A Challenge to Those Working in the Field of Genocide Prevention and Response11. This paper was originally written as a background piece for the conference, “The Way Forward,” sponsored by Wellspring Advisors, Bridgeway Foundation, Humanity United, and the U.S. Holocaust Memorial Museum.

Bridget Conley-Zilkic



United in the belief that civilians should not suffer from intentional infliction of widespread and systematic violence and the assumption that special measures are necessary to prevent and protect groups from such violence, a diverse group of scholars, educators, journalists, activists, advocates, policymakers, diplomats, and military leaders have raised their voices against genocide and mass atrocities. This group has grown exponentially over the last decade and can be understood as an emerging field in its own right. This essay explores some of the conceptual and practical challenges facing this field as it furthers its professional development.


| | |

• • •

1. A Moment Ripe for Self-Reflection

Today’s genocide and atrocity prevention efforts emerge from a long history of mass murder of civilians being accepted or deemed a lesser concern than negotiation processes, political allegiances, or the need to win a conflict. There is no shortage of examples of terrifying assaults against societies’ most vulnerable groups. The most recognized cases like the Holocaust, Rwanda and Srebrenica, dominate discussions, but there are also many less known cases like Guatemala.

The civil war in Guatemala (1960-1996) was among the bloodiest of Latin America’s Cold War conflicts. An estimated 200,000 people were killed or disappeared. Two specific years in the 1980s stand out as the most lethal. Between 1981-1983, some 100,000- 150,000 Guatemalan Maya were killed by the national armed forces (JONAS, 2009, p. 381). As part of a scorched earth counter-insurgency plan, governmental forces killed, raped, tortured, and forcibly displaced Maya in the rural mountain regions. Beginning in 1983, the army undertook measures to control the survivors, ushering in a second phase of assault marked by a combination of amnesty and intensified militarization of surviving communities. In the worst hit community, Rabinal, 14.6% of the population was killed and 99.8% of the victims were from the Maya population (HIGONNET, 2009, p. 27).

For the Guatemalan government, the offensives were deemed necessary to finally end the long-running civil war (1960-1996) and enable modernization of the state. For the key outside countries that supported the government, particularly the U.S., the most salient feature was Marxist insurgency, only one example of the perceived global threat. Human rights activists from Amnesty International and Americas Watch (now Human Rights Watch) were targeted as communist sympathizers for even documenting the atrocities. The U.S. government accused such groups of being part of “a concerted disinformation campaign […] by groups supporting a left wing insurgency” (SIKKINK, 2004, p. 167).

It was not until the United Nations-sponsored Commission on Historical Clarification in Guatemala published its final report, Memory of Silence, in 1999, was the term “genocide” applied to the violence. The Commission’s report describes what happened as “acts of genocide against the Maya people that live in the Ixil region, Zacualpa, northern Huehuetenango and Rabinal” (HIGONNET, 2009, p. 131).

Much has changed since the 1980s – and particularly since the failures of Rwanda and Bosnia-Herzegovina and the exponential growth of the field in response to Sudan. Today, the value of genocide and atrocity prevention work is recognized by global leaders and it is undertaken from a wide range of disciplinary and organizational approaches. While some of the most vocal and visible actors are in the U.S., individuals, organizations, and coalitions from around the globe, including places that have experienced past violence, are tackling atrocity prevention and response issues.

The field of genocide and atrocity prevention work is united in the ethical belief that entire groups of civilians should not be assaulted and the normative assumption that special measures should be created to protect against and respond to this violence. But beyond the ethical and normative consensus, much more discussion is needed. For the field to consolidate its progress and continue to grow, it must strengthen its capacity for self-reflection and criticism.

Today’s genocide prevention movement is marked by four signature characteristics. First, the field is emerging and not understood as coalesced.1 This is a time of great creativity and experimentation. This means that the basic practices, assumptions, tools, and vocabularies are up for debate. Multiple goals exist, and the differences between them lack clarity. One area where this is particularly noticeable is in the wide range of terminologies used to describe the phenomenon at hand: genocide, mass atrocities, crimes against humanity, ethnic cleansing, and so forth.

Second, in the shift from an emphasis on response to one of prevention, many people and organizations in the field (although certainly not all) have chosen to focus on structural concerns, both in terms of the conditions that enable violence to occur and in the agencies and forums that might respond to occurrences of violence. Both areas are born out of study of patterns across cases, a perceived imperative to engage before lives are lost, and the need to have stronger response measures queued up (within the “toolbox”) and ready to go. The shift can also be seen in the efforts of grassroots activists who are trying to realize a permanent and sustained constituency of engaged citizens, rather than creating new interest with each new individual case.

But crisis still drives the policy discussions and there is no low tide to allow for careful construction of new systems. Hence, the bureaucratic changes are undertaken in an environment where it is easy for longer-term needs to be overshadowed by today’s most pressing concerns. Further, improving response necessarily demands understanding case-specific dynamics, which is an entirely different set of skills and knowledge. Finally, introducing normalcy into response mechanisms for extraordinary violence threatens to lower the bar for when extraordinary measures can be undertaken.

Third, there is a shift from a historical human rights posture of opposition to governments to that of working cooperatively with governments and multi-lateral or international organizations to create stronger, more diverse, and attuned response mechanisms. This does not mean that advocates refrain from criticism of governmental policy; even the most cursory glance at recent reports would quickly refute that assertion. It is rather a subtle attitudinal change towards viewing government as a largely positive partner, even if one that needs goading at times, that should further assert its global power. The shift means that non-governmental actors have found allies inside governments and are choosing strategies that aim to result in real policy changes. This positive impact should not be underestimated.

However, such a strategy only works if there is a government that is willing to engage and bend to such pressure, hence it offers a model for action in only certain societies and on certain issues. It may also deepen national biases by prioritizing conversations within pre-established national political communities rather than compelling a search for international coalitions with diverse partners. It means that the field is developing around actions that are perceived as more possible because of their potential appeal to key governments, rather than necessarily being guided by circumstances in places at risk of or experiencing mass violence. And it invites governments to use coercive measures—an invitation that, once issued, may be difficult to control and/or recall.

Today’s genocide prevention activists have expanded further than many previous human rights campaigns to engage the general public in applying pressure on their own governments, particularly in the U.S. As a result, the movement has shown considerable creativity and tapped into a passion for its issues among a broader public. The presence of an audience for this work is doubtless and a great accomplishment for the field will be if it continues to grow and professionalize without losing the spark of ingenuity that characterized its most compelling public outreach efforts. Drawing in a larger public contributes to the field’s capacity to make its policy recommendations stick and to sustain political focus on the issues.

The first challenge of this trend is that maintaining a strong public outreach effort can consume an organization’s resources and promote policies and measures of success that shift the focus more to what people as advocates can do, and away from what will improve conditions for the individuals at risk. Second, members of the general public are not currently well educated enough to be informed actors in the movement. There is very little understanding of the dynamics of foreign policy making, case-specific background, and/or even the most basic concepts at play (i.e., human rights versus humanitarian organizations). While this may be a broader foreign policy concern, it has specific ramifications for organizations that emphasize the role of a public movement.

