Leading from the front: America,Libya and the localisation of R2P

Introduction

At the time of the NATO-led implementation of United Nations Security Council (UNSC) Resolution 1973 (UNSC, 2011b) authorising force to be used to protect civilians at risk from Gaddafi’s forces, the Obama administration allowed the world to believe they were ‘leading from behind’. This beguiling phrase elides the fact that United States diplomacy critically shaped the character of the United Nations (UN) mandate and that the US worked actively with other Security Council members to ensure the vote was won inside the Security Council chamber. In short, the US gave diplomatic leadership both inside and outside the UN to help resolve a humanitarian crisis that threatened to spiral out of control.

Should we be surprised that the world’s hegemon – a liberal hegemon as G. John Ikenberry (2011) reminds us – is prepared to take the risks and pay the costs of a humanitarian intervention? One notable author of a history of America’s response to genocide would answer ‘yes we should’ and without qualification: ‘the United States had never in its history intervened to stop genocide and had in fact rarely even made a point of condemning it as it occurred’. These words were not written by an angry political exile; they are of course Samantha Power’s words, President Obama’s long-time advisor (2003: xv). Power has subsequently been credited with pushing hard to develop and implement a policy on atrocity prevention and response – suggesting that this administration is seeking to confront past policy failures.

The US government’s focus on genocide and mass atrocity prevention/protection (GMAPP) is a domestic adaptation of the global norm known as ‘the responsibility to protect’ (R2P). Norms are often open to significant variance in interpretation and application – particularly when the state in question is a world power. To illustrate this dynamic, Part 1 of the article draws on Amitav Acharya’s (2004, 2009) work on ‘localisation’ to reveal what properties of the R2P norm are being grafted onto the United States’ atrocity prevention framework and analyses the early institutionalisation of the GMAPP norm inside the United States government (USG) and wider civil society. All norms have an identifiable ‘DNA’ that can be observed in different institutional domains – we examine how far the patterning of the norm’s code can be identified in the domestic political realm.

Part 2 of the article considers how far United States diplomacy and military action against Libya can be understood in relation to the localised variant of R2P. In so doing, we are not seeking to advance a causal story as this would imply that such claims could be tested; instead, we regard the configuration of the R2P norm – set out in Part 1 – as being primarily of heuristic value. US military action in Libya is not significant in and of itself; the US has intervened militarily in other crises when vital interests were at stake and justified these interventions (partly or wholly) on humanitarian grounds. Where the norm is significant, however, is that it is uniting various policy-tools, preferences, interests and values to expand the parameters for conceptualising and responding to atrocity situations.

Norm localisation and the international–domestic linkage

Although the extent to which a norm resonates domestically is influenced by how closely it ‘fits’ pre-existing social and normative structures, its content is not static (Krook and True, 2010) and international norms often purposefully lack specificity in order to facilitate their initial adoption (Sandholtz, 2008). This allows for considerable variation in how norms are then internalised and implemented at the domestic level. We should therefore not be surprised that the content of international norms may change as they encounter the domestic social and normative context. As Acharya highlights, the complex process of building congruence is dynamic and bottom-up as well as top-down. He calls this process localisation, whereby ‘norm-takers build congruence between transnational norms … and local beliefs and practices’ (2004: 241). For Acharya (2004: 247–250), several processes facilitate localisation: firstly, the belief that existing norms are inadequate for specific challenges; secondly, efforts by credible insider proponents who champion and actively seek to reshape the norm in ways that facilitate its internalisation; and finally, the belief that the new norm will enhance the legitimacy of domestic actors and institutions without undermining crucial elements of the domestic social and normative order, such as domestic institutions or national identity.

Norms, in turn, shape behaviour both constitutively and instrumentally. In terms of the former, as a norm is localised and becomes embedded in domestic institutions, processes, discourses and identities it influences the parameters for conceptualising and responding to situations to which it is applicable. There are, however, also instrumental reasons for acting in accordance with norms and which make it more difficult to deviate from. Rhetorical entrapment and the desirability of consistency between rhetoric and behaviour, for instance, make norm accommodation advantageous (Risse, 2000; Rublee, 2008: 427–428).

Given that R2P lacks formal legal standing among sovereign states, ¹ the best that can be claimed for its status is that it is an international norm. We regard a norm as a standard of appropriate conduct that has regulative and constitutive properties. Because of these properties, endorsing a norm generates an impetus to behave in a manner consistent with it and failure to live up to agreed standards of conduct invites scrutiny and criticism. The R2P norm is regulative in so far as it prohibits certain actions (genocide, ethnic cleansing, war crimes and crimes against humanity) and constitutive in so far as it assigns roles and responsibilities to different actors within international society such that membership of that society presupposes being a ‘responsible’ sovereign.

