The ambiguities of self-determination: IGAD and the secession of South Sudan

This paper examines the 1994 Declaration of Principles (DoP) for the resolution of the Sudanese civil war, adopted by the Inter-Governmental Authority on Development (IGAD). This was the only occasion on which an African inter-state organization included separation as an option for resolving a civil war. It was the basis for South Sudan's independence in 2011. The DoP was drafted by the Ethiopian government, and imposed on belligerent parties, both of which were, at the time, unionist. The paper identifies two concepts of self-determination within the DoP — independence for colonial territories and the Marxist-Leninist idea of self-determination for national groups. The rationale for including both arose from Ethiopian leadership within IGAD. The paper also examines the diverse Sudanese debates on self-determination, including several strands of nationalism, Islamism, and the ‘ New Sudan ’ of the Sudan People's Liberation Movement (SPLM). There was radical disagreement among Sudanese on national identity and self-determination, creating ambiguities that ironically facilitated the exercise in southern self-determination in 2011. Drawing on documentation of Sudanese negotiations, the paper examines how the DoP unlocked the Sudanese debate on the issue, and how the different concepts fared up to the time of the independence of South Sudan. vibrant contributor to Sudanese political discourse. An implicit critique of the Sudanese army leadership's obsession with defending territorial integrity, the Islamists kept the door open for claims to self-determination to be voiced and debated.

DoP-independence for colonial territories and the Marxist-Leninist idea of self-determination for national groups. The rationale for including both arose from Ethiopian leadership within IGAD. The paper also examines the diverse Sudanese debates on self-determination, including several strands of nationalism, Islamism, and the 'New

| INTRODUCTION
An essential step in South Sudan's road to independence was the affirmation, by neighbouring countries, of the right of self-determination for the people of southern Sudan. Meeting in Nairobi on 20 May 1994, the foreign ministers of Eritrea, Ethiopia, Kenya and Uganda-all members of the north-east African regional organization, the Intergovernmental Authority on Development (IGAD)-adopted a declaration of principles (DoP) for the resolution of the civil war in Sudan. Although the Government of Sudan and the Sudan People's Liberation Movement/Army (SPLM/A) both had pro-unity positions, the DoP contained the principle of self-determination-a radical and unexpected step. The drafters of the DoP did not, however, intend it as a charter for southern Sudanese separation. Instead, the dominant articulation of self-determination in the document-and the rationale for its inclusion-derives from the Ethiopian government's Marxist-Leninist theory of nations and nationalities. That formulation was focused on recognizing the rights of all the constituent nations of a multinational state as the foundation for unity. However, this had little traction in Sudan, which had its own vibrant debates on identity and unity. Nonetheless, the DoP unlocked a path towards a referendum in which southern Sudanese could vote for secession.
Almost 17 years later, former South African President Thabo Mbeki, representing the African Union (AU) speaking in Juba on the eve of the referendum that led to the creation of the independent Republic of South Sudan, reminded the southern Sudanese 1 that, 'it was IGAD that conceived and nurtured the agenda of selfdetermination as a right for the people of southern Sudan and as the central component in a resolution of the Sudanese crisis' (Mbeki, 2011b, p. 4). The IGAD DoP was the first time-and to date the only time-that an African interstate organization had recognized that a postcolonial state could be divided by the will of its people (or some of them). It was the foundational text for the peace process. Mbeki simplified. There was a well-established southern Sudanese demand for southern secession long before 1994. But without the explicit recognition of the right of southern Sudanese to self-determination by an African intergovernmental organization, that independence would simply not have occurred. The IGAD DoP is therefore an historic document. Remarkably, it has gained little attention. For example, Douglas Johnson's otherwise-authoritative history of self-determination in Sudanese political discourse mentions IGAD and the DoP only in passing (D. Johnson, 2013, p. 146). Hilde Johnson's 'inside story' of the Sudanese peace talks repeatedly mentions the IGAD DoP as the foundation of the right of self-determination but does not investigate the remarkable fact that it existed in the first place (H. Johnson, 2011, pp. 14, 18, 24 and 29). Others fail to mention it at all (Christopher, 2011;McNamee, 2012;Schomerus & de Vries, 2017;Sheeran, 2011;Weller, 2009).
The first task of this paper is to restore the IGAD DoP to its rightful place in the historical narrative of Sudan and South Sudan. The second task is to explore the Sudanese ideological debates and political processes concerning national identities and self-determination. These debates included radical disagreement on both the question of unity versus secession and on the terms of the debate itself. Systemic indeterminacy about the identity of the Sudanese nation created space in which different formulations of self-determination could be raised, and their implications obscured. The intersection between the DoP, which drew on Ethiopian political thinking, and Sudanese debates contributed to an outcome-South Sudanese independence-that the DoP drafters and Sudanese political leaders of the time neither foresaw nor wanted.
The DoP was drafted by Ethiopia's transitional president Meles Zenawi and finalized in the office of Foreign Minister Seyoum Mesfin. This was a remarkable reversal of the role played by Ethiopia in the African territorial dispensations of the colonial and the decolonization eras (see the introductory essay to this collection). But it is no accident that the same Ethiopian Foreign Ministry that had 30 years earlier so assiduously promoted the principle of the sanctity of inherited colonial boundaries would be the one to overturn that principle. The leaders of the Ethiopian People's Revolutionary Democratic Front (EPRDF) saw Sudan as a conquest state-cum-multinational empire, with a history of being ruled with an iron-fisted centralizing political ideology. This critique is identical to the EPRDF critique of Ethiopia's imperial statehood, and the proposed solution-a political settlement for a nation of diverse selfgoverning peoples-was also the same.
For each successive Ethiopian regime, policies towards immediate neighbours including Sudan were an extension of domestic politics. Haile Selassie's role in ending the first Sudanese civil war was an outcome of his interest in the country and the issues. In the 1972 Addis Ababa peace talks, he insisted that one country should have one army and therefore that the Anyanya forces should be absorbed into the national army. The government of Colonel Mengistu Haile Mariam was fiercely committed to maintaining the territorial integrity of Ethiopia, and its antiseparatism extended to its policy towards Sudan. When former southern Sudanese rebels congregated in western Ethiopia in 1983 to foment rebellion, Ethiopian security officers welcomed them but soon discovered that most were intent on resuming a secessionist war. One of the few who was a truly committed unionist was Colonel John Garang, and with that essential qualification, the Ethiopians backed him against his rivals as leader of the SPLA.
Eight years later, Garang was initially hostile to the EPRDF government, and his SPLA fought against the EPRDF inside Ethiopia (De Waal, 2004). The new EPRDF government and the EPLF, now in power in Eritrea, were also grateful to Khartoum for assistance given in the last 2 years of the war. When the Sudanese peace talks in Abuja broke down in 1993, the Sudanese government proposed that IGAD serves as the new mediator, expecting that the new Ethiopian and Eritrean regimes would lean its way.
The Sudanese were disappointed. By early 1994, Sudanese Islamist sponsorship of Eritrean jihadi groups had so antagonized the Eritrean government that President Isaias Afewerki was now committed to military support for the Sudanese opposition. The Ethiopians were also becoming hostile but took a less military stand. The EPRDF leaders also believed that they had discovered the political solution to unresolved questions of national identity in a large, diverse country: self-determination.

