Recovering African contestation and innovation in global politics: Francis Deng and sovereignty-as-responsibility

Abstract

Against both liberal narratives and postcolonial critiques, this article argues that sovereignty-as-responsibility – the theory of sovereignty embraced in the responsibility to protect (R2P) – is part of a problem space that emerged with decolonization, rather than the end of the Cold War. The internally displaced person (IDP), the vehicle which Francis Deng used to critique Westphalian sovereignty, had to be theorized against the rise of the postcolonial state. In recovering the questions motivating Deng, we find a stark politics driving his work on IDPs and sovereignty. Against the claim that the heart of R2P is armed coercive intervention for humanitarian purposes, Deng used sovereignty-as-responsibility to promote a profoundly political critique of the colonial legacy and the postcolonial state, which was taken up by states of the Global South in debates on the ratification of R2P. Recovering Deng's work on IDPs and sovereignty-as-responsibility highlights R2P as itself a site of contestation, and offers a case for how ideas emerge ‘from below’ in global politics.

Introduction: Africa and the responsibility to protect

When Russia invaded Georgia in 2008, Vladimir Putin cynically invoked the responsibility to protect (R2P) to justify the ‘protection’ of South Ossetians and Abkhazians from the Georgian government.¹ In 2014 and again in 2022, Russia claimed the protection of ethnic Russians in Ukraine to justify military invasions, initially calling the invasion of Ukraine in 2022 a ‘peacekeeping’ mission. The night of Russia's 2022 invasion of Ukraine, Martin Kimani (representing Kenya at the United Nations [UN] Security Council) spoke to the questions of conflict resolution and state building without resurrecting the logic of the nation-state:

Today, across the border of every single African country, live our countrymen with whom we share deep historical, cultural, and linguistic bonds. At independence, had we chosen to pursue states on the basis of ethnic, racial, or religious homogeneity, we would still be waging bloody wars these many decades later…We chose to follow the rules of the Organisation of African Unity [OAU] and the United Nations charter, not because our borders satisfied us, but because we wanted something greater, forged in peace. We believe that all states formed from empires that have collapsed or retreated have many peoples in them yearning for integration with peoples in neighboring states. This is normal and understandable. After all, who does not want to be joined to their brethren and to make common purpose with them? However, Kenya rejects such a yearning from being pursued by force. We must complete our recovery from the embers of dead empires in a way that does not plunge us back into new forms of domination and oppression.²

Kimani's position is an interesting vantage point from which to make this argument: Kenya accepted African Union (AU) mediation in 2008, led by Kofi Annan who later claimed to have been implementing the R2P framework, to resolve post-election violence stoked by Mwai Kibaki's supporters against Raila Odinga's supporters.³ Rather than asserting the primacy of non-intervention – a pillar of Westphalian sovereignty and an enumerated principle of the (by-then disbanded) OAU – Kenya became the first state subject to an R2P action.⁴

A prime argument for the legitimacy of the nation-state in the 19^th and early 20^th century was protection and unification of the nation. In a global context of resurgent right-wing nationalism, Russia's adoption of the language of protection of co-nationals to justify its armed incursions into neighbouring sovereign states⁵ may not be simply an attempt to expose the ‘hypocrisy’ of the West, but may itself be a legitimation strategy meant to appeal to right-wing nationalist groups in and outside the West. By contrast, a central aspect of both R2P and sovereignty-as-responsibility is the rejection of the belief that the nation-state is the only reliable institution for protection. Sovereignty-as-responsibility is a concept in which sovereignty is constituted by positive responsibilities to protect (understood broadly) a population, rather than in terms of non-interference.⁶ It is the definition of sovereignty upon which R2P rests. R2P is a UN doctrine, ratified in 2005, which recognizes a positive responsibility of the international community towards vulnerable populations. While under the R2P framework the state retains primary responsibility for protection of its population, the project of sovereignty-as-responsibility is to rethink state building and conflict resolution untethered to the ideology of the nation-state.⁷

By accepting outside assistance under the framework of R2P, Kenya also draws our attention to the breadth of the framework – the logic of R2P is not simply a formula for (Western-led) armed coercive intervention into warring states of the Global South. Importantly, Kimani's argument is distinctly, and temporally, postcolonial. He rejects the romanticism of restoring a pre-imperial world, instead accepting ‘the embers of dead empires’ as a messy starting point which necessarily forecloses some possibilities.

But much of the discourse surrounding R2P and sovereignty-as-responsibility frame these simply as paths to armed humanitarian intervention.⁸ Many liberal accounts offer a genealogy of R2P that focuses on the International Criminal Tribunal for the Former Yugoslavia (ICTY) and NATO's bombing campaign over Kosovo,⁹ presenting R2P as a legal basis for armed coercive intervention in cases of atrocity crimes. By contrast, postcolonial and critical theory accounts tend to treat sovereignty-as-responsibility and R2P as neoimperial assertions of hierarchy in global politics. These theorists¹⁰ treat R2P and sovereignty-as-responsibility as examples of direct Western domination legitimated through ‘ethics’ or ‘humanitarianism’. According to these critiques, R2P and sovereignty-as-responsibility are manifestations of the liberal impulse to define its own prerogatives as apolitical and deny political agency to actors in the Global South. While positing the imperial origins of ‘protection’, they accept the dominant institutional genealogy of R2P, which begins in the 1990s.

The postcolonial and critical theory critiques build on and extend critical accounts of humanitarianism¹¹ and human rights,¹² and deepen our understandings of how some liberals attempt to use the spectre of atrocity to de-politicize armed coercive intervention and other forms of interventionist politics. However, I argue that rather than easy cases for demonstrating neoimperial or Western dominance, R2P and sovereignty-as-responsibility are themselves sites of contestation and offer possible futures for reordering the sovereign states system. Recovering an alternate genealogy of R2P through African contexts from which sovereignty-as-responsibility emerged to provide a foundation for R2P, I hold, offers important insights for international theory about how ideas in global politics emerge ‘from below’.

Sovereignty-as-responsibility is part of a problem space that emerged with decolonization. The internally displaced person (IDP), a key concept in the genesis of sovereignty-as-responsibility, was theorized against the rise of the postcolonial state, during which those who appropriated the state apparatus legitimated themselves by repurposing colonial myths. Recovering Francis Deng's work on IDPs and sovereignty-as-responsibility helps us to recognize R2P as itself a site of contestation, rather than simply a neoimperial imposition. Although some liberals¹³ note the importance of Deng's diplomatic work, they do not critically reconstruct his intellectual contribution.¹⁴ Critical and postcolonial theorists¹⁵ treat Deng primarily as a fellow of the Brookings Institution, the author of several texts from the mid-1990s in which he ostensibly argues for sacrificing Global South sovereignty to Western ‘protection’. This essay, by contrast, treats Deng as central to understanding an alternate genealogy and vision of R2P, and why the doctrine receives strong support from many states of the Global South.

This essay proceeds as follows: first, I outline critical and postcolonial critiques of sovereignty-as-responsibility and R2P, and argue that an alternate genealogy and conception of R2P challenges these critiques. In the second section, I demonstrate Deng's importance to the study of sovereignty-as-responsibility, locating sovereignty-as-responsibility as part of his larger intellectual project, emerging from a twin critique of the colonial legacy and the postcolonial state. Then, I argue that two important moments in the history of R2P – the 2005 UN World Summit and the invocation of R2P as a response to Kenyan election violence in 2008 – challenge both liberal and left (critical and postcolonial) accounts of R2P treating the doctrine as reducible to armed intervention by hegemonic Western powers. I conclude by drawing on South African diplomacy for pointing towards a future for R2P after the 2011 Libya crisis.

R2P as the new imperialism?

