Building Knowledge around Genocide Prevention

It is an opportune time for genocide prevention. In 1948, in the aftermath of the Holocaust, the newly formed United Nations declared genocide a crime under international law, which they undertook ‘to prevent and to punish’.Footnote ¹ The United Nations Convention on the Prevention and Punishment of the Crime of Genocide was adopted by the General Assembly on 9 December 1948, coming into effect in 1951. Under the Convention:

For decades, however, the idea of genocide prevention attracted little attention or resources. It was only in 2004, in the aftermath of the UN failures in Rwanda and Srebrenica, that the UN launched an Action Plan to Prevent Genocide.Footnote ³ The position of the Special Advisor on the Prevention of Genocide was created. In the subsequent two decades, there has been a vast increase in the resources dedicated to genocide prevention globally. These include the Office on Genocide Prevention and the Responsibility to Protect at the United Nations, and dedicated capacity within regional organisations. Countries such as the US, UK and others have established formal strategies and groups focused on genocide and atrocity prevention, and created institutional capacity dedicated to prevention within their administrative bureaucracies. At the same time, civil society organisations dedicated to genocide and mass atrocity prevention have proliferated.

These increased resources have led to a new practitioner field of genocide prevention. Intergovernmental organisations now incorporate explicit focus on prevention. The UN Office on Genocide Prevention and the Responsibility to Protect, for example, works actively to reduce the risk of genocide and mass atrocities. Amongst their programs, addressing hate speech figures prominently. One program focuses specifically on religious leaders, working with them ‘to develop context specific strategies to prevent incitement that could lead to atrocity crimes’.Footnote ⁴ Governments in many countries also allocate resources for prevention. In the US, for example, the Elie Wiesel Genocide and Atrocities Prevention Act of 2018 mandates a whole-of-government approach to atrocity prevention and annual reports to Congress on government-led atrocity prevention efforts. The Biden Administration implemented the ‘2022 United States Strategy to Anticipate, Prevent, and Respond to Atrocities’.Footnote ⁵ While much of the infrastructure and expertise for atrocity prevention within the State Department has been disbanded in the subsequent Trump Administration, the underlying legislation remains in place. A multitude of civil society organisations (CSOs) have adopted a diverse range of approaches to genocide prevention. The Auschwitz Institute for the Prevention of Genocide and Mass Atrocities provides technical assistance and specialised training to policymakers and leaders, aiding them to develop and implement strategies to prevent identity-based violence.Footnote ⁶ Aegis Trust undertakes peace education, such as its work in Rwanda, for example, along with international advocacy on behalf of vulnerable groups.Footnote ⁷ The Sentry works to disable the financial networks that ‘undermine peace and good governance’ and fund the commission of genocide and mass atrocities.Footnote ⁸ Each of these types of practitioners must grapple with where to direct their limited resources, how to address differing levels of risk in different situations and how to navigate complex and dynamic conditions in attempting to stabilise and reduce the risk of genocide.

Practitioners rely on the research in the fields of genocide studies and genocide prevention to inform their approaches. Through such research, scholars have made great progress in understanding the antecedents and causes of genocide.Footnote ⁹ There is broad agreement around several key preconditions, including:

• The existence of an ‘outgroup’. One or more groups experience discrimination and disadvantage within a plural, divided society.Footnote ¹⁰

• The society experiences internal stressors, which may include war, political crisis, economic crisis or other major challenges.Footnote ¹¹

• A genocidal ideology develops or emerges from the fringes to become mainstream.

• Propaganda spreads, attempting to justify the ideology.Footnote ¹²

• Perpetrators have sufficient power and capacity to plan and organise the genocide.Footnote ¹³

Preconditions for genocide can also be grouped into broad categories, such as James Waller’s mass atrocity risk factors. These include governance factors, such as regime type and state-led discrimination; conflict history, including past intergroup tensions and genocides; economic conditions, including economic weakness and deterioration and social fragmentation, including identity-based social divisions and political instability.Footnote ¹⁴ Numerous scholars have also examined the role of additional factors, such as the level of centralisation of power, modernity and nationalism.Footnote ¹⁵

This research has led to the development of several models of the causes of genocide, or antecedents to them.Footnote ¹⁶ A prominent example is Gregory Stanton’s ‘Ten Stages of Genocide’, which includes stages such as Classification, Discrimination, Dehumanisation and Polarisation.Footnote ¹⁷ According to Stanton, genocide ‘develops in ten stages that are predictable but not inexorable. At each stage, preventive measures can stop it’.Footnote ¹⁸ Reducing the risk of genocide, according to this paradigm, becomes a matter of addressing its causes. This ‘root cause approach’ has been adopted by practitioners within the field.Footnote ¹⁹ For example, the Framework of Analysis of Atrocity Crimes: A Tool for Prevention, published by the UN Office on Genocide Prevention and the Responsibility to Protect, states: ‘If we understand the root causes and precursors of these crimes, and can identify risk factors that can lead to or enable their commission, it follows that we can also identify measures that can be taken by States and the international community to prevent these crimes’.Footnote ²⁰

Risk Assessment Models

Knowledge of risk factors for genocide has also led to the development of risk assessment models. Models use a constellation of risk factors and antecedents of genocide to attempt to rank and/or categorise risk of genocide onset (or a related measure) in particular countries in a given time period.Footnote ²¹ For example, the Early Warning Project, part of the United States Holocaust Memorial Museum, issues an annual ‘Statistical Risk Assessment for Mass Killing’.Footnote ²² In the 2024–2025 risk assessment, the top five ‘at-risk’ countries were identified as Chad, Sudan, Myanmar, Pakistan and India.Footnote ²³ The Atrocity Forecasting Project focuses on analysing the risk of genocide and politicide.Footnote ²⁴ In its forecast for 2024–2026, the top five countries at risk were identified as Afghanistan, Pakistan, Somalia, Tajikistan and Nigeria.Footnote ²⁵ Amongst the models, a quantitative, ‘big data’ approach to forecasting dominates, used by the Early Warning Project, Atrocity Forecasting Project and previously by Peoples under Threat, for example.Footnote ²⁶ Practitioners and policymakers may utilise these risk assessments to determine where to focus their attention and resources in efforts towards genocide prevention.