| THE IGAD DoP
The IGAD DoP contains two concepts of self-determination. The operative paragraphs read: 2 The rights [sic] of self-determination of the people of South Sudan [sic] to determine their future status through a referendum must be affirmed; and 3 Maintaining unity of the Sudan must be given priority by all the parties provided that the following principles are established in the political, legal, economic and social framework of the country: 3.1 Sudan is a multi-racial, multi-ethnic, multi-religious and multi-cultural society. Full recognition and accommodation of these diversities must be affirmed.
3.2 Complete political and social equalities of all people in the Sudan must be guaranteed by law.
3.3 Extensive rights of self-administration on the basis of federation, autonomy, etc., to the various people of the Sudan must be affirmed.
3.4 A secular and democratic state must be established in the Sudan. Freedom of belief and worship and religious practice shall be guaranteed in full to all the Sudanese citizens. State and religion shall be separated. The basis of personal and family laws can be religion and customs.
3.5 Appropriate and fair sharing of wealth among the various peoples of the Sudan must be realized. de WAAL 3.6 Human rights as internationally recognized shall form part and parcel of this arrangement and shall be embodied in Constitution.
3.7 The independence of the Judiciary shall be enshrined in the Constitution and laws of the Sudan. 4 In the absence of agreement on the above principles referred to in 3.1-3.7 the respective people [sic] will have the option to determine their future including independence, through a referendum.
The first concept, in Paragraph 2, is straightforward: it defines the people entitled to exercise self-determination with reference to South Sudan (though the use of the plural 'rights' is intriguing). That territory was defined as the southern provinces, according to the colonial administrative boundaries at independence on 1 January 1956, and as adopted in the 1972 Addis Ababa Agreement that brought Sudan's first civil war to an end. This concept of selfdetermination is based in the idea of granting sovereign independence to a colonial or non-self-governing territory, as set forth in United Nations General Assembly resolution 1514(XV) of 1960. It derogates from the principle of inviolability of colonial boundaries adopted by the OAU in the 1964 Cairo Declaration.
The second concept, in Paragraphs 3 and 4, is more complex and revolutionary. It is also ambiguous: Subparagraphs 3.1 to 3.3 refer to multiple peoples of Sudan, awarding them rights of self-determination without specifying the right to independence. Moreover, the right of secession is conditional on failure to agree a democratic constitutional order dedicated to unity in diversity. This is a similar formulation to that found in the Constitution of the Federal Democratic Republic of Ethiopia (FDRE), which was drafted in 1994 and adopted the following year.
The rights of self-determination in the IGAD DoP were radical and without precedent in Africa. The document not only provided a route towards secession for southern Sudan but also included another, less clearly defined, set of self-determination rights for peoples throughout the country. The political intent of the drafters was clear: the DoP was intended as a means of ensuring that peace brought about a democratic government committed to maintaining unity. Southern secession was conditional. Indeed, the provisions for self-determination beyond southern Sudan contained in Paragraph 4 were intended to enhance the incentives for unity. Over the following 15 years, the Ethiopian position was to support the unity of Sudan, fearing that should the country divide, an embittered northern Sudan would be drawn into the Arab camp and an independent South Sudan would be unstable. Meles Zenawi was clear on this in February 2002. Speaking to this author, the man who drafted the DoP said that southern self-determination was a matter of principle, as well as a political impossibility to discard. But Meles then went on to say that a grave mistake would be committed if the secession option were put on the same level as the unity option.
In common with the Ethiopian constitution, he argued, the right of self-determination was a precondition for, and a guarantee on unity. Thus, southern Sudan should have the right to secede as a last resort, an exit option in case of a catastrophe-implicitly treating the southern Sudanese case (Paragraph 2 of the DoP) on a par with that of other Sudanese peoples (Paragraph 4). Meles feared a referendum a few years after a peace agreement would just mean another round of war. He thus envisaged a 'one country two systems' approach, with the southerners' rights guaranteed through rights to language, culture, acknowledgement of a distinct history, self-government and equitable representation at the centre.

