A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to...

of 32 /32
APRIL 2018 A Poisoned Well: Lessons in Mediation from South Sudan’s Troubled Peace Process ZACH VERTIN

Embed Size (px)

Transcript of A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to...

Page 1: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

APRIL 2018

A Poisoned Well:Lessons in Mediation fromSouth Sudan’s Troubled Peace Process


Page 2: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


ZACH VERTIN served as Director of Policy for the USSpecial Envoy to Sudan and South Sudan from 2013 to2016. During this period he was intimately involved in theSouth Sudanese peace process both as a US diplomat andas a de facto adviser to the mediation led by the Inter -governmental Authority on Development (IGAD). He iscurrently a visiting lecturer at the Woodrow Wilson Schoolof Public and International Affairs at Princeton Universityand the author of a forthcoming book on South Sudan, A Rope from the Sky: The Making and Unmaking of theWorld’s Newest State.


The author would like to thank those interviewed duringthe course of this study, as well as those colleagues whoserigorous reviews strengthened the final product. Specialthanks also to Delphine Mechoulan, Jake Sherman, AlbertTrithart, and the entire team at the International PeaceInstitute.

IPI owes a debt of gratitude to all of its donors, who makepublications like this one possible. IPI would particularlylike to thank the government of Finland for its support toIPI's lessons from mediation project.

Cover Photo: President Salva Kiir signs

the agreement on the resolution of the

conflict in South Sudan at a ceremony

in Juba, South Sudan, August 26, 2015.

UN Photo/Isaac Billy.

Disclaimer: The views expressed in this

paper represent those of the author

and not necessarily those of the

International Peace Institute. IPI

welcomes consideration of a wide

range of perspectives in the pursuit of

a well-informed debate on critical

policies and issues in international


IPI Publications

Adam Lupel, Vice President

Albert Trithart, Editor

Madeline Brennan, Associate Editor

Suggested Citation:

Zach Vertin, “A Poisoned Well: Lessons

in Mediation from South Sudan’s

Troubled Peace Process,” New York:

International Peace Institute, April 2018.

© by International Peace Institute, 2018

All Rights Reserved


Page 3: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2



The Mediation Context. . . . . . . . . . . . . . . . . . . . . . . . . . 4







The Negotiations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12



A Theater for Regional Competition . . . . . . . . . . . . . 18




Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Page 4: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,
Page 5: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

Executive Summary

In 2013, the world’s newest nation—the Republicof South Sudan—descended into civil war. Theyoung country had been a widely celebratedsuccess story just two years earlier, havingovercome generations of war and neglect to declareits independence, peacefully, from Sudan. Thepartition of Africa’s largest state had offered SouthSudan the chance to determine its own future. Butjust thirty months into their state-makingenterprise, its leaders dragged a war-weary peopleback into conflict, erasing the promise of liberationand squandering an enormous reservoir of interna-tional goodwill.External actors moved quickly to convene peace

talks under the auspices of the IntergovernmentalAuthority on Development (IGAD)—one of EastAfrica’s preeminent regional organizations. Twoyears of acrimonious talks ensued in neighboringEthiopia as regional states and internationalsupporters attempted to negotiate an end to theviolence and a blueprint for sustainable peace.Meanwhile, war ravaged the country, claimingcountless lives, deepening ethnic fault lines, anddisplacing more than two million civilians.The peace process unfolded in two phases: Phase

I focused on a cessation of hostilities agreementbetween two warring parties, while Phase IIbroadened the agenda and participants, aimingtoward a comprehensive political settlement andstructural reforms. Under mounting internationalpressure, South Sudan’s two most polarizingfigures—President Salva Kiir and Vice President-turned-rebel-leader Riek Machar—ultimatelysigned a comprehensive peace deal in August 2015.But the agreement unraveled just a year later,before it could be implemented, and the warmetastasized.South Sudan’s leaders bear primary responsi-

bility for the conflict, the troubled nature of peacenegotiations, and the devastation borne by millionsof their fellow South Sudanese. But as the principalentry point for international actors, the IGAD-ledpeace process also merits critical review.Drawing on principles of mediation best practice,

an analysis of the peace process demonstrates anunusually fraught mediation context, includingdeficits in five fundamental areas: preparedness,consent, impartiality, inclusivity, and strategy.Further analysis reveals a second layer of dynamicsthat complicated the task of the mediators, fromintra-group tensions and forum shopping tosummit diplomacy and empty threats. Together,these issues offer important insights into thepeacemaking effort, the challenges confronted, andthe environment in which an ill-fated peaceagreement was forged.Finally, the role of regional actors—and wider