In each of these areas, however, there are conceptual limitations and methodological shortcomings that have had significant consequences as to the way in which the risk of genocide and addressing that risk have been understood. Our knowledge of the causes of genocide, for example, has largely been developed through one dominant methodological approach.Footnote ²⁷ Scholars have typically selected a small number of case studies of prominent genocides, such as the Armenian genocide or the Holocaust, and then carefully examined their antecedents. Risk factors common across multiple case studies have been identified as ‘necessary and sufficient’ preconditions.Footnote ²⁸ Yet while this approach has offered valuable insights into understanding the risk of genocide, little attention has been paid to its inherent limitations. Limiting case study selection to cases in which genocide has occurred has biased models of genocide towards assuming that risk progresses in a linear or cumulative manner.Footnote ²⁹ Moreover, it has excluded cases in which some, or even many, of these risk factors have been present, but in which genocide has not occurred. This has profound implications for the field of genocide prevention. The case study selection process has limited case studies to those in which factors promoting resilience have been absent or have wholly failed. These have thus been rendered insignificant by this methodological approach and have attracted little scholarly attention.Footnote ³⁰ A more inclusive methodological approach has the potential to lead to very different research findings. Genocide is relatively uncommon, and many countries that exhibit some risk factors for genocide are nonetheless reasonably stable over long periods of time. This suggests that factors promoting resilience to genocide may play a crucial role in stabilising and mitigating risk. Building our knowledge of factors promoting resilience may thus be critical to an evidence-based approach to genocide prevention. Doing so, however, requires examining a wider range of at-risk societies to understand the interplay of risk and resilience in determining a country’s vulnerability to genocide.

Greater knowledge of the interplay of risk factors and factors promoting resilience can also address current limitations of risk assessment models for genocide and mass atrocities. Currently these models, while broadly useful, have very substantial shortcomings. There are stark differences between the Early Warning Project and Atrocity Forecasting Project’s list of the five countries at highest risk in 2024–2025/26 (outlined previously), for example, with only one country appearing in both lists. This highlights current limitations with respect to their accuracy. The lists typically have very high false positive rates, overpredicting risk of genocide onset. The vast majority of countries identified as at grave risk will not go on to experience genocide within the forecasting period.Footnote ³¹ While this can be indicative of future risk, this level of overprediction is a significant challenge to practitioners who must prioritise their limited resources. This also relates to the imprecision of current risk assessment models.Footnote ³² Rankings appear to only broadly approximate risk level.Footnote ³³ Evidence as to the accuracy of these models does not clearly demonstrate that a country ranked first in terms of risk is more likely to experience genocide or mass killing onset than a country ranked fifth or ninth, for example.Footnote ³⁴ An even deeper challenge for practitioners relying on risk assessment models is that of underprediction, or false negatives. Risk assessment analyses have repeatedly failed to forecast risk of major cases of genocide and mass atrocities, including those in China, Libya, Central African Republic and Ethiopia. Such high false negative rates undermine their efficacy as a predictive tool, and their associated usefulness to practitioners and policymakers in the field of genocide prevention.Footnote ³⁵ Potentially, greater knowledge of the role of factors promoting resilience, and their impact on a country’s vulnerability, can contribute to improving the accuracy of risk assessment models. Currently, however, there is insufficient research on mitigating factors to adequately incorporate them into these models.

Genocide Prevention as a Process

In order to deepen our understanding of the processes that lead to genocide, and how to disrupt them, a new approach is needed. Fundamental to this approach is recognition that preventing genocide is a process, not an event. When US policymakers were responding to the genocide in Rwanda, prevention (or response) was falsely characterised as a choice between ‘doing nothing and sending in the marines’.Footnote ³⁶ Fortunately, in recent years, a much greater understanding has developed of genocide prevention as a process. This recognises that there is scope for preventive measures at multiple points along the spectrum of risk, and that many of the best opportunities for prevention exist well before that risk is imminent. As James Waller has remarked, ‘Upstream prevention is the “before” analysis of the longer-term governance, historical, economic, and societal factors that leave a country at risk for genocide and other mass atrocities and the inoculation avenues open to mitigating those risk factors’.Footnote ³⁷ Similarly, the Auschwitz Institute for the Prevention of Genocide and Mass Atrocities has recognised ‘Early preventive action requires first identifying and then reducing the risk factors that make this type of mass identity-based violence more likely over the long-term in a given country’.Footnote ³⁸ Importantly, this conceptualisation recognises that genocide prevention is often a long-term process. Evidence indicates that risk of genocide typically decreases relatively slowly, and thus a long-term commitment is required to ensure sustained and meaningful reductions in risk over time.Footnote ³⁹ It also recognises the importance of early intervention. Conceptualising prevention as a potentially long-term process therefore maximises the range of potential preventive actions and the possibilities for their success.

Structural and Operational Prevention

Current scholarship offers valuable insights into how to manage and mitigate risk of genocide. Typically, analyses consider prevention from two perspectives: that of long-term, structural prevention and that of more imminent crisis prevention and response.Footnote ⁴⁰ Structural prevention has focused on addressing the underlying causes of violence, or ‘root causes’. A key approach to structural prevention, for example, has been the promotion of good governance.Footnote ⁴¹ This includes measures such as strengthening the rule of law, security sector reform, addressing corruption, promoting democracy and building political stability.Footnote ⁴² Addressing poverty and economic instability, and promoting trade and economic growth form an important component of structural prevention.Footnote ⁴³ Measures promoting social cohesion, including education for genocide prevention, also feature prominently.Footnote ⁴⁴ Structural prevention may include disarmament, demobilisation and reintegration programs to address the risks posed by the ongoing presence of militias or insurgents.Footnote ⁴⁵ It might also include mechanisms to hold perpetrators to account for past atrocities, to promote deterrence or as part of a wider undertaking to promote reconciliation in a post-conflict society.Footnote ⁴⁶ As Stephen McLoughlin has noted, however, there are limitations in the way in which structural prevention has been conceptualised and operationalised. The ‘root cause’ or ‘upstream prevention’ approach implicitly assumes that there will be an inevitable progression towards violence, as ‘upstream’ inevitably flows downstream. Secondly, there has been a tendency ‘for international actors to decide, in general and global terms, what counts as root causes and how to ameliorate them, downplaying the role of local contexts and actors’.Footnote ⁴⁷