The norm developed in a series of diplomatic debates in the midst of significant negotiation around key principles, thresholds and distributions of responsibility. What emerged in the 2005 World Summit outcome document (WSOD) – and was later endorsed in UNSC Resolution 1674 (2006) and 1894 (2009) – reflected the evolving diplomatic consensus about how the international community should respond to mass atrocities (Bellamy, 2009: 37–41; Thakur, 2011; Weiss, 2007). Paragraph 138 of the WSOD begins with a clear statement about the sovereign state’s responsibility to protect, before progressing to the more loosely defined duties that are ascribed to the international community if the target state is unwilling or unable to provide for security from mass atrocity crimes. Despite this consensus, however, the meanings attributed to the norm vary considerably. Even liberal states – whose self-identities include the promotion of values such as human rights, political freedoms and freedom from persecution – vary considerably with each other in how they align their responsibility to protect with domestic thresholds and identity.

Although the US has long been a proponent of democratic values and human rights, its poor record of responding to mass atrocities in Rwanda and Darfur and slow responses to atrocities in Bosnia and Kosovo weakened its credibility and legitimacy as a supporter of humanitarian values. These situations demonstrated that existing priorities and tools could not sufficiently address the challenge posed by mass atrocities in places where the US did not possess strong strategic interests. Moreover, the catastrophic consequences of failing to respond in Rwanda made a mockery of the US’ repeated pledge of ‘Never Again’ and has since provoked retrospective apologies from Bill Clinton (1998) and other members of his administration, including Susan Rice (2011a).

The norm of R2P taps into these strong sentiments of regret and has galvanised Rice, Power and other members of the Obama team as well as elements of civil society and academia to consider how the international norm could be best accommodated within the domestic context. ² In addition, according to the incoming Obama administration, the US suffered from a legitimacy crisis, and a renewed commitment to multilateralism and mass atrocities prevention were two areas deemed important to improve the United States’ international standing (Obama, 2007: 7, 10–12). R2P touched on these issues but was sufficiently compatible and indeterminate enough to adapt to US identity, values and interests.

The localised norm (GMAPP) endorses much of the R2P framework, shares its broader normative agenda to respond to genocide and mass atrocity crimes and largely adopts the UN Secretary-General’s implementation strategy (Ki-moon, 2009). However, GMAPP is situated in the context of US national security priorities and its identity. As such, R2P is localised in ways that make it more appropriate to the domestic context. Some features of GMAPP suggest innovation, which may ultimately enhance both domestic and international variants of the norm and their implementation, while others reproduce the tensions and challenges present in the international norm. Figure 1 maps the international/domestic configuration of the R2P norm as it has been shaped during the first Obama administration. The diagram also maps where the localised R2P norm is endorsed or resisted. The processes by which institutionalisation ³ occurs are not discussed in detail below – although they will be picked up in our discussion of the Libya case in Part 2.

OPEN IN VIEWER

Figure 1.

United States Localisation and Implementation of R2P as GMAPP. ⁴

According to R2P, each state has the primary responsibility ‘to protect its populations from genocide, war crimes, ethnic cleansing and crimes against humanity’, and crucially this pertains to all states (United Nations General Assembly (UNGA), 2005: para 138). Pillar 1 of the Secretary-General’s implementation strategy requires all states to monitor and defuse domestic social tensions and human rights violations that are pathways to mass atrocities (Ki-moon, 2009). The localised GMAPP norm is blind to the possibility of a mass atrocity situation occurring within the United States, in part because of long-standing checks and balances on government branches, constitutional guarantees and domestic laws which arguably diminish the likelihood that the USG could be an R2P offender (at least in its territory). The absence of US consideration of its domestic responsibility to protect is indicated in Figure 1 with the shadow typeface. Instead GMAPP focuses exclusively on addressing situations at risk abroad, despite the fact that some of the early precursors for mass atrocities, such as social cleavages, ethnic tensions and hate speech do exist domestically (Scutari, 2009). ⁵

Yet governments are not the sole perpetrators of mass atrocities or importantly their incitement, and incidents that could potentially be characterised as incitement have occurred in recent years, such as the release of a film vilifying Islam and the burning of a Koran by a Florida-based pastor (Banks, 2011; Hunter, 2012). Both incidents sparked mass protests and violence internationally, but neither triggered a domestic debate about the United States’ responsibility to protect. The constitutional guarantee of free speech adds greater complexity to these ‘domestic’ issues, yet this too is absent from the GMAPP agenda. Implementation of Pillar 1 also involves joining and strengthening relevant institutions, including instruments of international law such as the Rome Statute of the International Criminal Court, which the USG remains opposed to doing.

Pillar 2 in R2P is dedicated to international assistance and building the capacity to protect based on the logic that earlier detection, before large-scale loss of life, will help to resolve potential crises before conflicts become intractable. Implementation of Pillar 2 in R2P includes monitoring and developing early warning systems, assisting foreign counterparts, providing international assistance and helping build capacity in ‘at risk’ states (Ki-moon, 2009; UNGA, 2005). In contrast, assistance and capacity-building in GMAPP, are primarily directed inward and focused on overcoming internal deficits, which is at odds with the international norm’s focus on developing the capacities of at risk states, although the relationship between the two is logically sequenced: GMAPP’s focus on domestic coordination and planning ought to better promote and consolidate R2P’s Pillar 2 (The White House, 2012a). The USG shows some innovation on this front, including evidence of early institutionalisation (Jentleson, 2012).