| THE CONTEXT OF THE DoP
Ethiopia provided both the intellectual framing and subsequently much of the political and military muscle to impose it on the GoS and the SPLM, both of which were formally opposed to self-determination. In 1994, Ethiopia was closely allied with Eritrea, which showed no interest in the IGAD negotiations. Isaias was convinced that Khartoum would fall militarily. He had expelled the Sudanese ambassador earlier that year and handed over the embassy to the National Democratic Alliance (NDA), and his concern was in reaching agreement among the opposition parties, which was achieved with the 1995 Asmara Declaration. Ethiopia moved more cautiously. Addis Ababa maintained diplomatic relations and the possibility of a negotiated agreement with Khartoum.
At the session of the peace talks that followed, the GoS delegate, Ghazi Salah el Din Attabani, fulminated against the DoP. He correctly insisted that it was contrary to the OAU's Cairo Declaration (although that does allow states to deviate). The GoS held out for 3 years, until its weak military situation and diplomatic isolation compelled it to sign.
The leader of the SPLM/A, John Garang was also unpleasantly surprised: during the previous 11 years, he had done all he could to keep the term 'self-determination' off the political agenda, well aware that it risked leading to secession for southern Sudan. He was a unionist who knew that the majority of his fighters were separatists. But at this point in the war, SPLA units were holding the last defensible position in southern Sudan before the Uganda border, at Parajok, with the assistance of troops and tanks from Uganda and Ethiopia. Reliant for survival on the troops of IGAD member states, Garang had no option but to agree. Over the following 3 years this situation changed, with strong military pressure exerted on Khartoum, primarily by Ethiopia and Eritrean forces which directly intervened in the war, and secondarily by the Ugandan army, the United States and the SPLM/A itself. By early 1997, when Congolese rebels and Rwandese forces, backed militarily by Uganda and Eritrea, were racing towards Kinshasa to impose regime change by force of arms, the GoS justifiably feared that it was next in the sights of the 'frontline states'.
From this position of military ascendance, Ethiopia and Eritrea went to war with one another in 1998, dramatically changing the regional balance of power. Although at war with one another, the Ethiopian and Eritrean governments maintained their positions on Sudan unchanged, each of them sceptical about southern Sudanese separatism.
Their respective representatives at the IGAD peace talks refused to speak to one another, but their interventions in the sessions expressed a common position. Nonetheless, when the IGAD peace talks were reinvigorated by the United States with Kenyan leadership in 2001, Ethiopia and Eritrea had far less influence over the content of the mediation documents. Meanwhile, the other two IGAD countries involved in the DoP-Kenya and Uganda-were politically sympathetic to the southern Sudanese and leaned towards southern independence, provided it could be made to happen.

| WHO ARE THE SUDANESE?
The dominant question in northern Sudanese postcolonial discourse has not been framed as one of selfdetermination but rather, 'who are the Sudanese?' This debate has been conducted on the axes of race, faith and class-with the last element the least prominent but perhaps the most pervasive. The term sudani emerged in the colonial period with two meanings. The first was a simple descriptor, referring to the people who were indigenous to the Sudanese territory. The second referred to black and detribalized people, chiefly from the south and the Nuba Mountains. In the latter usage, it was imprecise or deliberately indeterminate, an opposite to the comparably ambiguous term 'Arab'. 2 Over the decades, the term sudani migrated up the social scale to the point at which it referred to the economic, cultural and political elite, whose members considered themselves 'Arab' and the process of 'becoming Sudanese' referred to the adoption of the norms and habits of this elite (Doornbos, 1988). The Anglo-Egyptian treaty of 1936 was the first occasion on which Egypt recognized 'the Sudanese' as an entity (Chasapis-Tassinis & Nouwen, 2019).
The question who are the 'southern/South Sudanese' is similarly fraught. It is striking how the definitions have shifted between ancestry, physical appearance, language and territory, and back again. The 'southerners' simultaneously possessed a sense of being the original or true Sudanese and also being excluded from the Sudanese nation. Southerners have been defined in opposition to the north, and at the time of independence, southern Sudanese media stressed the themes of building the new nation on the memory of the SPLA struggle and the leadership of the late John Garang and overcoming tribalism (Frahm, 2012). An intrinsically sociopolitical identification became explicit during the process of registration for the 2011 referendum. As Nikki Kindersley observes of the process of defining 'southern-ness' , '[t]his huge issue, sensitive and difficult both emotionally and politically, was hardly de WAAL discussed by the general population or political elite. Instead, controversies in the run-up to the referendum focused more on accusations of rigging and political influence on both sides' (Kindersley, 2015, p. 83). The processes of identifying South Sudan's citizens and borders were based more on vernacular notions of identity and belonging, than on legal documentation.
The official languages of the Republic of South Sudan are English and the numerous indigenous languages of the country. Arabic is not an official language, and in fact, when the criteria for South Sudanese citizenship were drawn up, a linguistic criterion was used, which explicitly excluded those whose native language was Arabic.
Nonetheless, most of the July 2011 independence ceremony was conducted in Arabic and when President Salva Kiir Mayardit wants to address the South Sudanese nation, the language he chooses is Arabic. South Sudan is officially a secular state, with no religion given preferential status in the nation's constitution. However, the political processes of dismantling the Islamization of public life and embracing political solidarity with Christian groups that actively promoted southern secession have been more prominent than the process of separating state from religion (Salomon, 2014).