The second major approach to prevention is often termed ‘operational prevention’. This kind of prevention work is undertaken when there is a serious, escalating and/or imminent risk of genocide. It may be a response to an outbreak of violence, a major crisis or concerns that the early stages of a genocide may be underway. Diplomatic measures often feature prominently at this stage, with regional and international stakeholders working to stabilise the situation.Footnote ⁴⁸ Diplomatic intercessions often proceed within an environment of confidentiality, but at times diplomacy can also be a very public affair. Indeed, bringing international publicity to the situation can be utilised as a tactic to pressure belligerents to participate in talks or agree to particular outcomes. Facilitated negotiations between belligerents are often an important part of diplomatic measures, and may include direct and indirect negotiations. The use of economic measures is also a core component of operational prevention.Footnote ⁴⁹ More temperate approaches may involve economic inducements, such as the promise of economic benefits in exchange for desired outcomes, or alternatively invoke a threat to ongoing economic cooperation if change is not enacted. In other cases, more robust economic measures may be employed, including targeted economic sanctions, trade embargoes and financial restrictions. Arms embargoes may seek to partly or completely restrict parties to the conflict from having access to arms. Finally, peacekeeping and peace enforcement missions may be deployed in response to the imminent threat of genocide.Footnote ⁵⁰ One of the major challenges of operational prevention, however, is that as risk escalates and the likelihood of genocide increases, the range of measures available narrows, and the prospects of their success diminish. Political leaders may have embraced extremist ideology, for example, becoming resistant to negotiation and compromise. The prospects of intervention, moreover, often depend on external factors such as the political will of the permanent members of the United Nations Security Council and other UN member states.

A third approach within the scholarship has been that of ‘lessons learned’. Scholars and practitioners have analysed the successes and failures associated with managing particular crises, and reflected on the insights they contribute to addressing future challenges.Footnote ⁵¹ Two particularly prominent examples include the inquiries into the UN failure to prevent genocide in Rwanda and Srebrenica, each of which issued a raft of recommendations.Footnote ⁵² There have been a multitude of ‘lessons learned’ reflections on the genocide in Rwanda.Footnote ⁵³ Some, such as that written by UNAMIR peacekeeper Brent Beardsley, bring a deeply personal perspective.Footnote ⁵⁴ Others have focused variously on particular lessons that might be learned, such as the need for early warning, or the need for critical assessments and reform of UN institutions and human rights mechanisms.Footnote ⁵⁵ There have also been many ‘lessons learned’ analyses of the Holocaust.Footnote ⁵⁶ Harff’s prominent analysis, for example, utilised the Holocaust as a case study to examine antecedents of genocide and use that knowledge to develop a risk assessment model.Footnote ⁵⁷ ‘Lessons learned’ scholarship concerning Darfur has focused largely on the opportunity for better international responses to genocide, through humanitarian intervention and action.Footnote ⁵⁸ Some ‘lessons learned’ scholarship has also sought to apply lessons from a past genocide directly to a conflict underway.Footnote ⁵⁹ Collectively, this scholarship has made a substantial contribution to identifying key opportunities for advancing genocide prevention. This includes identifying major areas of failure around early warning, political will and humanitarian intervention. Nonetheless, the primary methodological approach, of extrapolating lessons learned from a single case study of genocide, has inherent limitations. Additionally, both scholars and practitioners have identified that ‘lessons learned’ often fail to be heeded, and there can be substantial barriers to effectively implementing change based on ‘lessons learned’.Footnote ⁶⁰

Evidence-Based Approaches

Despite this valuable scholarship in genocide prevention, much more remains to be done. Essential to this endeavour is a focus on evidence-based approaches. Policymakers and practitioners, in order to maximise their impact and effectiveness, should be able to draw upon a body of scholarship that provides clear evidence as to best practice. It should also be of sufficient depth to offer insights across a diverse range of at-risk countries, and insights useful at different points along the spectrum of risk. Currently, that evidence base is very limited. Only a select few measures have a strong evidence-base to justify their implementation. For example, there is a substantial body of research identifying the protective effects of democracy on reducing the risks of genocide (and other mass atrocities). There is also evidence that peacekeeping missions can be effective in reducing the risk of further warfare, with the latter significantly increasing the risk of genocide and other forms of large-scale violence.Footnote ⁶¹ Other measures commonly used have much weaker evidence in support of their effectiveness. There is little evidence to support the effectiveness of economic sanctions in contributing to genocide prevention, for example, and some evidence of their potential capacity to cause harm in certain contexts.Footnote ⁶² Overall, however, there is very limited evidence to support many measures for genocide prevention, with some entirely lacking an evidentiary basis.

There is a slightly larger, albeit still very small, scholarship exploring the evidence base in the broader field of atrocity prevention. The United States Holocaust Memorial Museum’s ‘Tools for Atrocity Prevention’ project, for example, has conducted a review into research on the effectiveness of twelve commonly used tools for atrocity prevention.Footnote ⁶³ Of these, it found the research on most revealed mixed results. Four measures appear to be effective when used in particular circumstances, or in concert with other measures. These include arms embargoes, mediation, peace operations and prosecutions.Footnote ⁶⁴ Other research, however, calls into question the efficacy of arms embargoes.Footnote ⁶⁵ A World Peace Foundation study assessing the effectiveness of atrocity prevention tools also found similarly context-dependent results.Footnote ⁶⁶ It nonetheless highlighted the potential effectiveness of diplomatic tools, particularly for prevention.Footnote ⁶⁷ In 2024, a Special Issue of Genocide Studies and Prevention focused on ‘Evidence-Based Approaches to Preventing Genocide and Mass Atrocities’.Footnote ⁶⁸ Several articles focused on particular measures, such as the use of memory spaces to contribute to prevention, improving mass atrocity forecasting and dangerous speech as an early warning indicator.Footnote ⁶⁹ A Special Issue in the Journal of International Peacekeeping also explored several case studies to reach some broad conclusions around atrocity prevention.Footnote ⁷⁰ Additionally, some focused scholarship on particular measures, such as preventive diplomacy, contains findings likely to be applicable to preventing genocide and mass atrocities.Footnote ⁷¹ Nonetheless, in 2024, it remained true that ‘In short, the literature offers no clear answers about “what works” to prevent mass atrocities in general.’Footnote ⁷² Current efforts at genocide prevention are therefore, at best, evidence-informed.

