Genocide Prevention and Western National Security: The Limitations of Making R2P All About Us

The case for turning R2P and genocide prevention from principle to practice usually rests on the invocation of moral norms and duties to others. Calls have been made by some analysts to abandon this strategy and “sell” genocide prevention to government by framing it as a matter of our own national interest including our security. Governments’ failure to prevent atrocities abroad, it is argued, imperils western societies at home. If we look at how the genocide prevention-as-national security argument has been made we can see, however, that this position is not entirely convincing. I review two policy reports that make the case for genocide prevention based in part on national security considerations: Preventing Genocide: A Blue Print for U.S. Policymakers (Albright-Cohen Report); and the Will to Intervene Project. I show that both reports are problematic for two reasons: the “widened” traditional security argument advocated by the authors is not fully substantiated by the evidence provided in the reports; and alternate conceptions of security that would seem to support the linking of genocide prevention to western security—securitization and risk and uncertain—do not provide a solid logical foundation for operationalizing R2P. I conclude by considering whether we might appeal instead to another form of self interest, “reputational stakes”, tied to western states’ construction of their own identity as responsible members of the international community.


Introduction
One of the central concerns of many genocide studies scholars and activists is turning the well-worn phrase "never again" into reality. A central component of encouraging genocide prevention is the Responsibility to Protect doctrine that sets out the responsibility of states to uphold the human rights of populations under their own control and in other societies in which the local state cannot or will not protect its own people. Upholding basic human rights and human security around the world is framed as both a moral good and duty that applies to us all. But while most states and in-ternational organizations have rhetorically expressed their support for R2P and genocide prevention, we have yet to see states put the principle into practice in more than a handful of cases.
In the last few years some analysts have tried to foster the political will for genocide prevention by making what they hope will be a more persuasive argument to policy-makers. Instead of relying on moral claims based on responsibility, duty, and the indisputable moral wrongness of genocide, calls have been made to see prevention as a matter of western states' national interest, including their national security broadly construed. This argument speaks to our na-tional and rational self-interest by suggesting that failure to prevent atrocities abroad imperils western societies in very real ways.
My question in this study is whether genocide prevention conceptualized as a necessary policy to ensure western societies' own security is a logical foundation for asserting that western states live up to their avowed responsibility to protect vulnerable populations around the globe. This is a different, albeit related, question from whether appeals to our national security will be successful in motivating effective genocide prevention by policy-makers.
On the face of it, the genocide prevention-asnational security thesis appears sensible. After all, purely humanitarian appeals have not produced consistent and robust genocide prevention. To get the attention of political leaders and policy-makers do we not need to "speak their language" by appealing to national security? What better way to conceptualize the need for planning and action than to tie genocide prevention to warding off threats, one of the most basic functions of the state? Indeed, in his address announcing the creation of his administration's Atrocity Prevention Board, President Obama emphasized that "[p]reventing mass atrocities and genocide is a core national security interest" of the United States (Presidential Study Directive on Mass Atrocities, 2011, p. 1).
If we look at how this argument has been articulated in policy reports over the past few years, we can see that this position is not entirely convincing. To demonstrate this point I turn to two policy reports that have made the case for genocide prevention based in large measure on national security considerations: Preventing Genocide: A Blue Print for U.S. Policymakers by the Genocide Prevention Task Force co-chaired by Madeleine Albright and William Cohen and the Will to Intervene Project by the Montreal Institute of Genocide and Human Rights Studies co-chaired by Frank Chalk and Lt. Gen. (Ret.) Romeo Dallaire. The central problem is that both reports fail to offer a sound foundation for prevention grounded in security, first, because the logic of the "widened" traditional security advocated by the authors does not fit comfortably with the evidence provided in the reports; and second, alternate conceptions of security that prima facia would seem to accommodate conceptualizing atrocity prevention as a security issue-securitization and risk and uncertainsimilarly do not provide a logical foundation for operationalizing prevention.
The question then becomes, how do we make the case for turning R2P from an agreed upon principle into common practice? In the last part of the article I draw on the reports' general appeal to the self-interest of would-be interveners in the west and contemplate whether operationalizing R2P might be more effectively grounded in appeals to reputational stakes tied to western states' construction of their own identity as responsible members of the international community. I then briefly examine whether the codification of R2P as a legal obligation might facilitate atrocity prevention and conclude that entrenching R2P in international law may not facilitate this process of norm internalization and that norm internalization in itself cannot guarantee consistent atrocity prevention. In short, while western governments may want to be seen to be doing good, they may not regularly acting to do good in the international system even if they were to be required to do so under international law.
Throughout the article, I am also trying to grapple with hard cases in which western interests-security, economic, geo-political-are not at play. These cases suffer from what I call the "Goldilocks problem". These cases constitute one of two kinds of circumstances in which prevention and intervention is unlikely to occur since they fall outside the "Goldilocks zone", that is, the parts of the world that are of direct interest, for a variety of reasons, to western states. Non-Goldilocks cases of atrocity occur in locations that are of either little strategic value or of "hyper" strategic value such that intervention is considered to be too dangerous a proposition for all involved.
My interrogation of the Genocide Prevention Task Force and Will to Intervene reports is admittedly quite narrow. Although the reports are designed to offer a set of policies for policy-makers, my focus is on the underlying security logic found in the reports since the authors ground their policy prescriptions in a logic of national interest tied largely (although not exclusively) to national security broadly construed. I proceed in a fashion analogous to a legal scholar examining the legal reasoning of a judicial decision, and as such, my analysis does not address the potential effectiveness of the prescriptions outlined by the authors, nor do I engage with the very important literatures on genocide prevention or R2P.

Preventing Genocide: A Blue Print for U.S. Policymakers (The Genocide Prevention Task Force Report)
As its title suggests, the Genocide Prevention Task Force Report is a set of recommendations aimed exclusively at American leaders, policy-makers, and institutions. Noting that R2P is in part the inspiration for the report, the authors suggest that "there is a growing understand-ing…that states have a basic responsibility to protect their citizens from genocide and mass atrocities" and that "[n]o government has the right to use national sovereignty as a shield behind which it can murder its own people. The challenge for the world community is not only to state this principle, but to implement it" (The Genocide Prevention Task Force, 2008, p. xxi).
