For just war doctrine to apply, the last resort requirement to exhaust all measures short-of-war must be fulfilled. Because of research and policy developments in the last fifteen years, the international community is now equipped with a richer understanding of how wars and atrocities evolve through time, improved precision about trigger points and risk factors that may accelerate that evolution, growing consensus on what prevention and mitigation steps to look for in that process, and new technologies for ascertaining these steps in order to intervene when mitigating action might deflect the escalation. It is thus argued that the responsibility of the international community to intervene in a timely and appropriate fashion has become increasingly clear and inescapable. It is further argued that the alert engagement of civil society in crafting this body of research and policy places a heavy public burden on government leaders to demonstrate that indeed all measures short-of-war have been exhausted. We now have at our collective disposal many more measures to deploy and many more witnesses to raise the alarm. Accordingly, the threshold for declaring that last resort has been reached has now become much higher.

Just war doctrine pivots on six requirements that must all pertain in order to make a moral case for going to war: a just cause, the right intention, conduct by a legitimate authority, a high probability of success in accomplishing the intended outcome, respect for the principle of proportionality, and a commitment to using force as a last resort. Despite this crisp summary list, nothing is simple in the application of these elements to a particular war or fact case. Over the centuries, most of these elements have received the close attention of religious scholars, moral philosophers, authorities on military strategy and history, and academic political scientists.

In current crises, great powers may seek to intervene in atrocity-laden conflicts where civilians and perhaps combatants are at grave risk of death or torture. The military undertaking in these crises is not one of self-defense, but rather is intended to promote international peace and security. In this context, there appear to be several reasons not to invoke just war as the legal, normative, or policy justification for international response. The grounds of just cause may prove unreliable, there may be insufficient assurance of success, or there may prove to be problems with proportionality.

The first part of this essay advances another reason why the notion of just war may not be applicable: insufficient documentation that last resort has been reached. Evidence demonstrating that all other measures short-of-war have, in fact, been exhausted is usually lacking. This gap in documenting the path to last resort may in the past have been due to disagreement over what public rationale was required, what information to privilege, or what options for intervention short-of-war might have seemed feasible. The gap may also have persisted because the international community – and civil society in particular – may have been less clamorous than they are now in seeking a full explanation for the decision of one country or countries to embark upon armed conflict in just war terms.

Based on efforts over the last fifteen years, the policy community is now equipped with excellent tools and technology for establishing actionable early warning and for selecting targeted early response based on the early warning supplied. The most precise and empirically grounded pathways for making these connections lie within the frameworks for early warning of atrocity crimes and for conflict prevention and response. These pathways offer a significant opportunity for the international community to mobilize and document the work that has been done to restrain the warrant for war.

Yet, as will be discussed in the second part of this essay, there also remains a substantial piece of work to be done to explicitly link inputs from the early-warning realm to appropriate specific interventions for prevention and response. Accomplishing this final step will require commitment from experienced analysts and authorities. The goal would be to make both sides of the current sets of documents – the early-warning framework and early-response toolkits – talk to each other in strategic and real-time operational terms. This section will provide a vignette of the complexities that lie ahead in making practical and timely connections between these two sets of activities – early warning and early response – and explore the remaining major challenges of shaping the response agenda to meet the nuances of early warning and the pressures of acting in time.

The underlying argument in this essay is that in the context of action aimed to protect people in other countries from atrocity crimes committed by state or nonstate armed actors, more policy attention must be given to ensure that all feasible political and diplomatic measures have been taken prior to embarking on armed conflict. The more useful and accessible the frameworks for warning and response, the stronger the cautionary tug on invocation of last resort should become.