Without a strong evidence base upon which to draw, practitioners are at risk of implementing projects that are not effective at reducing risk of genocide in at-risk societies, or potentially even increase risk. Indeed, in some cases this has already happened. Enough Project’s Satellite Sentinel Project, for example, sought to elicit an international response to imminent risk of genocide in Sudan through the provision of near real-time information to policymakers. Despite the substantial resources that went into the project, however, this outcome did not eventuate, and it had little, if any, impact on genocide prevention.Footnote ⁷³ A similar project conducted by Amnesty International highlights even more starkly the risks of interventions that lack an evidence base. In June 2007, Amnesty International launched its ‘Eyes on Darfur’ project to great fanfare.Footnote ⁷⁴ Using cutting edge satellite imagery technology, Amnesty announced, it would be monitoring a selection of villages in Darfur, Sudan, in an attempt to deter the Janjaweed (Sudanese Arab militia) from targeting these villages as part of its ongoing genocide in the region.Footnote ⁷⁵ The following year, after receiving anecdotal evidence of increased violence in the monitored areas, the project was quietly dumped. It was not until an independent researcher investigated several years later, however, that it became apparent that the project had actually led to the villages under surveillance – and their surrounds – becoming targets. ‘Eyes on Darfur’ was not just ineffective but, according to Grant Gordon, actively harmed those it was trying to protect.Footnote ⁷⁶ This highlights the importance of evidence-based approaches to genocide prevention. It is only through utilising evidence-based approaches that policymakers and practitioners can ensure they abide by the ethical principle to ‘do no harm’, and focus their resources in the best possible ways to contribute to reducing the risk of genocide.

Focus on Factors Promoting Resilience to Genocide – They are Essential to Prevention

The first key finding of the book is that factors that promote resilience to genocide matter. In fact, they can be critical to preventing it. In every single case study, a multitude of factors promoting resilience to genocide influenced the country’s vulnerability to genocide onset and the trajectory of that vulnerability over time. In some cases, factors promoting resilience proved insurmountable barriers to risk escalation. In other cases these factors contributed to the stabilisation of risk, and a reduction in vulnerability over time. Of particular importance is that in many cases studies, it was not the reduction of risk, or the diminution or disappearance of a particular risk factor, that drove change at key inflection points, but rather the impact of factors promoting resilience. These factors reduced the influence of risk factors, obstructed attempts at escalating violence, rendered untenable genocide onset, and effectively counterbalanced risk. Importantly, many factors promoting resilience have a swift impact. In cases of potentially imminent genocide onset, they may have the capacity to change the dynamics much more rapidly than attempts focused on addressing risk factors. This highlights the utility and effectiveness of incorporating factors that promote resilience to genocide equally alongside attempts to address risk as part of genocide prevention.

Two case studies highlight the importance of factors promoting resilience particularly well. As I explore in Chapter 3, in March 1943 in Bulgaria, the Jewish population came so close to being deported to the death camps that the cattle cars were literally waiting at the station. What halted the planned deportations was not a sudden reduction in risk, but a combination of several factors promoting resilience. These factors led to repeated delays in the deportation plans until eventually further deportation attempts became too politically risky. Had these factors not been present, almost certainly the deportation and extermination of the Jewish population would have proceeded – as it did elsewhere in Nazi Europe. In Chapter 7, focused on the Dominican Republic, the volume also identifies the importance of factors promoting resilience in managing long-term vulnerability. Across a time period of decades, factors promoting resilience effectively prevented increased persecution of Haitians, and Dominicans with Haitian ancestry, on a number of occasions. Moreover, they not only contributed to stabilisation, but over the longer term led to a substantial decrease in the vulnerability of Haitians and Haitian–Dominicans in the Dominican Republic. While a number of risk factors remain present, the risk associated with these factors appears to be largely offset by the presence of counterbalancing factors promoting resilience. That is, despite entrenched discrimination against Haitian and Haitian-Dominican minorities there, risk of genocide in the Dominican Republic is now considered low. This highlights the power of factors promoting resilience in reducing risk of genocide.

Genocide Prevention Requires Redundancy Not Efficiency

A second key finding is that preventing genocide requires a multitude of concerted actions to address vulnerability in any given at-risk country. In order to be effective, genocide prevention efforts must standardly adopt a systems approach – that is, a multifaceted, comprehensive and sustained approach. Focusing on one or two factors – whether to reduce risk or promote resilience – is unlikely to make a critical difference. Rather, genocide prevention efforts are most likely to be successful when there are multiple factors promoting resilience working in tandem to reduce vulnerability, and in conjunction with efforts to reduce the prevalence of risk. Any one factor may create a temporary barrier to risk escalation, but those with vested interests in such escalation will be actively working to undermine such efforts, and will likely have the capacity to do so. Very often some, if not most, factors promoting resilience to genocide fail in the presence of determined perpetrators. But in doing so, they create delays and difficulties that impede perpetrators and impact the power imbalance. The presence of a multiplicity of factors thus creates manifold barriers that can ultimately prove insurmountable as wider conditions change. A multitude of factors promoting resilience can also work synergistically to meaningfully change the trajectory of an at-risk state over the longer term, as they become the dominant influence on that trajectory. The power of factors that promote resilience thus lies at least in part in their multiplicity. Ensuring as many barriers as possible, through an approach that endorses redundancy over efficiency, can make a crucial difference.