In the past two decades, theorists influenced by critical and postcolonial theory have treated R2P and sovereignty-as-responsibility as straightforward examples of direct Western domination legitimated as ‘ethics’ or ‘humanitarianism’. Drawing primarily on the International Commission on Intervention and State Sovereignty's (ICISS) 2001 advisory report on R2P and the 2011 armed intervention into Libya under the authorization of UN resolution 1973, they read R2P as a straightforward manifestation of the liberal impulse to define its own prerogatives as apolitical and deny political agency to actors in the Global South.

R2P is asserted by these critics to be simply a new guise for colonial approaches to global politics: ‘the doctrine of the Responsibility to Protect continues to bear the weight of [imperial] legacies. While it appears to transform the external façade, it continues to embody colonial path-dependence’.¹⁶ Such theorists are suspicious of claims to use force for moral ends. Powerful states ‘prefer to [use] reassuring labels such as “humanitarian intervention,” “protection of the population,” or “defense of human rights.” Here, again, the current form of political messianic zeal recalls that of the colonial period, when conquests were justified by the fight against cannibalism, slavery, and other savage practices, and when war against insurgents was called “pacification”’.¹⁷

These are powerful critiques of liberal human rights rhetoric and important insights into how some actors attempt to use R2P for arguably neoimperial ends; however, how ethical imperatives are pioneered goes unexamined in these critiques. Sovereignty-as-responsibility and R2P become spectres haunting the postcolonial world, emerging from a colonial logic and enforcing a hierarchical order by denying the political agency and claims of postcolonial citizenship. In such critiques, these doctrines cannot but embody the wills of neocolonial states and institutions that do not believe these emancipated former subjects are ready, or could ever be ready, to rule themselves. In essence, they offer teleological accounts in which sovereignty-as-responsibility and R2P must eventually become part of the neoimperial arsenal. Thus, the politics of sovereignty-as-responsibility are hidden behind categorical prohibitions against certain practices. In clamouring for a ‘return to politics’, these authors obfuscate the politics that got us here.

Adom Getachew grounds her analysis in ‘The Limits of Sovereignty as Responsibility’ in the critique of the ethical turn, arguing that, ‘the relationship between principles and practices are viewed as a form of applied ethics where norms are first elaborated and agreed to and then applied and implemented in practice’.¹⁸ This, she holds, creates a ‘one-size-fits-all’ approach designed by ‘experts’ to circumvent the nitty gritty of politics. State capacity building, one of Deng's central concerns in Sovereignty-as-Responsibility: Conflict Resolution in Africa, comes in for especially harsh critique because rather than allowing the people to decide through (presumably democratic) political channels how the state should be organized, this approach relies on ‘expertise’ to dictate state form. Providing protection to its population, as sovereignty-as-responsibility and R2P advocates insist a state must do, is placed in opposition to politically empowering its people (the goal Getachew hopes a new state might strive for).

Even the use of ‘population’, Getachew holds (while acknowledging the significance of this term in genocide studies) is a de-politicizing move that enervates the possibility of a state responsive to its citizens, the more politically active term. Referring to populations ‘depoliticizes those entitled to protection. As populations to be protected rather than political subjects to be engaged, they are rendered passive recipients of the state or international community’.¹⁹ Her critique echoes Philip Cunliffe's²⁰ work placing ‘sovereignty-as-responsibility’ in tension with ‘popular sovereignty’.

Cunliffe argues that sovereignty-as-responsibility ultimately negates political responsibility: ‘For power to be truly responsible, it needs to be at least potentially accountable. Sovereignty as responsibility, however, makes the exercise of power unaccountable, and therefore ultimately irresistible’.²¹ Similar to Getachew, Cunliffe posits that the responsibility of ‘sovereign responsibility’ is to a nebulous international community; the ideal of popular sovereignty by necessity rejects international responsibility to make power always (theoretically) accountable only to ‘the people’. ‘Sovereignty preserves the freedom of a people to be self-determining, not the impunity of the state apparatus. To erode or call into question “absolute sovereignty” is to erode or call into question the idea of representative government and the self-determination of nations…under the banal talk of “states responsibilities” the responsibility to protect doctrine is calling into question people's capacity for and rights of self-determination’.²² In this way, sovereignty-as-responsibility erodes both the political accountability of weak states (which become accountable to powerful states), as well as powerful states (which become accountable only to their ‘international responsibility’).

In Cunliffe's account, sovereignty-as-responsibility is simply an idea that emerged from elites: ‘The fact that panels of eminent persons can set about redefining sovereignty in the space of under 100 pages, on the back of a series of elite consultations, indicates just how far sovereignty has been taken beyond the mass politics that once inspired it’.²³ He describes Deng et al.'s 1996 report Sovereignty-as-Responsibility: Conflict Resolution in Africa as ‘a way of asking African states to pull themselves up by their own bootstraps’.²⁴ In this account, sovereignty-as-responsibility is not an idea that emerged from a historico-political process, but something dictated by elites who cared not for mass politics and popular sovereignty.

Getachew and Cunliffe²⁵ frame international bureaucrats under sovereignty-as-responsibility and R2P as quasi-Rousseauvian lawgivers, foreigners dictating a framework. But sovereignty-as-responsibility in these accounts goes even further than Rousseau; the state guided in such a manner is always ultimately accountable to the international bureaucracy that moulded it, as opposed to Rousseau's lawgiver who disappears after the founding moment. Because the responsibility of ‘responsible sovereignty’ is not to citizens but to a de-politicized ‘population’, by this logic outside experts determine whether the state is upholding its responsibility. This is a perceptive critique of some liberal human rights advocates, such as Emilie Hafner-Burton²⁶ who claims that ‘human rights norms are settled’ and the UN is an insufficiently efficient institution for coercively enforcing those norms across the globe. However, R2P and sovereignty-as-responsibility are not reducible to human rights, so this critique ultimately has limited purchase.

Getachew and Cunliffe both aim to resurrect an ideal of popular sovereignty in a context of decolonizing and postcolonial states. But it is important to ask: what are the boundaries of political deliberation? There is a distinctly postcolonial political problem of ‘who are the people?’ in the aftermath of colonialism. It was not uncommon after decolonization for a dominant identity group to assert that the presence of another identity group was simply a residue of colonialism – and call for either forced assimilation or expulsion. Asserting the political and postcolonial identity of the Myanmar/Burmese people led to identification of the Muslim Rohingya population as itself a residue of colonialism. Idi Amin identified South Asians in Uganda as a continued colonial imposition. And successive governments in Khartoum viewed ‘the Southern Problem’ in Sudan as created by past British colonial governments. Although none of these states were (consistently) democratic, violence to define a unified national identity is an important part of the postcolonial political context. Sovereign responsibility is concerned with this real fact in the world.

Mahmood Mamdani frames his critique of R2P and sovereign responsibility within the critique of humanitarian reason. He also begins from the argument that R2P denies agency to postcolonial subjects:

Whereas the language of sovereignty is profoundly political, that of humanitarian intervention is profoundly apolitical, and even sometimes anti-political…The international humanitarian order…is not a system that acknowledges citizenship. Instead, it turns citizens into wards…The new language refers to its subjects not as bearers of rights – and thus active agents in their own emancipation –but as passive beneficiaries of an external ‘responsibility to protect’.²⁷

This, he holds, is not in fact novel – Western colonial powers always claimed to have acted selflessly to defend the victims of native tyrants.²⁸ R2P, then, is merely an updated embodiment of a colonial ‘right to punish’. But Mamdani adds an important claim: by criminalizing actors in localized conflicts, the international community may circumvent the possibility of a political settlement.

Building on his critique of R2P Mamdani examines ‘the criminal model’ of political violence which ‘is invested in this depoliticization of violence. Criminalization configures state-related violence as excess, something that occurs outside the realm of politics’.²⁹ Transforming political questions into humanitarian questions allows political factions to be criminalized; once criminalized, ‘foundational violence’ which might otherwise lead to negotiation, reconciliation, and re-founding the political community becomes the object of international responsibility. Outsiders can thus invalidate political settlements on the basis of the criminal status of one (or more) party.