peace process supporters—in any mediation effortcan be hugely consequential, as interested statescan variously support, shape, or spoil a peaceprocess. South Sudan was no exception. Mostnotably, the potential value brought to the SouthSudanese mediation effort by IGAD’s frontlinestates—Ethiopia, Kenya, Sudan, and Uganda—wasultimately outweighed by their competing nationalinterests and stakes in the outcome.Despite notable flaws in the process, the success

or failure of any mediation effort depends first andforemost on the political will of the partiesthemselves. South Sudan’s principal combatantsnot only lacked the will to make peace—they wereoften hostile to the very idea of a negotiated settle-ment. As such, IGAD and the wider constituency ofpeace process supporters faced a political andmoral dilemma often confronted by outside actorswhen a conflict is not “ripe” for settlement—whentradeoffs are made between ideal solutions and theimperative to stop the violence.In critically assessing the IGAD-led peace

process (2013–2015), singular conclusions are hardto draw. The process may have helped to slowSouth Sudan’s civil war and provided a platform toconfront the fundamental changes required totransform state and society. But inherent flawsmeant the peace deal lacked the political will, broadnational ownership, and implementing authoritiesnecessary to make it stick. As IGAD member statesand international partners now attempt to“revitalize” the peace process, they would be wise toevaluate, and build upon, its lessons.


Page 6: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


This paper examines the peace process for SouthSudan led by the Intergovernmental Authority onDevelopment (IGAD) from 2013 to 2015.1 It is nota history of the civil war, nor a detailed chronologyof the process. Viewed through a prism ofmediation best practice, it is a critical assessment ofthe attempt to negotiate a settlement of the conflictand, ideally, a distillation of lessons learned.Few outsiders had insight into what was often an

opaque peace process. This paper sheds light on themediation effort, the environment in which it tookplace, and some critical dynamics that shaped thefirst and second phases of the process. It assessesthe strengths and weaknesses of the mediationarchitecture and the roles played by individuals,institutions, and a wider constituency of peaceprocess supporters. In this context, it revisits thecomplicated political, moral, and resourcequestions presented by “subsidiarity” normsbetween regional and global institutions.Special emphasis is placed on the role of IGAD

member states, especially Ethiopia, Kenya, Sudan,and Uganda, and the impact of their interventionson the mediation effort. South Sudan’s belligerentparties were responsible for the devastation visitedupon their country, but the competing nationalinterests of IGAD states and other foreign actors—all playing out on a South Sudanese canvass—madematters worse. This study aims to serve as aresource for diplomats, policymakers, andmediators as they undertake future peacemakingefforts in East Africa and beyond.The first section introduces the reader to the

South Sudanese context through snapshots of thewar and the peace process. The second sectionframes the mediation context by identifying six“fundamentals” of mediation and analyzingwhether or not they existed in South Sudan. Thethird section spotlights the decisions and dynamics

that shaped the first and second phases of negotia-tions. The final section further unpacks thetroublesome regional dynamics that poisoned themediation effort, compromised regionalimpartiality, and ultimately weakened the 2015accord.



When the euphoria of South Sudan’s independencesubsided, deep fissures in its ruling party, theSudan People’s Liberation Movement (SPLM),were laid bare. Like so many movements before it,the SPLM struggled to transition from liberationfighters to governing party. Rather than designinga plan for achieving South Sudan’s developmentalaspirations, SPLM elites became consumed by anincreasingly contentious struggle for power. Thatdispute turned violent on December 15, 2013,exposing both fragile state institutions and theethnic divisions of an unreconciled past.2

President Salva Kiir alleged that recentlydismissed Vice President Riek Machar hadattempted a coup d’état. The unsubstantiatedcharge was a pretext to crack down on Machar andother party opponents but was also loaded withethnic connotations.3 Dinka forces hunted Macharand pushed ethnic Nuer troops out of the capitalcity before turning their guns on Nuer civilians.Machar escaped, South Sudan’s national armyfractured in two, and Nuer communities organizedin search of revenge. An impromptu rebellion wasborn. What had begun as an elite political disputequickly morphed into an ethno-regional conflict.Each side mobilized supporters by manipulatingethnic fears, and a cycle of massacres and revengeattacks left thousands dead, most of them civilians.4