Not everyone will agree with this presentation of the challenges facing the field or how the field of genocide and atrocity prevention should respond to its challenges. However, the strength of a field is not measured solely by its points of consensus, but also the vibrancy of its debates. This paper attempts to outline both areas of consensus in the field and the knowledge base that informs it, as well as areas of contention. To this end, it aims to be provocative in highlighting debates that are already underway in the field of genocide and atrocity prevention. The questions raised in this paper do not lend themselves to easy answers nor necessarily to consensus, and this may not be desirable. Instead, it is a hope that they contribute to the field’s capacity for self-criticism and reflection, while also challenging it to reach out to other fields to share insights and join forces.


2. What Do You Really Want To Achieve?

Debates over goals are valuable for a field, not because they produce consensus, but because they provoke discussion of the strengths and weaknesses of organizations as they continually grow in relation to evolving circumstances.

One example of this comes from the human rights profession as it debated whether and how to expand to include social and economic rights in addition to political and civil rights. “Poverty is the world’s worst human rights crisis,” former Amnesty International (AI) Executive Director Irene Khan emphatically states in The Unheard Truth: Poverty and Human Rights (KHAN, 2009). In June 2009, Amnesty launched their campaign, “Demand Dignity,” which aims “to end the human rights abuses that imprison people in poverty” (AMNESTY INTERNATIONAL, 2011). Mary Robinson, then writing as UN High Commissioner for Human Rights, agreed, noting that during the Cold War western countries tended to argue solely for political and civil rights, whereas Soviet bloc countries emphasized social and economic rights. The result too often was that the two sides spoke past each other. She argued, “The time had finally come to take the two sets of rights equally seriously” (ROBINSON, 2004, p. 866).

Human Rights Watch (HRW) took a different approach. Executive Director Ken Roth argued that “naming and shaming,” HRW’s hallmark methodology, depended on the organization’s ability to document a clear case of violation, violator and remedy. While agreeing that social and economic rights are critical for the well-being of a population and can be addressed in some ways via naming and shaming, he argued promoting them was not the central role for HRW given the organization’s strengths and limitations:

Moral capital does not accumulate through our voice alone (why should our opinion count more than others?), but through our investigative and reporting methodology. It is a finite resource that can dissipate rapidly if not grounded in our methodological strength. (ROTH, 2004, p. 65).2

This discussion offers some guidance for the genocide and atrocity prevention field. How can organizations balance their work between the nature of abuses they focus on and their unique organizational capacity to contribute to improvements? What is the most sustainable and rigorous strategy for an organization? It is a discussion that must assess both organizational strengths and weaknesses and grapple with the evolving external environment. And it is unlikely that the answer will be the same for all actors.

Within the genocide and mass atrocity prevention field, important distinctions in potential goals have not yet been debated. Below are several examples of ways to articulate a core objective. This is not by any means an exhaustive list. There are trade-offs with each, and none provides an obvious or easy option for actors within the field:

(1) Protecting vulnerable groups (defined by group identity or simply civilian identity) from the threat of death in moments of extreme crisis. The goal is defined in relation to a clearly visible threat or on-going violence that marks its victims for death (killing). It is a goal that most makes sense in contexts of fast-moving offensives against civilian groups and focuses on conditions that will stem the death toll. As the most narrowly defined goal on this list, it more readily lends to consistency in application. It also best corresponds to the extreme ethical stakes implicit in the term “genocide” and the various extraordinary response mechanisms in development by the field. However, it is also a rare form of violence, often develops with lightning speed, and can end or shift just as quickly.

(2) Decreasing the capacity of armed forces to commit and instances of large-scale violence against civilians. This goal begins in reference to the moment of crisis, defined either solely by a threat of killing or potentially expanded to what is called indirect death or the intentional creation of “conditions of life” that cause death. However, this goal also extends to engagement on longer-term solutions that alter the balance of power that enabled atrocities to occur. Inevitably, such a goal implies engagement with situations even beyond the moment of extreme crisis; a political vision—often partisan—of the most desirable realignment of power; and a much greater number of cases to potentially address, complicating efforts at consistency in case selection and focus.

(3) Increasing the civilian protection capacities of international and national agencies. The broadest goal introduced in this section, it adapts to a range of circumstances in which civilians suffer from violence. Such a goal is not defined solely in reference to a crisis, but also to how conditions functionally produce different forms of violence. It is less easily defined than either of the two above goals, but it might provide a framework for engaging situations characterized by long-term, low-level violence that appears in spikes, more isolated instances, or with multiple perpetrators who may function more like criminal networks.

This third type of violence, as we will see, may be increasingly common in our world today and over the long-term it can impact a great number of civilians–sometimes even more than episodic violence. In 2010, for example, 3,111 people were murdered in one Mexican city, Juarez, related to the activities of criminal gangs. This figure is more than the estimated 2,421 killed in Afghanistan (MORE, 2011) or the 2,321 violent deaths in Darfur, Sudan (DARFUR death…, 2011). Death tolls are often subject to dispute and should not be the only factor for consideration within the field, but these numbers suggest that a broader civilian protection agenda might significantly alter the standard array of cases the field focuses on.3

2.1 “Genocide” by any other name is not “genocide”

Much ink has already been spilled over the definitional challenges of “genocide.” What began as a scholarly debate over the major elements of the crime of genocide – the need to demonstrate an “intent to destroy,” the question of what constitutes “in whole or in part,” the articulation of the protected groups and the constitutive acts–has proven unhelpful in many ways for those interested in developing response mechanisms or improving policy. Over the past two decades, the courts have provided greater legal guidance for when the term might apply, but its power does not reside in its legal standing, but rather in its ethical and political significance.

Therefore, “genocide” will likely remain a highly contested term whenever applied.
Attempts to create or deploy other labels that retain the sense of significance embedded in “genocide,” but which are not similarly restricted in definition (see Table 1.1 for an overview) run into difficulties that cannot be resolved through terminology. The ethical force of genocide resides in the perception of its uniqueness. Using different language to cover a broader set of acts or using “genocide” in a loose fashion diminishes some of the conceptual clarity of the defined crime, the coherency of arguments for creating exceptional response mechanisms, and the power to mobilize a public.

Thus far, the field has developed largely through identification of extreme international failures in response to episodes of intensive, intentional killing, like the Holocaust, Rwanda, or Srebrenica, which are more readily identified as “genocide.” As such, the starting point of “genocide” is the exception, and it has worked backwards from there to identify rules that govern risk assessment, early warning and response mechanisms. At the moment of crisis–and particularly extreme crisis as witnessed in the examples above–there may well be no meaningful difference between the ideas and vision set out in the various objectives and terminology discussed in this section. However, as we will explore in the remainder of this paper, when one looks at questions surrounding prevention, models for understanding violence, and termination, the complications multiply.