In making the case for genocide prevention as a foreign and defence priority for the United States, the co-authors assert that their report is inspired by three key considerations. The first is the immorality of genocide and other atrocities that constitute "a direct assault on universal human values, including, most fundamentally, the right to life." The report immediately goes on to evoke a second and much more central concern that "[g]enocide and mass atrocities also threaten core U.S. national interests" since genocides "feed on and fuel other threats in weak and corrupt states, with dangerous spill-over effects that know no boundaries." The third concern is reputational. If the United States does not plan for and engage in successful genocide prevention around the globe U.S. "credibility and leadership" may be at stake (The Genocide Prevention Task Force, 2008, p. xv).
As a security threat, genocide is said to fuel instability in weak and undemocratic states. These kinds of states engage in terrorist recruitment, human trafficking, and experience civil strife, all of which have "damaging spill-over effects for the entire globe." Further, the report identifies the monetary costs and logistical challenges of humanitarian responses to refugee flows sparked by genocidal violence, noting that the United States often ends up footing much of the bill to feed, house, and care for refugees. It is in the United States' own interest to pay less up front to prevent genocide than more later to deal with its aftermath. Finally, the report links the United States' national interest and security to America's international reputation, warning that if the country does not establish the capability and will to prevent genocide, the international community will come to see the United States as "bystanders to genocide" which would in turn undermine the United States' ability to be a "global leader" and "respected as an international partner if we cannot take the necessary steps to avoid one of the greatest scourges of mankind" (The Genocide Prevention Task Force, 2008, p. xx).
Having made the security argument, among others, for genocide prevention, the report outlines institutional and funding reforms designed to integrate genocide prevention into US foreign, defence, and development policy-making. Political leaders, specifically the American president, must make genocide prevention a top priority and relevant Congressional committees should do the same. The United States government must develop early warning strategies and intelligence capabilities within existing departments and agencies to identify possible outbreaks of genocidal violence and implement a broad set of development assistance policies focused on democratization (including the protection of human rights and minority rights) and economic development in vulnerable states as an early prevention strategy. Government should also use a carrot and stick "preventive diplomacy" strategy with po-tentially genocidal regimes, develop plans for either the unilateral or multilateral use of force to stop genocide where the killing has begun, and cooperate with allies and international organizations to strengthen anti-genocide norms and institutions designed to prevent and punish atrocities.

Mobilizing the Will to Intervene: Leadership and Action to Prevent Mass Atrocities
Much like the Genocide Prevention Task Force, the authors of Mobilizing the Will to Intervene: Leadership and Action to Prevent Mass Atrocities (W2i, 2009) argue that we must build the will and capacity to stop mass atrocities since genocide and other gross human rights violations constitute threats to ourselves and to the world. An ideational and policy orientation that was sorely lacking in an earlier era of traditional statecraft, the "will to intervene" can be fostered by appealing to our own national interest grounded in a "widened" conception of national security in a post-Cold War, globalized world. Based on case studies gleaned mostly from interviews with American and Canadian officials that try to account for why the international community failed so miserably in Rwanda but managed to act decisively in Kosovo (W2i, 2009), the authors argue that if we want governments to get serious about genocide and mass atrocity prevention we need to stop appealing exclusively to the injustice of such acts and instead frame them through the lens of national interest grounded in national security.
Mass atrocities and the "chaos and loss of life" they visit on their victims in turn "produce shock waves" that act as "seismic wrecking balls" that destabilize the world far beyond the regions in which they occur (W2i, 2009, p. 4). The threats posed by atrocities abroad are identified by the authors as "costs" including: medical, health, and social costs flowing from pandemics that may originate and then spread from atrocity-torn areas; national security costs resulting from the creation of safe havens for piracy and terrorism in countries and regions that experience atrocities; financial and social costs produced by refugee flows; economic costs resulting from loss of access to strategic resources; and the political cost of alienating electoral constituencies at home. To eliminate or diminish these costs, and thus protect ourselves from the fall-out of any or all of these scenarios, governments and civil society must work together to prevent atrocities in the future and stop on-going mass atrocities in order to ensure our own health, security, and economic prosperity (W2i, 2009, pp. 9-17).
The report identifies four pillars around which the "will to intervene" should be mobilized: enabling leadership in government particularly at the Presidential/Prime Ministerial and cabinet levels; enhancing coordination between government departments and ministries; building institutional capacity for effective prevention and intervention within government; and fostering knowledge through a bottom-up process in which Canadian and American civil society groups as well as the media act to inform government of and press for the need to engage in humanitarian intervention (W2i, 2009, pp. 17-61). Similar to the Genocide Prevention Task Force Report, the first three pillars of W2i involve the introduction of new roles, structures, and processes in the executive, legislature, and civil service to make atrocity prevention a priority both in terms of policy importance and capacity to act.

The Problem of Appealing to a Logic of Western National Security
There is a certain attractiveness to the idea that we need to speak to government in the national interest security-oriented cost-benefit language to which it presumably is most accustom and which the Genocide Prevention Task Force and W2i authors hope it is more likely to listen. But what, exactly, is the conception of security to which the reports appeal? Both say that in a globalized world they take security not just to include military security but also economic, physical/health, and even reputational security. The W2i report, for instance, notes that governments once only thought of security as "the defence of territorial borders against invasion and attrition". Now "the meaning of security has expanded…beyond state centric concerns related to defence" to include a "wider variety of international and transnational threats affecting states and their citizens" (W2i, 2009, p. 9). But beyond this the key concept upon which the reports' authors rest their case is left largely undefined. To be fair, the reports are not intended to be scholarly exegeses on conceptions of security or an exercise in the redefinition of security. Nonetheless, as a scholarly reader of these reports I am interested in interrogating the logic of atrocity prevention grounded in national interest as mostly national security. To do this I now turn to three (admittedly not exhaustive) conceptions of security found in the security studies literature. The first is the conception seemingly used by the authors themselves, what sometimes is called in the security studies literature "widened" security. The second and third are two conceptions of security that would appear to be the most friendly to the idea of motivating atrocity prevention and intervention by appealing to national and global security: the process of securitization in which the inter-subjective identification and definition of threats by securitizing actors and an audience define what constitutes a threat and why; and the sociologically informed notions of risk and uncertainty in an interdependent globalized world.