| AN AMBIGUOUS INDEPENDENCE
Sudan achieved an ambiguous nationhood on 1 January 1956. The raising of the Sudanese flag followed a decade of intrigue and contestation among Sudanese political leaders and the two imperial co-domini, Britain and Egypt. The path to independence was laid by two agreements: the 1952 Self-Determination Agreement between the Sudanese legislature and the British, and an agreement between Britain and Egypt which allowed for a 3-year transition, followed by an internationally-supervised referendum on independence or unity with Egypt, scheduled for 1957.
The process was swift, with fundamental problems left unaddressed-notably the status of southern Sudan. As one of the leading historians of the period observes, 'A major reason for the unconcern over the problems of government was the belief that the whole situation was provisional. The outstanding question was self-determination and when that was settled attention could be turned to other matters' (Woodward, 1979, p. 160).
In December 1955, Prime Minister Ismail al Azhari pre-empted the promised referendum and rushed through a parliamentary vote to declare independence. He did this partly to forestall a challenge to his precarious coalition government (which indeed fell a few months later) and on the assurance to his patrons in Cairo that it was a steppingstone to unity with Egypt.
For the Umma Party and especially its core members, the Ansar-followers of the Mahdi-independence was Sudan's second liberation, and a return to the country's independence of the 1885-1898 period, when it existed as a Mahdist state. The Umma MPs voted for independence without precondition.
Azhari won the votes of the southern MPs with a promise of a federal constitution-a promise that he failed to keep (Alier, 1990). The position of southern leaders, adopted at a conference in Juba the previous year, was that, in the absence of guarantees on federalism, the southerners would claim their own right of self-determination. In fact, members of the Equatorial Corps of the Sudan Defence Force mutinied rather than be transferred to northern Sudan, foreshadowing the first Sudanese civil war.
For South Sudanese, the date '1/1/56' came to stand as a marker of a false start. In 1963, the Sudan African National Union-its name a deliberate echo of the parties that were steering East African countries to independence-petitioned for South Sudan's independence on the basis that the nation was suffering 'internal colonization' (Oduhu & Deng, 1963). From that point onwards, the 'southern question' , alongside the agenda of political Islam, dominated Sudanese political debate.
The lack of a unifying narrative and symbolism is reflected in the muted annual celebrations of independence day and the fact that sub-Saharan Africa marks Ghana's independence in 1957 as the beginning of decolonization.
Even more conspicuously missing from the shared nationalist story is the 1924 revolution and its leader Ali Abdel Latif. As Elena Vezzadini writes, 'The revolution has been lost twice: first in 1924, when it was violently quelled; and then historically, because it never really achieved the status of a stable element of the national narrative' (Vezzadini, 2015, p. 7). One reason for this is surely that Ali Abdel Latif was born of a Nuba father and a Dinka mother. He was a 'Sudani' in the sense of a detribalized 'black' Sudanese. Another reason is the careful ambiguity in the White Flag League's call for self-determination. While ostensibly emphasizing the ties of brotherhood between Egyptians and Sudanese, their message could equally well be read as putting Egypt in the position of having to support Sudanese independence (Vezzadini, 2015, pp. 160-162). Divergent opinions were hidden under opaque language: political unity of purpose was concealed by refusing to resolve key questions. Sudanese political debate remains marked by contingency and ambiguity to this day.
In a famous essay 'The Dilemma of the Southern Intellectual', written in 1961, Joseph Ukel Garang-a senior member of the Sudan Communist Party (and no relation to John Garang)-identified three categories of southerners.
There were secessionists (rightists), Marxist anti-imperialists (a very small group) and the 'perplexed' majority. He wrote: 'The mistake of the rightists lies in ignoring the major contradiction (the struggle against imperialism) or at best subordinating it to the minor contradiction, namely the South-North differences, while the mistake of the perplexed intellectuals lies in their putting the two contradictions on an equal plane' (Garang, 2010, p. 280). Writing at a time when Kenya and Uganda were still under British dominion, and Congo was in its post-independence crisis, Joseph Garang foresaw that an independent South Sudan would be a pawn of the colonial powers.