Mamdani notes that the International Criminal Court (ICC) – not party to the conflict – rejected a settlement between the warring Lord's Resistance Army and the Ugandan legislature, declared the government ‘incompetent’ and thus held the agreement to be invalid.³⁰ In doing so, the ICC sided with the Ugandan president in what Mamdani reads as a domestic political fight, and scuttled the peace agreement.

Mamdani paints with a broad brush, reading the ICC and R2P as emanations of the same neoimperial ‘right to punish’: in Mamdani's account, the first pillar of R2P (the responsibility of the state to its own population, based on sovereignty-as-responsibility), is flattened such that it is merely pretextual for the eventual punishment meted out by foreigners. Postcolonial states can effectively be declared incompetent by international institutions, invalidating any political decisions undertaken by their governments or approved by their peoples – regardless of any actual capacity of the state. In this way, ‘incompetence’ echoes the colonial idea that a people ‘are not ready’ for self-governance, and must be administered by outsiders guided by humanitarian concerns.

Interestingly, the concern about the use of the ICC as a punitive measure which can disrupt political reconciliation is a point on which Deng and Mamdani are in cautious agreement. Regarding the ICC indictment of Sudanese President al-Bashir in 2009, Deng writes that it ‘is too late to prevent genocide in the South [of Sudan]. In any case, punishing those responsible would have run against negotiating and consolidating the peace agreement, as it would do in Darfur. Priority should therefore be placed on ending Sudan's proliferating wars by addressing the crisis of national identity and the marginalization and gross injustices associated with the crisis’.³¹ For Deng, conflict resolution takes priority over punitive action against individual leaders. Because sovereignty-as-responsibility seeks to address root political causes, rejecting both dominant human rights and humanitarian approaches, it may be in tension with other international institutions.

The path Deng proposes here – towards reconciliation, against punitive international approaches – does not fit well with the portrait painted by Cunliffe and Getachew. I argue that reading Deng's diplomatic work closely, informed by an understanding of his legal anthropology, reveals sovereignty-as-responsibility and R2P as sites of contestation rather than neoimperial impositions. The emergence of the postcolonial state brought with it a unique set of problems, and sovereignty-as-responsibility recognizes the failure of the nation-state model in a postcolonial world.

Recovering an alternate vision: reading Deng on postcolonial sovereignty

Deng, a South Sudanese legal anthropologist and diplomat, is a central but under-theorized figure in the emergence of R2P and sovereignty-as-responsibility. Postcolonial and critical theorists who critique him³² view him simply as a fellow from the Brookings Institution, bent on denying or circumscribing sovereign authority for postcolonial states. Liberal supporters of R2P³³ acknowledge his role in co-authoring Sovereignty-as-Responsibility, and as an important figure in convincing reticent states of the Global South to sign onto R2P as part of the 2005 World Summit document, thus ‘refuting’ the postcolonial critique that R2P is a Western idea. But neither side delves deeply into his lengthy record of published work.³⁴

Deng's work provides an alternate genealogy of R2P, focusing on the legacy of indirect colonial rule and its consequences for the postcolonial state. Rather than being primarily concerned with establishing a legal basis for armed coercive intervention (as the liberal genealogy of R2P that focuses on ICTY and Kosovo does),³⁵ this vision of R2P focuses on the responsibility and capacity of states to protect their populations, and denies that the phenomenon of ‘internal displacement’ is simply a ‘domestic problem’ that international institutions can ignore. It is this vision of R2P, I argue, that was enshrined in the 2005 UN World Summit document – rather than that document being simply the result of a ‘messy compromise’ that eliminated the coercive ‘core’ of R2P.³⁶

Because Deng's work spans multiple decades and fields there is a danger in taking up a single text or single moment in his career and grouping him with those seeking to assert a Western ‘responsibility’ to powerless victims; this may obfuscate a powerful alternative way of thinking about problems in the postcolonial world.³⁷ His view of community, interaction, change, and contact, all reflect an engagement with and rejection of the ideology of indirect rule colonialism. Indirect rule theorized tribes and cultures as unitary and separate. But this isn't Deng's view of how communities and peoples interact – rather, it is a colonial fantasy made into an administrative reality.

Deng as anthropologist: the colonial legacy in the postcolonial state

Recent political and social theorists illuminate the broad ranging influence of late imperial ideologies. Mamdani³⁸ introduced the term ‘late imperialism’ and defined its logic; more recently Karuna Mantena³⁹ and Mamdani⁴⁰ looked explicitly at the influence of Henry Maine's social and legal theory on late imperial ideology and its legacy in the postcolonial world. Taking a different tack, Lauren Benton⁴¹ traced how late imperialism imagined terrain and geography's relationship to sovereignty. Central to these studies is the insight that late imperialism created forms of governance and control which sought to define and empower representatives of ‘authentic’ cultures. Mamdani, Mantena, and Benton draw our attention to the groups who were empowered to enforce the ideas of authenticity that imperial observers wished to elevate.

Forty years earlier, Deng wrestled with these questions in his legal anthropology of the Abyei region of Sudan. Ngok Dinka were a people privileged by British colonialism's quest to preserve ‘authentic culture’, and Deng's father, Deng Majok, was a chief who centralized authority under himself with colonial administrators' blessings. This experience of late imperialism, I argue, informs Deng's vision of the colonial legacy. In studying the effects of late imperialism on both a colonially privileged tribe and a postcolonial state, Deng's work anticipated postcolonialism's critique of a unified colonized identity and informed his critique of Westphalian sovereignty. Further, studying these communities allows us to see them as more than just collaborators with colonial rule. Stripping Deng from this context denies that Deng is in a lifelong conversation with the colonial legacy.

Deng did not study Ngok Dinka practices in order to ascertain a ‘true’ or ‘authentic’ culture which should be imposed or embraced. His work was a rejection of both Maine's approach to ancient law (which guided the indirect imperial rule Deng critiques), and (certain) romantic anti-colonialist attempts to recover a pre-imperial ‘authenticity’ which would bring peace and prosperity.

To be self-contained, as indirect rule made colonized societies, is to cut them off from how communities have always interacted. This has two implications: first, part of the colonial legacy is artificially stunted growth, which necessitates (a form of) modernization or development. Therefore, state capacity building (a major theme in sovereignty-as-responsibility and R2P) is a way of grappling with the colonial legacy, not a prescription for neoliberal structural adjustment. Although critics⁴² see it as a stalking horse for ‘underdevelopment’ and the empowerment of international bureaucrats, relating it to Deng's work we can see capacity building as an escape from paternalist notions of ‘tradition’.

There is a parallel between the objection to indirect rule's empowerment of ‘tradition’ and Deng's emphasis on the importance of state capacity building. Deng⁴³ documents that Dinka elite came to recognize the irony of their empowerment by British colonial administration – by remaining indirectly involved, the British did little to build the capacity of Dinka tribes to interact in a modern world. This would prove problematic after Sudanization: out of 800 administrative posts in the new nation, only four junior positions went to Southern Sudanese⁴⁴ Indirect rule as a system (as Mamdani later notes) stigmatized anything that was not defined by the British as ‘traditional’, and treated ‘traditional societies’ as encased in amber, resistant to ‘corrupting’ influences.

The embrace of sovereign equality after decolonization reifies the colonial fantasy of isolation and authority. Thus, although most postcolonial critiques of sovereignty focus on sovereignty's evolution in the assumed or legislated absence of non-European sovereignty,⁴⁵ and therefore claim that sovereignty is a colonial concept, for Deng Westphalian sovereignty has colonial resonances because it encourages that fantasy of self-contained communities.

Capacity building, reframed as against colonially imposed stasis, becomes a potentially emancipatory project which may involve a transnational community. The state as a unit in a larger system is not overthrown, but the community of states interacts as communities have always done – against the fantasy of hermetically sealed units, they cross-fertilize through interaction. Further, Deng critiques the postcolonial state's ‘overreliance on the state’ which attempted to centralize civil society, the economy, and other aspects through an all-encompassing state, but a state that simultaneously did not have the capacity to carry out the duties it jealously guarded.