"Widened" Security
Traditional understandings of security identify threats that are "out there" in the real world which are discernible through rational calculations of what does and does not objectively pose a threat to the survival of the state and society. Once identified, these threats must be neutralized through the application of (mostly military) power and other resources. By adopting a "widened" conception of security the reports say that objective security threats include not only external military or political threats but also the negative effects of a changing climate, the spread of epidemics and pandemics, and the actions of non-state actors such as terrorist organizations and other armed groups (Buzan & Hansen, 2009). The reports thus contend that in this kind of general security climate the follow-on effects of genocide, regardless of where the crime is committed, poses pressing objective threats to Canada and the United States. Further, both reports see genocide not as a military threat or a threat to military assets (although military assets and personnel may be put in harms way to stop genocide and therefore may impose costs on military institutions and personnel) but as a threat to other sectors such as the economy, the health and well-being of citizens, or the political fortunes of Canadian and American politicians. Let us now examine how the logic of linking genocide prevention to an expanded understanding of objective security threats fits with the arguments and evidence provided by the authors. Since W2i provides the most detailed security argument and empirical evidence, I will concentrate on this report specifically.
Under "security costs" the authors of W2i cite the fact that regional and global insecurity is often produced by "failed states" such as Somalia. The authors correctly note that the collapse of the Somali government in 1993 gave rise to warlordism within the country, piracy in the waters off the Horn of Africa, and we might add since the publication of the report, regional terrorism in the form of the now Al-Qaeda affiliated Al-Shabab. Failed states, however, are not necessarily synonymous with or measures of mass atrocities. The two are frequently related phenomena but the former does not always lead to the latter. Al-Shabab's activities have had an entirely negative effect on Somalia's internal security and that of its neighbours, but Somalia's status as a failed state has thus far not produced global or western insecurity. And while the authors assert that western policymakers are beginning to take seriously the link between development, human rights, and security in places like Somalia, the authors leave out how western states actually dealt with the Somali piracy problem. Rather than addressing the problem of lagging development, state failure, and human rights abuses, western states applied hard military power through extensive naval patrols in much the same way that the very traditional security threat of piracy has been countered for centuries.
The authors also recount that weak and failed states have created the conditions for the emergence of epidemics, such as the typhus epidemic in Burundi in 1997, and that these kinds of regional health crises driven by instability, conflict, and atrocities increase the possibility of global pandemics in an era of global commercial air travel. But as with the Somalia example, the fact situations used in W2i as evidence are not cases in which mass atrocities were at play. To make the claim that mass atrocities are security threats to us because air travel may transport diseases to western countries, we would need evidence from a case where this has already happened or nearly happened. Although the suffering on the ground in refugee camps was enormous, we did not see global disease transmission after Rwanda, or over the many years of the Darfur conflict, or currently coming out of the destruction in South Kordofan or the Blue Nile regions of Sudan.
The ebola crises of 2014 did not spawn global disease transmission despite the contagious nature of the disease, considerable fears in the west of such an outcome, and the fact that the health care systems in Sierra Leone and Liberia are still labouring under the damage done by years of conflict.
In terms of political costs, W2i suggests that in Canada and the United States citizens are concerned with human rights at home and abroad and that these increasingly cosmopolitan societies have diaspora communities that assert their cultural, ethnic, religious, and national identities in part by lobbying government to support their foreign policy priorities. While this is an accurate characterization of North America's demographics, the report offers little direct evidence that diaspora politics influences foreign policy and electoral fortunes. Sri Lankan Tamil protests in 2009 in Toronto and Ottawa are mentioned, as are the disruptions they caused, but no real evidence is provided to show whether or how these protests changed Canadian government policies with respect to the then on-going conflict in Sri Lanka. And of course we now know that this conflict ended with government forces, unrestrained by Canadian or other western governments, crushing the Tamil Tigers. Diaspora communities have not had an appreciable effect on electoral outcomes in either the United States or Canada whether or not they vote according to their preference for greater humanitarian intervention in their countries of origin. Voting behaviour in general is not usually motivated primarily by foreign policy or international humanitarian issues. And even when it is, most diaspora and immigrant communities are too small and thinly dispersed across geographically defined electoral ridings, districts, or the Electoral College to affect federal and/or presidential elections in Canada or the United States. By the logic of the W2i report, refugees and already existing expat Rwandans in the United States, for instance, might have been expected to take out their frustrations on President Bill Clinton during his re-election campaign in 1996. As it turned out, Clinton's share of the popular vote increased nation-wide from 43% to 49% between the two elections and Clinton carried four out of the five states (New Jersey, New York, Florida, California, but not Texas) in which most recent immigrants to the United States live (http:www.pewhispanic.org/files/20 13/02/PH_13.01.23_55_immigration_06_states). Similarly the governing Liberal Party of Canada that failed to act during the Rwandan genocide was re-elected in 1997, taking all but two seats in immigrant heavy Ontario and holding all of their ridings in the city of Montreal, home to many Rwandans who fled the genocide (http://www.parl.gc.ca/About/Parliament/FederalRidin gsHistory/hfer.asp?Language=E&Search=Rres&ridProvi nce=10&submit1=Search).
The two cases examined in the W2i report-Rwanda and Kosovo-also unwittingly illustrate the "Goldilocks problem". Countries that experience genocide and are of geo-strategic or economic value to the west lie within the "Goldilocks zone". Not unsurprisingly they are much more likely to capture the attention of the international community (e.g. Libya) while countries that are not strategic or are too strategic, in that they have very powerful friends and neighbours, lie outside the Goldilocks zone. The former cases are either unlikely to be seen as pressing national security issues because of their geo-political remoteness to the west and/or their marginal role even in a globalized political economy, while the latter are considered to be humanitarian crises that cannot be stopped through outside intervention without risking a much wider and destructive conflict no matter how badly behaved the regime or the other parties to an atrocity may be (e.g. the on-going Syrian civil war). It is made painfully clear in the W2i report that neither the American nor Canadian governments were motivated to intervene in Rwanda for moral or strategic considerations. The report repeatedly references comments by American and Canadian officials that Rwanda was of no value and therefore of no strategic interest, or by the logic of the report, not in the national interest and a national security threat. A senior US government source told the authors of W2i that the US "did not have massive strategic interests" in Rwanda (W2i, 2009, p. 91) while in the highest government circles in Canada there was "little interest in Africa (W2i, 2009, p. 69) because it "was not a priority and lay outside of Canada's traditional zone of interest" (W2i, 2009, p. 70). By contrast, as we will see later, the report cites officials and politicians in both countries claiming that Kosovo required intervention because of its strategic location in Europe, the need to make NATO effective and relevant in a post-Cold War world, and in the Canadian case, the need to appear relevant within the NATO alliance and to be seen as a small but important team player. None of the reasons offered by interviewees in the Kosovo case suggest a wider conception of security advocated in the report.