Arriving at a more nuanced discussion of the goals demands an examination of the language developed to name violence. However, while greater clarity and consistent application of terminologies is helpful, debates over language will not resolve the challenge of more clearly articulating goals.


3. Genocide is not inevitable; but is it preventable?

If you were a policymaker focused on Kyrgyzstan in 2009 or 2010, what would have been your most important concern? For U.S. policymakers, high on the list was the threatened eviction from the Manas military base, a critical route for reinforcing and supplying troops in Afghanistan. On February 3, 2009, Kyrgyz President Kurmanbek Bakiyev ordered the US base to close; although he later reversed his position.

That same year, the country ranked at number 42 on the Failed States Index,4 it occupied roughly the same positioning in 2010, somewhere in the warning range, at less risk than neighboring Uzbekistan or Tajikistan, but at greater risk than Kazakhstan or Turkmenistan. Kyrgyzstan did not figure among the top 20 countries in Barbara Harff’s genocide and politicide warning list 2009 (HARFF, 2009).

Amid increasing government oppression and economic tensions, in early April 2010, popular protests that resulted in at least 75 individuals dead and over 400 individuals wounded in the capital and other northern centers overthrew the president. Analysts at the time speculated that the leadership change could aggravate the country’s existing north/south tensions (TRILLING, 2010). The interim government under Roza Otunbayeva immediately began to develop plans to consolidate the new government and institute democratic changes; she also cast new doubt on the agreement allowing the U.S. to continue using the Manas base.
Then, in violence concentrated over four days in June 2010, largely in the southern cities of Osh and Jalalabad, Kyrgyz mobs attacked the minority Uzbek community: setting homes aflame, murdering an estimated 470 people, and displacing (both refugees in Uzbekistan and internal displacement) some 400,000 (KYRGYZSTAN INQUIRY COMMISSION, 2011, p. ii). Journalists reported the grisly details of assaults against women, children, and men, clearly targeted because of their ethnicity.

The example raises critical questions for prevention: What concerns should have claimed priority for analysts and diplomats focused on Kyrgyzstan? Should the atrocity prevention agenda be able to anticipate this level of violence? How could a risk assessment or early warning system be finely tuned enough to pay attention to the multiple risks within a single country? When is a risk of turmoil (and the type of turmoil) realized? When has it only just begun?

Genocide and mass atrocities do not appear spontaneously. Research into past cases supports the view that genocides develop incrementally (VALENTINO, 2004). Activists and policymakers have recognized that the human and financial costs increase exponentially the longer one waits to take action. As the report of the Genocide Prevention Task Force rightly asserted: “In its popular conception, early warning is often equated with an alarm bell sounded just before disaster strikes. This notion is much too limited” (GENOCIDE PREVENTION TASK FORCE, 2008, p. 17). However, translating these observations into finely-tuned, accurate and timely prevention activities is not a straightforward endeavor.

3.1 Do we know enough to substantiate a unique atrocity prevention agenda?

To begin this discussion, we must make a distinction between risk assessment and early warning for genocide and mass atrocities and then we will quickly review some of the tools used for both, before we raise questions about the limitations of an atrocity prevention agenda.
Ted Gurr (2000) defines the two areas thus.

Risk assessment “identifies situations in which the conditions for a particular kind of conflict […] are present […]. Whether or not risks are realized depends on whether the preconditions remain unchanged and on the occurrence of accelerating or triggering events.”

Early warning is “derived from monitoring the flow of political events, with special attention to actions that are likely to precipitate the onset of conflict in high-risk situations [… these] early warnings are interpretations that the outbreak of conflict in high-risk likely and imminent”
(GURR, 2000).

We will begin with looking at risk assessment. The factors that analysts commonly use to assess the level of risk in a particular country can be broadly grouped into five categories, see Table 1.2. The list below consolidates the work of several researchers-—see, for instance, the work of Barbara Harff, Ted Gurr, Montgomery Marshall, Lawrence Woocher, Benjamin Valentino, Jay Ulfelder, and Scott Straus–and highlights some significant differences between variables favored by individual researchers. For example, Benjamin Valentino and Jay Ulfelder (VALENTINO; ULFELDER, 2008) advocate using infant mortality rates as a means to capture “a variety of dynamics in the political economy, including not only the accumulation and production of wealth but also the ways in which governments and citizens use (or misuse) that wealth and the effectiveness of state agencies charged with executing policy” (VALENTINO; ULFELDER, 2008, p. 15). There is disagreement about the relative strength and weakness of the state and how that correlates to violence. Another point of dissension is the centrality of state ideology; for Harff, it is of utmost relevance, not so in Valentino and Ulfelder’s model. An important area of consensus is that armed conflict significantly increases the potential for atrocities. Given the high correlation of atrocity with conflict, we will explore conflict trends in more detail later in this section.

Moving quickly into early warning, there is a range of “triggering” events that could help further focus prevention activities. The Genocide Prevention Task Force outlines several potential triggers: contentious elections; high-profile assassinations; battlefield victories; and environmental conditions (for example, drought); deadlines for significant policy action, legal judgments, and anniversaries of highly traumatic and disputed historical events. In addition, Alex Bellamy argues that a range of shifts in armed conflict – outside intervention, broken agreements, and a surge to “end” a conflict — can cause escalation in atrocities. Unconstitutional regime changes, state incapacity, and rise of ideologically revolutionary regimes can also significantly increase the potential for widespread violence (BELLAMY, 2011, p. 12). But even here, the landscape of potential cases remains too broad to effectively define imminent threats.

Monitoring these shifts in circumstance and whether they are likely to produce or increase violence requires a high-degree of case specific knowledge. The work of international monitoring groups like International Crisis Group or Human Rights Watch is crucial here. Experiments in new technologies are also offering models for early warning. SwissPeace and the Alliance for Peacebuilding, for instance, have developed a project called “Before,” that uses a variation of crowd sourcing to gather information about threats in Guinea. Sudan Sentinel uses satellite surveillance as a way to “watch” what is occurring in otherwise difficult to access areas like Sudan’s Nuba Mountains.

It is possible and perhaps even likely that risk assessment and early warning work will improve over time. But across the board in the work of risk assessment and early warning, there are some significant challenges. Both identify significantly more countries at risk than those that result in atrocities. Neither can yet adequately distinguish between different types of risks: instability, which can be either positive or negative; atrocity understood as 1,000 deaths perpetrated by a government; genocide; war, etc. There is insufficient evaluation of both quantitative systems and expert analysis. In short, there is a lot of room to improve the core tools and strategies used to inform the prevention agenda.

3.2 You’ve been forewarned: now what will you do?

Even if researchers were able to identify a place where the risk assessment indicated a threat and where a commonly cited triggering event was imminent, there is the next hurdle: conveying this information to the right policymakers who would then implement appropriate response measures. The process of assessment, warning, communication and implementation are rife with difficulties such as insufficient or contradictory information, competing agendas, resource challenges, access to key decision-makers, and availability of appropriate and feasible response mechanisms.