For Deng, the state cannot be based on the colonial constructions that empires used to divide and rule. Tradition and isolation were colonially imposed doctrines; there is nothing inherent about treating each community as a closed, self-governing unit. Thus, ‘nation-building’ in Deng's parlance should not be misunderstood as merely a reference to development, but to the creation of a less-ethnicized imagined community.

Central to this challenge of constructing the nation, in Deng's estimation, is recognizing the degree to which ‘myth overshadowed reality’.⁴⁶ For Dinka, central to that myth is the extent of oppression from Arab neighbours, which began prior to Turko–Egyptian rule. ‘The Dinka were in contact with the Arabs long before colonialism, but the hatred caused by slave raids prevented profitable acculturation and disposed the Dinka to reject Arab ways’.⁴⁷ Not only did this mean Dinka were not isolated, but some assimilation of cultural and political influence did occur. Later, ‘the Turko–Egyptian administration collected taxes, permitted slavery and other forms of exploitation, but otherwise left the “natives” alone. It was never a model for change nor was it interested in change, yet it activated the assimilation of Arab political ideas and practices’.⁴⁸ Although Dinka remained a separate identity, their practices – and thus culture – evolved because of contact.

Deng's emphasis on the state allowed him to articulate an account of internally displaced Southern Sudanese in the early 1970s.⁴⁹ It is important to highlight this contribution: importantly for analysing his later diplomatic and political work, he articulates the problem of internal displacement in 1972, two decades before his reports on internal displacement for the UN. As Representative of the Secretary-General of the UN, he was neither a figurehead nor simply reporting the work of others; Deng was a prime driver in thinking about this problem.

The problematics of sovereignty led Deng to contemplate what the postcolonial state was lacking: social justice, welfare and self-help, the fair distribution of services.⁵⁰ Rather than valorizing communities as the proper sources of these (as postcolonial theorists often do), the state remains the vital partner in Deng's writings. Thus, between state capacity building as an emancipatory project, and recognizing the plight of internally displaced Southern Sudanese, Deng outlined sovereignty-as-responsibility more than two decades before the Brookings Africa Project presented Sovereignty as Responsibility. Deng shows us that sovereignty-as-responsibility is not simply an implementation of applied ethics principles, but tied to a longer-standing problem in the postcolonial world. When we connect Deng's work on colonialism and decolonization, we see that for Deng we live in a world made by colonialism – but we still must live in it. It's not that he believes postcolonial problems are homegrown, as critics such as Whyte⁵¹ charge; rather, postcolonial actors retain agency in this world shaped by the colonial legacy.

Further, although decolonization resulted in a newly sovereign nation-state, Dinka continued to lose out in the national model.⁵² Under successive regimes – Turko–Egyptian, Mahdist, the Anglo-Egyptian Condominium, and finally independent Sudan – rule and domination by outside authority was constant for Dinka peoples. The new Sudanese state ‘assumed that [Northern Sudanese] identity was the national model, and what prevailed in the South was a distortion that the colonialists had imposed to keep the country divided’.⁵³ To resist the national model of the state, but still take part in the modern state, Deng sought to promote a syncretic approach that could promote values centred around persuasion and positive obligations to the people.

Deng as diplomat: re-thinking the postcolonial state after the Cold War

The story of sovereignty-as-responsibility and R2P told by many liberals⁵⁴ as well as political theorists influenced by postcolonial and critical theories⁵⁵ begins in the early 1990s. I have posited a longer theoretical history of these doctrines by drawing on the legal anthropology of Deng and his rethinking of the state. Deng's anthropology guided his diplomatic work in the 1990s to influence how the international community (specifically via the UN) reconceptualized sovereignty and structured relations among states. Understanding the history and politics of these doctrines challenges the claims made under the parallel banners of ‘critique of the ethical turn’⁵⁶ and ‘critique of humanitarianism’⁵⁷ – these are intensely political debates, not the anti-political elevation of ethics over all else. The sense of ‘responsible sovereignty’ promoted by Deng was not a (neo)liberal attempt to deny responsibility for past harms, but rather a recognition of what was required to govern in light of past harms. He uses the IDP as a liminal figure to demonstrate the shortcomings of contemporary humanitarian and human rights approaches, and as a starting point from which to introduce the idea of sovereignty as entailing positive responsibilities. Focusing on sovereignty allows him to rethink both that which is ‘below’ (the state) and ‘above’ (the state system).

Deng was neither the first nor the only figure to identify postcolonial conflict as having roots in the colonial world. However, there were important differences to his approach: although anticolonial thinkers in the 1960s and 1970s commonly argued that the roots of conflict lay in colonial legacies, these formulations generally sought to blame an ongoing (neo)colonial structure for crises in the decolonized world.⁵⁸ They held that international institutions drew upon colonial constructs to perpetuate domination beyond the end of de jure colonialism. In contrast, Deng argued that the colonial legacy reshaped the agency and identity of decolonized peoples – but he regarded them as the ultimate agents in their political fates. Deng's early UN work highlighted the colonial construction of these postcolonial identities as a way of confronting – and escaping – the colonial legacy rather than naturalizing it. Further, Deng focused on the colonial state, rather than the colonial international structure – which meant he was attuned to the ways in which imperial governance strategies (re)produced difference and hierarchy through identity, rather than positing them as simply structural problems. Importantly, Deng made the argument for the colonial state's role in postcolonial violence as part of an analysis for the UN bureaucracy. This meant that Deng's work was an attempt to make this analytical claim central to how the UN understood – and dealt with – problems, rather than a critique or protest of the institution itself. In essence, he presented an ontological challenge rather than an immanent critique.

Deng draws directly upon his research in Sudan on both the colonial state and its legacy for governance through ethnic and racial identity. Against the sorts of reductionist approaches that guided Western audiences, Deng is at pains to lay out what these identities mean and, importantly, what they do not mean. Writing against the ‘Arab Muslim versus Black African’ framing which would later cause so many Westerners to misperceive conflicts in Darfur and South Sudan, Deng argues that ‘Diversity is by no means limited to that dualism…regional and ethnic diversities reflect vast distances from Khartoum in physical, political, and socioeconomic terms that explain the separation, if not alienation, of the national leadership from the rural populace’.⁵⁹ It is not simply racial, ethnic, or religious differences and hatreds that define conflict in the postcolonial state, which also means that dividing postcolonial states into smaller, more homogeneous units would not necessarily resolve these conflicts.

These communities are caught in a double bind: on the one hand, their ‘identity’ was hardened as a strategy of colonial rule, and on the other colonial rule enervated these communities and destroyed their ability to rule themselves. This formulation is developed in Deng's legal anthropology throughout the 1970s and 1980s, and we see here how that work frames Deng's diplomatic work.

The issue of internal displacement became a way for Deng to foreground the reverberations of the colonial legacy, and African experience in particular. Deng insisted that ‘resolving the problems of the internally displaced must ultimately mean addressing the causes of displacement’, and ‘guaranteeing the rights of citizenship without discrimination’ rather than adopting an apolitical humanitarian approach.⁶⁰ Treating internal displacement as a political problem would mean that, rather than simply offering aid, political solutions would be necessary. This was a dramatic departure from the dominant view of humanitarianism in the 20^th century, which was that humanitarian aid must be apolitical.⁶¹

The accusation of ‘apolitical protection’ is a central aspect of Getachew and Mamdani's critiques of R2P; that ‘citizens’ are political agents, but ‘populations’ are not. This accusation gains currency from attempts to include ‘protection of populations’ in the World Summit Outcome Document of 2005. This shares affinities with, but ultimately remains distinct from, a longer history on the question of the status of ‘citizens’ versus ‘populations’. Revocation of citizenship was a question that was at the heart of sovereign authority: as Mira Siegelberg⁶² recounts, in the aftermath of the First World War the question of whether a state could revoke citizenship, and whether other states could refuse to recognize another state's revocation of citizenship, was central to whether ‘stateless’ could become a recognized category. A powerful faction within the League of Nations bureaucracy opposed the attempt to deny a right of revocation, arguing that citizenship was central to the institution of sovereignty, and to place internationally mandated limits on that would be to deny a distinction between national and international realms. The League could argue in favour of protecting ‘national minorities’ by treaty, by contrast, because minority status was a protection not an individual right.