As the two warring parties laid waste to thecountry, each sought weapons and matériel fromexternal backers. Neighboring Uganda joined thewar on behalf of Kiir’s government, while Sudan

2 Zach Vertin

1 The Intergovernmental Authority on Development (IGAD) is comprised of seven member states (Djibouti, Ethiopia, Kenya, Somalia, South Sudan, Sudan, andUganda), which engage in cooperative action on a range of collective peace, security, developmental, and environmental concerns. IGAD and its member statesplayed influential roles in bringing about a peace deal between Sudan and Southern Sudan in 2005 (the Comprehensive Peace Agreement), and in supportingSouth Sudan’s referendum on self-determination and subsequent independence in 2011.

2 The root causes of South Sudan’s post-independence war, including structural problems inside the SPLM, are outside the scope of this paper.3 President Kiir is a Dinka and Riek Machar a Nuer. Kiir’s December 2013 allegations against Machar included provocative references to divisions that emergedbetween Dinka and Nuer communities during Sudan’s civil war, when Machar broke away from the SPLA, and the painful legacies that remain.

4 No definitive estimate of the number killed in South Sudan’s war exists.

Page 7: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


funneled weapons to the opposition.5 As theviolence intensified, diplomats from Africa, theUnited States, and Europe flooded the region,hoping to contain the conflict before it spiraled outof control.6 Within just two weeks, their collectivediplomacy forced government and oppositionteams to the negotiating table in Addis Ababa.There was no mistaking that the parties werereluctant to talk peace and had been coerced intomediation, but diplomats resolved that the alterna-tive was far worse.SNAPSHOT OF THE PEACE PROCESS

The IGAD-led peace process for South Sudan tookplace in Ethiopia from January 2014 until theAgreement on the Resolution of the Conflict inSouth Sudan (ARCSS) was signed in August 2015.Additional negotiations on a range of implementa-tion modalities continued into 2016. A team ofthree mediators from IGAD countries (Ethiopia,Kenya, and Sudan) was selected to lead the process.Most negotiating rounds took place in Ethiopia’scapital city, Addis Ababa, though the mediatorsaimed to shake things up on several occasions bymoving the talks to other locations in the country,including Debre Zeit and Bahir Dar.7

Phase I of the process (January 2014) focused ona cessation of hostilities agreement and the releaseof a group of high-profile SPLM leaders who hadbeen arrested and remained in governmentcustody. The participating stakeholders were thetwo main warring parties: the government ofPresident Salva Kiir and the recently christened“SPLM/A in Opposition,” a loose constellation ofanti-government elements led by Riek Machar.8

Phase II (February 2014–August 2015) attempted

to expand the peace process and its agenda, withthe goal of a comprehensive political settlementand structural reforms. In addition to the govern-ment and opposition, five other stakeholder groupswere involved at various stages of Phase II,including a group of high-profile SPLM leadersknown as “the former detainees,”9 other politicalparties, civil society, faith leaders, and women’sorganizations. Despite sustained efforts by theseparties, the mediators, and peace processsupporters to concretize a “multi-stakeholder”political dialogue, these groups were never allowedto engage as full participants.The process ebbed and flowed for more than a

year while fighting continued and the human tollmounted. Neither incentives nor pressure didenough to alter the parties’ calculations, nor did atanking national economy or the prospect offamine. Meanwhile, after poisonous regionaldynamics undermined the mediators and nearlyparalyzed the peace process, the IGAD mediationwas officially reconfigured as “IGAD Plus” in2015.10 But this attempt to expand the format didlittle to mitigate the problems at the core of IGAD’smediation effort.Regional heads of state convened six extraordi-

nary IGAD summits during Phase II of the peaceprocess, and US President Barack Obamaconvened his own in July 2015. Shortly thereafter,the mediators circulated a draft peace agreementamalgamating eighteen months of inputs fromSouth Sudanese constituencies on matters oftransitional governance, security, reconciliation,the economy, and institutional reform.11 In theensuing weeks, three stakeholder groups—the

5 While important, Sudan’s support in terms of weapons and matériel was limited by comparison. In addition to the weapons they retained when defecting, opposi-tion leaders acquired arms and ammunition from a variety of foreign sources.

6 In response to the crisis, the UN Security Council also adopted Resolution 2132 on December 24, 2013, authorizing the deployment of additional peacekeepers tothe existing UN Mission in South Sudan (UNMISS).