But perhaps the greater challenge resides in the very logic of a unique atrocity or genocide prevention strategy. In a comparison of the agendas for peacebuilding and atrocity prevention, Alex Bellamy notes few differences between the two. This leads him to conclude than a entirely separate atrocity prevention paradigm is not necessary: “what is needed is an atrocity prevention lens which informs and, where appropriate, leads policy development and decision making across the full spectrum of prevention-related activities” (BELLAMY, 2011, p. 2).

Granting that the prevention agendas, as laid out in and acknowledged by, for instance, Responsibility to Protect or Genocide Prevention Task Force Report, do not offer new approaches to established peacebuilding agendas, it should not be surprising that the reports’ military response components have garnered the most attention. It is arguable that instead of infusing an atrocities-prevention lens into pre-existing development and democratization efforts, these and other efforts to promote early action to prevent atrocities or genocide have unleashed an ill-defined paradigm for military intervention. The paradigm begins with a vaguely defined conceptual framework (genocide, mass atrocities, civilian protection, etc) and further blurs the lines into a generalized categorization of “prevention”.

How would the work of the field have to change if instead of focusing on building support for preventative military operations, actors in the field opted to prioritize increasing the funds available for development aid and crafting the prevention lens that would accompany such funds, for instance?

3.3 Respond to realities, not projections

Understanding the larger context in which atrocities become possible and how they develop is crucial. However, given the challenges of accuracy, communication and response to predictions, there is a strong argument to be made that response mechanisms have to be undertaken in relation to the distinct features of conflict or violence as they manifest, rather than what might yet occur. Framing engagement with countries around preventing the worst from happening may lead to policies that ignore or misrepresent the very real problems outside atrocities.

For those who do not agree that something worse is on the horizon, it is simple to discount warnings. Or if action is taken, it is similarly easy to discredit strong response mechanisms (particularly the more coercive ones) as politically motivated. And, of course, there is always a measure of politics involved in military deployments, as there should be.

Further, while there are a great number of similarities between peacebuilding and atrocity prevention, one important difference remains. Peacebuilding defines a positive goal (improving a situation) and atrocity prevention implies a negative goal (ensuring something does not happen). Engaging countries around a prevention agenda means getting locked into the basic logical trap of trying to prove that something would have happened if action had not been undertaken. This is always going to be a weak argument.


4. Conflict patterns are changing: are you ready?

Recognizing that atrocities and genocide frequently occur within armed conflict, the study of conflict trends is central to anticipating how atrocity trends might develop in the future and what new response mechanisms might be necessary. Not all conflicts result in mass atrocity, but given that atrocities frequently occur in the context of armed conflict, it is worth exploring this topic in greater detail.

First, some good news. Since the end of the Cold War era, conflicts are less deadly and there are fewer international conflicts.5 This is likely caused by a number of factors: end of the Cold War-fueled proxy conflicts, fewer conflicts involving the major powers, exponential increase in peacemaking and peacebuilding activities, development and expansion of international norms, global economic interdependence, increased number of democracies, overall decreased mortality rates, and rising national incomes. This does not mean conflict has ended.

Today, conflicts tend to occur inside poor countries, geographically clustered together (CENTER FOR SYSTEMIC PEACE, 2011, figure 8) that are neither democracies nor autocracies, but an unstable mix of both, termed “anoncracies” (HUMAN SECURITY REPORT PROJECT, 2011, p. 76). Since the end of the Cold War, while there are fewer autocracies and more democracies, there has also been a sharp increase in the number of anocracies. That number has held fairly steady ever since (CENTER FOR SYSTEMIC PEACE, 2011, figure 12). In other words, the number of societies at particular risk seems to have become a stable part of the global horizon. Conflicts today also re-start at higher rates than in the past, although there are differences among various researchers as to the resurgence rate (GENEVA DECLARATION, 2008 p. 58).
The key actors have changed. International and transnational actors play significant roles today, both as agents of conflict, but also in terms of responders and interveners. According to the Human Security Report (HSR), between 2003 and 2008, there was a 119% increase of non-state conflicts, defined as “confrontations between communal groups, rebels, or warlords that do not involve the state as a warring party” (HUMAN SECURITY REPORT PROJECT, 2011, p. 10). Mary Kaldor has argued that global trends towards the de-centralization and privatization have altered who has the finances to enable, communications to organize, and means to enact violence. She describes the old wars as conflicts of state-building, and the new wars as wars of state “un-building” (KALDOR, 2007, p. 16).

These trends culminate in new vulnerabilities for civilians. Notably, there has been an increase in smaller-scale assaults against civilians. Here, the numbers are alarming: between 1989 and 2002, the number of such campaigns increased by 70 percent (HUMAN SECURITY REPORT PROJECT, 2011, p. 177).

The perpetrators of the greatest number of casualties from one-sided violence remain mostly governments, but include non-state actors. They are, in order of ranking: Rwandan government (1994), government of Bosnia Serbs (this government was a breakaway faction, not the legitimate government of Bosnia), Government of Sudan, AFDL (DRC), government of Afghanistan, government of DRC, LRA, government of Burundi, Janjaweed, and the Islamic State of Iraq (HUMAN SECURITY REPORT PROJECT, 2011, p. 183). It will surprise no one in the atrocities prevention field that Rwanda, Bosnia, Sudan, DRC, Lord’s Resistance Army, and Burundi are on this list. The presence of Afghanistan and Iraq (dates not specified), two cases that have not been on the anti-genocide agenda (at least not in the U.S.) raises questions about how the field defines its scope of work.

These factors taken together lead us to question whether the greatest threats to civilians today have changed over time from large-scale offensives against civilians to more geographically and demographically contained cases increasingly involving non-state actors. How these changes impact an atrocity prevention agenda depends on the goal of actors in the field – the broader the mandate, the more today’s shifts in armed conflict are relevant. For those who wish to remain tightly focused largely on swift, potentially overwhelming killing offensives, the changes in conflict may be of more limited relevance.


5. What rules govern the exception?

Raphael Lemkin, who coined the term genocide in his 1944 publication, Axis Rule in Occupied Europe, is arguably also the first person to systematically study the phenomenon (LEMKIN, 1944). Key to his concept of “genocide” are two insights that would later be significantly altered in the legal definition. First, Lemkin specifically did not limit “genocide” to killing.

While the Nazi assaults against European Jews were the most radical manifestation of the regime’s genocidal policies towards occupied populations, but genocide enveloped much more than this extreme form. He described multiple objectives of genocidal plans, including the “disintegration of the political and social institutions, of culture, language, national feelings, religion, and the economic existence of national groups, and the destruction of the personal security, liberty, health, dignity, and even the lives of the individuals belonging to such groups” (LEMKIN, 1944, p. 90).