Of course, one of the goals of the W2i report is to change perceptions of what constitutes a challenge to national security. But to make this argument, the report needed to offer direct evidence that the costs to the west the authors associate with genocide, particularly in parts of the world considered non-strategic, really do exist and that governments have in the past responded to what policy-makers consider to be security threats, foreign and potentially domestic, from mass atrocity situations.

The Securitization of Threat
That a widened traditional conception of security as articulated and illustrated in the reports does not seem to effectively support the case for atrocity prevention may not mean that security considerations cannot provide a logical foundation for upholding R2P. Perhaps we can make a security-focused argument for atrocity prevention by looking at another understanding of security that would seem to accommodate the imperative of rethinking atrocity prevention as a western security issue. Here I turn to the critical security approach known as "securitization". As formulated by Buzan, Waever, and de Wilde in their seminal book Security: A New Framework for Analysis (1998) securitization eschews an understanding of security as exclusively military "objective" threats "out there" posed mostly by other states in favour of security understood as an intersubjective discursive process by which perceived threats are constructed as threats to the military, political, economic, or social sectors of a state. The process of securitization involves the articulation of a perceived threat through a "speech act" or security discourse in which a particular issue is said to pose an existential threat to a "referent object" (e.g. the state, society, or a country's territory). Responding to such a threat must involve extraordinary measures beyond the usual rules and procedures of "normal" politics (Buzan et al., 1998, p. 21). While an almost infinite variety of nontraditional security issues across any sector of society can be articulated as existential threats to a referent object (e.g. drug trafficking) a threat is not successfully "securitized" until an audience accepts that the issue is a threat. Without the acceptance of the audience we only have a "securitizing move" (Buzan et al., 1998, p. 25). Since threats are not objective and there is no metric by which we can measure objective threats, securitization is all about the construction of shared meaning. "In security discourse, an issue is dramatized and presented as an issue of supreme priority; thus, by labelling it as security, an agent claims a need for and right to treat it by extraordinary means". (Buzan et al., 1998, p. 26).
There are three discernible areas in which the Genocide Prevention Task Force and W2i reports conceptualize genocide prevention as a security threat such that it appears to fit the securitization approach. As we have already seen, the reports, like the securitization school, adopt an expanded view of security in which threats range across several sectors beyond the military-political. Second, both reports can be read as an exercise designed to encourage the Canadian and American governments to securitize genocide prevention. The authors of both reports call on the Canadian Prime Minister and U.S. President to use their respective offices to engage in "speech acts", in some cases quite literally in the Speech from the Throne and the State of the Union Address, to communicate the threats associated with genocidal violence abroad to their respective governmental and public audiences. Moreover, the bevy of institutional, procedural, and funding reforms meant to embed the will and capacity to prevent genocide in government institutions mirrors the idea that securitization may be institutionalized when a threat is securitized as persistent or likely to reoccur with some regularity. Finally, securitization would seem to dispense with the criticism I raised earlier about the lack of objective threats posed by genocides in regions outside the "Goldilocks zone". As a constructivist framework, successful securitization need not, and in fact cannot, constitute an "objective" measurable threat. So long as Canadian and American leaders and policy-makers frame genocidal violence elsewhere as an existential threat to their own states, economies, or societies, and other policy-makers, institutions, and the public accept this message, then genocide, wherever it happens, is a threat. But within this very same process of the intersubjective construction of meaning lie two compelling problems with reading the reports through the lens of securitization.
The first problem is posed by the requirement that threats be securitized as "existential". Consciously drawing on traditional security studies, securitization scholars assert that international security is about survival. There is, however, no universal standard for assessing whether a threat is existential or not since threats "can only be understood in relation to the particular character of the referent object in question" (Buzan et al., 1998, p 21). What constitutes an existential threat varies across different sectors: the survival of the state or the armed forces, for example, in the military sector; the sovereignty of the state, or international regimes in the political sector; the viability of a sector of the economy; collective identities in the societal sector; a habitable planet in the environmental sector (Buzan et al., 1998, pp. 22-23). Despite the fact that threats are perceived and constructed, threats cannot be of any magnitude; they must be about survival. As noted in my earlier critique of the W2i report, it would be hard to credibly argue, or in the words of the securitization approach to "securitize" through a speech act, the consequences of genocidal violence abroad as reasonably constituting an existential securi-ty threat to the military, political, economic, social or environmental security of Canada, the United States, or the international system as a whole, particularly in non-strategic countries that fall outside the Goldilocks zone. Indeed with respect to the military sector, Buzan et al. specifically say that peacekeeping and humanitarian intervention cannot be constructed as existential threats because they do not imperil the survival of the state or its armed forces and because they occur as "support for routine world order activities" (Buzan et al., 1998, p. 22).
This observation brings us to the second problem. Aside from constructing threats as existential, securitization also requires that the response to securitized threats be "extraordinary," going beyond the established rules and procedures of normal politics, some of which may become institutionalized over time. Sensibly, neither report argues that the Canadian or American governments should pursue genocide prevention through some equivalent of the Bush Administration's legally suspect approach to the Global War on Terror. The authors of the Genocide Prevention Task Force Report and W2i firmly ground their recommendations in either existing agencies and procedures or the introduction of relatively few new institutions and policies designed to regularize R2P and genocide prevention as standard, not extraordinary, operating procedures. What the reports seem to be recommending is not so much securitization (existential threats and extraordinary responses) but what securitization scholars call "politicization", that is, making an issue "part of public policy, requiring government decision and resource al-locations…" (Buzan et al., 1998, p. 23).
That the Genocide Prevention Task Force and W2i authors do not call for extraordinary or extralegal measures is not a failing of the reports. Rather, it is a strength as it shows great respect for the democratic process and the rule of law. It does, however, undermine our ability to make sense of the security logic contained within the reports according to a securitization model. It also means that the "politicization" of genocide prevention recommended in the reports could force prevention to compete for attention and resources alongside other security concerns. Stripped of the argument that genocide prevention is an extraordinary moral imperative, prevention would be "normalized", and ironically given the goal of the reports, rendered just another foreign policy issue among many.

Risk and Uncertainty
Given the inability of traditional and critical conceptions of national security threats to make the security logic of the GPTF and W2i reports comprehensible, perhaps it would be best to abandon the notion of threat altogether and instead think about the reports as grounded in risk and uncertainty. While the reports themselves speak of "threats" and "costs" linked explicitly to national "security", is it possible to frame the need for robust genocide prevention as a way of reducing "risk" and "uncertainty" in an increasingly interconnected world?