The 2014 release of the United Nations' Framework of Analysis for Atrocity Crimes opened up new potential at the highest level of international governance to mobilize efforts to protect populations in wars and atrocity situations.1 Atrocity crimes refer to three legally defined international crimes: genocide, crimes against humanity, and war crimes.2 The study and experience of atrocity crimes have yielded deep understanding of their origins, features, and effects; produced a densely articulated set of legal definitions; and spurred the development of a risk and escalation framework derived from historical excavation and practical experience. By articulating this robust framework for early warning of atrocity crimes, international policy-makers have now provided analytically and empirically derived categories of information – or risk factors – that serve to structure determinations regarding local or regional trends in number, kind, and intensity of these impending or escalating crimes. The risk factors that apply across all atrocity crimes are:

  1. Situations of armed conflict, crisis, or other forms of instability;

  2. Record of serious human right violations and related impunity;

  3. Weakness of state structures;

  4. Motives or incentives to commit atrocity crimes;

  5. Capacity to commit atrocity crimes;

  6. Enabling or preparatory action to commit atrocity crimes;

  7. Absence of mitigating factors; and

  8. Triggering factors.3

The UN Framework also identifies risk factors and indicators that may be unique to a specific atrocity crime (such as genocide or crimes against humanity).

This accumulating fact picture, employing assessment of risk factors and presence of particular indicators of intensity and escalation in a specific context, permits what the UN terms “early warning” of atrocity crimes.4 The warning is “early” in that the level of identified atrocity in a country or region may not have reached the international legal standard of atrocity crimes, and it is a “warning” in that it is constituted to provide the international community (including the UN, national governments, and civil societies) with an informed and timely depiction of instability and potential paths of escalation in a given situation.

An early-warning analysis does not require that all the risk factors be present or that all the indicators within them be fulfilled. The historical, geographical, political, or cultural features of a specific country will make it susceptible to events or trends that in another context might not be likely to recruit social hatreds or exacerbate deadly competition for resources.5

By identifying and describing key “trigger events,” the Framework provides analytical depth to the dynamic process that observers must monitor in order to assess the potential for atrocity crimes.6 In a landscape described by risk factors, vigilance given to new emerging features permits early attention to be given to possible trigger events. These events might signal a rapid shift in social dynamics or circumstances that serve to destabilize existing social, economic, or political relationships among subgroups who exist in uneasy equilibrium with each other.

Many historical instances of such trigger events that precipitate major conflicts or atrocities have earned each of these risk factors the status of classic disrupters. The comprehensive list includes: sudden deployment of security forces; spillover of armed conflicts or territorial tensions; measures taken by the international community perceived as threatening to sovereignty; abrupt or irregular regime changes or transfers of power; attacks against the life or security of leaders of opposition groups or terrorist acts; incitement or hate propaganda targeting specific groups of individuals; census, elections, or other developments relevant to the process of power or governance; sudden changes to the economy or workforce, such as a financial crisis, major disaster, or epidemic; discovery of natural resources or their exploitation that has an impact on the livelihoods of groups or civilian populations; commemoration of past crises or collective traumas or historical events; and acts related to accountability processes, if seen as unfair.

The various stakeholders and civil-society groups who have worked on atrocity and war have developed a powerful consensus on the key elements and relationships among these elements that could, were an “atrocity prevention lens” applied,7 allow for identification of patterns and critical inflection points that characterize an escalation scenario for atrocity crimes. Applying the Framework in local and national settings reinforces and organizes the steps of careful attention that are required of local actors who seek to participate in these critical assessments of the potentially negative dynamics in their own societies and who aim to stay alert to the slow or rapid sequence of events that can erupt into atrocity crimes.8

The categories of assessment and alert outlined in the Framework can provide crucial guidance to policy-makers on what kinds of information should be gathered and shared. Essential to the process of gathering and sharing information is the timeliness of ascertainment and transmission.

Timeliness has a complex profile in that the reported information must arrive in time for it to be relevant to the choices that might need to be made, but also at the time when the policy-maker is ready, willing, or prepared to receive it. In some instances, this issue of readiness can be directly linked to political will, but in others, it involves problems of coherence and distraction in the bureaucratic chain of command.