In 2009 President Obama appointed Power as director of Multilateral Affairs and Human Rights (MAHR) in the National Security Council to lead on R2P issues. In August 2011, Presidential Study Directive 10 (PSD-10) authorised an interagency review of preparedness and response capabilities. Reiterating a claim made in 2010, PSD-10 found that capabilities exist in government agencies but they are not ‘integrated and focused on the problem in a sustained way’ (QDDR, 2010: 122–123). The Atrocities Prevention Board (APB) – comprised of high-level representatives from various government agencies – was created to overcome this hurdle and coordinate a whole-of-government strategy. Internal capacity-building efforts centre on developing military doctrine and a civilian training component for mass atrocities response (Genocide Prevention Task Force (GPTF), 2008: 30; see also Sewall et al., 2010).

In relation to Pillar 3 responsibilities, the US variant of R2P reflects many of the disjointed policy formulations of other world powers. To date, it has failed to clarify the conditions under which coercive instruments would be used to respond to mass atrocities. For instance, it does not answer divisive questions about who should intervene (unilateral or multilateral), when (threshold, conditions), where (selectivity) or how (strategy, doctrine, command and control). As argued elsewhere, the indeterminacy of R2P was due to political necessity in the lead up to the World Summit and the nature of the UNSC (Bellamy, 2009; Evans, 2008; Luck, 2006). Statements on US prevention/protection policy also invoke the weak wording of the WSOD, which emphasises the determination of action on a case-by-case basis. This allows the United States to preserve its freedom of manoeuvre, which remains important domestically to help build consensus on the fledgling norm.

Tension between humanitarian values and strategic interests also remains evident in the implementation of GMAPP. ⁶ The USG tacitly acknowledges that, in the past, responses to mass atrocities were impeded by lack of domestic political will due to competing interests that won out over a ‘protection of civilians’ agenda (Darfur 2004–2006) or lack of interests (Rwanda 1994). The first Obama administration attempted to address the problem of political will by recalibrating the relationship between interests, values and identity. Central to the rationale of the domestic norm is that ‘when tensions threaten to escalate to mass atrocities, [United States] core values as well as our security interests are deeply threatened’ (QDDR, 2010: 122). Likewise, PSD-10 declares that ‘preventing mass atrocities and genocide is a core national interest and a core moral responsibility of the United States’ (Obama, 2011b). In other words, the localised norm embraces a broader view of national interest, which destabilises what Obama terms as the ‘false choice’ between interests and values (The White House, 2010: 5).

This recalibration is not a rejection of the idea that the international community has a responsibility, which is the foundation of R2P, but rather that atrocity prevention and response is not only about international responsibility but that it is also a core goal of US national security strategy. Advancing this moral argument is critical to reducing domestic resistance and creating domestic political will: even so, reconceptualising mass atrocities prevention as a core interest and linking it to national identity and values does not eliminate prudential concerns. Nor will it necessarily prevent competing interests from prevailing. In a speech endorsing R2P, for instance, Ambassador Rice (2009) acknowledged that ‘humanitarian requirements will often jostle with other legitimate policy concerns’.

Continuities remain between where other administrations set the parameters for humanitarian intervention and those specified in the evolving domestic norm (Clinton, 2011b; Gates, 2011b; Obama, 2011e). ⁷ Decisions to intervene militarily for humanitarian and partially humanitarian reasons have and will continue to balance reasonable hope of success, costs and its foreseen impact on other interests. Thus, there will be times when military action is deemed unacceptable or inappropriate, as the crisis in Syria continues to demonstrate. However, this does not mean that by-standing is a viable policy option; instead, both R2P and its domestic variation attach great importance to broadening responses to mass atrocities to enable earlier and possibly preventative action and to ensure that other policy tools (besides the use of force) are available.

One of the tasks of the APB is to develop a wider range of policy tools to address mass atrocity situations. Don Steinberg, deputy administrator for the United States Agency for International Development (USAID) and APB member, reminds us that the USG has a ‘toolbox’ for responding to genocide and mass atrocities, with military action being a credible last resort. But, diplomacy, travel bans, technological deterrents, humanitarian relief, accountability mechanisms, justice capabilities and sanctions are all preferred options (The White House, 2012a, 2012b). R2P/GMAPP are changing the parameters of engagement so that the choices are not narrowed to either military intervention or allowing atrocities to happen. Domestically, this has helped to minimise resistance to the norm though certainly not overcome it.

For both R2P and the localised GMAPP norm, collective action is vital when force is deemed necessary, but the rationale for prioritising collective action is different. Collective action taken through the Security Council is the cornerstone of Pillar 3 in R2P because it confers international legality and legitimacy and because it enabled international consensus on the WSOD by serving as a check on powerful states. Legality and international legitimacy are also characterised as important considerations for the USG in GMAPP. The desirability and prioritisation of action taken through the Security Council has been repeatedly highlighted by the Obama administration, but there is a risk that action may be blocked by a veto from one of the Permanent Five members (GPTF, 2008: 96).

