mass atrocity—will be described as both all those individually injured by the violence and “society at large.” That recurring abstraction—conceptually muddy, yet seemingly inescapable here—is now understood as both the totality of its individual members and yet also as something ineffably more than their sum. A public apology is always directed to the individual victims or their immediate family members. Yet there must also be some official formulation of the way mass atrocity impacted the life of the nation. That was often through the fear permeating everyday life, the pervasive sense of mutual distrust and helplessness created by the shared knowledge of ongoing atrocities in the society’s midst, committed and suffered by one’s friends and neighbors.

Australian PM Kevin Rudd apologies to Aboriginal Stolen Generations. (Courtesy: KMJPhotography Australia/Flickr.)

Australian PM Kevin Rudd apologies to Aboriginal Stolen Generations. (Courtesy:
KMJPhotography Australia/Flickr.)

The quality of public governance will also have suffered greatly under the kind of authoritarian rulers usually responsible for mass atrocity. Thus, in both respects—in their private lives as friends or neighbors, and in their public life as citizens—people in general were victims as well. They too, thus, warrant an apology, apart from apologies to the individuals actually murdered or tortured. Wording to this effect hence always finds its way into these official pronouncements. In the discourse of official apology, then, mass atrocity is chiefly a wrong that the state—one type of collectivity—does to society and the public at large, other such types.

Civil Compensation

When victims seek monetary compensation for mass atrocity, the wrongdoer is nearly always a collectivity, if only because scarcely any individual is wealthy enough to pay so large a judgment. The deemed wrongdoer is generally the sovereign state accused of large-scale human rights abuse, of aggression (as a civil “delict”), or of grave breaches of the Geneva Conventions. Multinational corporations were occasionally sued, with intermittent success, for complicity in atrocities, notably under the Alien Tort Claims Act, and corporations are obviously collective actors. However, only in the United States did the law of civil compensation recognize this form of collective liability, and only for a time; it relied upon a dubious interpretation, ultimately rejected by the federal courts,4 of the relevant domestic statute.

How the law conceives the victim varies according to whether the wrong occurs in war or during peacetime. With peacetime violations of civil and political rights protected under international human rights law, the victims are perceived as individuals, and so receive remedies tailored to their individual losses, when possible. Those to benefit from remedial measures must also be the very same individuals harmed by the defendant’s illegal conduct. Aggregating victims into a single class remains the very rare exception, and does not abandon the individualist premises, the implicit social ontology, underlying more conventional civil litigation.

By contrast, misconduct in war more often involves state violations of international humanitarian law, rather than human rights law. And though monetary compensation remains the goal, the victim is conceptualized quite differently, as the state.

Official Memorials

Historically, public monuments in honor of a country’s dead began with war memorials, but today more often concern deaths from mass atrocity, whether in war or peace. In war memorials, the assumed wrongdoer—cause of the nation’s suffering—is always the enemy, the military antagonist on other side of the conflict, a collective actor. This remains the case with atrocity memorials; the perpetrators, even when they can be identified, are not individualized.

Hall of Names, Yad Vashem, Jerusalem. (Courtesy: Reinhardt König/Flickr)

Hall of Names, Yad Vashem, Jerusalem. (Courtesy:
Reinhardt König/Flickr)

What has changed over the last generation is how such monuments conceptualize the victim. The victim used to be rendered very abstractly, as the country at large or simply as all those who fought and died for it, who may be collectively represented in the indistinct form of an “unknown soldier.” This statuary figure is ‘individualized’ only in a necessarily conjectural, almost spectral way. Today, though, victims of war are depicted as concrete individuals, whether specific soldiers or—now increasingly—civilian atrocity victims too. Maya Lin’s Vietnam Veterans Memorial revolutionized prevailing iconography to this effect, though a few prior monuments did adopt a similar point of view.5 The new approach, since then broadly embraced, abandons the collectivizing premises of the earlier symbolic stance, which posited the nation as one body, united in the war effort, uniformly convinced by the justice of its cause. That older iconography also intimated that the people as a whole shared at least vicariously in the suffering of the troops, so that the more immediate pain of particular soldiers could be fairly subsumed under a collectivizing aesthetic.

With today’s atrocity memorials, of which there are now hundreds throughout the world, there is nonetheless some ambivalence about whether to represent the victims as individuals or groups, either state or nonstate in character. In nearly all cases, individual victims were chosen on the basis of their group membership—as Rwandan Tutsis, for instance. Yet some today hold (explained below) that depicting atrocity victims in group terms as these reproduces—paradoxically and unwittingly—the very same dehumanizing forms of collectivization which wrongdoers themselves acted upon in designing their crimes.


This is a legal response to mass atrocity in the sense that it is designed to set limits on other such responses, notably criminal prosecution, civil liability, and personnel screening. Some granting of amnesty is nearly always an indispensable component of a transitional justice regime. Wherever the number of individual wrongdoers is very large, the potential societal disruption from and political backlash against more aggressive justice seeking is considerable. Amnesty then becomes essential to ending the violence and restoring social peace.

The wrongdoers who formally receive amnesty are invariably individual persons, though one could argue that business entities sometimes also receive a de facto pass. The hard questions arise over whether to grant so-called “blanket” amnesty to all those directly involved in mass atrocity or to restrict amnesty’s scope to those whose contribution was relatively limited, as by murdering a single Bosniak neighbor. There is some indication that the interpretation of international law has shifted over the last decade from encouraging broad-scale, collective amnesties—at least for wrongs other than grave war crimes—to now favoring more individualized approaches, such as that pioneered by South Africa and thereafter employed in Timor Leste, among other places. Yet state practice has not actually much followed these pronouncements of principle. An intermediate position has also emerged: to immunize only those beneath a certain echelon in the administrative apparatus, thereby treating them collectively, while taking a more individualizing angle when evaluating those of higher rank.

Anything more precise than a very broad amnesty requires careful procedures for fairly distinguishing the actual contributions of particular individuals. Broad-based amnesties are far simpler and hence also less expensive to administer than more individuated approaches. With blanket or echelon-linked amnesties, there is no need even to identify the beneficiaries by name. But this approach offers only a rough cut at justice, often much rougher indeed.