A relatively new approach in security studies, the risk and uncertainty literature is a direct response to two developments. The first, and most general development, is globalization characterized by increasing interconnections and interactions between states and other international and transnational actors across a number of different sectors, and a blurring of the line between domestic and international economics, politics, cultural, and social practices. In this global context, multiple actors can be effected, positively and negatively, by changes and crises elsewhere in the world. The second development is the advent of transnational terrorism perpetrated by diffuse networks of non-state actors operating with or alongside so-called "homegrown" terrorists, both of whom operate in and exploit the interconnectedness of a globalized world.
What we face in the post-9/11 world is not "threats" or "insecurity" in the present but what sociologist Ulrich Beck has coined a "risk society" confronted with an almost limitless array of incalculable risks in the future. These risks simultaneously result in and constitute uncertainty. "Risk" in a globalized late modern post-Cold-War world transcends time and space, forcing actors to "foresee and control the future consequences of human activity" (Beck, 1999, p. 3). The empirically and policy-oriented literature on risk sees risk as harmful outcomes ranging in severity, irreversibility, uniqueness, numbers affected, and temporal, spacial, and knock-on effects (Inter-Governmental Liaison Group on Risk Assessment, 2002). For constructivist risk and uncertainty scholars, risk cannot be objectively defined. What constitutes a risk, and the probability of that risk, is constructed through meaning attached to the interaction of actors and intersubjective knowledge grounded in cultural beliefs, norms, and biases (Williams, 2008) or epistemic communities (Kessler & Daase, 2008). Risk and uncertainty may also be the prelude to catastrophe (Kessler and Daase, 2008, p. 225). A mass casualty terrorist attack, for example, remains a risk, the certainty of which we do not know or sometimes cannot even conceive, until the attack occurs. Once the attack happens it ceases to be a risk and becomes a catastrophe.
The concept of risk is inextricably linked to uncertainty to such an extent that the two concepts cannot meaningfully stand alone. Since the risks the world faces in the post-Cold War, post-9/11 period are fed by the interdependencies of a globalized world, calculating what risks we face and how likely they are to occur involves a considerable amount of uncertainty. This is what former US Secretary of Defence Donald Rumsfeld likely meant when he referred to "known unknowns" and "unknown unknowns". Such "unknowns" are produced by the unknowability and thus uncertainty of a complex globalized modern society (Giddens, 1998, p. 23) in which late modernity itself produces untold benefits and risks. How we respond to risk and uncertainty is equally difficult to calculate because responding to risk may itself be risky since we cannot know with certainty the results of our actions. Moreover, the way in which we think and talk about risk can itself be dangerous, turning fairly improbable events, so-called "wild cards" or "discontinuous scenarios", into the possible or even probable (Kessler & Daase, 2008, pp. 225-226).
The GPTF and W2i reports do not consciously ascribe to a risk and uncertainty approach. Instead, the authors identify the threats, present and future, that they think are the by-products of unfettered genocidal destruction. Nonetheless, there are implicit echoes of the risk and uncertainty approach in the reports. The list of negative outcomes for the west associated with genocidal violence abroad includes many of the general risks identified in the risk and uncertainty literature. Further, the risk and uncertainty approach and the GPTF and W2i reports all trace the ability of these negative outcomes to reach North America's shores to the structures and technologies of late globalized modernity such as affordable transnational air travel that can spread refugees, terrorists, and germs around the globe.
As with the traditional and securitization approaches, however, the security logic of the GPTF and W2i reports does not conform particularly closely to the logic of risk and uncertainty. First, the reports do not see the negative effects of genocide on Canadian and American society to be potential risks in the future but actual tangible threats in the here and now. This is so despite the dearth of current or historical evidence in the reports that genocides outside of the Goldilocks zone have such an effect. If the reports had adopted the risk and uncertainty perspective they would each stand on a firmer logical foundation since much of what the reports say by way of the effects of genocidal violence on Canada and the United States is speculative. The authors wager that in a globalized world there are future risks to North America of the negative effects of genocide in far off lands without knowing if the risks are really there, how serious the risks are, or the ability to calculate the effects of Canadian and American prevention and intervention strategies and whether they will ameliorate or exacerbate these risks.
Second, risk and uncertainty scholars are clear that risk in a globalized world transcends time and space. But as I have already noted, the W2i report's own studies of Rwanda and Kosovo unwittingly demonstrate that when it comes to genocidal violence and its effects on the Global North, space does matters. Western states only see genocide elsewhere as a threat or risk to themselves if the society in question is geographical-ly close enough or strategic enough for atrocities to warrant a response. The 2011 UN-backed NATO intervention in Libya is a case in point. Although all of the intervening members of NATO professed genuine humanitarian concern that the civilian population of Bengazi be protected (Lindstrom & Zetter, 2012) the intervention was also motivated by a variety of other security and reputational factors. France was driven by the geographical proximity of North Africa, the opportunity in a live operation to try out its reintegration into NATO's command structure, show-off a new fighter aircraft, and allow President Sarkozy to realize his personal political goal of asserting his foreign policy prowess in advance of the 2012 presidential election (Lindstrom & Zetter, 2012, pp. 20-24). At the behest of the US, the UK accepted a leading role in the operation to demonstrate the health of the UK-US "special relationship", as well as counter the spread of terrorism in the region, and bolster Prime Minister David Cameron's bona fides as an international leader (Lindstrom & Zetter, 2012, pp. 32-38). The United States, meanwhile, was slower to respond and did so primarily to reconfirm its commitment to its European allies, to payback some of those allies for fighting alongside the US in Afghanistan, and to maintain the credibility of the UN Security Council (Lindstrom & Zetter, 2012, pp. 46-48). Canada was keen to show its continuing relevance to NATO and its ability to be a fighting, not peacekeeping, middle power (Domansky, Jensen, & Bryson, 2012). NATO also reckoned that, because of Libya's desert geography and Mediterranean coastline, the use of force without boots on the ground (i.e. without risking NATO casualties) would be "doable" (Lindstrom & Zetter, 2012).