But the move to addressing ‘populations’ is not necessarily a replay of League-era protections. Rather than debating whether sovereignty carries the authority to revoke citizenship, ‘population’ can address the results of such a decision. Thus, Deng recognizes that stripping people of their citizenship can be used as an excuse by governments to deny responsibility for them: ‘governments that are unwilling or unable to provide equal protection to all nationals’⁶³ will resort to invoking their sovereign right to refuse outsiders access. In this quote, Deng uses ‘all nationals’, not ‘all citizens’, such that the denial or revocation of citizenship cannot absolve a state of its responsibilities to the displaced. In arguing that revocation of citizenship cannot absolve a state of its responsibilities to those who were citizens, Deng is affirming the individual rights of those persons, rather than proposing a League of Nations style ‘minority protection’.

Deng's answer to internal displacement was not that postcolonial states did not ‘deserve’ ‘full sovereignty’ – which was a common claim throughout the 1990s and 2000s.⁶⁴ Rather, Deng pressed the system as a whole towards a different understanding of sovereignty. Deng recognizes the colonial legacy as a structural and epistemological conditioning of postcolonial actors. But these actors, in recognizing their situatedness, can work to remake the structures in which they live.

The task Deng sets forth is the remaking of the postcolonial state.⁶⁵ In challenging the postcolonial state, Deng posits a vision of sovereignty around which the state system can be remade. Central to this project is recognizing and rejecting the colonial logics guiding state form. But this does not mean rejecting the state in toto: ‘Until a replacement is found, the notion of sovereignty must be put to work and reaffirmed to meet the challenges of the times in accordance with accepted standards of human dignity’.⁶⁶ The state remains the (flawed) essential partner, never fully reducible to the community.

Although the state was a colonial imposition, Deng argues, the postcolonial state must remake itself as a syncretic entity. Thus, a pre-colonial past is useful for thinking about the world after colonialism, not a template for political organization.

Deng is not content to assign blame to colonialism for the origins of postcolonial problems: ‘however external their sources or continued linkages, the primary responsibility for solutions, especially in the post-cold war era, fall first on the Africans themselves’.⁶⁷ This is a stark statement, seemingly at odds with the portrait I draw of someone continuously wrestling with the colonial legacy. Taken in isolation, this might seem to confirm Whyte's thesis⁶⁸ about R2P as Neoliberal responsibilization – making postcolonial states ‘responsible’ for violence within their borders, with no consideration of colonial legacies.

Deng's statement, however, is different in important ways from Whyte's critique of the Neoliberal erasure of the colonial past in service of making postcolonial states responsible. Deng and Whyte employ two different meanings of ‘responsibility’. For Whyte, Neoliberalism makes postcolonial states ‘responsible’ by claiming that the past is over, and that conflicts and catastrophes in the postcolonial world are homegrown and not the legacy of centuries of oppressive imperial rule. Deng identifies postcolonial states as responsible, but ‘responsibility’ is not because there is no past. Rather, postcolonial states are responsible because they are agents and must work to escape the legacy of the colonial state. Anticolonial and postcolonial theorists also sought to portray these actors (states, civil society actors, communities, intellectuals, etc.) as agents, but theorized that escaping the colonial legacy would take the form of resisting continued domination from the Global North. Internal aspects of the colonial legacy, prior to Mamdani's pioneering work,⁶⁹ were frequently associated with a Western-educated colonial elite.⁷⁰

For Deng, the internal structure of the postcolonial state – its reliance on ‘strict’ sovereignty, its approach to ethnicity, and its jealous centralization of all functions of society under its own power – is part of that colonial legacy, but has been adopted by postcolonial state actors to further their own power.⁷¹ Although state borders were the most outwardly recognizable colonial aspect of the postcolonial African state, its internal structure also preserved colonial organization.⁷²

In most African countries, the determination to preserve national unity following independence provided the motivation behind one-party rule, excessive centralization of power, oppressive authoritarian regimes, and systematic violations of human rights and fundamental liberties. The participatory decisionmaking in African society was later alluded to by nationalist leaders to justify the one-party system, the rationale being that since Africans traditionally sat and debated until they all agreed, the multiparty system was antithetical to African culture.⁷³

Salient identities also became shorthand for graft and a substitute for state capacity building. Rewarding supporters was a substitute for building state institutions. This tied power even more closely to identity, further raising the stakes in conflicts over identity. And it made the state less responsible to its broad population.⁷⁴ Even where corruption was not a motivator, patronage government could still be a rational response of the state in the face of scarcity.⁷⁵ When the state does not have the capacity to ensure an equitable distribution of resources for all its citizens, the ‘responsibility’ felt by some leaders will be not to the state as a whole, but to those with whom they most closely identify.

Consignment to minority status could thus mean immiseration or persecution.⁷⁶ When the postcolonial state enshrined an identity as the ‘authentic’ identity of the state – even where it was not a numerical majority – the lack of institutions which might facilitate political contestation exacerbated conflict. By mythologizing the notion of national unity, these states did not develop political mechanisms for resolving conflict resulting from competing demands from different identity or civil society groups and could only recognize such conflict as a threat to national unity. The postcolonial state, in this account, had no means for successfully managing ‘internal’ conflict in a political context. Thus, there was simultaneous pressure on the state to engage in repression to ‘unify’ the population, and on ‘minority’ (or simply alternative) identity groups to pursue their interests through violent means. State capacity building, in Deng's formulation, is centrally concerned with developing mechanisms to facilitate political conflict, rather than allowing conflicts to metastasize into ‘ethnic conflicts’. In this way, it is more than the ‘prevention’ pillar of R2P. A common form of capacity building occurs through conflict mediation and reconciliation; outside actors may commit to help build institutions that can channel disagreements into political fora, to avoid the resumption of violent conflict.

Remaking the postcolonial state as ‘a nation [based] on indigenous identities, structures, values, institutions, and heritage does not mean a return to the past. Instead, the concept envisages a dynamic reinterpretation and transformation of tradition as a foundation and a resource for promoting a transitional integration that is self-reliant and self-sustaining. This would make development a process of self-enhancement from within’.⁷⁷ Africans as agents can work to create the new reality of states, and this process of creation is an emancipatory project. ‘There can be no question that Africa stands to benefit from cross-cultural fertilization, but that should mean what the word says: synergizing the positive aspects of the interactive culture and related value-systems’.⁷⁸ Importantly, syncretism is not simply derivative.

In Deng's argument, restructuring the state may mean accepting certain universal ideals for states but being open to different institutional and procedural norms informed by local histories and customs: ‘all African countries strive in varying ways and degrees to transcend the simplistic Eurocentric model constitutions and principles of constitutionalism which assumed a degree of homogeneity with hardly any regard to the specificities of the African context, its cultural values, institutions, and patterns of behavior’.⁷⁹ Contra Getachew's critique, this is the opposite of a ‘one-size-fits-all’ approach to state making.