Amnesty not only benefits former wrongdoers, but imposes further harm on their victims. Given the considerable global momentum for transitional justice, individual victims of mass atrocity today have fair reason to anticipate that those legally responsible for their suffering will meet some form of public condemnation, usually entailing punishment or compensation. As a collectivity, moreover, all the victims’ fellow citizens who favor meaningful sanctions against such wrongdoers—often their former political and military leaders, after all—are likely to feel “victimized” themselves, in a nontrivial sense, by any such step in law or policy, especially when the amnesty is comprehensive in scope, extending to highest echelons. In fact, when the wrongs in question constitute major international crimes—the case with any form of mass atrocity—then impunity for wrongdoers imposes an independent injury on the entire “international community.” For this community—certainly an otherwise elusive animal—is often and most credibly defined precisely by the company of those for whom this very class of crimes “shocks the conscience,” people therefore committed to their prosecution.

“Restorative Justice” Initiatives

One mode of atrocity accountability receiving ever greater attention throughout the world involves localized efforts at “restorative justice.” It is often the central government that imposes these initiatives. They nonetheless employ forms of dispute resolution originating in longstanding cultural practices and are enforced by traditional institutions of justice at the municipal or neighborhood level. The immediate goal is to restore harmony between the individuals immediately-affected, as victim and wrongdoer, people who in many cases maintained close relations before the atrocities. Both parties are therefore so conceptualized. The ultimate aim and hope, however, is that by inducing many such interpersonal reconciliations, these policy initiatives, in their aggregate effect, will advance nationwide reconciliation between social groups, aggressor and aggrieved.6

The Rwandan experience of gacaca, from 2001 through 2012, provides the archetype, though it has since inspired comparable initiatives in northern Uganda and the Democratic Republic of the Congo. Today, in places as far afield as Colombia and Northern Ireland, policy makers and nonstate organizations employ roughly analogous approaches. In Uganda, the DRC, and Colombia, in particular, these methods, though unorthodox from a juridical point of view, are proving especially congenial to the task of “reintegrating” thousands of former child soldiers—by then, battle-hardened adolescents or young adults—into the local communities from which they were kidnapped years before.

Gacaca Trial, Rwanda. (Courtesy: Scott Chacon/Flickr)

Gacaca Trial, Rwanda. (Courtesy: Scott Chacon/Flickr)

In Rwanda, the victims and perpetrators involved in gacaca have mostly chosen to interpret “peace” and “reconciliation” quite modestly, to entail scarcely more than non-violent cohabitation of the same small piece of geography. A wary modus vivendi often accompanies this physical co-presence. Even that small measure of progress generally proves sufficient, however, to ensure that surviving members of victims’ families refrain, most importantly, from taking vengeance on those returning home from prison. In quest of this elementary coexistence, the law and practice of gacaca shrewdly conditioned grants of amnesty from further incarceration on satisfactory gestures of personal repentance, apology, and material compensation. Individual perpetrators had to make these commitments during solemn ceremonies before a community assembly, a practice designed to maximize the upward reverberations—from micro to macro—of harmony between individuals to reconciliation between Tutsi and Hutu.

The aspect of gacaca most striking for present purposes is the state’s concerted effort to eliminate any reference—within both the proceedings themselves and in public life more generally—to Hutu and Tutsi ethnicity. All citizens would henceforth be officially conceived simply as fellow Rwandans. Their former identities—the basis of prior genocidal animus—would entirely dissolve, it was hoped. These would be subsumed into the nation, this larger collectivity, now purposefully reinvigorated. There could surely be no more radical way to deny any role to collectivities in either the causing or suffering of mass atrocity.


Mass atrocity and “aggressive” war-making are the chief inspirations for imposing sanctions. The sanctions levied by both international organizations and particular states treat the victims of wrongs as collectivities. If the wrong consists of waging aggressive war, as by Russia against Ukraine, then the victim is the state which has suffered armed attack. And if the wrong entails massive human rights abuse by a state against its own people, then the victim will be a subnational group defined in terms of some legally protected characteristic, whether religious, racial, ethnic, political, or otherwise.

More controversial is whether to understand the wrongdoer here as a collectivity, or simply as a set of discrete individuals. Global opinion has moved from the first to the second of these understandings, though with certain reservations, described below. The longstanding approach until the late 1990s was to target an entire country, via “comprehensive” trade sanctions. The rationale was that the majority of citizens must be somehow complicit in the state’s wrongs, and therefore bear nontrivial moral responsibility for them. In any event, sanctions that target society as a whole will more likely induce the people as a whole to resist the oppressive policies prompting the foreign measures now harming everyone.

It gradually became clear that both arguments were unfounded. Accumulating experience with sanctions against such countries as Iraq, Cuba, Libya, and Pakistan clarified that such measures could not induce domestic revolt against a highly repressive regime, no matter how gravely its people suffered their economic consequences. Most citizens in such regimes were simply unable to nontrivially influence major policy-making. From this ineffectualness one would have to infer the unfairness of holding the general population responsible for atrocities orchestrated entirely by ruthless political overlords.

A decisive shift the occurred away from sanctioning collectivities to targeting the individual elites responsible for a state’s most repressive policies. This meant elites both within the state itself and in sectors of the business community, like the arms trade, who profited directly from these policies. These so-called “smart sanctions,” deployed against such countries as North Korea, Angola, and Iran, inspired great hopes for nearly a generation. Yet careful empirical assessment now suggests that they are no more successful than traditional, comprehensive measures, and sometimes less so. Smart sanctions were also imposed through a process that did not—until very recently, under pressure from international human rights courts—offer any opportunity for the sanctioned individual or company to challenge the accusation of its complicity in state repression (or terrorism). In this respect, individuated sanctions simply were not individuated enough, that is, regarding individuals’ rights of due process.

These difficulties with an entirely individualizing approach have prompted a move toward an intermediate position: sanctioning neither entire states nor only specific individuals, but rather particular industries, those on which the regime most depends for hard foreign currency. An industry, like fuel exports, need not be directly implicated in the repression.7 Industries are collectivities of a sort—collective actors, in fact, since they are usually represented by a trade association. Yet specific industries are much smaller in scale than a whole country. Even so, the harmful impact of sanctioning the entire telecom, energy, or financial sector can be far-reaching, impairing the routine operation of society at large—including the regime’s organized domestic opponents. The recent experience with sanctions on Iran illustrates such unintended consequences.