Third, as noted, some scholars in the risk and uncertainty literature argue that risk is the prelude to harm which, when it actually occurs, is catastrophic. While the GPTF and W2i reports tell us in varying degrees of specificity what they think the harms are for Canada and the United States of genocide committed abroad, sensibly neither report claims that these harms would be a "catastrophe". Genocide is literally a "catastrophe" for the victims but it is not a catastrophe for those of us in the west, or if it is, it is a moral catastrophe because of our own lack of response. As such, unfettered genocide cannot be credibly framed as a catastrophe beyond the society or possibly region in which it occurs. The logic of the risk and uncertainty literature in this regard is too tightly tied to mass casualty terrorism aimed at the west to apply to the aftereffects of genocide executed far from home.
Finally, while the authors of the GPTF and W2i reports offer several well-meaning policy prescriptions for genocide prevention, they spend little to no time contemplating the possibility that acting to ameliorate risk may unintentionally create greater risks and harm. Both reports, but particularly the GPTF, suggest that long-term prevention to avoid outbreaks of genocidal violence requires foreign development assistance focused on democratization, robust protection for human rights as well as economic development strategies to foster economic growth, the redistribution of wealth, and equitable access to economic opportunities for all groups in society (The Genocide Prevention Task Force, 2008, pp. 35-53). While these recommendations are laudable they essentially require that societies vulnerable to genocide must be remade not only in the west's imagine, but better. Most western countries, including the United States and Canada, would not live up to the proposed standards set out in the GPTF report.
More importantly, the steps required to bring about these wholesale changes would effectively amount to slow-motion regime change or what could turn out to be, or at least look like, serial meddling in the internal affairs of other states in the name of longterm genocide prevention. Not only do these kinds of prevention strategies risk actual or perceived neocolonialism, as the risk and uncertainty literature points out, we cannot be sure that our efforts will produce the outcomes we hope for and may in fact risk unintended consequences beyond our control that either do not prevent genocide in the long-run or may make genocide more likely. As genocide scholar Michael Mann has argued, the beginning phase of democratization can increase the potential for genocidal violence, particularly if the democratic idea of the "demos" (the people") becomes intertwined with the "ethnos" (race or tribe) (Mann, 2005, pp. 2-4). Shortterm diplomatic, economic, and military interventions face a similar problem in that they may unintentionally inflame potential or actual genocidal situations. Exits too are risky since there is no guarantee that the destruction will not begin anew once international forces leave.

Prevention Tied to Self-Interest without Security: Reputational Stakes
We seem to be faced with a conundrum. On the one hand, security and risk variously defined do not appear to provide a secure foundation for fostering genocide prevention when genocide is perpetrated beyond the Goldilocks zone of western national interest. On the other hand, the invocation of moral duties to protect vulnerable populations from predatory regimes and non-state actors continues to fall on deaf ears. How, then, do we make the case for prevention? Perhaps the answer is to retain the idea found in both reports that appeals should be made to national self-interest but decouple national interest from security. To explore this possibility I now turn to another source of national interest: concern for maintaining a state's international reputation.
As a widely agreed upon principle, but one that is not codified in public international law, R2P is analogous in some ways to unwritten customary international law (CIL). Neither is explicitly linked to a set of enforcement mechanisms or sanctions for noncompliance. Studies of the role of reputation in international politics, specifically compliance with customary international law, suggest that states' decisions to follow international rules, norms, and principles are grounded at least in part in states' concern for establishing and maintaining a positive international reputation. In his compliance-based theory of customary international law, Andrew Guzman argues that states comply with CIL in the absence of robust enforcement measures or coercion when they calculate that their good reputations would be at stake if they were to violate these kinds of rules. As such, reputation is tied to "the existence of an obligation in the eyes of other states" (Guzman, 2002(Guzman, , p. 1825. A state's commitment to international law "is only as strong as its reputation. When entering into an international commitment, a country offers its reputation for living up to its commitments as a form of collateral". A state's reputation thus "has value and provides that country with benefits" which it will not wish to jeopardize (Guzman, 2002(Guzman, , p. 1825Keohane, 1984, p. 26).
As rational actors that make cost-benefit calculations about whether or not to comply with international rules, states must calculate whether the possible loss of reputation will outweigh the costs of complying with a rule, or if the benefits of maintaining ones reputation outweighs the costs of compliance. The primary cost of non-compliance that results in damage to a state's reputation is likely to be an inability on the part of the reneging state to convince would-be partners that it can be trusted to honour its obligations in the future. This is particularly true for economic agreements in which states engage in repeat interactions (Guzman, 2002(Guzman, , pp. 1851(Guzman, -1853. A state that fails to live up to its customary international economic commitments suffers "reputational sanctions" that take the form of tangible economic costs to the reneging state's economy. Reputational sanctions are most severe when the severity of the violation is high, the reasons for the violation are strategically or morally indefensible, when other states know of the violation, and when a state publicly commits to upholding a customary law but then clearly violates it (Guzman, 2002(Guzman, , pp. 1861(Guzman, -1865.
If we take states' commitment to the R2P doctrine and atrocity prevention as analogous to CIL we could then argue that appeals to national self-interest in upholding R2P should be linked to reputational stakes and the possibility of reputational sanctions if the US, Canada, and other states do not comply with their responsibility to protect. Some of the evidence presented in the W2i and GPTF reports suggest that the authors implicitly make this argument as do some of their interviewees.
In their review of Canadian government decisionmaking in the Kosovo case, the authors of the W2i report mention the perceived importance of showing NATO's relevance and capacity to act in the immediate post-Cold War world and Canada's concern that its own and NATO's reputations were at stake. Minister of Defence Bill Graham recalled that "intervention in Kosovo became a 'Canadian imperative' by virtue of European and U.S. interests in the Balkans" while another cabinet minister suggested that since the conflict in Kosovo was occurring in NATO's backyard, "interven-tion…was a matter of protecting the prestige of the alliance. NATO's reputation would have suffered from a failure in Kosovo" (W2i, 2009, p. 79). Concerning Canada's own reputational stakes, the Canadian government felt compelled to support and participate in NATO's bombing campaign over Serbia because senior Canadian politicians "also wanted Canada to be seen as a reliable international ally, to strengthen alliance solidarity, and to guarantee Canada a seat at the postconflict negotiations" (W2i, 2009, p. 85). As for upholding human rights norms, cabinet ministers and senior bureaucrats expressed genuine humanitarian concern that Kosovo not become a repeat of the massacre at Srebrenica or the genocide in Rwanda, the latter of which one official lamented "engendered a sense of shame" (W2i, 2009, p. 83). Once NATO came to frame Serbian actions in Kosovo as yet another instance of Balkan ethnic cleansing, NATO took the stand that "we couldn't allow this to continue" (W2i, 2009, p. 84).