The USG has not specifically ruled out taking action outside of the council in a mass atrocities situation or the unilateral use of force when America’s vital or core interests are at stake. National Security Strategy (NSS), for instance, states that ‘The United States must reserve the right to act unilaterally if necessary’ (2010: 22) and that ‘No international order can be supported by international institutions alone’ but must require bilateral, multilateral, and global strategies to ‘build new spheres of cooperation’ (p. 40). The careful phrasing of the NSS suggests that the USG will continue to avoid making statements or policies that could rhetorically entrap it or restrict its choices (see, e.g., Risse, 2000). Despite the attempt to reframe atrocity prevention as a national interest (Obama, 2011b), there is little to suggest that a humanitarian interest alone would be sufficient to justify the use of force taken without a covering Security Council resolution and international participation. Indeed, in relation to Libya, a covering UNSC resolution was deemed critical for the military operation to proceed (Corn, 2012: 208).

For the USG, legitimacy is more broadly associated with multilateralism; collective action through the council is simply the preferred means of achieving it. With that in mind, the NSS commits to enhancing cooperation with the UN but also with regional organisations which ‘can be particularly effective at mobilising and legitimating cooperation among countries closest to the problem’ (p. 46). Collective action through the UN and multilateralism are also prioritised as a means of burden-sharing (Andrews, 2011; Mitchell, 2011; The White House, 2010: 1). In a telling commentary on US engagement with the UN, Esther Brimmer (2011), Assistant Secretary at the Bureau of International Organization Affairs, argued that ‘by working through the United Nations, we help bring security to countries where U.S. military operations aren’t feasible or desirable – at far lower cost to the United States – and where U.S. leadership can leverage important contributions by other states.’ The administration’s position on multilateralism surfaced in deliberations over military intervention in Libya, with representatives repeatedly stating that unilateral (military) action was not an option because it would lead to unintended and detrimental consequences (Andrews, 2011; Inskeep, 2011). In addition, the role of (selective) regional organisations was critical, as Part 2 will discuss.

The legislative branch is also cautiously moving the GMAPP agenda forward. In December 2010, the Senate unanimously passed senate concurrent resolution 71, which supported efforts to develop a government-wide strategy for preventing and mitigating mass atrocities. It also affirmed ‘that it is in the national interest and aligned with the values of the United States to work vigorously with international partners to prevent and mitigate future genocides and mass atrocities’ (S.Con.Res. 71, 2010).

As we have argued, there are multiple areas of convergence between the international and localised norm. Yet other aspects of the international norm and its implementation are excluded from domestic level dialogue. As Part 2 will show, certain aspects of the localised norm were applied unproblematically to the situation in Libya, including: the US could not be a bystander as such behaviour runs contrary to its identity; earlier (preventative) action would be less costly in the long-run; a range of responses was appropriate; a multilateral strategy was necessary to diffuse the economic and reputational costs of action; and no US ground troops should be deployed.

The norm of R2P and its localised variant should not be viewed as ‘causing’ particular actions in the Libyan case. This suggests a deterministic model that is contrary to most constructivist accounts of normative change. However, the norm was clearly ‘at work’ in shaping the expectations and parameters for acceptable behaviour in the Libyan crisis. Within these parameters there remained a lot of room for strategic, moral and mixed motives and actions. We argue below that the evolving normative structure created opportunities for early and wide-ranging responses that were consistent with the localised norm. This suggests at minimum that policymakers were able to draw on the tenets of the localised norm to strengthen their case for when and how to act.

The United States, Libya and R2P

By the beginning of February 2011 there were initial signs of growing support for the localised version of R2P within the executive and legislative branches of government as well as civil society (Obama, 2012; Power, 2011; S.Con.Res. 71, 2010; e.g. GPTF, 2008; The Stanley Foundation, 2008). The crisis in Libya, although preceding much of the institutionalisation of GMAPP within government agencies, nevertheless helped to consolidate the fledgling norm and gave the Obama administration an opportunity to live up to early campaign promises to act on American principles because, as Obama stated in the second presidential debate of 2008, ‘we have moral issues at stake’ and the US is diminished ‘when we stand idly by’ (NBC News, 2008). For administration supporters of GMAPP/R2P, Libya was an obvious test-case for the USG’s commitment to atrocity prevention and response.

The Libyan case provides fertile empirical grounds for understanding to what extent the ‘DNA’ of the GMAPP norm is evident in the United States’ response to a mass atrocities situation. The norm shaped the parameters of the USG’s response in relation to the decision to take principled action and also pragmatic considerations for how this would be done. These features suggest an alignment between USG responses and the localised atrocity prevention norm, and we locate the norm’s presence in the following three places: first, norm resistance (mindful that contestation requires recognition of the norm’s potential influence on behaviour); second, the norm’s domestic legitimacy; third, leadership in relation to GMAPP.