“State Responsibility for Internationally Wrongful Acts”

The wrongdoers here are again sovereign states, hence collectivities of course. A state deemed legally “responsible” for wrongdoing is not free to respond that it bears no responsibility for the crimes of an individual leader or group of leaders who act beyond the scope of their domestic legal authority.

Though counter-intuitive, the victims too are collective actors. In war, when one state wrongfully harms the citizens of another, it is the second state that is deemed the legal victim, not those citizens themselves. The second state may therefore sue the first in the International Court of Justice, alleging grave breaches of the Geneva Conventions against its own soldiers or civilians. Those soldiers and civilians themselves, though, cannot sue. To this day, international humanitarian law affords the immediate human victims of war crimes no right of action for compensation from the state responsible for their mistreatment. The same is true for victims of crimes against humanity.

Additional Protocol II to the 1949 Geneva Conventions does formally require the offending state to compensate the particular individuals it has wronged. Yet that treaty does not create a private right of action in such persons, either individually or as a class. Nor does it require this of states, and none do so. The result: individual victims can only hope that—in post-war diplomatic maneuverings—the state of their nationality chooses, among its more pressing priorities, to assert their claims in its negotiations with the offending state. In short, for purposes of attributing responsibilities to states, international humanitarian law still construes war’s individual victims as mere “property” of sovereign entities, the only “legal persons” to whom it accords standing.

Counterterrorist Use of Force

Terrorist attacks often yield atrocities on a large-scale. The intended victims of contemporary terrorism are national and ethno-religious groups. This is the case, for instance, when The Islamic State targets Shiite populations in Iraq and Syria, or when Al Qaeda construes its “far enemy” as Western states and their general publics. It is individuals who are killed, of course, but as proxies for these various collectivities, of which such individuals are sometimes leaders, but more often merely members.

In defining the wrongdoers, Western statesmen emphasize that their countries are in immediate conflict chiefly with a small number of hateful individuals, the specific people who lead certain radical Jihadist groups. American diplomatic discourse, in particular, makes every rhetorical effort to disentangle these persons from the general population of the national societies spawning them, and certainly from the umma at large.8 Counterterror policy encompasses far more than drone strikes, of course, with their individualizing conception of chief wrongdoers and, by implication, of where proximate solutions lie. Yet it has been a particular source of Western consternation that international humanitarian law struggles so unconvincingly to conceptualize even the undoubted terrorist leader as a legitimate “military objective.” The problem stems from the fact that this body of law historically understands all armed conflict, by its very nature, as occurring only between sovereign states and their agents, chiefly comprised of soldiers in their armed forces (sometimes also nonstate militias, but only when under a state’s effective control). The more we wish to train our military force upon the heads of particular individuals, on terrorist leaders not under any state’s thumb, the less help we therefore find in international humanitarian law.

In one key respect, however, the U.S. government—under both the present and preceding presidential administration—seeks to collectivize the understanding of relevant wrongdoers, by contending that all members of designated terrorist groups constitute “unlawful combatants,”9 who may therefore be targeted at any moment, whether or not they pose an immediate threat of violence. This approach is more forgiving of targeted killing than that preferred by other Western states and relevant NGOs. Those entities instead contend that members of nonstate organizations remain civilians, and may therefore be attacked only at times when, as individuals, they “take direct part in hostilities,” in the applicable Geneva Convention wording.10 Yet even the Obama administration initially defended its more indulgent approach to the scope of legitimate targeting with the qualification that such attacks would occur only after a particularized factual inquiry into whether a given individual not only “belongs” to such a group, but exercises significant responsibilities within it and cannot be apprehended alive.

Thus, on the one hand, presidential directives consistently describe terrorist networks as collective actors whose individual participants may therefore be lawfully killed at any instant, once their mere “membership” is determined; in this respect, the U.S. would treat individual terrorists, leaders and followers, no differently than the individual soldiers of sovereign states at war.11 On the other hand, the President also insisted that the U.S., before placing any individual on a secret “death list,” would conduct a more detailed, elaborate investigation into his precise role, relative significance, and potential capture. These twin policies made for a novel and curious hybrid of individuation and collectivization. Its creators hoped somehow to harmonize the concepts and commitments of international humanitarian law with those of international human rights law. The former body of law is essentially collectivist in character, since sovereign states remain its core actors, whereas the goals of international human rights law are almost wholly individualist. The sphere of operation within which each legal enterprise traditionally travelled was therefore always quite different.

In any event, Obama soon found it undesirable to honor his initial commitments. The attempted compromise between legal standards for the use of lethal force in war and in peacetime hence proved short-lived. Its individualizing elements were deliberately designed to leaven its collectivizing component (and thereby mollify critics), yet quickly succumbed to it. The more particularized approach to target identification imposed such onerous evidentiary requirements as to foreclose drone strikes against probable terrorists identifiable only through their “signature.” This now-familiar term refers to a consistent pattern of activity on a suspect’s part, as by his frequent entry into a well-established training camp, or buildings clearly identified—through aerial reconnaissance, data-mining, and human intelligence—as locations where Al Qaeda or ISIS members frequently meet. The exact identity of the individuals therein may—without apparent moral qualm—remain forever unknown, unless the targeted organization chooses to confirm their deaths. Their personal identities are immaterial, in any event, if one conceives the ultimate wrongdoer as a collectivity, the terrorist network itself. One then construes its individual members, even its leaders, essentially as component parts, whose particularity one should not exaggerate when choosing suitable targets.