Technologies for information aggregation and transmission can sometimes help override the indifference or hostility of the presiding authorities and raise a more public early-warning alert to marked effect. Newspaper stories from war reporters at the Crimean battlefields about the deplorable health conditions of the troops outraged the British public and galvanized strong support for the hospitals and public health measures undertaken by Florence Nightingale. News from war reporters of British atrocities during the Boer War did little to change official policy at the time, but has served to damage the official version of events in the historical record. News reports and especially photographs coming in rapidly from the war in Vietnam mobilized public opinion against the war – to the fury of civilian and military officials. Alert and intrepid correspondents in the 1992–1993 period of the Yugoslav Wars, not versed in atrocity crimes but appalled by the civilian carnage, were the first to raise the concern that what was occurring was not a civil war, per national and international claims, but a brutal ethnic conflict.9

Although we have acquired increased understanding of atrocity crimes in the legal frameworks elaborated after World War II, and especially with the creation of the International Criminal Court in 2002, the new UN Framework has now shaped the questions and delineated the zones of practical relevance to permit real-time identification of these crimes as they unfold. This precision on what to look for accelerates the potential for information communication technologies (ict) to provide more actionable information on local escalation and trigger events to key actors.

Crucial to supporting rapid decision-making is relevant information to enable assessment of rapidly unfolding events at the level of groups or subpopulations. Local assertions that people are fleeing an attack or have decided to move out of the area to seek safety can now be robustly tracked in relatively real time by satellite, cell phone geolocaters, or unmanned aerial vehicles (uavs). Crowdsourcing allows insight into behavior, perceptions, and patterns of movement of large populations. Information on issues that could be uniquely provided by ict capacities relate to gathering and aggregating data on social moods and perceptions (social media analysis, crowdsourcing) and data on shifts in social behavior (markets, religious festivals, crowds, demonstrations) by satellite data and geospatial mapping of large group population movements in urban and rural areas.

Deployment of the full array of ict capacities in disaster settings is now routine in the civilian response community, and military reliance on these capacities is well known.10 The applications of ict potential in alert and monitoring modes for atrocity crimes are now under active study.11

In recent conflicts, a recurrent early warning of trigger events is a reported discovery of new mass graves. Whether identified through satellite (as in Srebrenica in July 1995) or by close observation on land (as with the Physicians for Human Rights in Afghanistan in January 2002), the import of the finding may be enormous, but action depends upon the receptivity of the policy community. In the instance of the Srebrenica report, arriving a few days after the event in the midst of intense international debate about the conflict, the findings contributed to a tipping point in the international decision to move definitively to bring an end to the conflict.12 Yet verified news of mass graves in Afghanistan in the early days of the U.S.-led war fell on unresponsive ears: the United States was preoccupied with military objectives and information implicating U.S. allies in the region was unwelcome; the United Nations was in emergency mode and feared such news would further aggravate ethnic tensions; and the nascent Afghan authorities, composed of leaders of several rival groups, had neither the capacity nor the interest in maintaining site security for future investigation.13

The information on possible trigger events arriving at the desks of policy-makers, who almost by definition will be somewhat distant from the action, must be precise, accurate, valid, and empirically linked to the key questions at hand. Human rights observers in Kosovo from 1998 to 1999 would frequently report to the authorities that smoke was coming from the hills in the early mornings after Serbian night attacks on Kosovar villages. Ascertainment capacities (geospatial coordinates and satellite photos) were nascent and even verified photographs, if they existed, were unavailable. As a result, absent further information about exact location or identifiers of presumed assailants, the monitors in the Organization for Security and Cooperation in Europe could not immediately take action.14

Given the advances in technological capacities as well as in understanding of atrocity crimes, these obstacles have, in many instances, been resolved. Now the challenge is to link these focused and real-time early warnings to targeted early responses.