7 Moving participants away from the capital city, the alternate venues were selected to focus efforts and try to create a different dynamic. But talks returned toAddis Ababa following these alternate rounds, as the mediators determined that logistical and communications challenges outweighed any added value to theprocess.

8 “SPLM/A in Opposition” will hereafter be referred to as “opposition.”9 These eleven individuals were arrested by the government shortly after the outbreak of conflict, having been accused of conspiring with Riek Machar to overthrow

the government. The political prisoners were all senior SPLM party members and included national ministers, a state governor, and the party’s secretary-general.While allies of President Kiir during the run-up to South Sudan’s independence, these elites were among those who had begun to openly criticize the president in2013. Many of them were dismissed from government posts in July 2013, six months before the conflict erupted. When released, they opted to join neither thegovernment nor the opposition, hoping instead to chart a “third way” at the peace process. They were largely unsuccessful.

10 The expanded mediation format—“IGAD Plus”—included IGAD members as well as five African Union member states, the United Nations, the Troika (US, UK,Norway), the European Union, and China. For more detail on IGAD Plus, see the fourth section of this paper, on regional competition (p. X).

11 In the end, the ARCSS included chapters on: (1) a transitional government, (2) a permanent cease-fire and transitional security arrangements, (3) humanitarianassistance and reconstruction, (4) economic and financial management, (5) transitional justice, (6) a permanent constitution, and (7) joint monitoring and evalua-tion, as well as eight corresponding implementation appendices. “Agreement on the Resolution of the Conflict in the Republic of South Sudan,” Addis Ababa,Ethiopia, 17 August 2015.

Page 8: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

4 Zach Vertin

opposition, the SPLM former detainees, andeventually the government—bowed to focusedinternational pressure and signed the accord.Despite its flaws, a comprehensive deal had been

inked that could conceivably end the fighting,frame a post-conflict transition, and begin the tasksof reconciliation and reform.12 But it was adocument that significant constituencies—some ofwhom were opposed to any compromise—refusedto accept. A Joint Monitoring and EvaluationCommission (JMEC) was established in October2015 to oversee implementation. Headed by aformer African president, JMEC’s membersincluded the South Sudanese stakeholders, IGADmember states, the African Union (AU), theUnited Nations, and the wider set of nations thathad formed IGAD Plus. Implementation was slow,however, and because IGAD heads of state werereluctant to cede authority to, or robustly back, theoversight body, it struggled to enforce the deal orhold the parties accountable.The UN Mission in South Sudan (UNMISS),

with some 12,000 troops in-country and a multidi-mensional Chapter VII mandate, was to be animportant implementing partner.13 It played nodirect role in the preceding mediation process,however, as it was consumed with critical humani-tarian tasks, including, above all, the protection ofsome 200,000 civilians who had sought refuge onUN bases since the conflict began.Machar returned to Juba in April 2016 per the

terms of the accord’s power-sharing protocol, buttensions simmered. Just three months later, theyboiled over and the capital again descended intoviolence. Machar was driven out of the country in adramatic manhunt, and the deal, while not whollydead, was effectively suspended. More than a yearof fighting and large-scale displacement passedbefore IGAD and the wider internationalcommunity would attempt to “revitalize” the peaceprocess.14 In the meantime, conflict dynamicsevolved and fighting spread to new parts of thecountry—areas that had not, to date, been directlyimpacted by the violence.

The Mediation Context

Peace processes are often complex, messy, andnonlinear affairs. As such, few mediation effortsmeet the aspirational principles outlined in theUnited Nations’ Guidance for Effective Mediation.15This guide is intended to inform the design andmanagement of mediation processes; based onlessons learned around the globe, its principlesreflect the ideal environment for third-partymediation. Such environments are rare, however,and mediators must adapt to imperfect circum-stances, make tradeoffs, and in some cases breakthese rules to advance their objectives. Nonetheless,these “mediation fundamentals” remain usefulbenchmarks for framing any particular third-partymeditation effort, allowing practitioners to assessdecisions made and dilemmas confronted, andconducting comparative analysis across mediationefforts.As such, before further examining the peace

negotiations in South Sudan, this section firstassesses the mediation context. Drawing on the UNguidance, it identifies six principles of mediationbest practice and demonstrates whether or not eachwas applied to the process in South Sudan. Theseshort analyses should illustrate what opportunitieswere missed and what circumstances precluded themediators or peace process supporters fromadhering to these first principles of mediation. Asevidenced here, as well as in the third section of thepaper on the negotiations, deficits in five of thesesix areas demonstrate an unusually fraughtmediation context.MANDATE, ARCHITECTURE, ANDCOHERENCE OF EFFORTS