| Al-Sudan al-Jadid
At the other end of the political spectrum, Islamists sought to reinvent the Sudanese nation and Sudanese people, at one point adopting the slogan al-Sudan al-Jadid, 'the new Sudan'. The Islamist philosophy underpinning this enjoins the creation not only of a new country but a new people. The existence of an Islamic state is premised upon the Utopian paradox that only truly virtuous Muslims can create an Islamic state, but Muslims cannot achieve true virtue unless they live in such a state. Another challenge to Islamists is how to establish a constitutional order for the affairs of people, while asserting that sovereignty belongs only to God. How are the spiritual and temporal aspects of authority to be reconciled? How are people-whether the universal umma of believers or a section of the umma defined by nationality-to act legitimately on God's behalf? Sudanese Islamists have failed to resolve not only these problems-inherent in Islamist political theory-but also dilemmas specific to the Sudanese predicament.
Sudanese Islamism is inextricably linked with the enigmatic figure of Hassan al-Turabi. His political thought and practice constituted a bundle of paradoxes, combining liberalism and authoritarianism, radicalism and conservatism.
Nothing was straightforward in al-Turabi's writings and speeches: he often appeared to believe opposites, and what his critics derided as hypocrisy, his followers praised as creative reinterpretation and the jurisprudence of necessity.
Perhaps the final verdict on Sudan's leading Islamist was that he was a 'torn soul' (Berridge, 2017, p. 313). The language of the 1997 Constitution obscures as much as it illuminates, but it opens up the possibility that national or territorial sovereignty is conditional rather than absolute, and so raises the further possibility of permitting the separation of non-Muslim communities as well as even more radical reconfigurations of state territorial dispensations.
The option of excluding non-Muslim communities from an Islamic state and its territories was a common topic of debate among Sudanese Islamists in the 1980s. After taking power in 1989, President Omar al-Bashir publicly raised the possibility of southern secession. Twenty years later, Bashir's uncle Tayib Mustafa ran a newspaper, al Intibaha, that was both consistently racist and intermittently separatist, arguing that only when southern Sudanese have separated would it be possible to establish a truly Arab Islamic state.
This highlights a second tendency within Islamist thinking, which is to equate the Islamic community with Arab identity and civilization. Sudanese Islamists have long suffered from racist attitudes in Egypt, Lebanon and the Gulf States, with many people in these countries regarding people with darker skin as second-rate Arabs. This has in turn coloured Sudanese Islam (Troutt-Powell, 2003).
The Sudanese Muslim Brothers, originally a junior branch of the parent Egyptian organization, found that they had to contend with a very different sociopolitical context to that in Egypt. First, northern Sudanese political life was never fully secular: the dominant parties were based on Muslim sects. Consequently, the Islamists found themselves competing with parties that already drew their legitimacy from Islam. Second, Sudan was marked by great ethnic, racial and religious diversity. In 'discovering the South', the Islamists faced a trilemma: whether to pretend that diversity did not exist and follow the then-conventional centralist state-and nation-building model, whether to accept that Sudan was two countries and not one or whether to adopt an agenda of colour-blind 'African' Islam (El-Affendi,-1990). They never resolved the issue.
The Islamists' default option was to identify Islamism with Arabism, assert the civilizational superiority of Arab-Islamism and pursue a political agenda of conventional territorial nationalism alongside the demands for Islamic law. This is evident in frankly racist attitudes towards southern Sudanese and Darfurians. For example, Hassan Makki dismissed African cultures as a tabula rasa and considered resistance to Islamization as a missionary conspiracy (including what he called 'anthrobiology'). (Ahmed, 1989, p

| THE NEW SUDAN
The Islamists' nemesis, Garang, Chairperson of the SPLM, similarly was opaque and shape-shifting in his political philosophy. Today, he is revered as the father of an independent South Sudan, but he himself espoused, variously, a united socialist Sudan, a united democratic Sudan, and a 'New Sudan'-a concept immune to precise definition.
Garang wanted power in Khartoum, not separation. But beyond that, as Bona Malwal notes, 'Garang's real reasons for fighting the long civil war with the North were never clear because he never spelt them out' (Malwal, 2014, p. 157).
Francis Deng, the most articulate and sensitive writer on the subtleties of Sudanese and South Sudanese identity issues, has also shown agonizingly divided sentiments on whether to espouse unionism or separation. He observed that the real programme of the SPLA rank and file 'was reflected in the Dinka saying popular among fighters: Ke tharku, angicku, "What we are fighting for, we know"' (Deng, 2005, p. 6).

| THE SUDANESE DEBATE AND THE IGAD DoP
Possibly, the Sudanese could have wrestled indefinitely with the enigma of their country, unable to resolve their contradictions, and even delighting in them; indeterminacy might have been the formula that kept the country together.