Two strands of research and policy have come together here: one is linked to the analysis of atrocity crimes,15 and the other is linked to the efforts of the responsibility to protect (r2p).16 Both of these structured analyses have acquired the status of international policy instruments; and the UN offices charged with implementing their underlying strategies (early warning for the Framework and early action for r2p) have now merged in order to strengthen and streamline the application of these robust analytic and policy tools.

Discussed here are how these efforts have largely succeeded in merging two rather different analytic paradigms and response architectures. Also discussed are significant challenges ahead in bringing the broad and somewhat cumbersome menu of r2p interventions into tighter alignment with the kind of targeted measures needed for effective response to trigger warnings of impending atrocity crimes.

The three pillars of r2p17 deliberatively and sequentially lay out a menu of initiatives (development, legal, diplomatic, humanitarian, and coercive nonmilitary) that have been used in many instances, and often to good effect, to ward off or restrain conflict or prevent attacks against civilians.18 Recent examples of effective use of the r2p doctrine can be found on the website of the Global Centre for the Responsibility to Protect,19 and the toolkit for implementing r2p is available on the website of the International Coalition for the Responsibility to Protect.20

With these training manuals and toolkits, the r2p community (state and nongovernmental) has elaborated a taut range of mitigating interventions to respond to the emergence of many different signs of instability and deepening social divides. In parallel, within the community working against atrocity crimes, various stakeholders and civil society groups have studied the evolution of atrocity-laden conflict and have developed a powerful consensus on the key elements and relationships among these elements that could, were they applied to an early warning system, allow for the identification of patterns and critical inflection points that characterize an escalation scenario for atrocity crimes.

In a major shift that recognizes the latent power in the merger of these two communities of analysis and practice, the un Special Advisers on the Prevention of Genocide and on the Responsibility to Protect have supported the establishment of offices of national focal points appointed by each member state and have guided the development of national policies and programs aimed at atrocity prevention and protection of at-risk populations. This national focal point works by interacting with civil society and local governance throughout the land to discern early signs of social and economic antagonisms between groups that might develop into the active commission of atrocity crimes.21

These developments have now allowed those undertaking atrocity prevention efforts at international, national, and local levels to work directly with r2p colleagues to instill the awareness and skills needed to assess the practical dynamics of social and political situations, as well as teach what is known about how to restrain tensions that arise within a particular community or segment of the population.22 Now embedded in the r2p architecture and operations are people who are deeply familiar with what to look for as well as others who are familiar with options for responses at local, national, and international levels.

Trigger events are not familiar terms in r2p language. Preventive action to take in the context of the Framework's risk factor language maps moderately well with the conflict prevention approaches of pillar two. Yet trigger events rise as the pivotal moment for definitive response along any constellation of risk factors in the Framework approach. For most trigger events, military intervention would, in most cases, be inappropriate and damaging; yet the time for most slow-moving prevention measures would have passed.

The early response that a trigger event requires has not been specifically delineated in the new merged documents. This delineation is what is now urgently needed because the r2p/atrocity community might well be positioned to create the documentary pathway that leads to last resort. In the foundational r2p report, when speaking about pillar three, which has at its extreme the invocation of just war, the precautionary measure of last resort is defined as follows:

4.37 Every diplomatic and non-military avenue for the prevention or peaceful resolution of the humanitarian crisis must have been explored. The responsibility to react – with military coercion – can only be justified when the responsibility to prevent has been fully discharged. This does not necessarily mean that every such option must literally have been tried and failed: often there will simply not be the time for that process to work itself out. But it does mean that there must be reasonable grounds for believing that, in all the circumstances, if the measure had been attempted it would not have succeeded.