The norm provided a compelling frame through which to view the crisis, which contrasted with other possible, less persuasive interpretations, such as a terrorist action. Gaddafi’s alienation of much of the international community, lack of compelling US economic interests in Libya (although other states including China, India and Italy had significant economic ties), as well as incidents such as the Lockerbie bombing all gave the regime a pariah status that served as an enabling factor. Yet, this alone is insufficient to explain US and international action against the regime. Indeed, a strong regional predisposition towards non-intervention and the association of interference in the domestic affairs of another state with neo-imperialism were countervailing factors that, especially in the African Union (AU) case, opposed military intervention.

R2P and the notion of sovereignty as responsibility were normatively and rhetorically crucial in helping to overcome countervailing factors and to establish a compelling cause to act in Libya, prior to large-scale loss of life. Within the USG, the norm was also at work in linking rhetorical commitments to R2P and promises of ‘Never Again’ with strong memories of past failures to respond to the Rwandan genocide and the fear of damage to the United States’ identity as a world leader and a proponent of responsible sovereignty and human rights if the USG once again failed to prevent mass atrocities.

The desirability of prevention and early response also played a fundamental role in the way international engagement unfolded. In his remarks to the Security Council on 25 February, the Secretary-General reported over 1000 regime-attributed deaths in Libya (UNSC, 2011a). While this figure is considerable, much higher fatalities in other conflicts had not provoked similar or concerted international action. Momentum against the regime developed as much in response to Gaddafi’s defamatory rhetoric, public incitement and promises to wipe out protesters as it developed in response to the facts on the ground. In essence, the international community mobilised to prevent mass atrocities primarily on the basis of speech-acts which were treated as if they were reality.

The norm also set the parameters of acceptable behaviour. Importantly, by-standing or waiting for large-scale loss of life before taking action was not an option. As Obama (2011d) subsequently argued:

the core principle that has to be upheld here is that when the entire international community almost unanimously says that there’s a potential humanitarian crisis about to take place, that a leader who has lost his legitimacy decides to turn his military on his own people, that we can’t simply stand by with empty words; that we have to take some sort of action.

Early on the USG accepted it had a responsibility to protect Libyan civilians, and officials widely agreed that a range of policy tools should be used to change Gaddafi’s behaviour, including: public condemnation, economic sanctions and diplomatic actions vis-a-vis the Libyan regime to halt the violence and vis-a-vis other states to isolate the regime. This constellation of non-military responses as well as the escalation of responses is consistent with Pillar 3 of R2P and GMAPP.

Though there was some criticism that USG responses were not strong enough or quick enough, consensus within the administration did not break down until peaceful instruments were deemed insufficient and the use of force became a serious consideration. Even then, resistance in the administration and the wider public sphere focused primarily on prudential concerns, namely that the situation on the ground was sufficiently uncertain that military action generated undue risks, that proposed responses were not proportionate to the crisis, and that the scope and mission were poorly defined. There was also some resistance to the view that intervention for humanitarian protection purposes should be elevated to the level of a core national interest. Senior officials in the Department of Defense, for instance Tom Donilon, argued that Libya was not ‘a vital strategic interest’ (European Affairs, 2011). This constitutes a deep challenge to a central tenet of the localised norm while at the same time not calling into question the standing of R2P in the international community.

Early on NATO Secretary-General Rasmussen argued that the crisis in Libya did not pose a threat to NATO or its allies, and thus it had no plans to intervene (VOA, 2011b). In contrast, on 25 February, Obama (2011a: 1, emphasis added) issued Executive Order 13566 which framed the crisis as a national security priority:

the prolonged attacks, and the increased numbers of Libyans seeking refuge in other countries from the attacks, have caused a deterioration in the security of Libya and pose a serious risk to its stability, thereby constituting an unusual and extraordinary threat to the national security and foreign policy of the United States, and I hereby declare a national emergency to deal with that threat.

Consistent with the GMAPP norm, this executive order framed the crisis in internationalist language whereby a humanitarian catastrophe in Libya posed a security problem for the United States and its allies.

Executive Order 13566 also authorised the seizure of assets belonging to specific regime agencies, officials and family members of Muammar Gaddafi and restricted financial transactions between United States citizens and named individuals. Not only did USG sanctions predate those imposed by the council under Resolution 1970 on 26 February, they were also more comprehensive (Lindsay and Lyons, 2011). The decision to impose unilateral sanctions and support multilateral sanctions in the council was not contested within the administration.

Consensus building through the Human Rights Council (HRC) was an additional avenue of USG response. According to Brimmer in early February, the US delegation had, in her view, ‘become the most active delegation on the Council’ and contributed to timely action by the HRC in response to human rights crises in Kyrgyzstan and Cote d’Ivoire in 2010 (Brimmer, 2011). On 25 February, the HRC, with the support of the US and 20 other member-states, convened a special session on Libya. The resulting resolution added to the momentum against the regime. The following day Resolution 1970 passed unanimously in the Security Council, and the USG supported – for the first time – a referral to the International Criminal Court (ICC). On 28 February, Secretary of State Hillary Clinton (2011c) spoke at the HRC urging member-states to suspend Libya’s membership in the HRC. Consistent with the approach outlined in GMAPP, the USG used multiple and varied policy tools and multilateral fora to influence regime behaviour.