Several case studies highlight the importance of redundancy, rather than efficiency, in preventing genocide. In Chapter 2, we see the clear impact of multiple factors in promoting resilience to genocide in Nazi-occupied Denmark. Factors promoting resilience contributed to the Danish population’s willingness to aid the Jews and the Danish government’s insistence they not be deported. They meant attempts to persecute and deport the Jews were delayed for a significant period. When cooperation between the Nazis and the Danish government broke down, factors promoting resilience also proved crucial to the success of the vast majority of Danish Jews in evading and escaping Nazi roundups. Any one factor would not have proved sufficient, but collectively the multitude of factors promoting resilience proved sufficient to prevent an imminent genocide. In Bulgaria, too, as I explore in Chapter 3, a multiplicity of factors combined to delay the deportation process repeatedly, until the changing tide of the war reduced risk significantly. In cases of much more moderate risk, such as that of the Dominican Republic, evidence also indicates the importance of several factors promoting resilience to stabilising and reducing risk over the longer term. Some, such as the country’s democratisation and the strengthening of its democratic institutions, have and continue to exert a protective influence over time. Other factors have had a decisive impact at specific inflection points. Collectively, they have meaningfully and substantially reduced risk over a period of decades. Adopting a multifaceted approach, with a focus on redundancy rather than efficiency, is thus an effective approach to genocide prevention at multiple points along the spectrum of vulnerability. These case studies provide powerful insight into the need for sustained and comprehensive action to build resilience in at-risk societies.

It is also worth reflecting that an approach advocating redundancy rather than efficiency in efforts to prevent genocide is markedly different from the ‘toolbox’ framing that has proliferated in discourse around the responsibility to protect. The ‘toolbox’ approach implies prevention is about selecting the right tool/s and implementing them at the right juncture. First used as a framework within the International Commission on Intervention and State Sovereignty report on the responsibility to protect, this analytical heuristic prioritises discrimination in the selection of tool/s over deployment of a multitude of tools. A relatively small range of tools are perceived as in the ‘toolbox’ for both structural and operational prevention, including political/diplomatic, economic, legal and military instruments.Footnote ⁸⁷ A situation might escalate if the ‘wrong’ tool is employed.Footnote ⁸⁸ This analytical approach has subsequently become popular amongst academics, policymakers and practitioners.Footnote ⁸⁹ As highlighted earlier, however, evidence for the efficacy of tools implemented in isolation is equivocal at best. By contrast, the case studies examined herein provide strong evidence of the value of a multifaceted approach, which incorporates a much wider range of factors that promote resilience to genocide, and which builds multiple barriers to escalation. Redundancy, rather than efficiency, has proven effective in preventing and reducing the risk of genocide.

Early and Robust Opposition to Persecution is Vital to Promote Resilience

Evidence clearly indicates that early and robust opposition to persecution is critical to promoting resilience. That is, the process of strong, clear and public condemnation itself directly impacts a country’s vulnerability to genocide, and contributes to determining subsequent events. Crucial to the effectiveness of this measure is that such condemnation must come early, and preferably from the earliest opportunity, as this is the inflection point where this process has the strongest possibility of influencing the outcome. Such condemnation must also be public, unequivocal and sustained. A very powerful source of robust opposition is internal. When public figures employ this tactic, whether they be political figures (from the ruling party or opposition), religious leaders, community leaders or others, they are sending an important message not only to the perpetrators, but to the entire community. This has multiple effects. It puts perpetrators on notice that the cost of their actions may be higher than they anticipated. It also alerts the wider community to strong opposition to this course of action, providing clear cues that supporting the perpetrators may be inadvisable. It may also embolden those opposed to the persecution to seek out like-minded individuals or take further action to strengthen opposition. External opposition can be similarly powerful. Early and strong international condemnation can change perpetrators’ perceptions around the opportunity costs of escalating persecution, at a point before decisions have been made on further escalation. Currently, many governments and international organisations refrain from early, public and robust condemnation of escalating persecution in at-risk states. Indeed, often limited outbreaks of violence are largely ignored. Yet the evidence presented herein indicates the power of such early and robust opposition for promoting resilience to genocide.

The evidence demonstrating how strong, early opposition to persecution reduces the likelihood of that persecution escalating to genocide is particularly pronounced in the case of Denmark during the Holocaust. As I explore in Chapter 2, from the outset government, religious and community leaders made their opposition to the Nazi persecution of the Jews clear. This directly impacted Nazi policies in Denmark. Even when the Nazis were at the height of their powers, and were occupying Denmark, they were responsive to this opposition. This was instrumental to the survival of the vast majority of Danish Jews during the Holocaust. In Bulgaria, pronounced public opposition to the persecution and planned deportation of Bulgarian Jewry also proved critical to their survival, as examined in Chapter 3. What we see in each of these cases is that early and sustained opposition influenced Nazi policy as it was being formed, and did so despite the superior power of the Nazis in each case. This testifies to the power of this factor promoting resilience. In Chapter 7, the impact of international condemnation of persecution is also clear. At several points in the Dominican case study, international (and particularly US government) condemnation of repression contributed to its stabilisation or amelioration. By contrast, when the international response to persecution was lacklustre, new discriminatory measures became entrenched. These case studies clearly demonstrate how the process of early and robust opposition to persecution has a substantial influence, raising the opportunity cost of genocide for the perpetrators and promoting resilience to genocide.