4.38 If the crisis in question involves a conflict between a state party and an insurgent minority, the parties must be induced to negotiate. Ceasefires, followed, if necessary, by the deployment of international peacekeepers and observers are always a better option, if possible, than coercive military responses. The long-term solution for ethnic minority conflict or secessionist pressures within a state will often be some kind of devolutionist compromise that guarantees the minority its linguistic, political and cultural autonomy, while preserving the integrity of the state in question. Only when good faith attempts to find such compromises, monitored or brokered by the international community, founder on the intransigence of one or both parties, and full-scale violence is in prospect or in occurrence, can a military option by outside powers be considered.23

This quote from 2001 conveys to current readers a major difficulty with the menu of options in the r2p portfolio. In the Framework approach, response to trigger events must be specific and rapid. It is not evident in the current r2p menu, however, that options have yet been shaped to be as focused, nimble, and as easily malleable to specific context conditions as the unfolding situation might require. The r2p response toolkit, although now affirmed in the last six years to apply in a narrow and deep way (to atrocity crimes),24 still straddles uneasily a menu of broad conflict prevention options that, when evidently exhausted, move strategically and briskly into just war activities.

As Edward Luck, a leading authority on r2p, pointed out in 2010, “The hardest questions about the value of the responsibility to protect as a policy tool have revolved around its utility in spurring and shaping an effective response, not in encouraging preventive measures.”25 Looking at the following vignettes from this perspective, what elements are there in the current r2p toolkit to respond to these latent or increasingly obvious situations in which trigger events begin to be discernible against a background of known (but not irredeemably terrible) risk factors? In many instances, atrocity can begin to unfold and very few people outside the local area may take notice. Mounting hate speech against the Roma in Hungary from 2010 occurred against such a heavy background of disdain for this population that it took several years for the international community to raise an alert.26 Mob violence has become so common (as in Kampala now) that the local population and the government have become inured to the phenomenon.27 Ordinary people or officials may not have the knowledge or experience to realize that what they are witnessing may be an atrocity crime and it is in the interest of the authorities not to disclose details.

The sudden explosion in Syria of violent government repression of Sunni demonstrators in March 2011 is recognized as the onset of the massive crisis of war and forced migration that now engulfs vast areas of the Middle East. In hindsight, had a focused response apparatus been linked to the early warning provided by a lens of atrocity crimes, it would have been possible to discern elements in the lead-up to the fierce assault by the Syrian government against its people in 2011 that might have prompted a robust international response aimed at mitigation. The different outcome in part derives from failure to pay close notice to contextual risk factors, intensification of indicators, and a trigger event (the unprecedented 2006–2010 drought in the Northeast) that marked a sudden rupture in brittle communal accommodations relating to location, livelihoods, and voice.28

Interpretation of possible trigger events requires not only excellent quality and quantity of information, but also the judgment of seasoned members of the society in question. For instance, three potential trigger events lie immediately ahead: the death of the aged autocrat, President Mugabe of Zimbabwe, the possible reelection of an increasingly autocratic leader, President Kagame of Rwanda, and the accelerating popular contestation over presidential campaigns in Burundi, which are becoming increasingly communalized.29 Deciding what indicators to watch, how to get reliable and accurate information about them, how to assess their valence in the current situations, and how to choose mitigating interventions constitutes a formidable analytic task. Trusted local interlocutors will be essential but not sufficient. The national focal points in the African Great Lakes region will play important liaison roles, but also must be protected from unwelcome scrutiny.

Assessing trigger factors in the complex and shifting social movement now underway in India could benefit from judicious targeted application of ict capacities. Prime Minister Narendra Modi has encouraged his right-wing Hindu base by not restraining a mounting series of atrocity crimes against Muslims and Christians in various regions of the country. The tensions across ethnic and class lines are mounting and, in protest, many prominent Indian scholars and artists are returning prestigious national prizes and awards they have received over the years from the Indian federal government. Against the distant history of the 1947 partition and in the recent shadow of several large-scale communal massacres, these developments in India are raising alarms within the country as well as internationally.30

Knowing where these instances have devolved into mass atrocities or anticipating through an atrocity lens where they might spiral into grave threats to subpopulations, the question is: Does the substantive and targeted menu of mitigating measures now exist to respond effectively to these trigger events? Or is that the challenge ahead? Without such an operational menu, it will be difficult, first, to impede the escalation that the triggers portend and, second, to attest with confidence that the responsibility to prevent has been fully discharged.