Second, Lemkin identified genocide as composed of two phases: “one, destruction of the national pattern of the oppressed group; the other, the imposition of the national pattern of the oppressor” (LEMKIN, 1944, p. 79). This two-step process reflects Lemkin’s expression that one could document “genocide” against both a population that had been removed (in any number of ways) and also against one that remained subject to colonization by an oppressor.

For decades after Lemkin’s pioneering work, there was very little study of genocide. A small group of scholars took up this project beginning in the 1970s. Their work forms the starting point for what became an explosion of interest in the late 1990s and into the new millennium.

The study of genocide was for many years, and still is to some extent, overdetermined by the dominance of one case: the Holocaust. This history has cast a long shadow over the development of explanatory theories and response mechanisms. While the Holocaust remains a logical part of any study of genocide, it is marked by some specific characteristics that must be appreciated in order to make any meaningful comparisons. The international scale of both the armed conflict and the genocide, the attempt to murder all of the targeted group, and the use of industrialized killing mechanisms combine to create a unique circumstance. Genocide studies has tended towards cases and processes that more closely adhere to a model of the Holocaust.

The research dynamics are now changing, but the field developed with little attention to negative cases, that is instances that demonstrate similar characteristics but did not result in genocide. Further, the dominant theories rely heavily on study of the nation-state during episodes of mass violence, yet rarely connect specific episodes of violence to larger political, economic and social processes, including global processes, which impact the state development (MOSES, 2008). Even within the state model, many theories place particular emphasis on the role of national elites (LEVENE, 2004a, 2004b). The study of genocide, and now mass atrocities, has too often developed without reference to the literature of political violence. Further, only recently have studies in this area included the impact of variations in contexts of mass violence: for example, cases that occur as part of colonial or settler violence, during coups or counter coups, communal violence, revolutionary social transformation, secession, partition, or counter-insurgency.

But many of the assumptions of the early development of research on genocide have found their way into the work of prevention and response. More recent work that expands the array of cases and contexts in which atrocities occur has yet to be integrated into the policy-oriented discussions.

5.1 The importance of studying local dynamics

By 2001, major fighting across front lines had come to a standstill in the Democratic Republic of Congo (DRC). But violence escalated in the east as local groups, known as Mai Mai, took up arms (STEARNS, 2011, p 251-266). The fought for a range of reasons: in opposition to Rwandan and Rwandan-allied forces, notably the RCD; sometimes to protect their own villages and exact revenge for attacks they suffered; their own enrichment, often by “taxing” local populations; settling old disputes over land or power; or punishing civilians associated (however loosely) with an opposing side. With weapons flowing from both the Congolese and Rwandan governments, Mai Mai groups formed quickly and without needing to rely on local populations for support.

It was in this context that some 1,000 people were massacred at Kasika, a small jungle village one hundred miles west of the Rwandan border. The road through Kasika eventually leads to a gold mine, hence its strategic value. The chief of Kasika, Francois Naluindi, was widely popular and respected. But nearby was a more militant leader, known as Nyakilibi, who had begun arming youth declaring he would defend his country against the Tutsi invaders, although locals thought Nyakilbi’s real interest was in expanding his land rights.

When a group of Rwandan and RCD soldiers passed through the town, Naluindi offered them food and welcomed them. As they exited the town, Nyakilibi and his men fired shots at them. Soldiers searched for the Mai Mai and then moved on. A week later, the RCD and the Rwandan group once again passed through the village. Again, the Mai Mai shot at the party, killing a high-ranking and legendary Rwandan leader, Commander Moise.

This time, the Rwandans and RCD soldiers stayed in the village overnight. The next morning RCD troops attacked a group at the Church, bludgeoning them to death. They succeeded in killing the village priest and several nuns. They also killed the chief, his pregnant wife and most of Naluindi’s extended family, who had sought shelter at his house. The victims were not only killed–many were disfigured and torn apart. One survivor commented: “It was like they killed them, and then they killed them again. And again” (STEARNS, 2011, p. 257). A Congolese researcher, Floribert Kazingufu, also notes that the murder of the chief later set off another conflict over succession that further divided the village (KAZINGUFU, 2010).

Among those who committed the massacre at Kasika, were Banyamulenge, Congolese youth of Rwandan background but who had lived in Congo for years – decades for some. They had long suffered discrimination in Congo, and like the Mai Mai, had many motivations for joining militias. Stearns enumerates a few:

The longing to be accepted as Congolese citizens, to obtain land rights, and to be represented in local and provincial administration. Of course, many of the youth also wanted to succeed, to obtain power and fame […] the careers of many ambitious Banyamulenge had been blocked by the discrimination and favoritism fostered by Mobutu.
(STEARNS, 2011, p. 264).

This story demonstrates that not all patterns of violence can be explained at the national level. Long-standing, unaddressed political and social claims coupled with the means and license to resort to violence, as well as unpopular external actors all contributed to the larger context of armed conflict. From there, each act of violence set in motion new grievances and further legitimized violence as a means to resolve claims.

A small amount of research has been done on the variations of when, where, and how violence occurs inside a single case study, but the work that exists suggests that response mechanisms not currently understood or even visible may be embedded in the knowledge of these patterns (KALYVAS, 2006, p. 14). Other studies, like Scott Straus’ work on hate radio in Rwanda (STRAUS, 2007) suggest that some of the response mechanisms considered part of the “toolbox” may be much more limited than often considered.

Critical to extrapolating policy insights from this level of inquiry is a sense of the dynamic evolution of violence. This implies a marriage of rich, case-specific knowledge with understanding of how international response mechanisms function not only to respond to given conditions, but also as productive forces in the complex interplay of the international, national and local dynamics.

Outside of increasing recognition of the need for more work at this level, there is little consensus. For instance, in her work on the local dynamics of violence in the DRC, Severine Autesserre urges international actors to focus on interventions at the local level to address the political claims of local actors. Only then, she argues, will the international players see real sustainable gains resulting from their peacemaking efforts. Alex de Waal (DE WAAL, 2010), however, has argued that such interventions are unlikely to be successful. Local disputes in conflict-ridden societies often follow a logic of the marketplace, a process of continual renewal of agreements based on the going price – be it actual funds, access to power or resources, or other terms of negotiation. International interventions into this process momentarily inflate the “price” of a settlement, and once the international engagement diminishes, the market will “correct,” often violently.

5.2 Who does the killing?

In the last week of February 2007 at the trial of Vujadin Popovic et al before the International Criminal Tribunal for the former Yugoslavia, a witness for the prosecution told one of the most astounding stories of the genocide at Srebrenica that has yet come to light. The witness worked as a truck driver with the Bosnian Serb army, on the day in question he delivered drinks and food to soldiers working the execution squads.