Strong Leaders Contribute to Genocide Prevention

The role of strong, individual leaders in preventing genocide emerges as a key theme from multiple case studies. In some cases, an individual leader has played a critical role in stopping the onset of genocide. In others, leaders have worked together to achieve such an outcome. Very often, the judicious choices of strong leaders have shifted the balance from circumstances of escalating risk towards the increasing dominance of factors promoting resilience. These leaders typically possess some key characteristics. First, they are willing to speak loudly, publicly and often in defiance of the predominant narrative. Second, they are willing to take personal risks. Responding to a crisis takes precedence over concerns for their careers, or for adhering to convention. Third, over time they become increasingly invested, with relatively small and limited actions building momentum for larger and more consequential actions when required. Interestingly, strong leaders can emerge from diverse places. Sometimes it may be from places where one might hope and expect to see leadership, such as the UN or US. Religious leaders, leaders from within the vulnerable minority and leaders from governments with a strong relationship with the at-risk state, also have the potential to play a vital role. At times, however, leaders emerge from completely unexpected places, such as from within the perpetrating regime, or forces allied with it.

Several of the case studies within this book demonstrate the central role leaders can play in promoting resilience to genocide. Chapter 4 on East Timor, for example, demonstrates how a range of leaders working together were collectively able to change the international approach to addressing the violence there, ultimately resulting in international intervention. East Timorese leader Xanana Gusmao’s sensitive leadership, promoting unity and restraint at critical moments, had a decisive impact, despite his relatively limited power. In Denmark (discussed in Chapter 2), strong leadership from the King, Prime Minister and church created an environment in which Jews were relatively safe for a time, despite Denmark being regarded as a ‘model protectorate’ under German occupation. Perhaps surprisingly, leadership from a Professor of Church History in Denmark helped foster the cultural resistance that would later protect the Jews. In Bulgaria (discussed in Chapter 3), leadership also emerged from an unexpected place within the country’s parliament, despite Bulgaria’s status as a Nazi ally. Here, church and community leaders also played a vital role. Strong leadership also played a crucial role in the Yazidi case (discussed in Chapter 5) and that of the Baha’is in Iran (discussed in Chapter 6). In these cases, effective advocacy from diaspora communities facilitated robust international responses, such as the intervention to protect the Yazidis besieged on Mt Sinjar.

International Engagement Contributes to Prevention

It is no surprise that the role of international engagement emerges as a powerful factor promoting resilience to genocide. What is surprising is just how much impact concerted efforts by even relatively powerless groups can have. Certainly, system leaders such as the US can play a critical role. When leading international powers are invested in preventing genocide, their impact can be transformative. At times this may involve direct intervention, but relatively modest actions can also be highly effective. This volume, however, particularly highlights the capacity for middle powers to play a commanding role in promoting resilience to genocide. Middle powers, or countries typically regarded as having a modest influence on international relations, have been able to use their influence, working collaboratively and unilaterally, to prevent genocide even in circumstances of extreme risk. Increased international engagement by middle powers, particular in cases in which they may have a regional connection or other affinity, is thus a strong evidence-based measure promoting resilience to genocide. Diaspora communities, comprised either of members sharing identity markers with the vulnerable group or of broader communities supportive of them, have also proven highly effective advocates with the potential to have an outsize impact. Broadening our concept of international engagement to recognise and build the capacity and engagement of middle powers, diaspora communities and other invested groups thus has strong potential to promote resilience to genocide.

Several case studies in the book highlight the diverse ways in which international engagement can make a crucial difference in promoting resilience to genocide. In Chapter 5, I explore the dramatic rescue of the Yazidis trapped on Mt Sinjar and targeted for annihilation by ISIS. Here, the intervention of US forces played a key role. But so too did the expatriate Yazidi community, in providing crucial information to US government officials and working to promote the engagement of senior officials with the crisis. Additionally, the YPG, the Syrian affiliate of the Kurdistan Worker’s Party, was instrumental in opening up an escape route from Mt Sinjar. These multiple levels of international engagement enabled the extraordinary escape of tens of thousands of Yazidis at imminent risk of genocide. In East Timor, concerted efforts by middle powers Portugal and Australia, working with other invested countries and a determined UN Secretary-General, managed to build international pressure for a UN peacekeeping mission to address the crisis there. As I explore in Chapter 4, the Australian solidarity network on East Timor helped to build the public pressure that led to the Australian government’s robust stance on the issue. In Chapter 2, I also examine the role of Sweden in preventing the genocide of Danish Jewry during the Holocaust. Through its decision to open its border to Danish Jews, it provided a crucial escape route at a time of desperate need. This demonstrates the crucial role of international engagement, and in particular the role of system leaders, middle powers and engaged communities in promoting resilience to genocide.

The Importance of Swift Action at Key Moments

Swift action at key moments is vital for genocide prevention. Sometimes such action is needed to directly prevent genocide when onset appears imminent. Such action can be difficult, but even at these crisis points there are still possibilities for effective intervention, as this volume demonstrates. Here, measures that might delay or even only partially obstruct imminent genocide onset can still save lives. Evidence indicates that genocidal regimes can only maintain the capacity to commit genocide for limited time periods. A delay may prove sufficient to prevent genocide for just long enough until that capacity is reduced by another measure, or the changing wider context. Many genocides occur within the context of war, for example, which can be highly dynamic. There are also many key inflection points well before genocide onset is imminent, however, where timely action can have a marked impact. Timing is critical. There are points when regimes are actively considering whether to escalate persecution of vulnerable groups, and robust internal and external opposition at these points can be pivotal. Regimes persecuting minorities may be particularly vulnerable to financial pressure at certain points, and taking advantage of these moments to leverage that pressure can materially change trajectories of risk and resilience. Swift intervention to prevent a local hotspot erupting, or contain a limited outbreak of violence, can make a critical difference as to whether the situation stabilises or the crisis escalates. Providing international assistance when an at-risk country requests it, whether it be to manage a refugee influx or respond to a natural disaster, can also significantly influence the interplay of risk and resilience moving forward. Finally, reminding allies of the importance of respect for human rights, within the context of the maintenance of an alliance of mutual benefit, can also be impactful at the right moment. To take swift action at key moments requires sufficient depth of context-specific knowledge and capacity, highlighting the importance of genocide prevention infrastructure within governmental and intergovernmental institutions.