Public debate over the legitimacy of these so-called signature strikes focuses on the possibility that, by demanding less in the way of prior personal identification, higher numbers of innocent casualties are likely to occur than with the earlier “personality strikes,” as they are contrastingly called. Another important implication of the newer targeting approach, less often noticed, is its apparent supposition that despatching only identifiable personalities may well prove inadequate to the task of disabling the collectivity they serve. That conclusion is notably charier about ultimate prospects for strategic success than the prior practice of individuated targeting. In an odd sense of the word, personality strikes were guardedly “hopeful” in presuming that the names of even emergent terrorist figures could be readily discovered and their physical locations pinpointed to a near certainty.12

In their ethical implications, non-violent aspects of counterterror policy differ importantly, of course, from drone strikes of either sort. These policies—the increased regulation of global financial transactions—nonetheless resemble signature strikes in aiming to disrupt the criminal activities not merely of specific individuals, but of Jihadist organizations as such. The more proximate targets of such legal and administrative measures are still-larger collective actors. Certain international banks, in particular, have been implicated in money-laundering on behalf of terrorist groups and in illegal avoidance of trade sanctions against miscreant states. These financial measures receive much less publicity, though, than the drone strike, with its technological wizardry. This is not only because of their lesser “glamor.” They have also apparently, experts acknowledge, achieved very little. Underground forms of “Islamic finance” almost entirely elude these efforts, whereas drone attacks are often quite effective in their immediate tactical aims, at least.

Because media attention focuses chiefly on the flamboyant spectacle of successful drone strikes (especially against established terrorist leaders), the public face of counterterrorism continues to be highly individualistic in how it renders the central wrongdoers. This remains the case even as the less conspicuous, bureaucratic measures conceive the problem more holistically, in terms of worldwide networks of web-linked sympathizers engaged in a host of well-organized if sub rosa financial practices. These difficult, frustrating legal efforts in the commercial realm, scarcely noticed beyond specialized professional circles, extend the focus of counterterrorism policy, in its understanding of chief wrongdoers, from individual to collectivity.

The “lone world” terrorist is a recent development requiring that counterterrorism policy shift gears yet again and reconsider its answer to the recurring question: who is really the (most relevant) wrongdoer, individual or collectivity, that the law and policy should focus upon? And though the answer reflected in most aspects of counterterrorism policy appears to move subtly from individual to collectivity, that answer proves inadequate with respect to this new and unforeseen challenge. For with the lone wolf, the wrongdoer suddenly and surprisingly reemerges as the single human individual, albeit one now enmeshed in global networks through which he acquires his ideas and masters his methods.

Let us now summarize the preceding analysis in a single table, succinctly delineating the collective or individual character of both wrongdoer and victim, including changes over time from one such characterization to the other. Most revealing about the table is that no method of atrocity accountability is exactly the same as any other in how it represents the wrongdoer and victim, as individual or collectivity, over time.

Osiel Chart


In assessing the legacy of a given episode, we not only witness frequent conflicts between legal and political actors over whether to individualize or collectivize a given character. We are also very likely, in response to the question, to sense warring impulses within ourselves. Let me begin by indicating our conflicting impulses on the puzzle of how best to characterize the wrongdoer, and then proceed to identify our equally opposing inclinations over how to conceptualize the victim. The natural question to follow from such an analysis is surely: When does each of these inclinations come to the fore, assert itself most powerfully in real-life struggles over alternative modes of atrocity accountability? The brief analysis that ensues can only summarily gesture at the answers (which I assay at length in the larger, forthcoming work from which this article draws).

Imagining the Wrongdoer: Why Individuate?

Among those actively engaged in choosing among responses to mass atrocity, there is often a powerful proclivity to individuate the wrongdoer. This has three sources.

First, deterrence. The now prevailing view is that, in light of how mass atrocity comes about, we “get incentives right” when we threaten individual leaders with serious sanction, notably international criminal prosecution. This, at least, is the best way yet discovered to “align the interests” of leading wrongdoers in these episodes and those who would condemn (or suffer) their wrongdoing—that is, the rest of us. To prevent major misconduct by large collectivities—sovereign states, military bureaucracies, complex business entities—we have to activate the self-interests of their individual leaders. Those institutional behemoths can readily absorb the costs of sanction, more easily than “natural persons,” and pass them on to citizens, taxpayers, or stockholders. In the economic lingo, “agency problems” are too severe to address the matter from the opposite direction, by inflicting the costs of misconduct directly on these collectivities, expecting that this will somehow induce their leaders to responsibly serve the interests of those they notionally serve and represent.

The second reason for individualizing the wrongdoer is that the opposing path makes it virtually impossible to engage in retributive justice. There has been a longstanding fear that once we start down the path of plumbing the structural sources of wrongdoing, any notion of moral responsibility for a given incident of misconduct—any notion suitable for purposes of sanctioning, at least—soon dissolves into social scientific explanation. We resist this result especially with mass atrocity, because the magnitude of the wrong elicits in us strong moral sentiments, reactive attitudes, which make us unwilling to settle for a value-neutral diagnosis of impersonal sociological structures. The ethical enormity of the events and the measure of indignation they inspire strongly dispose us toward finding entities to blame and sanction, provided that we can conceive a convincing way to do so. This indignation gives expression to retributive sentiments that—our moral intuitions mostly tell us—are defensible and intelligible only when directed against someone or something with a capacity for moral will. We generally believe that collectivities lack that capacity, strictly speaking.13

Third, it is widely thought that, following large-scale horrors to which so many lent their energies, individuation contributes to reconciliation. This was the rationale that inspired international criminal law, at the very outset of its modern history, in the 1945 Nuremberg trial: to establish that individuals, not entire nations, are responsible for aggressive war and mass atrocity. There were wide hopes that this narrowing of blame and punishment, in contrast with the broader approach taken in the Versailles Treaty following World War I, would help overcome the poisonous nationalism that had both encouraged World War II itself and inspired so many war crimes during its course, on all sides of the battle lines. International tribunals would retell that story so that it was no longer interpreted, as it had been during its course, as one of bitter mutual hatred among whole peoples, but instead one of discrete acts by an evil few. Everyone else could then more easily return to harmonious relations with former antagonists.

These three reasons combine to ensure our inexpungible impulse to individualize the wrongdoer.

Why, Then, Sometimes Collectivize the Wrongdoer Instead?

Despite all this, we nonetheless also often have strong inclinations, and good reasons at times, to collectivize the wrongdoer, when our purposes are not chiefly those of deterrence or retribution.