The story begins in the days following the fall of Srebrenica, one of the last Bosnian government hold-outs in the eastern territory almost totally controlled by Bosnian Serb forces. In the month of July 1995, Bosnian Serbs launched their final assault on Srebrenica. They took Dutch peacekeepers hostage, and, receiving little international response from NATO or the UN, seized the town. They separated the men from the women and children and hunted down other men who had fled through the woods. All those captured were taken to execution sites where they were systematically murdered. In total, some 8,000 Muslims, mostly men, but including some women and children, were killed.

At one such killing site, Bosnian Serb soldiers had just fired their weapons at a line of blindfolded and bound Muslim men, when the above truck driver arrived. He testified to what he saw there, as the men fell dead:

In that heap, in that pile of dead bodies, who did not resemble people any longer, this was just a pile of flesh in bits, and then a human being emerged. I say a human being, but it was actually a boy of some five to six years. It is unbelievable. Unbelievable. A human being came out and started moving towards the path, the path where men with automatic rifles stood doing their job. And this child was walking towards them. All of those soldiers and policemen there, these people who had no trouble shooting — I shouldn’t judge them because I don’t know about their situation. Perhaps they did it because of the order they received and perhaps they did it because of their nature. There are all kinds of people, and some of them may have done it gladly. Some probably did it because they had to. And then all of a sudden they lowered their rifles and all of them, to the last one, just froze.

The truck driver explained that the commanding officer demanded that the soldiers shoot the boy, but not one of them would, not even the officer himself. Finally, they turned the boy over to the driver to bring back with the next load of victims. The driver instead took the boy to a hospital and he survived.

What changed in that moment such that men engaged in a killing operation suddenly refuse orders? What do we know about the individuals who kill? Those who stand on the sidelines, not taking sides when innocents are murdered? Those who resist? Many stories of survival from across the cases of genocide include examples of people refusing the logic of genocide – sometimes emphatically, sometimes only fleetingly. But even those cursory seconds provide us insight into the factors that impact individual’s decision-making in times of atrocity.

In 1950, Theodor Adorno, a leading intellectual from the Frankfurt School who fled Nazi Germany, together with Else Frenkel-Brunswick, Daniel Levinson and Nevitt Sanford published The Authoritarian Personality (ADORNO et al., 1950), an inquiry into the psychological profile of people who support authoritarian governments. Their conclusion was that certain personality characteristics tend toward fascist ideology. An authoritarian personality, they posited, is a form of psychological aberration.

A few years later, in 1963, Hannah Arendt, after watching the Adolf Eichmann trial, came to a very different conclusion. In Eichmann in Jerusalem: A Report on the Banality of Evil (ARENDT, 1963), Arendt argues that extraordinary evil is possible because it becomes the norm and regular people carry out its measures. Today, the consensus is much closer to Arendt’s position than to Adorno’s.

Across the range of cases, one clear insight has become apparent: perpetrators of atrocity are “normal”– they represent a demographic cross-section of their societies. In his signature exploration of the motivations of perpetrators, James Waller concludes:

As we look at perpetrators of extraordinary evil, we need no longer ask who these people are. We know how they are. They are you and I. There is now a more urgent question to ask: How are ordinary people like you and me, transformed into perpetrators of extraordinary evil? (WALLER, 2002, p. 133).

This question leads us back to some of the same questions presented in the early warning and risk assessment section about structural factors that impact the likelihood of mass violence.
But micro-level research involves more than perpetrators. Understanding the motivations, options, and strategies of a range of individuals– survivors, bystanders, rescuers — helps us better understand the phenomenon of genocide itself. However, motivations and patterns of participation within and across cases vary greatly and often change over time. This high degree of variation renders the wealth of narrative examples an endless and fascinating body of work to explore, but may make it difficult to arrive at broad-reaching conclusions related to prevention.


6. How would you measure success in ending genocide(s)?

On June 17, 2009, U.S. Presidential Envoy for Sudan, General Scott Gration stated that Darfur was experiencing “remnants of genocide,” thereby touching off a bitter disagreement within the Obama Administration, notably with U.S. Ambassador to the United Nations, Susan Rice. Two days earlier, Rice had described the situation as “genocide,” as had President Obama earlier that month (WONG, 2009). Journalists’ accounts of the disagreement used the adjective “furious” to describe Rice’s response to Gration’s comments.

By 2009, the scale of systematic assaults on civilians had significantly decreased and mortality rates in the refugee and displaced persons camps were largely back to normal levels. Yet there remained an enormous, vulnerable population of displaced civilians beset by a range of acts of violence in a context of civil war with a government that retained the capacity and had amply demonstrated the will to conduct organized campaigns of violence against civilian groups. Could this be defined as the end of genocide?

On October 19, 2009, debates within the Obama administration were resolved with the announcement of a new Sudan policy. They retained the Bush Administration’s use of “genocide” to describe the situation, and the policy consisted of three simultaneously–and apparently equally weighted–priorities: a “definitive end to conflict, gross human rights abuses, and genocide in Darfur,” implementation of the 2005 Comprehensive Peace Agreement between Sudan’s North and South, and efforts to ensure that Sudan would not again become a haven for terrorists.

But the debate between Gration and Rice was neither simply semantic nor was it purely a disagreement over policy options. It related to a fundamental question for the field: what constitutes an end to genocide and who determines the definition? How groups choose to answer this question is critical; it constitutes the ultimate measure of their success.

6.1 The shadow of what ought to be

The tension in the field today about what constitutes its objectives spills over into the discussion of endings6 and therefore what defines success. Is success defined as ending genocide or atrocities, as such, or the ending of discrete occurrences of genocide or mass atrocities? There has been little focused discussion of even what the latter, more contained goal, entails. The “moment of ending” that currently informs work in the field often contains multiple measures of success: an end to dying, an end to the circumstances that enabled the dying to occur (up to and including regime change), and the arrival at some form of justice for the victim group, be it judicial, monetary or symbolic.

However, actual endings rarely live up to this vision: mortality rates may decrease, but perpetrator regimes may remain in power, some form of conflict may continue, and victim’s needs may be unmet. Often, one incident or series of offensives may end, only for violence to reappear later or elsewhere. Nonetheless, these suboptimal endings may be more realistically achieved; and they may save lives. Disaggregating these components of endings enables a more nuanced understanding as well as a more frank discussion over which ending might constitute a goal for different actors, what actions might most support it, and how one would measure success.

6.2 The historical record

As an illustration, a small selection of past cases is presented in Table 1.3. Of these twelve cases reviewed, we find five cases where the genocidal episode coincided with the end of the armed conflict, in four of those cases the perpetrator regime was completely defeated (Ottoman Turkey, Nazi Germany, Khmer Rouge in Cambodia, and Rwandan government). In the one case, Bosnia-Herzegovina, outside interveners engaged largely on humanitarian rather than political or interested basis and the conflict ended through negotiations that kept much of the perpetrator regime intact in areas it controlled. World War I, World War II, Vietnam’s invasion of Cambodia, and the Rwandan Patriotic Front’s victory were all fought primarily to win a war, the results of which ended genocide. The capacity of armed forces associated with the victim groups to mount resistance played a role in ending atrocities in the Nuba Mountains, Rwanda and Bosnia. Although it is worth noting that armed rebellion is among the factors cited as a risk factor for atrocities to occur in the first place.