The importance of timely action is evident in multiple case studies within this volume. Chapters 2, 3, 4 and 5 all highlight, in different ways, the importance of a swift and robust response to the imminent threat of genocide. In Chapter 5, the window of time to save the Yazidis trapped on Mt Sinjar could be counted in days, and had the US not responded as rapidly as it did, the outcome would likely have been very different. In Chapter 4, the unfolding violence in East Timor in the aftermath of the referendum became a dire situation in which every day counted. The situations in Bulgaria and Denmark also demonstrate the vital importance of timely intervention in preventing genocide. Yet the case studies also demonstrate other opportune moments for promoting resilience, some of which were effectively utilised and others that became missed opportunities. Chapter 7 explores several such moments with respect to the Dominican Republic, while Chapter 6 does so in considering the vulnerability of the Baha’i minority in Iran. Taking advantage of opportune moments to build resilience to genocide, and acting swiftly to prevent escalation when crises erupt, is critical to genocide prevention.

Recognise and Facilitate the Agency of Vulnerable Groups

An evidence-based approach to genocide prevention is one that recognises and facilitates the agency of vulnerable groups. Vulnerable groups are deeply invested in their self-protection.Footnote ⁹⁰ They are often well aware of their vulnerability, of particular risks for escalation and potential pathways to reduce their vulnerability and build resilience in their communities. The agency of vulnerable groups has made a critical difference in promoting resilience to genocide, and in directly preventing genocide. Very often a key element of this agency has been through vulnerable groups alerting leaders and/or the international community to their plight. Doing so has sometimes initiated a causal sequence that has led to interventions to protect the group. Vulnerable groups have also utilised their agency to physically hide from or escape imminent risk of genocide or targeted violence (or the fear thereof). In these cases, they have used their own resources, networks and local knowledge to attempt to survive. Indeed, in many cases of genocide, those who have survived have often done so due to their own efforts to escape or evade the violence.Footnote ⁹¹ It must be recognised, however, that such groups, by their very vulnerability, are unlikely to have sufficient capacity to unilaterally ameliorate their circumstances.Footnote ⁹²

The role of the agency of vulnerable groups in promoting resilience to genocide is evident is several of the case studies presented herein. The case study of the Yazidis in Iraq (discussed in Chapter 5), for example, demonstrates the centrality of their agency to their survival. First, their ability to flee was essential for escaping immediate capture by ISIS. Very quickly, however, tens of thousands of Yazidis found themselves besieged and without sufficient access to water, in a hot desert environment. Here, their ability to use their mobile phones and networks proved critical to informing Iraqi officials, UN officials, international media and the Yazidi diaspora of their plight. This in turn enabled the interventions that led to their rescue. In Bulgaria (discussed in Chapter 3), too, Jews used their networks to bring their plight to the attention of community leaders, who then advocated for them. In Denmark (discussed in Chapter 2), when Jews found out about their imminent deportation, they used their agency to first hide and then find safe passage to Sweden. In East Timor (discussed in Chapter 4), the East Timorese used multiple strategies to bring international attention to their plight, over many years. East Timorese chose to risk their safety by voting in the referendum, believing the vital importance of the vote to their future merited the risk involved. Many subsequently sought safety by fleeing Dili, or seeking shelter in churches and other places, even before the post-election violence escalated. These case studies demonstrate the role of the agency of vulnerable groups in contributing to resilience to genocide, and of the potential to enhance that role through timely intercession in response.

Capacity for Movement across International Borders Aids Prevention

The ability for vulnerable groups to cross international borders to reach safety emerged as a protective factor in multiple case studies. Being able to escape the imminent threat of genocide by crossing a border can be literally a matter of life or death. In these circumstances, vulnerable groups able to flee to another country may be able to save themselves; those unable to do so can be at much greater risk of genocide. This is not a new finding, yet it is one that has only been partially heeded by the international community. Prior to the Holocaust, it was the inability of many Jews to obtain visas to other countries that preventing them from escaping Nazi Europe. Large numbers of Jews who might otherwise have fled were thus trapped and subsequently murdered. In the aftermath, the 1951 Refugee Convention and its 1967 Protocol recognised the right of refugees to flee persecution and seek safe haven in other countries. Yet this right remains far from assured. Many countries have erected physical, legal and other barriers to severely restrict the ability of asylum seekers to seek refuge. This research highlights the ongoing importance of allowing vulnerable groups to seek refuge as a means by which they can reduce their risk of genocide.

Two case studies in particular highlight the importance of the capacity to cross international borders as a protective factor. In Denmark (discussed in Chapter 2), when the Nazis commenced their efforts to deport the Jews, Sweden opened its borders to any Jewish person able to reach them. This proved critical to the survival of Danish Jewry, as several thousand Jews were able to take refuge there for the remainder of the war. In Iraq (discussed in Chapter 5), the ability of the Yazidis to escape Mt Sinjar was dependent on their being able to cross into Syria, even if only temporarily. The only feasible escape route was through Syria, and had this not been accessible, the group would have remained trapped and at extreme risk of genocide. In the Dominican Republic as well (discussed in Chapter 7), Haitians and Dominicans of Haitian descent have at times escaped targeted violence through flight to Haiti. The border has also functioned as a de-escalatory mechanism in other ways. The evidence highlights the need for vulnerable groups to have the capacity to cross international borders as a measure to protect themselves from the imminent threat of genocide.

Use Financial Leverage to Promote Resilience

Using financial leverage can be a powerful and feasible measure to promote resilience to genocide in the right circumstances. The focus here is not on sanctions, which have mixed results and have been analysed elsewhere. Rather, what emerges from the case studies in this volume is the capacity for bilateral trade relationships and financial connections to be leveraged to induce or even coerce positive behaviour change by regimes targeting vulnerable groups. When important trade partners communicate that trade relationships can only prosper within a framework of respect for human rights, this can significantly raise the opportunity cost of continued or escalating persecution of a vulnerable minority. When states are particularly financially vulnerable, or not yet committed to a genocidal ideology, this can be effective in promoting resilience. At lower levels of risk, inducement can play a major role. For example, proposals for preferential trade relationships, or the extension of such relationships, can be made contingent on reforms or safeguards to protect vulnerable groups. Pre-existing cooperative relationships between states, often providing mutual financial benefit, can be leveraged to build respect for human rights into the process. There is a great deal of merit to focusing on the potential of financial levers to promote resilience to genocide, rather than viewing trade relationships as separate from the wider context of a bilateral relationship.