Internal displacement due to violent conflict became a crisis in postcolonial Africa. The tendency of the international community to treat humanitarian crises and violent conflicts as isolated moments, rather than as manifestations of deeper rifts, meant that the causes of internal conflict in Africa were allowed to hide under the veil of strict state sovereignty.⁸⁰ The denial of access to the international community, Deng notes, is due to a particular vision of sovereignty. Sovereignty as identified with non-interference of all kinds, including giving aid to IDPs, insists on an absolute barrier between ‘domestic’ and ‘international’. Deng identifies this ‘traditional’ conception of sovereignty as the primary stumbling block in addressing the issue of IDPs.⁸¹

Deng's mandate to study internal displacement from the perspective of human rights created obstacles to tying the failures to deal with internal displacement to sovereignty as a ‘problem’.⁸² The UN Human Rights Commission was seeking an implementation strategy or set of reforms to enable the UN to address a particular form of immiseration that it had not previously recognized as part of its purview. Deng, however, used the assignment to bring bigger questions to the fore, arguing that implementation and reform could not address the problem in the way he was constructing it.⁸³ The IDP, for Deng, had to become a category – and thus a subject – in a way that existing approaches were not able to represent.

Deng's purview as representative of the UN Secretary General to the UNHCR did not include a rethinking of sovereignty as the basis for the international system. Thus, in order to introduce sovereignty into the discussion, Deng presented the question of sovereignty as already contested. After noting the possible need for ‘overriding’ or ‘forfeiting’ sovereignty, he noted that ‘world developments suggest that transcending sovereignty is no longer a forbidden territory for discussion’.⁸⁴ Although Deng first echoed the idea of sovereignty as a problem, the turn to ‘transcending’ signals how Deng would chart an alternate course. Although not using the phrase ‘sovereignty-as-responsibility’ yet, he is framing sovereign power around this notion of responsibility rather than non-interference.

Deng's framing of the international community differentiates him from liberal threats to use force to crack down on human rights abuses. For human rights liberals such as Hafner-Burton,⁸⁵ liberal democracies must enforce human rights. These norms are settled, and those who respect them (almost exclusively Western democracies) must protect the abused of the world. New institutions (for Hafner-Burton, a ‘concert of democracies’) may be necessary to avoid the gridlock of global participatory institutions such as the UN. Bellamy treats African norms and institutions instrumentally, claiming that ‘by the time of the 2005 World Summit, Africa had already developed a peace and security architecture that mirrored – indeed went well beyond – the norms espoused by R2P’.⁸⁶ What Bellamy wishes to do, however, is to appropriate that political and moral authority to legitimate a doctrine of armed coercive intervention emanating not from regional institutions themselves but from the UN.

Deng's approach understands that decolonized states are determined not to give up their hard-won authority. Furthermore, Deng's studies of indirect rule imperialism highlight the ways in which colonial powers used pacification and exaggerated threats from others to make populations pliant to alien rule.⁸⁷ Entrusting the enforcement of human rights to powerful states who can override the sovereign authority of others risks re-establishing this fear-based hierarchy. Thus, a major difference between Deng's approach and the Liberal humanitarian approach is Deng's emphasis on regional human rights instruments.⁸⁸ To be respected as responsibilities entailed by sovereignty, human rights cannot be treated as norms to be ‘diffused’ to or ‘enforced’ on the Global South.

This culturally syncretic approach distanced Deng from both emerging discourses around human rights and the democratic peace, both of which sought to make foreign-imposed regime change (FIRC) part of an international arsenal to enforce compliance. The responsibility of the international community Deng envisioned did not include FIRC nor treat armed coercive intervention in service of human rights as a primary goal.

Deng is an excellent case for examining the ways in which colonially informed sovereignty and the nation-state were being challenged across Africa in the 1990s and 2000s. Deng was not alone in his convictions, nor did he work as a solitary actor. The Conference on Security, Stability, Development, and Cooperation in Africa (CSSDCA) in conjunction with the African Leadership Forum (ALF) promoted the Kampala Document, in what Olusegun Obasanjo referred to as ‘a second wave of liberation struggle, this time against internal domination’.⁸⁹ The Kampala Document recognized the sovereignty of each African state, but in a marked shift away from the OAU's conception of strict sovereignty (which Salim Salim as OAU secretary-general already challenged), also recognized the interdependence of security, stability, and development of African states. According to this vision for Africa, ‘The key to security…is the responsible exercise of state sovereignty, in the absence of which cooperation among neighbors is required to deal with internal problems and conflicts’.⁹⁰ Although Deng was a member of the ALF, he was one of several on this ‘council of elders’, and not acting alone.

Further, the disbanding of the OAU and the establishment of the AU in 2002 demonstrated that the principle of non-intervention was losing popularity among African states. As noted above, the OAU was founded in part to secure the principle of non-intervention. By contrast, Article 4(h) of the Constitutive Act of the African Union established the legitimacy of an AU-led intervention into African states. Taken together, the CSSDCA and the AU point to ways in which Deng's ideas were both taken up in Africa, as well as demonstrating that he was not sui generis, but part of a larger movement from the Global South challenging the sovereign state.

R2P in action?

Critical theorist and postcolonial critics might see the recovery of Deng's anthropology and diplomatic work as ultimately nothing more than an exercise in the genealogical fallacy – the erroneous belief that to ‘discover’ the origin of a thing is to discover its true use or purpose – or simply far too credulous in accepting actors at their words. Cunliffe⁹¹ and Getachew⁹² focus on how power is ‘actually’ wielded through R2P – namely, the armed intervention into Libya authorized under UN resolution 1973. For them, a critical reading of origins (for Getachew, two texts on sovereignty-as-responsibility; for Cunliffe, the ICISS report) reveals that despite gestures towards ethics and humanity, the raw exercise of Western liberal power against a postcolonial state was always the telos of R2P. That Deng is African, or that the 2005 World Summit document received unanimous consent, are ultimately of little relevance.

But R2P is a framework, and intervention is one possible outcome. R2P challenges received accounts of sovereignty but also attempts to unify state capacity, conflict mediation, and armed coercive intervention under a new normative and institutional framework. When critics look for R2P operating in the world, they may only recognize instances that confirm their hypothesis. The intervention in Libya in 2011 should be treated as one case of R2P, rather than what R2P inescapably is. African-led attempts at conciliation in the aftermath of violence stemming from winner-take-all elections provide an alternate case for R2P. Kenya's 2007–2008 post-election violence is a contested ‘success case’ for R2P. There are several plausible reasons to not describe the post-election negotiation and settlement as R2P: the UN did not hold a vote ‘authorizing’ an R2P ‘action’, and the Kenya National Dialogue and Reconciliation (KNDR) was authorized by the AU, not the UN. Further, to discourage the possibility of ‘forum shopping’, Kofi Annan discouraged other figures and institutions – in particular, the ALF⁹³ – from offering assistance during negotiations.

However, there are other important aspects of the Kenyan case that indicate it should be treated as an R2P case. First, in a 2008 interview Kofi Annan himself stated, ‘I saw the crisis in the R2P prism with a Kenyan government unable to contain the situation or protect its people…I knew that if the international community did not intervene, things would go hopelessly wrong. The problem is that when we say “intervention,” people think military, when in fact that's a last resort. Kenya is a successful example of R2P at work’.⁹⁴ Further, the UN engaged the conflict as an R2P emergency without calling for armed coercive intervention. UN special adviser on R2P Edward ‘Luck, with the full support of the Secretary-General and the other organs of the UN system, [decided] to treat the situation as the first “test case” for R2P’.⁹⁵ Luck argues that the Kenya election crisis ‘was the first instance in which the United Nations employed a responsibility to protect (RtoP) lens in shaping its responses to an ongoing crisis. The Secretary-General decided, following consideration by the Policy Committee, that the world body's first goal in Kenya should be to prevent the further commission or incitement of RtoP crimes and violations’.⁹⁶ UN action, it is important to note, can also include deferring to and supporting regional institutions – such as the AU. Additionally, ‘R2P was also invoked by Francis Deng, the UN Special Adviser on the Prevention of Genocide’.⁹⁷ It is clear that many within the UN Secretariat considered this to be a case of R2P; what is important, then, is to consider what that means, rather than ‘debunking’ the claim that this was an R2P case.