The more frequent ending of the cases on our list is that the perpetrators remained in power, but after defeating an opposing force or subduing them, stopped short of total destruction. The communities suffered terribly, but physical elimination of a targeted group is not a common perpetrator goal. The deployment of violence, as Stathis Kalyvas has argued, can be exceptionally cruel even if its aim is to control a group, not to exterminate it (KALYVAS, 2006, p. 26-27).

Very little is known about the internal processes in perpetrator regimes across the historical record of atrocities. Some measure of internal dissent about the scale of killing appears to make a difference in the path of violence. But the decision-making process in many cases has not been sufficiently probed. Are there clues in such study that could inform efforts to alter perpetrators choices?

In cases where either the conflict continued or the regime remained in power, often the capacity and will to commit widespread abuses was unchanged, and violence began again against the same or new groups. This implies that even where instances of genocide or atrocities end, that further analysis is needed to see where additional internal threats might appear.

Finally, negotiations are not well suited to ending atrocity, even if they remain the only option for ending a conflict. Negotiation implies a measure of equality between sides and the ability of said sides to assert their interests; atrocities and genocide occur as asymmetrical assaults against civilian groups. While resistance, as noted above, can ultimately turn the tide of a conflict, it is, by definition, not an option for the civilians as the victims of the mass violence campaigns.

6.3 Who defines the end?

Even in 2004, some Rwandan women were still dying directly from the perpetrator’s actions during the period of genocide. Women who had been raped by HIV positive men were marked for death because the perpetrators knew that even if their victims survived the killing campaigns, that AIDS would curtail their lives. In the case of many women rape survivors for whom antiretroviral drugs were simply too expensive, they were correct.

Minority returns have been a problem that has plagued post-conflict Bosnia. Displaced people and refugees returned in significant numbers to areas where their ethnicity was a majority, but for individuals whose pre-war homes were in areas where they would now be an ethnic minority, the rates of return were significantly lower. The end result is a country where the effects of wartime ethnic cleansing have permanently altered the society.

Among the millions of displaced Darfuris living in camps–some for what will soon be a decade–at the edge of towns, blurring the line between camps and slums, are farmers. For them, the loss of a connection to the land is a vital blow to their identities and communities. Calculating an end that would somehow right these wrongs, or at least provide an acceptable address, is a worthy undertaking. However, it demands a long-term commitment to particular societies well beyond the end of killing or even the conflicts that enabled killing.

It is also unlikely to form the basis of an ending for policy discussions that focus on the techniques of a “toolbox” for response mechanisms. For organizations and individuals committed to ending genocide and mass atrocity, there needs to be a serious discussion based on what they can deliver–not necessarily all the time or according to a perfect agenda–but in realistic scenarios. What constitutes success for the field? How could said success be measured?

The crime dubbed an “odious scourge on humanity” by the drafters of the UN Genocide Convention is much easier to condemn then it is to transform. Efforts to do so will likely fall short of their ideal goal and the populations at risk suffer unimaginable pain; it is the nature of the problem. And many more civilians will suffer from targeted violence in ways that are difficult for the field to address in any coherent way. To some extent, none of these assaults has ever ended: in terms of the permanent scars communities and individuals pass on to subsequent generations, the land and other goods stolen, and the grossly unjust violence perpetrated. But it does not withstand the test of humility, sustainability or honesty to imagine that the “field” however constituted could deliver on all these promises.

Today, it is up to the field of genocide prevention and response to examine what, precisely and realistically it defines as success and what is necessary to achieve it.

• • •


1. Jill Savitt’s research and insights on the nature of the field today greatly influenced the development of this section of the paper.

2. Since 2004, HRW has increased the amount of work they do on economic, social, and cultural rights. See, for example: <>. Last accessed on: May. 2012

3. The cases most often cited, even a slightly expanded list, entail the deaths of close to 100,000 people, sometimes exponentially more. For example: assaults against the Herero, Armenian genocide, Holocaust, Nigerian Civil War, Bangladesh, East Timor, Cambodia, Guatemala, Burundi, Bosnia- Herzegovina, Rwanda, Sudan. Dropping this number to 5,000, as Alex Bellamy has done in a report for the Stanley Foundation significantly expands the list to 103 examples in just the post-WWII era. Dropping it further to 1,000, for instance, alters the nature of the phenomenon at hand. Different metrics make sense for different goals, inclusion of a broad number of cases is important for research but may not be a practical for dictating policy response under the banner of “mass atrocity” or “genocide” prevention.

4. The Failed States Index is an attempt to quantitatively measure state instability and is produced annually by the Fund for Peace. For more information, see <>. Last accessed on: May. 2012.

5. See, for instance, Human Security Report 2009/2010 (HUMAN SECURITY REPORT PROJECT, 2011), produced annually by the Simon Fraser University, Canada’s Human Security Report Project. The authors write that from 1984 to 2008, high intensity conflicts that result in 1,000 or more deaths a year, which tend to be between states and involving major powers — have decreased by 79% (157). Other researchers have argued that this decline reversed slightly between 2005 and 2007, largely due to the impact of armed conflict in five countries: Afghanistan, Iraq, Pakistan, Somalia, and Sri Lanka. See “Global Burden of Armed Violence” (GENEVA DECLARATION, 2008, p. 9).

6. This section draws on the work from a series of seminars held with Alex de Waal and Jens Meierhenrich. The seminars examined a number of cases, as well as disciplinary approaches to endings. More about the research project can be found here: <>. Last accessed on: May 2012.

• • •


Bibliography and other sources

ADORNO, T.W. et al. 1950. The Authoritarian Personality. New York: Harper and Row.

AMNESTY INTERNATIONAL. 2011. Demand Dignity. Available at:  http:/ . Last accessed on: 15 Oct. 2011.

ARENDT, H. 1963. Eichmann in Jerusalem: A Report on the Banality of Evil. New York: The Penguin Group.

BELLAMY, A. 2011. Mass Atrocities and Armed Conflicts: Links, Distinctions, and Implications for the Responsibility to Prevention. Washington, D.C.: The Stanley Foundation.

CENTER FOR SYSTEMIC PEACE. 2011. Global Conflict Trends. Available at: . Last accessed on: 27 Nov. 2011.

CHARTER OF THE INTERNATIONAL MILITARY TRIBUNAL (CHARTER). 8 Aug. 1945. Available at: . Last accessed on: May 2012.

DARFUR death toll mounts amid new war fears. 2011. Agence France-Presse, 19 January. Available at: . Last accessed on: 27 Nov. 2011.

DE WAAL, A. 2010. Dollarised. The London Review of Books, London, v. 32, n. 12, p. 38-41.