The capacity for the international community to utilise financial leverage to promote resilience to genocide is clearly demonstrated in three case studies in this volume. In Chapter 4, we see how financial pressure contributed to Indonesia’s decision to allow an international peacekeeping force into East Timor, for example. As the violence intensified in the aftermath of the referendum there, the World Bank and International Monetary Fund both placed significant pressure on Indonesia to resolve the situation in order to continue to receive funding. The US also indicated there would be significant financial consequences were the situation to continue. As Indonesia struggled to recover from the Asian Financial Crisis, this was a powerful lever to compel its cooperation. This type of financial pressure also proved persuasive in the case of Iran’s persecution of the Baha’i minority in the 1980s, as I explore in Chapter 6. Following the 1979 Iranian Revolution, Iran’s rapid economic growth swiftly reversed. As international pressure increased on Iran over its treatment of the Baha’i minority in the mid-1980s, Iran’s major trading partners also indicated their displeasure. Iran was heavily dependent on oil exports to improve its economic position, and this approach thus placed significant pressure on the regime to de-escalate its persecution of the Baha’is. Finally, Chapter 2, examining Denmark and the Holocaust, also offers interesting insight into the potential of financial leverage in promoting resilience to genocide. Denmark’s capacity to provide Nazi Germany with large quantities of agricultural produce and manufactured goods essential to the war effort gave it some power to negotiate the terms of its occupation. Here it leveraged its capacity to provide material benefits to Nazi Germany, to insist the latter refrain from persecuting Danish Jewry. These case studies highlight the potential impact of financial levers in promoting resilience to genocide.

Develop and Promote Inclusive Conceptions of National Identity

Promoting inclusive conceptions of national identity is a very powerful way to promote resilience to genocide. Genocide is fundamentally about identity. Indeed, the presence of an ‘outgroup’, that is a group outside the universe of obligation, is recognised as a foundational risk factor for genocide. Promoting an inclusive conception of national identity, one in which the entire population of the state ‘belongs’, and is included in the universe of obligation, fundamentally undermines this risk factor. When at least some of the population perceive a shared identity with a vulnerable minority, they are more likely to protest discrimination against that minority. Efforts at marginalising that minority become more difficult, as the narrative of exclusion is challenged and rejected. In this volume, inclusive conceptions of national identity, rather than other identity markers, emerged as a significant factor promoting resilience. Whether threats are internal or external, this reflects the power of the state with respect to addressing – or escalating – risk of genocide.

Two case studies convincingly demonstrate the power of inclusive conceptions of national identity in promoting resilience to genocide. In Chapter 3, I examine how ordinary Bulgarians rejected Nazi attempts to characterise Bulgarian Jews as a threatening other. Instead, they perceived Jewish Bulgarians as fellow citizens, entitled to the same rights and protections of the state as all Bulgarians. When the Bulgarian government attempted to implement discriminatory legislation, large numbers of ordinary Bulgarians vigorously protested. When the Bulgarian government, under Nazi pressure, attempted to begin deportations, their attempts were thwarted by the strong opposition of political leaders, religious leaders and the wider public. This case study has many similarities with that of Chapter 2, examining the Holocaust in Denmark. Here too, Jews were explicitly included in the ‘imagined community’ of Danish people.Footnote ⁹³ When the Nazi occupiers attempted to deport the Jews, many Danes were willing to help them hide and ultimately escape. While it’s important to note that in both cases support for the Jews was not universal, and a minority did embrace antisemitism, in each case the support of a substantial portion of the population proved sufficient to protect the Jews from deportation. A sense of common national identity proved critical to this outcome.

The Importance of Discursive Space in Contributing to Prevention

The presence of discursive space, in which competing views can be publicly presented and discussed, is vital to promoting resilience to genocide. Free speech (with careful attention to ensure it does not morph into hate speech) provides crucial opportunities to contest narratives justifying exclusion, discrimination and persecution and to promote alternative narratives. They allow opposition to genocidal ideology to be expressed, and for people holding such views to find and identify with others with similar views. Typically, discursive space is associated with democratic institutions, such as a parliament, a free press and the right to protest. While repressive regimes often curtail these institutions, the continued presence of some discursive space can have a strong influence on promoting resilience. This might include within religious institutions, where religious leaders can speak somewhat freely from the pulpit, for example. A relatively accessible underground press can challenge the dominant narrative and counter propaganda. Even when fraught with risk, protests can also call attention to the presence of active opposition to divisive ideology. In societies where risk is lower and violence less imminent, discursive space can be used to protest discriminatory policies and hold governments to account. The ability to contest narratives, protest against the treatment of vulnerable minorities and bring public attention to persecution is an important factor in promoting resilience to genocide.

Several chapters in the volume highlight the importance of discursive space. In Chapter 3, we see a dramatic depiction of this in Bulgaria, with a parliamentary petition to halt the deportation of the Jews there. While Bulgaria was not democratic at this point, the ability of parliamentarians to publicly speak out against the planned deportations in parliament, and to petition the Bulgarian Prime Minister, proved pivotal. In both Bulgaria and Denmark (discussed in Chapter 2) in the Holocaust, church leaders were able to publicly express their opposition to antisemitism, challenging propaganda and providing moral guidance. In Chapter 7, we also see the importance of discursive space in challenging discrimination against Haitians and Dominicans of Haitian descent in the Dominican Republic. Non-governmental organisations and the press could bring public and international attention to discriminatory policies, raising the cost of pursuing such policies for the Dominican government. The presence of such discursive space means that efforts to escalate discrimination and persecution can be publicly contested. It ensures that divisive ideologies are questioned and challenged, disrupting the impact of partisan narratives and ideologically driven propaganda. Discursive space thus provides an important function in promoting resilience to genocide.