First, collectivities always provide deeper pockets for civil compensation of victims than do individuals, even when it is possible to trace his hidden foreign assets, as often it is not.

Second, collectivizing the wrongdoer is appealing when there is serious danger of political backlash against transitional justice measures from the particular people most responsible for the wrongs, those likely to suffer sanction under an individualizing approach. It is easier for a truth commission to abstractly condemn a “culture of impunity,” even to criticize the armed forces and police in general terms, rather than the generalíssimo in particular, for very specific criminal acts on his part at a given date and place.

Third, if atrocities are not to recur, institutions will often require reform. To reform the “security sector” (that is, the police, armed forces, and intelligence agencies), for instance, requires that we first identify that entity itself as a wrongdoer, in its own right, in some intelligible sense. From this perspective, the individuals within it acted largely in service of its goals, due to the incentives it created for them, through its structure and policies. We must treat it as a collective actor if we are to effectively get at the source of the problem.

Nuremberg Trials. (Courtesy: Marlon Doss/Flickr)

Nuremberg Trials. (Courtesy: Marlon Doss/Flickr)

Fourth, it is impossible to delineate the relevant forces at work in a given episode of mass atrocity, for purposes of historical explanation, by examining only particular acts of misconduct by top leaders or, for that matter, by the lowest echelons proximately engaged in the immediate violence. It is necessary to contextualize events within a larger frame of social and political analysis, a task to which historians, social scientists, and journalists may importantly contribute. This is true even if we favor so-called “intentionalist” over “functionalist” accounts of what transpired—designations for competing schools of thought concerning origins of the Holocaust—and therefore choose to highlight “agency” rather than “structure” (in the idiom of social theory) within our preferred causal account. Offering the public an accurate, fair recounting of what took place is very often a central aim of transitional justice. This serves to undo the pervasive lies that oppressive prior leaders told the public in justification or denial of their misconduct.

Finally, in individuating the wrongdoer, we raise the evidentiary hurdles to holding anyone responsible for the misconduct higher than when we collectivize. This is apparent if we employ the table above to contrast the modes of atrocity response conceptualizing the wrongdoer in one fashion rather than the other. Burdens of proof are clearly more demanding, and concerns about due process more acute, with respect to international criminal law, personnel screening, and individualized sanctions than in regard to apology, truth commissions (when the latter don’t “name names”), and civil compensation, including compensation via findings of “state responsibility.” Thus, if policy-makers wish to level blame for mass atrocity, their odds of success are often better if they choose to construe the wrongdoer as some form of collectivity, rather than a set of discrete individuals.14

Imagining the Victim: Why Collectivize?

The impulse to collectivize the victim has four sources. First, this step too, like collectivizing the wrongdoer, makes for better historical explanation, since victims were invariably targeted because of their membership in stigmatized religious groups, races, ethnicities, or on account of their gender. Their individual deaths were, in a sense, “incidental” to the wrongdoer’s effort to harm a collectivity of some kind.

Second, the inclination to collectivize the victim, to treat it as a single entity, emerges when it proves impossible to compensate or otherwise redress individual injuries, despite relevant differences in what each person suffered. Thus, a collective remedy, like construction of a public monument or a new tribal headquarters, is the only remedy available, practically speaking, when many individual victims are dead; this is also true when the “information costs” of identifying and corresponding with all surviving ones are preclusive. These are practical issues, of administrative cost and effective implementation, not of first moral principle.

Third, collectivizing the victims, representing injured individuals as part of some larger social entity, often well serves to mobilize internal support among members of this broader group. This support may be material or symbolic, and designed to protect at once specifically injured individuals and the collectivity (which is usually ethno-religious in character). Activists and advocates for a given group therefore frequently find it advantageous to conflate and assimilate the harm directly experienced by certain of its members with wider, more indistinct threats plausibly posed by the same set of wrongdoers. A similar interpretive strategy finds occasional use by arch nationalists urging collective resistance to international institutions. Thus, for instance, militants among all major ethno-religious groups in the former Yugoslavia—Serbs, Croats, and Bosniaks—have striven, with considerable success, to characterize the criminal conviction of fellow members in international courts as a defamation not only of such individuals, but of the group as a whole; the conviction is received as no more than a libel perpetrated by the other two groups, through their insidious influence upon these tribunals, and by the world at large.

Finally, there may be genuine elements of harm to a group as such, to its communal life and institutions, harm not fully captured by remedies that “reduce” the distinctive experience of a shared social life to the sum of what particular individuals have suffered. When community institutions are destroyed by a genocide, for instance, the law’s remedial efforts must then defensibly concentrate, many believe, on rebuilding these very institutions. The law does so for reasons distinct from the fact that this will also benefit the particular individuals who suffered past harm and who may risk its prospective recurrence. For the remedial focus is here instead on facilitating the involvement of individuals in these reestablished institutions so as to rebuild the larger communities such institutions would serve.

We may even value the re-creation of these communities as an achievement from which we ourselves, though outsiders to them, derive pride and satisfaction. This is to regard their survival, once imperiled, as an end in itself, or more precisely, as a means by which to preserve and extend a salubrious form of cultural variety in the world, sustainable only in social groups. As members of an increasingly global society, we thereby become—individually and collectively—the secondary beneficiaries of such responses to atrocity, though we have not been its immediate victims.

Why We Sometimes, Nonetheless, Individualize the Victim

Finally, there is often a powerful intuition to individuate the victim of mass atrocity, for three reasons. First, some people fear that collectivizing victims implicitly reproduces the wrongdoer’s same moral error: denying their human dignity as individuals, and the irreducible particularity of their suffering, by subsuming them all into some broad, sociological category, the same category their murderers employed as reason to murder them. There is something distinctly uncomfortable, even perhaps dehumanizing, about conceptualizing the victims as an indistinct “mass,” lumping them into an amorphous remedial category, even if this obviously cannot be equated, even metaphorically, with dumping their bodies into mass graves.

A second reason for individuating the victim is that when the victim will receive major compensation for loss, fierce disagreements often arise over who should count as one. Those who must pay the compensation, or who simply endorse a highly individualist understanding of justice, naturally favor conceptions of the victim more particularized than those who seek to benefit from the anticipated remedial schema; this is especially so of those whose claim to victimhood, and hence to profit from such a program, is tenuous. Broader-brush sociological categories generally better serve these claimants than more precise legal tests requiring specific evidence of individuated harm.