As domestic and international attention turned to the possibility of a no-fly zone or other military options, domestic resistance coalesced around the argument that prudential concerns precluded effective military action. Secretary of Defense Robert Gates, for instance, underscored the undesirability of another intervention in the region: ‘we also have to think about, frankly, the use of the U.S. military in another country in the Middle East’ (Buel, 2011; Bumiller, 2011). This and other negative comments by Gates and senior military officers were later moderated by a Pentagon press secretary (Agence France-Presse (AFP), 2011). Yet, Clinton also downplayed the no-fly zone option, and it initially seemed that those opposing military involvement had the upper hand (Buel, 2011; VOA, 2011a).

Subsequent interviews given by Obama and administration officials clarified the preconditions for military intervention set by the administration, that: military action must be multilateral, in accordance with international law, avoid using ground troops, effective, and have a well-defined and achievable goal (Hastings, 2011). As argued in Part 1, the norm of GMAPP avoids setting out a specific criteria for when and how the USG will respond – to preserve as much latitude as possible for the executive and Congress. However, when it came to implementing GMAPP in Libya, the justification for the use of force came close to setting out preconditions which could become a precedent for future interventions (Sanger, 2012: 351).

The precondition that any military action be multilateral was to reduce the economic and political costs of intervention and to garner and preserve domestic and international legitimacy, but regional support was crucial in the decision to intervene (Obama, 2011b; QDDR, 2010: 20, 55–57; The White House, 2010: 12–13, 40–50, 2011). The administration pushed the criteria of multilateral support farther than other recent administrations. Strong regional support was identified as a priority instead of reliance on ‘like-minded’ allies, ⁸ which has left the United States open to claims that it is acting only in the collective interests of the Anglosphere. Gates (2011a) and others frequently emphasised US sensitivity to the region: ‘We are very mindful of opinion in the region… [We are] very sensitive to NATO being responsive to those organizations [Arab League, African Union, and UN] rather than taking an initiative on its own’ (Carney, 2012; Donilon and Rhodes, 2011). The administration also pushed for active participation rather than mere rhetorical support. According to Chairman of the Joint Chiefs of Staff, Admiral Mike Mullen (2011), the coalition ‘is not only a coalition of the willing; it is a coalition of the able with each nation bringing to the effort what they can in terms of knowledge and skill.’

In early March, support for a no-fly zone began to coalesce. On 1 March the US Senate unanimously passed a non-binding resolution condemning the violence perpetrated by Gaddafi’s regime and encouraging the UNSC to consider a no-fly zone (S. RES. 85, 2011). ⁹ On the same day the General Assembly – at the urging of the HRC and Clinton, among others – suspended Libya’s membership in the HRC. From 7–9 March the Gulf Cooperation Council, Organisation of Islamic Cooperation and Libyan Interim Transitional National Council all publically endorsed a no-fly zone. By 10 March NATO had also shifted position, with Rasmussen arguing that intervention by NATO was conditional on demonstrable need, clear legal basis and firm regional support (Ryan and Brunnstrom, 2011). Finally, on 12 March the Arab League also called for a no-fly zone. Discussions with EU and NATO officials began in earnest once regional support and the promise of active military and economic contributions by several Arab states became clear. France, Lebanon and the UK began lobbying for a second Security Council resolution which included provisions for a no-fly zone. This push by allies in Europe and key Middle Eastern countries enabled R2P supports inside the USG to support a ‘humanitarian’ military intervention.

Nevertheless, public support in the United States was split. When participants of a 14 March poll were asked whether they would ‘favor or oppose the United States and other countries attempting to establish a “no-fly zone” in Libya’, 56% favoured it, 40% opposed it and 4% had no opinion. ¹⁰ Polling data also showed the limits of public support for military involvement. A Pew Research Center (2011) poll conducted between 10–13 March asked whether participants supported sending ground troops to Libya: 13% of participants favoured sending ground troops, 82% opposed it and 5% were unsure. ¹¹ In keeping with public opinion on the limits of a military intervention, Obama ruled out ‘boots on the ground’ early on. The administration could have used the lack of overwhelming public support as an excuse to avoid escalating atrocity prevention measures. After all, much higher domestic advocacy and support was evident in relation to Darfur in 2005 yet the USG did not seriously contemplate military action, an ICC referral or many of the other tools in the atrocity prevention ‘toolbox’. In the case of Libya, the Obama administration deemed backing down to be more costly than escalation. This suggests at minimum that in broadening the tools and focusing on prevention, R2P/GMAPP is helping to change the parameters of the domestic decision-making process in relation to atrocities (hereby indicating the presence of domestic legitimacy, see Figure 1).