Mediation is most successful when practiced by alead mediator, ideally representing a single institu-tion and grounded in a clear mandate. Who ischosen to lead depends on the specifics of theconflict and an assessment of comparativeadvantages. Once determined, coordination amonga wider set of peace process supporters is thencritical to developing a coherent process, including

12 Donald Booth, “South Sudan’s Peace Process: Reinvigorating the Transition,” Chatham House, London, UK, February 9, 2016.13 In October 2015, the UN Security Council revised UNMISS’s mandate to include support for implementation of the peace agreement. UN Security Council

Resolution 2241 (October 9, 2015), UN Doc. S/RES/2241.14 The revitalization initiative was endorsed by IGAD in June 2017, and a new team of special envoys began consultations in August 2017. IGAD, “Communiqué́ of the

31st Extra-ordinary Summit of IGAD Assembly of Heads of State and Government on South Sudan,” Addis Ababa, June 12, 2017.15 United Nations, Guidance for Effective Mediation, 2012, available at https://peacemaker.un.org/guidance-effective-mediation .

Page 9: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


consistent political messaging, resource support,and a division of labor.16

On December 27, 2013, as the violence in SouthSudan escalated, the IGAD heads of state convenedin Nairobi for an emergency summit.17 Citingconcerns about the conflict, its increasingly ethnicovertones, and reports of widespread atrocities, theycalled for a cease-fire and an inclusive politicaldialogue. They resolved that “face-to-face talks byall stakeholders” should begin in ninety-six hours.The summit communiqué left the details to be filledin by the countries’ respective foreign ministers anda team of newly appointed “special envoys.”18

Former Ethiopian Foreign Minister SeyoumMesfin was one of the appointees and soon becamethe chief mediator of the IGAD peace process.General Lazaro Sumbeiywo of Kenya was the secondappointee, and while not explicit in the document, itwas later resolved that he would serve as deputy. Ina bizarre stroke that would prove common to IGADsummits, a third name was added to the docket,albeit after the communiqué was issued: GeneralMuhammad Ahmed al-Dabi of Sudan, whorounded out a three-headed mediation team.IGAD’s internal politics were at the heart of this

last-minute addition, as regional foreign ministerssought to appease Khartoum by balancing represen-tation. They believed it better to have the Sudanesegovernment inside the tent than out. Though itdrew little attention at the time, the haphazardpersonnel decision was an early signal that IGAD’sinternal politics would sometimes take precedenceover the needs of the peace process. Institutionalpolitics shape most mediation efforts, but in timeIGAD’s internal problems would infect themediation team and complicate the peace process.IGAD’s communiqué left something to be

desired in terms of structure and detail, but the

mandate for a forthcoming mediation effort wasclear enough to get started.19 And importantly, theIGAD leaders did articulate one clear request to theAU, UN, and wider international community:support the nascent IGAD process. Such momentsare critical; as international players attempt tocoalesce around an appropriate mediator,interested third-party actors and institutions oftenjockey for position.While the UN, AU, United States, and a range of

eminent personalities could conceivably have fit thebill, most diplomats and observers close to the crisisbelieved it appropriate for IGAD to lead themediation. The direct involvement of neighboringstates in a mediation can be a double-edged sword(as outlined later), but the rationale for IGAD’sleadership in this instance was widely shared. Theorganization’s most influential member states—Kenya, Ethiopia, and Uganda—all had a longhistory in South Sudan. Each had played a criticalrole in ending Sudan’s civil war and safeguardingthe South’s independence in 2011, and so each hadentrée with its political leadership. Moreover, eachof them shared a border with South Sudan andwould be most impacted by conflict spillover andrefugee flows, and so they had a shared interest inpreventing collapse. They enjoyed generally cooper-ative relationships with each other and had provenable to work together on regional security issues.Some African Union representatives appealed for

an AU lead, but most believed IGAD to have thecomparative advantage. In private consultationsbefore the IGAD summit, US Special EnvoyDonald Booth and UN Special Envoy HaileMenkerios made clear their support for an IGADlead, and European partners followed.20 Three dayslater, the AU Peace and Security Council endorsedIGAD’s leadership, and the UN Security Council

16 Ibid., p. 19.17 The summit was preceded by diplomatic outreach to Juba, including a visit by Ethiopian Prime Minister Hailemariam Desalegn and Kenyan President Uhuru

Kenyatta, a visit by IGAD foreign ministers together with AU and UN representatives, and visits by senior diplomats from the United States and Europe. Each delega-tion urged the government to stop the violence and engage in dialogue with its opponents.