Southern Sudanese leaders decisively put self-determination on the table of the peace process in 1992, at
Nigerian-convened peace talks ('Abuja I'). The Nigerians (who had their own troubled history of secessionism) and the Sudanese government did not want self-determination on the agenda, but the breakaway SPLM-Nasir faction insisted that it be discussed, and the SPLM-Torit ('Mainstream'-Garang's faction, but negotiating without its leader in the room) agreed. Six days into the talks, the two SPLM delegations made a Joint Declaration of a Common Position on Self-Determination (Wöndu & Lesch, 2000, p. 123). Not for the last time, Garang was furious with his delegation for raising self-determination. The talks broke down.
The Nigerians convened a second round of talks-'Abuja II'-in 1993, with only the Government of Sudan and the SPLM-Mainstream. This time, Garang made sure his own agenda prevailed and even travelled to the Nigerian capital to be on hand in case his delegates deviated from their instructions. The SPLM proposal was based on three possible models: a unitary state dominated by the north, a confederal system or separation (Wöndu & Lesch, 2000, pp. 56-57), of which the confederal system (including the 'three areas' within the southern entity) was the preferred option. The SPLM delegation mentioned secession only as the fallback option in case any formula for unity failed.
These talks broke down rapidly. Government delegates were confident they were winning the war and would not need to compromise. Not for the last time, Khartoum overplayed its hand and missed an opportunity.
In September 1993, at Sudanese invitation, IGAD took over the mediation. One of the first steps of the four mem- Thus, the IGAD DoP scripted an opposition platform that could have brought together the southern Sudanese and the marginalized people of northern Sudan. It held the potential for a creative approach to self-determination that would have been more consistent with Garang's 'New Sudan' political vision. However, Garang did not take this opportunity. He was, at the time, under serious pressure from the neighbouring states that backed him, his own officer corps and the wider constituencies of Sudanese. Hence, he was abandoning his Marxist-Leninist baggage and instead trying to build broader coalitions, both within southern Sudan and with northern opposition groups. Just the month before-April 1994-Garang had reluctantly convened the first ever SPLM National Convention, which was marked by strong calls for developing a civil authority, humanitarian and welfare programmes and a political structure with mechanisms of accountability.
During the 1990s, Sudan's formerly vibrant civil society moved into the diaspora, and the non-Islamist political dialogue was conducted among politicians, activists and academics who were sympathetic to the SPLM and NDA, while also highly critical of what they saw as their shortcomings (Jok, 2000;Rahhal, 2000). the British. Darfur's leaders had not seriously entertained a claim to self-determination because they felt that they were strongly integrated into Sudan, and although neglected, their future interests were better served by a bigger stake in power in Khartoum. The ethnic argument for self-determination for Darfur did not hold and was not made: Darfur is a multi-ethnic region in which the Fur, though the largest group, constitute only about one quarter of the total population. Darfur's ethnic map resembles a chequerboard, in which it would make little sense for the black squares to secede from the white. The Darfurians' favoured approach was 'unity in diversity' with a federal system.
The conference participants' talked around these issues at length, agreeing only that '[t]he varying claims of the Nuba, South Blue Nile people, Beja and others should be resolved as a matter of priority' (Abdel Salam & de Waal, 2000, p. 291).
Garang meanwhile revived his option of a confederation of two states, each with its own constitution, in permanent cohabitation. Complete with Garang's characteristic Venn diagrams, this was presented to IGAD in an April 2000 document, entitled 'Legal Framework for Peaceful Resolution of the Sudanese Conflict' . It is similar to a proposal issued a year later by the Center for Strategic and International Relations in Washington DC, co-authored by Francis Deng, which advocated 'one country, two systems' (Deng & Morrison, 2001).
This was also the preference of the George W. Bush Administration. The Special Envoy Senator Jack Danforth was a sceptic of secession: the first draft of his report to President Bush omitted any reference to self-determination whatsoever. He then rewrote it with a modest concession (Danforth, 2002, p. 23-24): Any peace agreement must address the injustices suffered by the southern Sudanese people.
Southern Sudanese have claimed the right of self-determination as a means of protecting themselves against persecution; however, there are different views of what self-determination would mean in Sudan's future.
The view that self-determination includes the right of secession is contained in the IGAD Declaration of Principles, and is supported by many Sudanese. However, secession would be strongly resisted by the Government of Sudan, and would be exceedingly difficult to achieve.
A more feasible, and, I think, preferable view of self-determination would ensure the right of the people of southern Sudan to live under a government that respects their religion and culture. Such a system would require robust internal and external guarantees so that any promises made by the Government in peace negotiations could not be ignored in practice.
Danforth's lack of enthusiasm for southern separatism was echoed by Egypt which, jointly with Libya, launched a parallel peace process with the intent of reconciling the Sudanese government and the NDA, in such a way as to rule out secession.