In most situations where just war doctrine has been invoked, the evidence is sparse that effective efforts at prevention and mitigation have been taken. As military analysts and policy-makers have long argued, for measures of mitigation to have much chance of success there must be actionable early warning and tested options for early response. Much of the current academic and policy discussion of early response also rests on the repeatedly observed failures to take up early warning and fashion it in a timely way into early response.31 Consequently, in the absence of real-time and focused warning, and without tested and targeted response options, the path to last resort would seem barren of alternatives and free of friction. Much time would not be needed along its route.

The time to populate that path is now upon us. Much of the hardest theoretical, political, and policy work has already been accomplished. The two conceptual structures and strategies have been carefully delineated. For the prevention of atrocity crimes, the historical record has been analyzed. The major ideas are now captured. The technologies for rapid ascertainment are getting ever more sophisticated. The major players and their roles have been specified. The international community must now assume the responsibility of linking these assets to produce a realistic menu of interventions in the face of risk factors and, most crucially in terms of averting just war, develop and test specific mitigating options in the face of discernible trigger events. Such responsibility is no less and no more than mobilizing the talent, political will, and resources of the r2p/atrocity community to specify and then inject timely and effective responses in advance and in anticipation of the devolution of fraught and fragmenting dynamics in a specific country at a particular time.

Within just war doctrine, evidence is always needed to establish the argument that the element of just cause has been shown to apply. There is an abundant theoretical and empirical literature on how to establish elements of just cause in any given situation. But one must always ask in any instance: How is that evidence amassed? What critical lens has been applied to that evidence? How diverse are the actors who are required to see and react to it? As has been mentioned in the introduction, there are significant empirical challenges to the assertion that elements of just cause can be sustained when embarking upon armed conflict in contemporary situations.

Just war doctrine insists that the determination of last resort be made by a legitimate authority. But on what basis is this determination made? There has been little discussion in the open literature about who constructs and analyzes a given fact picture that provides the assurances that all measures have been taken and the time for taking measures of last resort has been reached. How would the evidentiary brief for that extraordinarily consequential determination be organized and established for the record?

The argument here is that the record of early warnings given and early responses taken within the r2p/atrocity community could provide the grounds for that determination. When the next call for just war is issued, citizens and authorities outside the closed chain of national command can ask to see the record. What have been the early warnings? What have been the early responses? How have they been connected? Has the case been made that all reasonable efforts have been exhausted short of initiating war? How will we know that the time and space for last resort has arrived?

With these developments in early warning and early response, the burden of proof has intensified for proponents of just war theory. But the burden also implicates the entire international community. No longer can last resort be viewed simply as the official and closed assertion of an end state. It must be seen now as an extended and time-tethered process of early warning and early action (conducted in open as well as backdoor channels) that has reached its limits. All reasonable measures have been taken, targeted mitigation has been introduced to respond to trigger events, and some reasonable measures at the late stages of a disintegrating process have been deemed impossible to initiate or unable to accomplish their goals (with explanations provided). At these limits, military intervention must be presented as the result of the incapacity of the international community to restrain mass violence. Outside the context of the just war doctrine, last resort must be seen not as a declaration of war, but as a declaration of our collective failure.


United Nations Office of the Special Advisers for the Prevention of Genocide and the Responsibility to Protect, Framework of Analysis for Atrocity Crimes: A Tool for Prevention (United Nations: New York, 2014),


Ibid., 1, 25–32. The governing definitions of atrocity crimes are codified in International Criminal Court, Rome Statute of the International Criminal Court (Enschede, The Netherlands: International Criminal Court, 2011). The UN Framework of Analysis and all expert interlocutors using the Framework rely on these same definitions. See United Nations, Framework, 18–21.