Third, and ironically perhaps, it is sometimes advantageous to individuate the victim if one’s chief goal, as a political advocate, is to evoke international sympathy and support for one’s group. After all, the particular atrocity victim is often chosen on grounds of her group membership. And a photograph of her transparent suffering will inevitably inspire greater commiseration than a picture or writing rendering the true victim as the group from which she was perhaps indiscriminately plucked. This latter form of victim-representation plays best to nationalist sensibilities largely unique to members of the targeted collectivity. When the rest of us powerfully empathize with the sad fate of Iraqi Yazidis at the hands of ISIS, donating to charitable organizations combating their destruction, our empathy is surely that of “citizens of the world.” This is to say that we identify with these victims more as individuals—that is, as bare, culturally stripped-down homo sapiens—than with the distinctive ethno-religious group whose survival is thereby equally endangered.

It is then unsurprising that, whenever one visits a memorial honoring those killed in a given large-scale atrocity, one invariably observes that foreign visitors soon gravitate spontaneously to the now-obligatory wall listing the names and posting pictures of those murdered. We foreigners—members of national groups other than that attacked—prefer to apprehend these individuals simply in terms of their raw vulnerability, simply as frail, fellow members of our species. An adjoining portion of the memorial will frequently seek to register, in more abstract and monumentalizing form, the shared hardships of all and celebrate the survival of their collectivity. That portion of the site, however, will uniformly draw fewer visitors, for fewer moments of meditation. And when this less arresting section of the memorial depicts the victimized group as a unified “people in arms”—even if one united only in resistance to foreign aggression or imperial conquest—the compassionate attention of foreign visitors usually diminishes still further.

These patterns are readily observable and not lost upon clever advocates for the victimized group, concerned chiefly with advancing its members’ shared interests. Partisans of this sort now design memorials to themselves, shrewdly and self-consciously, so as to draw the greatest possible support of outsiders. It is hoped that the worldwide sympathies elicited through focus upon the indisputable agony of specific individuals may then, with the aid of further interpretive strategies, “migrate” and be imparted to that of the group at large.


Among the most important of decisions taken in the aftermath of mass atrocity is the implicit choice we necessarily make, when favoring one accountability mechanism over the rest, in conceiving the key characters as individual “natural persons” or as collectivities of various kinds. For that choice will determine who, in the ashes of these cataclysmic episodes, our law and policy will punish or reward, elevate or derogate in public stature. Distinctive to mass atrocity, in fact, is how it invariably conjoins collective and individualistic aspects in both the doing and suffering of harm. It does so, more precisely, to a much higher degree than other wrongs, so that the result is a veritable difference in kind. This is one reason why the choice between alternative responses to these large-scale events becomes so vexing, when only one or two such methods will be seriously pursued. In light of how they conceive the same events—differently, yet not incompatibly—several such approaches should usually be combined, whether at once or in sequence, as circumstances permit.

This frequently proves impossible, however, for a variety of reasons, some of these (resource limitations, for instance) entirely legitimate. Still, each method of atrocity accountability—in what it leaves out, whether individuals or groups, as wrongdoers or victims—all but invites some form of “correction” through recourse to another such method, and often more than just one. Some of these methods and their respective goals fit congenially with certain others. Some, however, are clearly in tension, even antithetical at times.

The different atrocity responses serve distinct purposes. We commit ourselves to advancing these purposes when we choose the corresponding method of legal and policy response, with each such method uniquely defining the relevant characters. We are thereafter “stuck” with its particular characterization of the parties only if we take it to reflect a hard and fast ontology, naively believing that our concepts must always “correspond” jot for jot with some such crystalline empirical reality. Once we make this move, we engage in self-contradiction, in fact, if we then pursue two different modes of atrocity accountability where each embodies a different conception of “the victim” or “the wrongdoer.”15

It is surely preferable to understand any particular characterization of the parties, adopted by a given mode of atrocity response, as something of a cultural construction, one credible way among others of interpreting and representing the world, capturing crucial elements of social reality without encompassing everything of significance to us. To say this is not to imply, of course, that all episodes of mass atrocity are susceptible to equally convincing classification in the ways employed by any of the atrocity-response methods here anatomized. Certain modes of response, in how they construe the characters, will clearly comport with the facts of a given episode and the circumstances of its aftermath more than other such methods.

There nonetheless remains an inescapable element of artifice and rhetorical contrivance involved in fitting a given accountability method, with its particular rendering of victim and wrongdoer, to a specific historical experience of such complexity. Because the fit is never perfect, there will always be an uncomfortable pinch of moral and factual remainder. There is therefore no reason, for instance, why they should not treat the wrongdoer as an individual for purposes of retribution, but—when their aims turn to compensation—as a collectivity, of some sort. Or the victim as a collectivity where this best advances the goal of historical explanation, yet as an individual when it comes time to memorialize.16

It would be wrong to imply that we must, at the very outset of deliberation, prioritize our purposes for transitional justice in some definitive rank-order; that these initial preferences alone should then govern our choice of atrocity responses; and that this choice will in turn conclusively settle the question of whether we will be deciding the fate of individuals or of collectivities. A responsible analysis of policy issues as complex as these can never be so simple. We should adopt a given purpose for transitional justice only if there is good reason to think we may attain it, through the methods of law and policy available to us. This choice requires sober analysis of what is politically realistic in present circumstances, to be sure. But it also demands close attention to what brought about these calamities in the past17 and what, by implication, would be conducive to their recurrence—often a serious danger at these politically precarious moments.

Prudent policy-makers will nonetheless rightly admit uncertainty about how much we can truly know concerning the historical causes of relevant wrongs, near and far, and hence also about what would likely avert further ones. This uncertainty should induce a healthy skepticism toward others claiming to ground their favored policy responses on “the clear lessons of history.” When assumptions about the past prove faulty or merely overconfident, the efficacy of policy measures adopted on their basis will suffer. It is therefore advisable that our choice among alternative accountability mechanisms draw inspiration not only from our genuine, hard-won knowledge of the horrid past—from the all-too-abundant “case studies” it offers up—but also in our very doubts about what we think we have yet learned from it.