Without going into the details of specific cases, the GPTF report also references the importance of upholding human rights norms and laws, flagging them in a final section of the report as central to effective genocide prevention (The Genocide Prevention Task Force, 2008, pp 93-110). The report also rues at various points the failure in many cases of the United States and its allies to protect the human rights of threatened populations. The authors characterize genocide as "unacceptable" and a crime that "threatens" not just American national security but "our values" (The Genocide Prevention Task Force, 2008, p. ix). Failure to stop it is said to undermine the United States' ability to be a "global leader" and "respected as an international partner" (The Genocide Prevention Task Force, 2008, p. xx). These sentiments where later echoed by President Obama in his Presidential Study Directive on Mass Atrocities. "America's reputation suffers, and our ability to bring about change is constrained", Obama argued, "when we are perceived as idle in the face of mass atrocities and genocide" (Presidential Study Directive on Mass Atrocities, 2011, p. 1).
While the reputations of the United States, Canada, NATO, and perhaps the UN appear to have been key factors in the Kosovo and Libya interventions, we may not be able to generalize from this case since we are dealing with a circumstance in which intervention ac-tually occurred. What role, if any, has reputational considerations played in cases of non-intervention? If the evidence presented in the W2i report on the response to the unfolding genocide in Rwanda in 1994 is any indication, the answer is, not much. With respect to Canada, reputational concerns are only mentioned once in the report in the context of a memo written by a senior official, Robert Fowler, in which he urged quick and decisive action, noting that history would harshly judge Canada's reasons for its lack of concerted response (Memorandum to the Minister of National Defence David Collenette, June 6 1994, quoted in W2i, 2009. Fowler recalled his widely-read memo "failed to substantively affect policy" although it "made people [in the Canadian government] feel guilty. That's all" (W2i, 2009, p. 78). As for the United States, W2i shows that reputational concerns played a more direct role in the Clinton Administration's response to the Rwandan genocide but not in a way that included cost-benefit calculations of how the administration's policy might affect the United States' reputation in the eyes of other international actors with whom the United States maintained on-going relationships. Instead reputation was considered only with respect to American public opinion and domestic human rights groups, albeit in rather contradictory ways: to show that the administration was sensitive to public concerns but also determined that any media reports showing the suffering of Rwandan victims would not push the administration into doing something that it did not want to do (The Genocide Prevention Task Force, 2008, pp. 101-102).
Second, as already noted, states' reputations are most at stake when rule violations are severe and when the nature of the commitment violated is clear. But even in these circumstances it is difficult to know how much reputational capital is at stake (Guzman, 2002(Guzman, , p. 1877. The serial ignoring of R2P in all but a few cases suggests that failure to uphold the principle is not seen by a state's allies as a serious violation. As such, states likely do not think less of each other when individual or collective humanitarian intervention is not forthcoming in the face of mass atrocities (what academics and the public think is another matter). And since R2P is a relatively new non-institutionalized principle, the commitment made by states in publicly endorsing R2P is weak relative to other international legal obligations that are codified in treaties and conventions.
Third, the violation of R2P by western states does not demonstrably harm the (mostly western) allies of these states, leaving little prospect that reputations will be damaged and that reputational sanctions will result. Reputational loss and reputational sanctions are likely to be exacted by one's allies only if these same states are the other parties to which a duty or obligation exists that has gone unfulfilled. In R2P most western states in the international system have made a commitment to protect the lives and physical integrity of populations in states outside of the western alliance. The obligation of R2P, then, is not to one's allies but to third parties who are often powerless populations unable to exact tangible sanctions on western states that fail to act to save them.
The one circumstance in which reputational calculations might be different, however, is if the violation of R2P is attached to an international organization rather than its individual member states and if such a violation is seen by members to fundamentally undermine an international organization that is important to their own national interest. As we have seen, this appears to have been at least part of the logic behind the purported necessity of intervening in Kosovo and later Libya.

Norms, Identity, and the Legal Codification of Atrocity Prevention
Although reputational stakes are unlikely to play themselves out in the way predicted by compliance theory when applied to upholding R2P and stopping genocide, how states think about certain elements of reputation tied to conceptions of their own identity and how they construct each others' identities, may be a way of retaining some of the insights offered by the literature on reputational stakes but seen through a constructivist rather than a liberal cost-benefit lens.
The central problem identified by the authors of the W2i and GPTF reports is how to turn a norm that most states have agreed to into action. Although both reports argue that self-interest linked to national security threats is the answer rather than appeals to moral duties and obligations, it is still the case that operationalizing R2P must include a discussion of how and why norms are upheld (or not). In this concluding section I would like to very briefly consider, theoretically at least, the relationship between norms and action and whether the legal codification of R2P would up the reputational states, as it were, for states that fail to uphold their responsibility to engage in robust atrocity prevention.
The translation of norms into policy involves two closely related processes: norm internalization and policy-making (Reus-Smit, 2004;Sandholtz & Stiles, 2009;Wendt, 1999). Norm internalization is necessary for policies to emerge that reflect norms to which actors have publicly committed themselves in some way to other actors, be it specific domestic constituents, the general public, other states, non-state actors, or international organizations. This step is linked to selfinterest, something the two reports imply but do not argue explicitly. For states to initiate the practice of complying with norms, compliance at first needs to be seen as a matter of national self-interest. But because national self-interest can in certain circumstances lead actors to calculate that non-compliance is in their own interest, national self-interest is an unstable founda-tion for norm compliance in the long run. For consistent norm compliance, norms need to become integral to an actor's self-conception, a component of one's own identity, and thus reflected in actions or policies in such a way that compliance becomes simply a matter of "this is what we do" because "this is who we are" rather than calculations about whether compliance with a norm furthers self-interest.
The problem is that norm compliance through policies designed to prevent and stop mass atrocities is unlikely to be internalized because atrocity prevention in non-Goldilocks cases-which constitute the majority of real-life cases-is not in the self-interest of western states, or the risks associated with norm compliance are calculated to be too high. As a result, the habit of consistently turning R2P into policy has not yet been developed and thus the first step on the road to norm internalization has not been taken by most states. In instances where R2P has been implemented it has been linked to meeting real threats to the real selfinterest of the interveners. This suggests that if the process of R2P norm internalization is happening at all it is stalled at the initial self-interest stage. Thus both Goldilocks and non-Goldilocks cases of self-interested compliance and non-compliance respectively seem to ensure that the R2P norm is destined, at least for the foreseeable future, to lack internalization in a way that a constructivist approach suggests would lead to the implementation of the R2P norm regardless of where mass atrocities occur. Even if the R2P norm were to be fully internalized there is no guarantee that norm internalization itself will always produce norm-compliant policies as has been the case, for example, of repeated non-compliance with the widely recognized norms concerning the treatment of refugees (Sandholtz & Stiles, 2009).