United Nations, Framework, 9–17.


Early warning is “the collection, analysis and communication of information about escalatory developments in situations that could potentially lead to genocide, crimes against humanity or massive and serious war crimes, far enough in advance for relevant UN organs to take timely and effective preventive measures.” See Lawrence Woocher, Developing a Strategy, Methods, and Tools for Genocide Early Warning (Prepared for the Office of the Special Adviser to the UN Secretary-General on the Prevention of Genocide, 2006), 7,,%202006-11-10.pdf.


United Nations, Framework, 6–7.


Ibid., 17–18.


Alex J. Bellamy, “Mass Atrocities and Armed Conflict: Links, Distinctions, and Implications for the Responsibility to Prevent” (Muscatine, Iowa: Stanley Foundation, February 2011),


Jennifer Leaning, “Early Warning for Mass Atrocities: Tracking Escalation Parameters at the Population Level,” in Reconstructing Atrocity Prevention, ed. Sheri P. Rosenberg, Tibi Galis, and Alex Zucker (Cambridge: Cambridge University Press, 2016), 352–378. Summary of Framework is an expanded version of the explanation used in an expert report filed in Sexual Minorities Uganda v. Scott Lively. See Jennifer Leaning, “Expert Report of Dr. Jennifer Leaning,” Sexual Minorities Uganda v. Scott Lively, Civil Action 3:12-CV-30051 (New York: Center for Constitutional Rights, 2016),


Roy Gutman and David Rieff, eds., Crimes of War: What the Public Should Know (New York: W.W. Norton, 1999), 8–12.


Sam Brophy-Williams, Nic Segaren, and Jennifer Leaning, “Digital Technology in Humanitarian Response: Harnessing Information to Support Best Outcomes,” in World Disasters Report 2013: Focus on Technology and the Future of Humanitarian Action, ed. Patrick Vinck (Geneva: International Federation of Red Cross and Red Crescent Societies, 2013), 163–193.


Jennifer Leaning, “The Use of Patterns in Crisis Mapping to Combat Mass Atrocity Crimes,” in Mass Atrocity Crimes: Preventing Future Outrages, ed. Robert I. Rotberg (Washington, D.C.: Brookings Institution Press, 2010), 192–219. See also Lloyd Axworthy and A. Walter Dorn, “New Technology for Peace & Protection: Expanding the r2p Toolbox,” Daedalus 145 (4) (Fall 2016).


Stephen Engelberg and Tim Weiner et al., “Massacre in Bosnia, Srebrenica: The Days of Slaughter,” The New York Times, October 29, 1995,; and Samantha Power, A Problem from Hell: America and the Age of Genocide (New York: Basic Books, 2002), 393–441.


John Heffernan and Jennifer Leaning, “Preliminary Assessment of Alleged Mass Gravesites in the Area of Mazar-I-Sharif, Afghanistan: January 16–21 and February 7–14, 2002,” internal memo, Physicians for Human Rights, 2002; John Barry, “The Death Convoy of Afghanistan,” Newsweek, August 25, 2002,; and Leonard S. Rubenstein, “Request Submitted Under the Freedom of Information Act,” letter submitted to foia officer on behalf of Physicians for Human Rights, July 21, 2006, These relate to a mass grave at Dasht-e-Leili in Afghanistan, allegedly the burial site for hundreds of surrendered Taliban fighters. See Stefan Schmitt, Afghanistan Forensic Support Project October 2007 (Geneva: Physicians for Human Rights, 2007),


Organization for Security and Co-operation in Europe Office for Democratic Institutions and Human Rights, ed., “Part III: The Violation of Human Rights,” in Kosovo/Kosova As Seen, As Told (Vienna: Organization for Security and Co-operation in Europe, 1999); and Human Rights Watch, Under Orders: War Crimes in Kosovo (New York: Human Rights Watch, 2001).