ECK, K.; SOLLENBERG, M.; WALLENSTEEN, 2003. One-Sided Violence and Non-State Conflict. In: HARBOM, L. (Ed.). States in Armed Conflict. Uppsala: Uppsala University, Department of Peace and Conflict Research Report. p. 133-142.

FUND FOR PEACE. 2009. The Failed States Index. Available at: . Last accessed on: 27 Nov. 2011.

GENEVA DECLARATION. 2008. Global Burden of Armed Violence. Geneva: Geneva Declaration Secretariat. Available at: . Last accessed on: 27 Nov. 2011.

GENOCIDE PREVENTION TASK FORCE. 2008. Preventing Genocide, A Blueprint for U.S. Policymakers. Washington, D.C.: U.S. Holocaust Memorial Museum, The American Academy of Diplomacy, The United States Institute of Peace.

GURR, T.R. 2000. Peoples Versus States: Minorities At Risk in the New Century. Washington: United States Institute of Peace Press.

HARFF, B. 2009. Barbara Harff’s risk assessments: 2009. Available at: . Last accessed on: 27 Nov. 2011.

HARFF, B.; GURR, T. 1988. Toward Empirical Theory of Genocides and Politicides: Identification and Measurement of Cases Since 1945. International Studies Quarterly, Bloomington, v. 32, n. 3, p. 359-371.

HIGONNET, E. (Ed.). 2009. Quiet Genocide: Guatemala: 1981-1983. New Brunswick and London: Transaction Publishers.

HUMAN SECURITY REPORT PROJECT. 2011. Human Security Report 2009/2010. New York and Oxford: Oxford University Press.

INTERNATIONAL COMMISSION ON INTERVENTION AND STATE SOVEREIGNTY (ICISS). 2001. The Responsibility to Protect. Ottawa: International Development Research Centre.

JONAS, S. 2009. Guatemala: Acts of Genocide and Scorched-Earth Counterinsurgency War. In: TOTTEN, S.; PARSONS, W. (Ed.). Century of Genocide. New York and London: Routledge. p. 377-411.

KALDOR, M. 2007. New Wars and Human Security. Democratiya, New York, v. 11, p. 14-35, Winter.

KALYVAS, S. 2006. The Logic of Violence in Civil War. Cambridge and New York: Cambridge University Press.

KAZINGUFU, F. 2010. Two Conflicts, One Village: The Case of Kasika. Insights on Conflict, 27 May. Available at: . Last accessed on: 27 Nov. 2011.

KHAN, I. 2009. The Unheard Truth: Poverty and Human Rights. New York: W.W. Norton And Co.

KYRGYZSTAN INQUIRY COMMISSION. 2011. Report of the Independent International Commission of Inquiry into the Events in Southern Kyrgyzstan in June 2010. Available at: Last accessed on: May 2012.

LEMKIN, R. 1944. Axis Rule in Occupied Europe. Washington, D.C.: Carnegie Endowment for International Peace.

LEVENE, M. 2004a. A Dissenting Voice: Or How Current Assumptions of Deterring and Preventing Genocide May Be Looking at the Problem Through the Wrong End of the Telescope. Journal of Genocide Research, London, v. 6, n. 2, p. 153-166.
______. 2004b. A Dissenting Voice: Part Two. Journal of Genocide Research, London, v. 6, n. 3, p. 431-445.

MORE, E. 2011. More civilians killed last year in one Mexican border town than all Afghanistan. CNS, 25 February. Available at: . Last accessed on: 27 Nov. 2011.

MOSES, D. 2008. Toward a Theory of Critical Genocide Studies. Online Encyclopedia of Mass Violence, 18 April. Available at: Last accessed on: 27 Nov. 2011.

ROBINSON, M. 2004. Advancing Economic, Social, and Cultural Rights: The Way Forward. Human Rights Quarterly, Baltimore, v. 26, p. 866-872.

ROTH, K. 2004. Defending Economic, Social and Cultural Rights: Practical Issues Faced by an International Human Rights Organization. Human Rights Quarterly, Baltimore, v. 26, p. 63-73.

RUMMEL, R.J. 1990. Lethal Politics: Soviet Genocides and Mass Murders 1917-1987. Rutgers, New Jersey: Transaction Publishers.

SCHEFFER, D. 2006. Genocide and Atrocity Crimes. Genocide Studies and Prevention, Toronto, v. 1, n. 3, p. 229-250.

SEMELIN, J. 2009. Purify and Destroy: The Political Uses of Massacre and Genocide. Translator: Cynthia Schoch. New York, New York: Columbia University Press.

SEWELL, S.; RAYMOND, D.; CHIN, S. 2010. Mass Atrocity Response Operations: A Military Planning Handbook. Cambridge: Harvard University.

SIKKINK, K. 2004. Mixed Signals: U.S. Human Rights Policy and Latin America. Ithaca: Cornell University Press.

STEARNS, J. 2011. Dancing in the Glory of Monsters: The Collapse of the Congo and the Great War of Africa. New York: Public Affairs.

STRAUS, S. 2007. What is the Relationship Between Hate Radio and Violence? Rethinking Rwanda’s ‘Radio Machete. Politics & Society, Thousand Oaks, CA, v. 35, n. 4, p. 609-637.

TRILLING, D. 2010. Kyrgyzstan: Experts Voice Concern about Possibility of Prolonged Strife. Eurasia Insight, 7 Apr. Available at: . Last accessed on: 27 Nov. 2011.


UN GENERAL ASSEMBLY (UNGA). 2002. Rome Statute of the International Criminal Court (last amended January 2002), 17 July 1998, A/CONF. 183/9. Available at: . Last accessed on: May 2012.

VALENTINO, B. 2004. Final Solutions: Mass Killing and Genocide in the 20th Century. Ithaca and London: Cornell University Press.

VALENTINO, B.; ULFELDER, J. 2008. Assessing Risk of State-Sponsored Mass Killing. 1 Feb. Available at Social Science Research Network: . Last accessed on: May 2012.

WALLER, J. 2002. Becoming Evil: How Ordinary People Commit Genocide and Mass Murder. Oxford: Oxford University Press.

WONG, K. 2009. US Tries to Walk Back From Comments Downplaying Darfur Genocide. ABC News Blogs, 18 June, Politics, Political Punch. Available at: Last accessed on: 27 Nov. 2011.


INTERNATIONAL CRIMINAL TRIBUNAL FOR THE FORMER YUGOSLAVIA (ICTY). 2007. Prosecutor vs. Vujadin Popovic et al. 22 February. Available at: . Last accessed on: 27 Nov. 2011.

Bridget Conley-Zilkic

Bridget Conley-Zilkic is Research Director for the World Peace Foundation at the Fletcher School. She previously worked as research director for the U.S. Holocaust Memorial Museum’s Committee on Conscience, where she led the Museum’s research and projects on contemporary threats of genocide.


Original in English.

Received in November 2011. Accepted in May 2012.