Could the codification of R2P in public international law overcome this norm internalization problem by raising reputational stakes through the codification of enforcement measures and sanctions that are applied to states that fail to engage in robust atrocity prevention? Would upping the reputational ante help transcend compliance with R2P from reputationally-motivated national self-interest to a matter of national identity and legitimacy; that is, the full internalization of the antiatrocity norm by foreign and defence policy makers?
There are arguments to be made that legal codification might have this effect. First, states that negotiate international treaties do so under the general principle of pacta sund servanda (bargaining in good faith). Thus states as a matter of principle, and I would suggest reputation, promise to bargain and uphold international law for the reasons they state publicly and not some pernicious ulterior motive. Second, states can agree to provisions in treaties that tie their good reputations to compliance in ways that are not possible in CIL or written declarations by including clear legal obli-gations, mandatory dispute resolution and enforcement mechanisms, and robust monitoring capacity. Research suggests that the presence of these kinds of provisions tends to increase compliance since states calculate that violations will be more obvious (there are explicit rules that would be seen to be broken) and be met with mandatory tangible punishments (Simmons, 2008, p. 195). Finally, international law legitimizes behaviours that are codified as "legal" and constructs the identity of actors that uphold international laws as legitimate actors in good standing in the international community (Simmons, 2008, pp. 196-197).
There are, however, very strong headwinds that would likely thwart or at the very least seriously diminish the effectiveness of a codified R2P treaty in terms of fostering norm internalization tied to reputational stakes. While scholars and legal practitioners have made compelling arguments concerning the moral obligations states owe to those requiring protection, as a practical matter, it is hard to conceive of exactly what enforcement mechanisms, particularly sanctions, states would realistically consent to include in such a treaty. We must remember that international treaty law only contains what signatory parties can agree to. While we know that robust monitoring and enforcement mechanisms can change state behaviour in favour of treaty compliance, an explanation for why states would voluntarily agree to include these measures in something akin to a global Good Samaritan law in the first place is less clear. It would likely require states to already have internalized the anti-atrocity norm; precisely the thing that the codification of R2P is meant to produce. To paraphrase Rousseau, "states would have to be prior to laws what they ought to become by means of laws" (Rousseau, 2003). If past human rights instruments like the International Covenants on Political and Civil Rights and Social and Economic Rights are any indication, we would end up with a strong statement of legal obligations to protect vulnerable populations but with weak monitoring through self or third party reporting, and even weaker enforcement.
But perhaps the biggest barrier to fulsome R2P compliance is that identified by Sandholtz and Stiles (2009) noted above. Even when, as in the case of the refugee regime, states universally sign on to international treaties that contain strong legal obligations to uphold a norm that they have fully internalized, violations of these laws and norms by states still occur. The unfolding refugee crisis in Europe is sad testament to this fact. Thus even if states were somehow able to agree to a treaty that clearly articulated their legal obligation to protect and they internalized this norm as an obligation they ought to uphold, there is no guarantee that widespread compliance with R2P would be the result. In the more realistic circumstance in which states would agree to a treaty with weak enforcement provisions, we would be no further along than we are now with R2P as a statement of principle akin to customary international law. Reputational stakes would still be at play in the self-interested way I described earlier with little threat of adverse reputational or material consequences for western and other states that shirk their humanitarian responsibilities.
One solution to this problem is to take from the W2i and GPTF reports the insight that atrocity prevention and intervention is only going to be saleable to states if it can be shown that national security and other tangible interests are at stake. But unlike in the reports we need to recognize that appeals to security threats and self-interest will only be persuasive for regional states directly and negatively affected by the outbreak of atrocities. In the cases of Kosovo, Libya, and the current attempt to degrade the Islamic State, the intervening states are western states acting to counter proximate regional and global threats and to safeguard their reputations. In cases outside the Goldilocks zone the "first responders" to atrocities are, or could be, regional states whose security and interests are directly at stake. These regional actors could be supported by western states that are committed in principle to the R2P norm as a self-interested reputational, but not a security or strategic, issue.
This solution rests on the principle of subsidiarity. Subsidiarity claims that "each human individual is endowed with an inherent and inalienable worth, or dignity, and thus the value of the individual person is ontologically and morally prior to the state or other social groups" (Carozza, 2003, p. 42). The primary responsibility for upholding human dignity and human security lies with associations, organizations, and authorities at the "lowest" level in a system.
A more explicit embrace of subsidiarity in the implementation of R2P might facilitate genocide prevention and intervention in such a way that action could become more likely and in a way that does not open the door for neo-colonial interference. First, if R2P is seen as primarily the responsibility of states and other actors most directly affected by the ripple effects of atrocities, the world will no longer wait, likely in vain, for just the west to intervene in instances where the west's fundamental national security is not at stake. The operationalization of R2P based on subsidiarity would thus make R2P a truly collective responsibility of the entire international system and not solely the United States and other western countries. This shift could not only spur on timely intervention by regional actors, it may also help diffuse concerns that R2P is really a mask for neo-colonial interference in the affairs of societies in the Global South. Second, western states might be more willing to lend material and other support to local or regional responders since doing so would allow western states to uphold and "own" the R2P norm as part of their national identity but without having to spend their own blood and treasure to do so.

Conclusion
In sum, a case can be made that genocide prevention and R2P more generally should be linked to western states' national self-interest. Prevention as self-interest should not, however, be couched in terms of the west's own security but rather concerns for safeguarding states' reputations grounded in conceptions of themselves as good global citizens willing to assist firstresponder states directly affected by atrocities. Although I have focused in this article on critiquing the prevention-as-security thesis and offered an alternative self-interest approach to fostering the consistent and robust operationalization of R2P, it is important not to dismiss out of hand the impact of humanitarian arguments. As the GPTF and W2i reports and the brief discussion of the Libyan intervention demonstrates, post-Rwanda, western policy-makers sometimes do genuinely believe that they have a responsibility to protect. Alas this belief is most likely to arise when the killing happens in the "Goldilocks" zone.