| MACHAKOS AND ITS DISCONTENTS
Just 3 months after Danforth's report, on 20 July 2002, Ghazi Salah el Din and Salva Kiir (Garang's deputy) signed the 'Machakos Protocol' . This is the basic text of the CPA: the other protocols that followed over the succeeding two and a half years consist in fleshing out the details. The chief mediator, the Kenyan General Lazarus Sumbeiywo, was pessimistic about reaching agreement and (in his own words) tried to distil the IGAD DoP down to a 'Single Negotiating Text'. His first draft did not include self-determination at all, and the SPLM delegates were furious (Waihenya, 2006, p. 85;H. Johnson, 2011, p. 46). Garang himself was absent at that session. Sumbeiywo revised the text to include self-determination. Salva was now ready to compromise on Ghazi's insistence that Islamic law remain in the north. To the surprise of almost all, they agreed and signed.
Two paragraphs provide the key provisions on self-determination: 1.3 That the people of South Sudan have the right to self-determination, inter alia, through a referendum to determine their future status.
2.5 At the end of the six (6) year Interim Period there shall be an internationally monitored referendum, organized jointly by the GoS and the SPLM/A, for the people of South Sudan to: confirm the unity of the Sudan by voting to adopt the system of government established under the Peace Agreement; or to vote for secession.
In the hands of the Sudanese negotiators, the original DoP vision had become an interim period of 'one country, two systems', with a solid get-out clause for the south. While the text emphasizes that it is derived from the DoP, its philosophy is different (Young, 2012, p. 95). The broader commitment to self-determination for all the peoples of Sudan had gone and with it the elaborate conditionalities on the southern referendum (replaced elsewhere in the text by a less specific commitment to 'make unity attractive'). A prominent separatist within the SPLA had achieved a fait accompli, with the active connivance of the mediator. Garang reportedly 'went berserk' (Young, 2012, p. 94). The Americans, assured by Garang that negotiations would be on the basis of unity, were also unhappy and felt blindsided by Sumbeiywo. Zenawi was unenthusiastic but was tied up in an internal power struggle. But there was no going back: Machakos was met with popular acclaim.
Why did Ghazi sign Machakos? The main reason was that Khartoum had already accepted the IGAD DoP 5 years earlier. In 1997, the GoS had returned to IGAD and signed the DoP. The main reason for this concession was the military pressure exerted by Ethiopia and Eritrea. A secondary reason was the GoS 'peace from within' strategy of making local security pacts with southern militia. One of the many ironies of Sudanese politics was that during the 1990s, the southern separatists sought patronage from Khartoum, while Garang was unionist. This was mostly simple expediency on both sides, but the alignment contributed to the GoS formally accepting self-determination.
Before returning to IGAD, the Sudanese government signed peace agreements with breakaway factions of the SPLM: first Riek Machar (Khartoum Agreement) and then Lam Akol (Fashoda Agreement). Both agreements included de WAAL self-determination clauses. Al-Turabi also included self-determination in Sudan's 1997 Constitution. Article 139 (g) stipulated that 'the exercise of the right of self-determination for the Southern States is one of the fundamental rights that cannot be amended save through a referendum all over Sudan' . The constitution provided the options of unity or secession and even gave a date: 4 years after the formation of the Coordinating Council for the Southern States.
Speaking to this author, Ghazi later made four additional points. One was that he envisaged power-sharing and wealth-sharing arrangements that incentivized unity-he criticized the formulae later detailed in the CPA as having an inbuilt incentive for separation. The second was that while separation was an event, secession was a process, which might take a long period to accomplish. Third, a referendum should be seen as an advisory action rather than a political decision. Finally, Article 1.3 contains the words 'inter alia': there were options other than a plebiscite. But these are details: Ghazi conceded the main event.
When signing the DoP, the GoS also stipulated its understanding of the scope of the IGAD peace talks: the agenda covered southern Sudan alone. Any issues within northern Sudan were strictly off the agenda. The IGAD foreign ministers never formally accepted this limitation, nor did Garang. The SPLM delegation to the IGAD talks in 1998 was headed by Yousif Kuwa, SPLM governor of the Nuba Mountains, who insisted that 'north' and 'south' were points of the compass, not distinct parts of the Sudanese territory. In the 'reinvigorated' IGAD talks after 2001, Sumbeiywo did not address the GoS attempt to restrict the agenda head on, but instead found ways to accommodate some additional issues in parallel negotiating tracks, with their resulting agreements folded into the CPA. However, this approach precluded addressing the diverse conflicts within northern Sudan in a comprehensive manner. In effect, Khartoum was trading concessions on the South for the right to maintain its Islamic identity and relegating the rest to marginalia.
The 'peace by piece' approach was controversial (Simmons & Dixon, 2006), most especially criticized for making it all-but-impossible to resolve the conflict in Darfur. At the time, it was defended as a necessity by its proponents, including the 'troika' of the United States, United Kingdom and Norway. A multiparty negotiation was, they said, simply too complicated. Better to resolve the principal problem-which they identified as north versus south-first and then move to the questions of democratization and the marginalized peoples of the north. While Garang was alive, there was a reasonable anticipation that these latter issues would be addressed in due course. After his death, that was not the case.
The most important and explosive issue remaining was what to do with the 'three areas', many of whose people had fought as part of the SPLA, but which excluded from self-determination. The failure to accommodate a resolution to their grievances sparked a new war at the time of the separation of the south.