Administration officials resolved that the United States would play a supporting role in any military intervention using its unique military capabilities at the front end but then passing leadership to NATO. Thus any attempt to develop a coalition would necessarily have to consider the position of possible allies on whether or not a covering Security Council resolution was necessary. In the case of Libya, EU and NATO allies made intervention conditional on legal authority (Fam, 2011; Ryan and Brunnstrom, 2011). A covering Security Council resolution would fulfil this requirement, and given the high level of support for a no-fly zone by regional actors and the Libyan opposition, the administration ultimately decided to push for this (Corn, 2012: 212–213).

On 15 March, Obama met with his advisors to discuss draft resolution 1973 and the broader situation in Libya. The proximity of GMAPP supporters to the President, including Susan Rice and Samantha Power, proved to be vital. A media report of the meeting explains the shift in the administration’s position; in the article Obama is quoted as saying ‘we knew that a no-fly zone would not save the people of Benghazi’. It was ‘an expression of concern that didn’t really do anything’ because the real danger to civilians was posed by ground troops not aerial bombardment (Lewis, 2012; see also Cooper and Myers, 2011; Mann, 2012).

Without a stronger resolution than was originally envisaged by other pro-intervention countries, Obama feared the action would be ineffectual, thus breaching one of the administration’s conditions for intervention. According to one participant at the meeting, Obama concluded ‘just signing on to a no-fly zone so that we have political cover isn’t going to cut it. That’s not how America leads’ (Hastings, 2011). For Clinton as well, the USG needed to focus on two things: a workable strategy that had a reasonable hope of protecting civilians and getting it passed in the council (Ghattas, 2011). Later that night, the administration came out strongly in support of an expanded Resolution 1973, which included a robust no-fly zone and civilian protection mandate.

Both Rice (2011b) and Clinton (2011a) reported the USG made intense diplomatic efforts to ensure Resolution 1973 passed. Ambassador Rice (2011a) was tasked with expanding the draft resolution in the UN to allow strikes against ground targets. President Obama discussed the United States’ position with President Sarkozy and Prime Minister Cameron, and the following morning he contacted President Medvedev to ensure Russia would not veto an expanded resolution (Lewis, 2012). He also sought support for the resolution from South Africa (Corn, 2012: 212). A 17 March update on council deliberations, noted the following:

The United States, which had not taken a clear position on a no-fly zone earlier, is now keen to include language that goes beyond a no-fly zone and provides for broader authorization for military intervention if necessary (UNSC Report, 2011).

The question of international authorisation was ultimately addressed when Resolution 1973 was adopted by the Security Council. One of the reasons the administration gave for its support of the resolution was the future credibility of the council and the wider international community. In a letter to Congress, Obama (2011c) describes Gaddafi’s failure to implement Resolution 1970 as ‘a lawless challenge to the authority of the Security Council’. Elsewhere, Obama (2011d) argued that ‘the writ of the United Nations Security Council would have been shown to be little more than empty words, crippling that institution’s future credibility to uphold global peace and security’ and that ‘this matters to us’ in part because ‘the words of the international community would be rendered hollow’ by a failure to respond (Obama, 2011f). US and presidential credibility were also on the line (Corn, 2012: 206–208; Mann, 2012).

The issue of domestic legitimacy proved more problematic and US public support remained ambiguous. Polls taken shortly after the decision to intervene showed a spike in public support, with 70% favouring a no-fly zone, 27% opposed and 3% unsure (CBS News, 2011; CNN/Opinion Research Corporation, 2011b). Yet, there remained clear limits for public support for military action for the protection of civilians. Public opinion favoured United States participation and support over leadership and disagreed that the United States possessed a ‘responsibility to do something about the fighting in Libya between government forces and anti-government groups’ (see, e.g., Polling Report, 2011). Only 27% of participants believed that the United States possessed the responsibility, 63% said the United States did not have the responsibility and 10% were unsure (Pew Research Center, 2011).

Even though the Obama administration set preconditions for a military response, it nevertheless surpassed public opinion in advancing the claim that the US had a national interest in protecting Libyan civilians from mass atrocities, an identity that demanded action in these circumstances, and a responsibility to protect. In a 28 March speech justifying United States support for the military intervention, Obama (2011e) stated:

To brush aside America’s responsibility as a leader and – more profoundly – our responsibilities to our fellow human beings under such circumstances would have been a betrayal of who we are. Some nations may be able to turn a blind eye to atrocities in other countries. The United States of America is different. And as President, I refused to wait for the images of slaughter and mass graves before taking action.

Obama’s statement re-affirms the linkage between international norms and domestic politics. The USG had a responsibility to act because of its identity as a liberal power – an identity that takes seriously the deliverance of duties beyond borders.

The important question for the administration was not whether a responsibility existed but how to respond and when to put military options on the table. Obama’s stated reasons for intervention in Libya focused on humanitarian concerns and civilian protection as well as concerns for regional security and international credibility (see, e.g., Obama, 2011c; Steinberg, 2011). As argued, however, there was some resistance to an expansion of national interests that framed atrocities prevention in Libya as a core national security priority, and even less agreement as to whether American military involvement was appropriate. This tension was notable even after the robust no-fly zone was established.