Consistent with Deng's argument about the role of regional partners under sovereignty-as-responsibility, regional partners did not accept initial claims by Kenya that this was a ‘domestic concern.’ ‘Suggestions that the crisis was purely an internal matter were…quickly dismissed by Kenya's neighbours, who felt the effects of violence almost immediately’.⁹⁸ The broader international community also rejected Kenya's claims of it being a domestic political matter, setting the stage for international involvement though not necessarily armed coercive intervention. Although there was a broad international consensus that violence had to be stopped, the variety of actors seeking to pressure or promote solutions hurt the effectiveness of international efforts: ‘Kibaki, in particular, seized upon what was initially an ad hoc and disorganized international response to evade talks’.⁹⁹ The AU emerged as the mediator for talks, headed by Annan. That UN officials, as well as Annan (who was no longer secretary general), identify this as an R2P case means we must understand R2P as a framework for thinking about stopping mass atrocity crimes, and that such a framework can include empowering or supporting regional institutions responding to mass atrocity crimes – R2P need not be defined by votes in the General Assembly authorizing one or another intervention.

Although some early commenters were quick to identify Kenya as a ‘success case’,¹⁰⁰ other analysts argue that the R2P framing did little to affect or enhance negotiations.¹⁰¹ Serena K Sharma holds that ‘Strategic considerations…proved to be a necessary enabling condition for action. In this regard, the crisis in Kenya wasn't so much a “test case” for R2P, but rather an ideal case’.¹⁰²

Both Sharma and Junk remain critical of the idea that R2P in the Kenya case was innovative or represented something particularly novel. ‘The requisite ingredients of successful mediation…are by no means new, or unknown…what was perhaps exceptional in the case of Kenya was the unprecedented unity exhibited by external actors’.¹⁰³ However, the question need not be ‘was R2P necessary to resolve this conflict?’ There is nothing ‘new’ about peace negotiations – just as there is nothing new about armed coercive intervention. It is not that R2P necessarily invents new remedies, but rather that R2P unifies responses to atrocities under a single framework. Thus, what is important about the Kenyan case is that it helped build the institutional framework of R2P. As Sharma herself notes, there was ‘unprecedented unity’ supporting action; material interests and strategic concerns are not unprecedented, thus the unity was likely driven by the emerging norm as well. Additionally, this negotiation was not, contra what postcolonial critics of R2P claim, merely pretextual for later armed coercive intervention.

Additionally, the KNDR laid out four agenda items: first, stopping violence; second, dealing with the post-election humanitarian crisis – in particular, 700,000 IDPs; third, adopting a power-sharing agreement; and fourth, significant institutional reforms to Kenyan government.¹⁰⁴ The focus on IDPs and institutional reform are consonant with Deng's work on sovereignty-as-responsibility and R2P; these are not pretexts to reach a ‘last resort’ of armed coercive intervention.

In this light, it is important to review how the 2005 World Summit document, rather than the 2001 ICISS advisory report or any other commenter, frames R2P. The sections read:

Responsibility to protect populations from genocide, war crimes, ethnic cleansing and crimes against humanity

138. Each individual State has the responsibility to protect its populations from genocide, war crimes, ethnic cleansing and crimes against humanity. This responsibility entails the prevention of such crimes, including their incitement, through appropriate and necessary means. We accept that responsibility and will act in accordance with it. The international community should, as appropriate, encourage and help States to exercise this responsibility and support the United Nations in establishing an early warning capability.

139. The international community, through the United Nations, also has the responsibility to use appropriate diplomatic, humanitarian and other peaceful means, in accordance with Chapters VI and VIII of the Charter, to help to protect populations from genocide, war crimes, ethnic cleansing and crimes against humanity. In this context, we are prepared to take collective action, in a timely and decisive manner, through the Security Council, in accordance with the Charter, including Chapter VII, on a case-by-case basis and in cooperation with relevant regional organizations as appropriate, should peaceful means be inadequate and national authorities are manifestly failing to protect their populations from genocide, war crimes, ethnic cleansing and crimes against humanity. We stress the need for the General Assembly to continue consideration of the responsibility to protect populations from genocide, war crimes, ethnic cleansing and crimes against humanity and its implications, bearing in mind the principles of the Charter and international law. We also intend to commit ourselves, as necessary and appropriate, to helping States build capacity to protect their populations from genocide, war crimes, ethnic cleansing and crimes against humanity and to assisting those which are under stress before crises and conflicts break out.

These two operative paragraphs provide no direct endorsement of armed coercive intervention. For some,¹⁰⁵ this is the unsatisfying result of messy institutional negotiation. At one level, this framing is obviously correct; negotiations in international institutions rarely result in a straight-forward acceptance of a particular party's initial position. And these two paragraphs in particular were the subject of fraught negotiations.

In the aftermath of the ICISS report, states of the Global South began from a position of scepticism; more importantly, almost all states were generally sceptical. Bellamy recounts that, at first, ‘the Non-Aligned Movement (NAM) flatly rejected the R2P’ and that India and Malaysia, in particular, worried that R2P was either ‘military humanism’ or ‘a potential reincarnation of humanitarian intervention, for which there was no basis in international law’.¹⁰⁶ The US ambassador to the UN, John Bolton, was a particularly vociferous critic as well. To overcome such widespread opposition required significant negotiation as well as institutional and bureaucratic manoeuvring.¹⁰⁷ A final draft was held back from broader view to avoid re-opening negotiations that would prevent its inclusion in the Outcome Document, allowing only for a final up-or-down vote. The final vote was unanimous, meaning both great powers and small states ultimately endorsed the paragraphs on R2P.

But the ‘obviousness’ of such a framing requires reading R2P as a project of liberal humanitarian actors. If we read actors from the Global South as stakeholders and innovators, rather than as needing to be persuaded to support the liberal humanitarians, then we can read this process as endorsing a different vision of R2P, rather than Global South actors simply being ‘spoilers’ in the negotiations. This is the struggle South Africa faced; the South African party the African National Congress (ANC):

stresses that critics [misunderstand] South Africa's interpretation of what R2P means and how it can be implemented in practice. The tendency of Western actors to see their values and interests as representing the only possible incarnation of R2P underpin their misrepresentation…both ANC heavyweights and career diplomats stress their commitment to the “norm”…and remind critics that they played an essential role in rallying developing countries behind the outcome document.¹⁰⁸

I posit that paragraphs 138 and 139 of the World Summit Outcome Document endorse Deng's views on sovereignty, not the more ‘muscular’ interventionist pose advocated by those seeking to supply the sort of legal basis that NATO's 1999 Kosovo bombing lacked.

Bellamy notes that between 2002 and 2004, Deng was key in convincing initially sceptical states of the Global South to support R2P¹⁰⁹ but does not expand beyond that – which can give the impression that Deng simply pressed for the liberal humanitarian vision of R2P. Erin D. Mooney recounts the way Deng's diplomatic work on supporting the Guiding Principles for IDPs during this same period helped defuse concerns over sovereignty: ‘At the outset of virtually every meeting he held over the course of his mandate, and especially in discussions with governments, Deng routinely would sketch out his view of sovereignty as responsibility before then raising the specific IDP protection concerns that arose in each country situation’.¹¹⁰ Thus, although sovereignty-as-responsibility was not formally enshrined in either R2P or the Guiding Principles on Internal Displacement, we know that Deng's vision of sovereignty-as-responsibility was crucial to his diplomacy. For this reason, we should look for echoes of Deng's vision in accounts by actors formerly sceptical of R2P.

At UN negotiations over the implementation of R2P in 2009, Maged A. Abdelaziz, speaking on behalf of the NAM, re-affirmed the adoption of R2P as UN doctrine through the 2005 World Summit document, while expressing concern about its potential for abuse. This was a significant stance: institutional opponents of R2P (e.g., General Assembly president Miguel d'Escoto Brockmann, and delegations from Cuba, Venezuela, and Iran), insisted that paragraphs 138 and 139 of the 2005 World Summit document did not establish R2P as UN doctrine. And as noted above, NAM emerged in 2002 as one of R2P's earliest institutional opponents. That such a shift took place is significant.