Philosophers will predictably urge at this point that we should, first and foremost, strive for greater analytical clarity regarding our most general, near-universal normative objectives (compensation, deterrence, etc…) and their relative significance. That intellectual exercise is undoubtedly helpful in initiating the necessary discussion about what is to be done. Yet these refined abstractions, derived from “ideal theory,” nowhere directly determine policy choices.18 Nor should they, and nor would most philosophers insist that they must. Those who make these decisions will often find themselves reshuffling their early rank-ordering of policy goals. Such shifts in relative priorities, in light of practical opportunities and constraints, will also reconfigure the dramatis personae, as by transforming preliminary conceptions of the wrongdoer from individual into collectivity, or in the opposite direction.

Mass atrocity represents an historically distinctive constellation of events, inviting a unique amalgam of policy responses. And because it plays out simultaneously at different levels, micro and macro, these responses will be satisfactory only if writ both large and small.19 We should in turn understand transitional justice as a field of thought and action at once singular unto itself, yet demanding several courses of action, often equally indicated by the facts of a given episode. Until we better fathom the connections—still glimpsed only gauzily—between these various modalities of law and policy, we are likely to overlook or devalue, unwittingly or intentionally, either the individual or collective dimensions of these extraordinary yet ubiquitous episodes.

My chief purpose here has been to clarify the demands and choices in play within a post-atrocity situation, without offering an explanation of why and when those choices go awry, and without suggesting how my “sociological” analysis might directly help in setting matters aright. It is nonetheless fair to infer that when a country pursues only a single mode of atrocity response, it will necessarily over-individualize or over-collectivize its treatment of wrongdoer and victim. And through this this act of over-simplification, it will impose additional injustices on, and stir new grievances among, those prioritizing different aims, preferring alternative methods of accountability. Experience suggests that policy-makers will then generally face mounting pressure to adopt further such methods. Not only will these more comprehensively redress initial wrongs—the atrocities themselves—but also assuage the secondary sources of moral indignation and resentment engendered by the earlier approach itself, now deemed unduly narrow.

In sum, it is true, in a sense, that each mode of atrocity response construes the characters to mass atrocity in a way “inconsistently” from the rest. Yet only when we bring a number of these accountability modalities at once to bear upon a given episode does it become possible for each of these to address the inevitable shortcomings and distortions introduced by others, standing alone. Only at this point, in fact, do we ourselves act consistently with requirements of justice. Conceived as a single, coherent enterprise, then, transitional justice ceases to provide only a specialized “toolkit” of sundry professional techniques that we may pick up or put down as we like, with none of these bearing any close, necessary relation to the rest, many of them therefore often left in the box, rusting away.


This article is based on June 2015 lectures at Oxford Transitional Justice Research and at the University of London, School of Oriental and African Studies, Centre for the International Politics of Conflict, Rights, and Justice. Early thoughts on these matters were also presented to the Legal Studies Workshop, University of Iowa, and to the International Law Workshop, University of Chicago. The author expresses his gratitude to organizers and audiences at all these venues. Ziv Bohrer, Jovana Davidovic, Paul Gowder, Pavlos Eleftheriadis, Jonathan Leader Maynard, Nathan Miller, and Maya Steinitz individually offered helpful reactions to drafts. For the longer, book-length elaboration of the present argument, the author welcomes comments and criticism, directed to [email protected].