The literature on genocide over the past twenty-five years has called upon the development of a capacity for early warning of indicators of genocide, with the aim of mobilizing the international community to intervene early in order to prevent the genocidal process from unfolding. See Roger W. Smith, ed., Genocide: Essays Toward Understanding, Early Warning, and Prevention (Williamsburg, Va.: International Association of Genocide Scholars, 1999); Carol Rittner, John Roth, and James M. Smith, eds., Will Genocide Ever End? (St. Paul, Minn.: Paragon House, 2002); and Rosenberg et al., eds., Reconstructing Atrocity Prevention.


The concept of the responsibility to protect grows out of the debate in the 1990s, in the wake of the wars in Rwanda and the former Yugoslavia, where in each instance the unfolding events provided identifiable points in the conflict escalation where international intervention against a sovereign state might well have protected the attacked civilians from outrages and death. The three pillars of the responsibility to protect, which detail prevention and mitigation actions that might be taken along a continuum of intensifying crises, provide empirical rationale for assertive protection measures that might be taken in particular circumstances. See International Commission of Intervention and State Sovereignty, The Responsibility to Protect: The Report of the International Commission on Intervention and State Sovereignty (Ottawa, Ontario: International Commission on Intervention and State Sovereignty, 2001).


For a full definition of the three pillars of r2p, see Global Centre for the Responsibility to Protect, “About r2p,”


Ibid., 47–126.


Global Centre for the Responsibility to Protect, “Ten Years of r2p,”


International Coalition for the Responsibility to Protect, A Toolkit on the Responsibility to Protect (New York: International Coalition for the Responsibility to Protect, 2013),


Global Action Against Mass Atrocity Crimes,


See the websites of the UN Offices of the Special Advisers for the Prevention of Genocide and the Responsibility to Protect; the Global Centre for the Responsibility to Protect; the International Coalition for the Responsibility to Protect; the Global Action Against Mass Atrocity Crimes (gaamac); the Genocide Prevention Advisory Network (GPAnet); the State Failure Project; the Stanley Foundation; the International Association of Genocide Scholars; and the Auschwitz Institute for Peace and Reconciliation.


The International Commission of Intervention and State Sovereignty, The Responsibility to Protect, 36–37.


Jennifer Welsh, “The ‘Narrow but Deep’ Approach to Implementing the Responsibility to Protect: Reassessing the Focus on International Crimes,” in Reconstructing Atrocity Prevention, ed. Sheri P. Rosenberg, Tibi Galis, and Alex Zucker (Cambridge: Cambridge University Press, 2016), 81–94.


Edward C. Luck, “The Responsibility to Protect: Growing Pains or Early Promise?” Ethics & International Affairs 24 (4) (Winter 2010),


Amnesty International, Violent Attacks against Roma in Hungary: Time to Investigate Racial Motivation (London: Amnesty International, 2012),


See World Justice Project, World Justice Project Rule of Law Index 2015 (Washington, D.C.: World Justice Project, 2015),; and “Angry Mob Slaughters Suspected Kampala Gangster,” Chimpreports, April 22, 2013,


Shahrzad Mohtadi, “Climate Change and the Syrian Uprising,” Bulletin of the Atomic Scientists, August 16, 2012,


“Crisis-Torn Burundi's Leaders Should Pursue Dialogue, Listen to Concerns of Their People,” UN News Centre, January 31, 2016,


Amartya Sen, “India: The Stormy Revival of an International University,” The New York Review of Books, August 13, 2015,; and Nirmala George, “Indian Intellectuals Alarmed by Attacks on ‘Science and Reason,’ and Modi's Silence,” U.S. News & World Report, October 31, 2015,


Micah Zenko and Rebecca R. Friedman, “UN Early Warning for Preventing Conflict,” International Peacekeeping 18 (2011): 21–37.

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license, which permits copying and redistributing the material in any medium or format for noncommercial purposes only. For a full description of the license, please visit