| THE 'THREE AREAS': THE NUBA MOUNTAINS, BLUE NILE AND ABYEI
The Machakos Protocol kicked the debate on who was entitled to self-determination into renewed life. The next month, the SPLM Governor of Blue Nile, Malik Agar confronted the IGAD mediator, saying: 'General, if you do not include the Funj people of Southern Blue Nile, we shall finish you' (Waihenya, 2006, p. 93). The Nuba and Blue Nile groups that had participated in the Kampala civil society conference began organizing a forum at which they planned to make their case for self-determination, and Garang set about a parallel conference aimed at obtaining the endorsement of the SPLM members in the 'two areas' for him to act as their sole representative in negotiating a peace agreement. Both conferences convened in late 2002 (Komey, 2010, pp. 134-136). The Nuba Mountains and Southern Blue Nile Civil Society Forum was held in Kampala in November. Forty participants from civil society attended. They expressed their dissatisfaction with the marginalization of the two areas in the peace process. In Point 4, they recognized the right of self-determination for the two areas, 'in accordance with international conventions' but also 'opted for the choice of safeguarding the unity of Sudan as the priority based on the principles of equality, justice and respect of basic human rights'. Mountains. His speech there was strong on rhetoric but short on details. He said, 'I want to reiterate the commitment of the SPLM/A to these areas. We will not let you down. Whatever agreement we reach in the IGAD we will include you' (quoted in Komey, 2010, p. 135 Convention in which it adopted a manifesto, including Article 5.16.3, which is remarkable in its reflection of the spirit of Paragraph 3 of the IGAD DoP: As a result of the failure of the successive regimes in Khartoum to manage the diversity of Sudan, and their continuous attempts to impose the exclusive and racist Arabo-Islamic identity on the ethnoculturally diverse peoples of Sudan through the state organs, and as a result of the lack of seriousness in implementing the signed peace agreements, the SPLM-N adheres to its position for the right of self-determination for all marginalized Sudanese groups. Self-determination shall be exercised either internally, by consensus on a secular democratic system of governance, which provides these marginalized peoples with the opportunity to exercise all of their rights within a united state that guarantees these rights or externally, by full independence to liberate themselves first and then to seek their ultimate voluntary unity. 4 Another question (the third of the 'three areas') was Abyei. This was not a question of self-determination as such

| CONCLUSION
Mbeki addressed two Sudanese audiences on the eve of the January 2011 referendum. In Khartoum, he insisted that the division of Sudan would not create separate 'African' and 'Arab' states but rather two African states, each marked by the challenge of governing diversity (Mbeki, 2011a). The Sudanese government did not rise to this challenge. On the eve of the separation of the south, the army forcibly occupied all of Abyei district, sparking a confrontation that almost brought about a new north-south war and which required the interposition of an international peacekeeping force. The generals tried to do the same in the Nuba Mountains, on this occasion meeting stiffer military resistance from the SPLA forces in the area, resulting in a war, that soon spread to Blue Nile.
Speaking in Juba, Mbeki reminded the South Sudanese that, notwithstanding the long years of struggle, the experience of the previous half century of postcolonial Africa showed that 'the work of freedom is just at its beginning' (Mbeki, 2011b). That lesson was also not heeded. In December 2013, unable to resolve their internal differences politically, South Sudan's political leaders launched a civil war, which led to atrocities and starvation comparable to the experiences under Sudanese rule that had so powerfully motivated the South Sudanese demand for separation. War and economic collapse led the Salva Kiir government to seek closer political, security and economic ties to Khartoum. Symbolic of this was the appointment of Tut Kew Gatluak as Kiir's national security advisor: Tut is an ethnic Nuer who as a young boy in the 1980s was adopted by Brigadier Omar al-Bashir and was brought up in the president's household before training as a security officer. In his new position in Juba, he reported both to Kiir and to his adopted father. As South Sudanese laconically observed, separation had entailed a shift from 'one country, two systems' to 'two countries, one system'.
At the time of writing (January 2020), the conflict in Abyei remains frozen, war has been fought in Nuba Mountains and Blue Nile for eight-and-a-half years, and the South Sudanese civil war is staggering towards a negotiated settlement under international pressure. Sudanese and South Sudanese hope that the 2019 popular uprising that overthrew the al-Bashir regime, with exemplary non-violence, will create conditions under which these wars can definitively be resolved. This may be too optimistic, but at least it represents another twist in the unending conversation about who are the Sudanese and South Sudanese and how their people should govern themselves.
According to its advocates, the IGAD DoP was both an innovative effort to reconsider the rigidities of postcolonial territorial politics in Africa and also a charter for the political transformation of all of Sudan. The outcome of the exercise in southern Sudanese self-determination points to less sanguine conclusions. Some identify the problem as a clumsy misinterpretation of the subtleties of the IGAD text and the political vision that lay behind it (e.g., Young, 2012). There is indeed some truth to this charge: the Machakos Protocol and the CPA represented a considerable simplification. But the fundamental problem with the DoP was that it was an Ethiopian document and not a Sudanese one. It drew from a political tradition that had little traction in Sudan. Thus, even those Sudanese leaders who were most sympathetic to its intent were not able to use it to generate debate among Sudan's political class, let alone use it as a charter for comprehensive transformation. The meeting point between the DoP and Sudanese political realities was southern separatism and in due course that was the outcome. The 1994, IGAD DoP was indeed a remarkable and unique document, which validated self-determination as a means of trying to resolve Sudan's civil war and therefore made South Sudanese independence a political possibility. But for it to have helped guide Sudan to any other outcome demanded a configuration of political factors that did not exist in Sudan, or internationally, then or since. Like so much of Sudan's history, South Sudan's independence was born of contingency, ambiguity and misunderstanding.

ENDNOTES
1 This paper follows the practice of using 'southern Sudan' to refer to the southern region within a united Sudan and 'South Sudan' to refer to the independent state after 2011. Note however that many documents, including the IGAD DoP itself, are not consistent in their usage of 'South Sudan' .