  1. This expression, which is not a legal term of art, shall here refer to episodes of genocide, crimes against humanity, and major war crimes (“grave breaches” of the Geneva Conventions) involving significant numbers of both wrongdoers and victims. The concept also encompasses large-scale, violent violations of international human rights by sovereign states.
  2. I employ the term collectivity in a capacious sense to designate significant numbers of persons widely perceived as constituting a single entity of some kind, in relation to which insiders or outsiders orient their thought and action. Defined this broadly, collectivities vary widely, of course, in size, degree of institutional stability, formal organization, regularity of interaction among members, self-identification as a group, extent and explicitness of shared understandings. These sources of variation are crucial for many purposes, but not present ones, which concern only the question of what happens when we take the step of enlarging the focus of analysis to view the individual victim or wrongdoer as part of something larger, whatever its more specific character. This increase in persons entails an elementary analytical move, yet is often far-reaching in ramifications, practical and conceptual. That is the case, for instance, whenever—as often with mass atrocity—the wrongdoer views his victims not as an aggregate of discrete individuals, but instead subsumes them under a rough, “generic” category, like “non-Serb residents of Republika Srpska.” Such a classification says nothing at all about any actual patterns of social interaction, if any, among those so pigeonholed, and it will not well align with their self-understanding of who “us” truly consists. A pejorative generic, imposed by others, is sufficient to create a collectivity, in the present sense, for purposes of victimizing its members. However, to convincingly describe the wrongdoer as a collectivity requires something more, some self-identification among its members as members, at the very least. Individual wrongdoers then constitute, in current terminology of the philosophy of social science, a “social entity” with “emergent properties” indiscernible by examining only its constituent parts in isolation. Especially germane among these properties is the causal power, arising from members’ “collective intentionality,” to shape the course of such large-scale events as mass atrocities. Still, when we shift attention to this more institutional level of analysis, we need not insist the collectivity at issue exists in ways irreducible to the conduct of its members. We need merely consider its effective workings and organizational structure more material to our immediate purposes than anything about the particular people now composing it. And because its leaders employ the organization’s rules and resources to coordinate its members’ conduct, that entity may simply provide the most congenial entrée and point of leverage for efforts to influence the behavior of individuals within. Thus, the present discussion does not wed itself to any particular stance, inevitably controversial, on what kind of entities, ontologically speaking, populate the social world. My approach is instead pragmatic and ‘methodological’: it concerns itself only with discovering optimal methods to redress past atrocities and prevent future ones, and thereby to improve the human condition.
  3. Resembling personnel screening in this respect is “humanitarian intervention” through the exercise of military force. Inescapably, this policy entails judgments about whether an end to mass atrocities requires full-scale “regime change”—directed at the reform or even abolition of certain state structures, collectivities by nature—or merely change in the identity of a few individual leaders. This may simply require the political pressure necessary to force abandonment of a single, morally objectionable policy.
  4. Kiobel v. Royal Dutch Petroleum, 621 F.3d 111, 119-121 (2d Cir. 2010).
  5. Maya Lin, Andrew Barshay, Stephen Greenblatt, et al., Grounds for Remembering: Monuments, Memorials, Texts (Berkeley, CA.: Townsend Center for the Humanities, 1995): 9-10.
  6. Phil Clark, The Gacaca Courts, Post-Genocide Justice and Reconciliation in Rwanda: Justice Without Lawyers (Cambridge, U.K.: Cambridge Univ. Press, 2011): 41, 44, 315.
  7. Sometimes, more individuated sanctions accompany this type of atrocity response. Thus, current sanctions against Russia target both influential individuals—via asset freezes and visa bans—and large commercial enterprises within economically vital industries.
  8. This discursive maneuver indulges more than a modicum of disingenuousness, insofar as millions of Muslims throughout the world share certain major goals of radical Jihadist groups and even endorse some of their criminal methods, notably the suicide bombing of civilians. See, e.g., Pew Research, Levels of Support for Suicide Bombing over Time (Jun. 30, 2014),
  9. Harold Hongju Koh, Legal Adviser, U.S. Dept. of State, Address at the Annual Meeting of the American Society of International Law 14, available at, p. 15 (Mar. 25, 2010).
  10. Protocol Additional to the Geneva Conventions of 12 August 1949, relating to the Protection of Victims of International Armed Conflicts (Protocol I) art. 51(3); see also Protocol II, Art. 13(3).
  11. In inter-state war, military attackers need only establish the formal membership of their intended targets within the enemy’s armed forces or its associated militias. This is relatively easy to determine in most circumstances, as through dress and insignia. It is generally far more difficult to clearly ascertain an individual’s close affiliation with a terrorist organization, unless he can be observed while directly engaging in certain kinds of violent activity. Perhaps the principal normative concerns raised by drone strikes are therefore distinctive to this method of warfare, that is, altogether absent when the target of attack consists of soldiers formally within the armed forces of a sovereign state.
  12. The older targeting method also presupposed that individuals who occupied the very top ranks of a terrorist organization, like Osama bin Laden over Al Qaeda in his heyday, maintained near total control over its activities and could be adequately replaced only with great difficulty. Serious doubts about these empirical propositions soon emerged, however, even well before the advent of so-called “lone wolf” terrorism.
  13. Of course, we routinely hold formal organizations, both governmental and nonstate, legally liable, notably in tort, for many types of wrong. The law does this, though, for purposes of compensation and (to lesser degree) deterrence, not retribution. Whether collectivities of particular sorts genuinely lack the sort of “group agency” necessary for attributions of moral responsibility is a subject of some dispute in the philosophical literature. See generally Christian List & Phillip Pettit, Group Agency: The Possibility, Design, and Status of Corporate Groups (Oxford, U.K.: Oxford University Press, 2011). My assertion in the text simply concerns our prevailing moral psychology.
  14. To reform the training of armed forces and the police—teaching their members the basics of international humanitarian and human rights law, respectively—requires nothing in the way of documented, institutional responsibility for prior wrongs, evidence of which will generally lie close at hand.
  15. It may be that no philosophically sophisticated person actually holds this view, and that my criticism therefore finds its target only among many laymen. In any event, I here seek chiefly to render more explicit and articulate the existing intuitions that literate people may already semi-consciously hold concerning the necessarily constructed character of relevant parties to these events.
  16. This stance is “pragmatic” in philosophical orientation. Support for a multiplicity of responses to mass atrocity nevertheless also derives from the “coherence” theory of truth. For even in ordinary legal life, when government officials violate the law, it is sometimes possible for victims, both individuals and organization, to civilly litigate against the state as such, while also suing the individual office holders deemed responsible, in their nonofficial capacity, for such violation. In these cases, the state is itself a victim to its officials’ illegalities. It may therefore criminally prosecute these persons for the conduct, and sue them civilly too, in order to recover public resources embezzled in the course of their wrongdoing. We thus find all possible variations: individual victim vs. individual wrongdoer, individual victim vs. collective wrongdoer, collective victim vs. collective wrongdoer, and collective victim vs. individual wrongdoer (in this final case, within both civil and criminal proceedings). It is nonetheless extremely uncommon that every such litigation scenario, each with its distinct permutation of individual and collective, will actually arise from a given incident of official misconduct. With mass atrocity, though, the full plate of possibilities lies nearly always within realistic range of imagination and potential adoption.
  17. Thus, if we believe the primary cause of recent misconduct in a given country is found in the ideological worldview of its officer corps, we will seek, when circumstances allow, to concentrate our preventive efforts on reforming the professional education of all its members.
  18. Even when an a given objective, initially atop their list, eventually prevails against the alternatives, competent decision-makers could never defensibly reach such results, nor adopt the corresponding policy measures, simply by way of logical deduction from it. A given normative objective clearly cannot be treated, in other words, as a fixed first principle, pursued indifferently to contextual considerations.
  19. An ideal set of coordinated responses would also need to address the expectations of disparate audiences: juridical, scholarly, as well as the interested lay public, both national and global, the concerns of the currently living as well as their successors, to some extent.
Facebook Twitter Email

Category: Online Exclusive, The Ethics of War and Peace

Comments (1)

Trackback URL | Comments RSS Feed

  1. Joanna Bryson says:

    This is an excellent article. However, there are a couple things overlooked here that occurred in the research of Ben Valentino

    1) One reason to identify individuals as culpable is that historically it is often a fairly individual decision. Valentino looked hard for structural explanations but found that very often it was individual personalities that made all the difference.

    2) Two “solutions” to the problem then came out of Valentino’s work — either assassination, or facilitating the flight of oppressed peoples from their oppressor until the regime changes naturally. This is one reason why figuring out how to facilitate rapid mass migration is critically important.

    Note that these observations are by no means more important than the excellent coverage in the article, I would just like to see them incorporated into the review.