The NAM statement included two important points for thinking about R2P beyond intervention and as the result of innovation from the Global South. First, noting the unease that some supporters harboured, it argued: ‘In order to build consensus on the way forward, there must be clarity on what needs to be done, based on our agreement that each individual State has the responsibility to protect its populations. Capacity building is key in this regard in order to allow States to shoulder this responsibility, and allow the international community under the umbrella of Chapter VIII of the United Nations Charter to support their efforts, as necessary and appropriate, and to assist those States which are under stress before crises and conflicts break out’.¹¹¹ State capacity building is not a matter of neoliberal structural adjustment but an emancipatory strategy for postcolonial states. Sovereignty-as-responsibility, in focusing on the question of sovereignty rather than intervention or human rights, addresses problems by simultaneously considering state capacity and responsibility, as well as the responsibility of the international community to every state's people. Although ‘sovereignty-as- responsibility’ is never named in official R2P doctrine, state capacity building is clearly an outgrowth of Deng's conceptualization of sovereignty.

Second, the NAM statement referenced African innovation: ‘the African Union is a pioneer in implementing R2P due to its particular historical experience. The conditions for implementation are clearly stipulated under Article 4 (h) and (J) of the Constitutive Act of the African Union …. To date, the African Union has dispatched two operations, both upon the decision of its Assembly’.¹¹² Locating the positive example of the genesis of R2P in the AU places both the innovation in sovereignty and institutional implementation in the Global South.

NAM's statement helps us understand the impact of Deng's work, and gives voice to a sometimes overlooked side of R2P. In carefully reconstructing Deng's work – first his legal anthropology, then on IDPs, and finally sovereignty-as-responsibility – we trace a significant innovation in the state system to its roots in the postcolonial world. Although other studies¹¹³ frame R2P as a particular approach to humanitarian intervention emerging from the aftermath of NATO's Kosovo campaign, such accounts cannot explain why the doctrine eventually drew widespread support in the Global South and downplay the role of state capacity building. NAM's statement in support of R2P, however, demonstrates that important actors from the Global South promote an understanding of R2P which is rooted in Deng's vision of sovereignty, not the ICISS nor that of the major powers' approach to intervention.

Conclusion: Re-thinking R2P after Libya

In 2011, the UN authorized R2P in Libya under resolution 1973. Initially authorized for the protection of civilians, NATO forces quickly chose sides in the conflict and treated the resolution as a mandate for regime change. After Qaddafi was ousted, the international community did little (and perhaps could only do little) to rebuild the Libyan state; Libya plunged into civil war, de-stabilizing several other states in the region such as Mali. For critics such as Getachew and Cunliffe, this was always the telos of R2P – whatever coalition might support it, whoever might be said to have ‘invented’ it, R2P was always going to become a doctrine supporting Western-led humanitarian intervention and regime change.

A coalition of states from the Global South – led by Brazil, India, and South Africa – proposed significant reforms to R2P doctrine after Libya. Postcolonial critics, though, might reasonably be sceptical; if R2P doctrine is neocolonial, then abolition, rather than reform, would be the proper approach since reform might simply be a legitimation strategy, rather than addressing a structural inequality.

A central argument of the present article is that such a framing of R2P is fundamentally mistaken; R2P is a site of political contestation, and actors from the Global South have competing and compelling accounts of the doctrine's purpose and function. This article has focused on Deng's contribution and influence on R2P doctrine; South African diplomacy in response to Libya offers one view of how to think about R2P outside of a neocolonial framework. Differing approaches to the meaning of ‘intervention’ in South African foreign policy ‘demonstrates that the R2P debate is not merely one in which emerging powers support or reject a “Western” norm of suspending sovereignty to protect civilians’.¹¹⁴

Post-Apartheid South Africa was a leader in pushing to replace the ‘dictator's club’ of the OAU with the AU. Article 4(h) of the AU Constitutive Act famously calls for African-led interventions in African crises – and South Africa has committed troops all over the continent for these purposes. It would be a mistake to view South Africa as simply a ‘middle power’ that aspires to project military power through humanitarian intervention. First, for many in Africa, the move away from ‘non-indifference’ represented by Article 4(h) ‘emerged in large part out of the disappointment in the dominant powers’ reluctance to effectively deal with Africa's crisis of underdevelopment, war, and authoritarianism’.¹¹⁵ Thus, attitudes towards South–South intervention are much more a matter of ‘continental self-help’, rather than being junior partners of great powers.

More importantly, and in keeping with Deng's broader vision of R2P and sovereignty-as-responsibility, ‘rather than conceiving intervention as solely a question of military operations, South African diplomats insist that it is a comprehensive concept that prioritises dialogue, capacity-building and national reconciliation’.¹¹⁶ South Africa voted in favour of UN resolution 1973, with US president Barack Obama speaking directly with South African president Jacob Zuma about the need to protect civilians, particularly in Benghazi. Zuma and other South Africans felt betrayed, then, by NATO's quick shift towards the goal of regime change, which NATO framed as the only effective way to protect civilians.

In the aftermath, Zuma was criticized at home for supporting UN resolution 1973. But this is not a clear-cut case of a leader out of step with anti-imperialist sentiment in a decolonized state; Zuma, Thabo Mbeki, Ebrahim Ebrahim, and Cyril Ramaphosa – all major figures in South African politics and foreign policy – have all supported forms of intervention and R2P, with Ramaphosa even serving on the ICISS committee. Rather, for many South Africans, NATO's support of a Libyan rebel victory was not R2P, but Western states taking revenge against an African leader. ‘Virulent criticisms of the inequitable international system and Western power politics should thus not be understood as a sovereigntist rejection of the duty to protect civilians and arrest genocide or crimes against humanity, but rather as a desire to strengthen the notion of an African responsibility to solve the continent's problems and shield its citizens from war and oppressive rulers’.¹¹⁷ The ideals of sovereignty-as-responsibility need not be in tension with an anti-imperialist politics; indeed, they can be complementary.

There is much to the argument about the anti-politics of atrocity, but for critical and postcolonial critics of R2P and sovereignty-as-responsibility, I argue, it obfuscates innovation from the Global South, and how ideas emerging from Africa have been taken up in international institutions. In reconstructing Deng's anthropological and diplomatic work, I push back against theorists such as Getachew and Whyte who reduce him to a Brookings Institution fellow – as though this affiliation signals that sovereignty-as-responsibility simply emanates from liberal institutions of the Global North. This reconstruction recovers sovereignty-as-responsibility and R2P as sites of political contestation and also recovers an alternate understanding of the colonial legacy as embodied in the postcolonial state.

In a critique of the postcolonial intellectual, the South African writer Fetson Kalua argues that after decolonization too many intellectuals became servants of regimes. Rather, the intellectual must be a transformative thinker: ‘Such transformations should focus on amelioration in realms such as physical development, education, health, human rights, identity, various freedoms, and several spheres of life. In a word, rather than being sterile or non-creative technocrats, genuine intellectuals are people who are imbued with and have learnt to cultivate the spirit of impartiality and justice as the founding principles or ideals of a stable society’.¹¹⁸ Rather than focusing on being an effective international bureaucrat in a structurally unequal world – and treating internal displacement as a matter for technocracy – Deng's challenge to sovereignty worked to upend colonial assumptions that fuelled modern conflict. It falls to the postcolonial political community to remake the state such that it can address the problems the colonial state created. A return to an idyllic pre-coloniality is mythic; these states must think syncretically. Ultimately, this argument is not the anti-politics critiqued by thinkers such as Getachew, Whyte, and Mamdani: Africans can remake their states, and in doing so remake the structure of the states’ system.