Less than two weeks after the United States deployed units to join the no-fly zone, Gates stated on national television that Libya was not a vital interest, though he went on to say that the United States had non-vital interests in Libya and the region. Clinton, also present at the interview, tried to deflect Gates’ remark, but even she stumbled over the question of whether the situation in Libya was a vital interest of the United States. Clinton argued that while Libya had not attacked the United States, the vital interests of NATO allies in Europe were impacted, thus it was incumbent on the US to go to their aid. Nowhere did she or Gates allude to Obama’s argument that mass atrocities prevention is both a moral priority and a national interest (Gregory, 2011).

President Obama was clearer than his two senior cabinet colleagues at Defense and State. Consistent with the evolving norm of GMAPP, when Obama explained the intervention, he framed his decision to protect Libyan civilians as both a national interest and a responsibility:

… when innocent people are being brutalized; when someone like Qaddafi threatens a bloodbath that could destabilize an entire region; and when the international community is prepared to come together to save many thousands of lives – then it’s in our national interest to act. And it’s our responsibility. This is one of those times (2011g).

In other words, when the conditions are right and the international community is mobilised, a military response to atrocities is in the national interest. Responsibility is important, but it cannot be the deciding factor alone. The President neatly captured this case-by-case logic, familiar to the Pillar 3 wording of R2P, that ‘the United States should not – and cannot – intervene every time there is a crisis somewhere in the world’ (Obama, 2011g).

R2P consistently balances competing interests with the normative priority to take action in cases of mass atrocities. Neither R2P nor the US variant would require action in cases without a reasonable hope of success. But, in the Libya case the norm helped to shape the parameters for action by framing the situation as an R2P crisis and by delineating what can and should be done to respond to mass atrocities. It also incentivised the Obama administration to act for fear that if the USG did not take reasonable action to respond to mass atrocities it would seen as callously indifferent rather than merely upholding state sovereignty.

Conclusion

In the longue duree of humanitarian intervention (Bass, 2008), rarely has there been a case as straightforward as Libya. Gaddafi found himself in the middle of a perfect storm: the Arab Middle East was undergoing a period of radical change led by social movements; as a leader, he had few allies left in the region; his relationship to Western countries was unstable; and he sent all the ‘wrong’ signals during the uprising, particular his intent to ‘cleanse’ towns and cities and go ‘house by house’ crushing the uprising. Throw into the mix the fact that Libya was militarily weak and located on NATO’s doorstep – and you have an undeniable casus belli.

Yet, this article has shown that the NATO-led action was not as straightforward as is often implied. To get a robust UNSC resolution that could protect civilians and tip the balance of power in favour of the rebel forces associated with the national transitional council, Obama and his administration had to lead from the front – not ‘from behind’, as insiders described their stance at the time. This meant pushing council allies to agree to a robust resolution that could protect Libyan civilians. Leading from the front also meant facing down framings of humanitarian intervention as a practice that was not in America’s vital security interests. And leadership meant taking risks; negligible in relation to the lives of US servicemen and women, but not insignificant politically especially given Obama’s strategy of drawing down troops and commitments following the two long wars of the 9/11 decade. This, after all, was the reason why Vice-President Joe Biden, and other senior figures in the cabinet, counselled the President not to take the military option in Libya.

Leadership inside the Obama administration was critical but so was the domestic and international normative context which meant intervention had greater legitimacy than previously. In the years following the 2005 World Summit, the international norm of R2P has cascaded throughout international society, forging a consensus on states’ domestic responsibilities even if there remain divisions over how to give effect to the international responsibility to protect.

We should not be surprised that international norms are indeterminate given the absence of a world authority capable of turning norms into laws backed by coercive power. Yet despite the lack of binding character, international norms matter in international society: although, as the comparison of Libya and Syria remind us, the way they matter is not the same across distinct cases. To show how they matter it is necessary to look at the grafting of international norms onto domestic political orders – and the mutations that take place during this process.

From the period of the 2005 World Summit to the present, the norm of R2P has been adapted to an American political context. GMAPP signals certain priorities and blind-spots in Washington’s thinking and policy formulation. No attention is paid to atrocity prevention or protection at home, or likewise to potential crimes against humanity committed by American forces aboard; similarly, despite its diplomatic leadership over Libya, the precise conditions under which the USG would be prepared to respond militarily to mass atrocities remains underdeveloped in terms of the policies and processes associated with GMAPP.

To date, what has been said on this issue leads to a consensus trap whereby either the UNSC or another multilateral institution can veto decisive action the USG might otherwise be prepared to take to prevent or mitigate mass atrocity crimes. On the upside, the priority in the government has been loosely connected to Pillar 2 (in R2P language) conceptions of prevention and assistance, achieved through improved policy coordination, planning and information sharing and analysis regarding at-risk countries and peoples.

These differences and convergences between R2P and GMAPP conform to the expectations set out in the literature on norm localisation. In this instance, however, it is not middle and small powers outside the liberal order that are experiencing localisation. It happens to world powers too.