18 The term “special envoy” was used throughout the process to refer to the designated mediators. For simplicity’s sake they are referred to in this paper as “mediators.”19 The IGAD communiqué of December 27, 2013, called on the South Sudanese parties to “undertake urgent measures in pursuit of an all-inclusive dialogue including

reviewing the status of the detainees.” It also called for broad participation and determined that “face-to-face talks by all stakeholders in the conflict should occur bythe 31st of December 2013.” To this end, it also called on the conflict parties to “liaise with IGAD envoys [mediators] and the Council of Ministers to support theprocess of dialogue and related political and technical reforms.” The communiqué also appointed the mediators and tasked the IGAD secretariat to “avail its goodoffices in support of the above process.” IGAD, “Communiqué of the 23rd Extra-ordinary Session of the IGAD Assembly of Heads of State and Government on theSituation in South Sudan,” Nairobi, December 27, 2013.

20 Former South African President Thabo Mbeki had led an AU panel originally focused on the conflict in Darfur (2009). His mandate was later amended and his AUpanel led negotiations over the partition of Sudan and South Sudan (2010–2012). In October 2012, his panel’s mandate was again amended to include “the promotionof the democratic transformation in Sudan and South Sudan.” Though Mbeki aides appealed for an AU lead, private concerns were expressed about the Mbeki team’ssuitability for the task and acceptability to the parties.

Page 10: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

6 Zach Vertin

followed suit.21 This organization of internationalactors around a single, coherent mediation effort isnecessary to give any peacemaking attempt achance to succeed, and South Sudan was noexception.An architecture was beginning to emerge:

IGAD’s designated mediators would lead the peaceprocess, buttressed by a small group of diplomatsfrom the AU, UN, and South Sudan’s biggestbilateral partners—the United States, selectEuropean actors, and, to a limited extent, China.Though the rationale was not stated in such terms,the principle of “subsidiarity,” whereby regionaland subregional organizations are the first torespond to matters of international peace andsecurity, had been reflected in practice.22 Initialfunding had been secured. All that remained wasfor IGAD to assemble a strong technical secretariatto support the mediators.MEDIATOR PREPAREDNESS

An able and experienced mediator seems anobvious starting point, but too often mediation isviewed not as a unique competency but as a taskthat can be assumed by any statesman ofconsequence. Gravitas alone does not make aqualified mediator, however, and in any case, aviable process requires more than one individual.Mediator preparedness combines the “individualknowledge and skills of a mediator” with both a“cohesive team of specialists” and sufficientpolitical, financial, and administrative supportfrom the mediating entity.”23

The Mediators

Seyoum Mesfin was deeply acquainted withregional politics, having served as Ethiopia’s chiefdiplomat for more than two decades. He was notonly a member of the ruling party’s centralcommittee, but a founding member of its minorityTigrayan constituency, which has dominatedEthiopia’s coalition government since it came topower, also by way of a liberation movement, in the1990s. Seyoum had been Ethiopia’s foreignminister during the Sudanese civil war and

throughout the peace process that ended it, and hewas deeply acquainted with the pivotal role theUnited States had played in negotiations. With thisin mind, Seyoum met with US Special EnvoyDonald Booth in the days before the talks began,asking again for political backing from Washingtonand a partnership with Booth’s team, as well asother Western supporters.The veteran diplomat would commit himself

admirably to a thankless process, but early tacticalmistakes made apparent that Seyoum, despite hiscredentials at home and abroad, had less mediationexperience than many had assumed. In time,Seyoum would also have to navigate difficult watersinside his own government, where senior political,military, and party officials took an interest in theprocess, its outcomes, and its implications forEthiopia’s standing in the region.General Lazaro Sumbeiywo was a retired career

officer in Kenya’s national army but was bestknown for his role as chief mediator of an earlierIGAD peace process that had ended Sudan’s civilwar in 2005. As a result, he was well known toSouth Sudan’s warring factions and to key bilateralpartners. Nearly a decade had passed, however, andSumbeiywo was not the same man. Once a closeconfidant of former Kenyan President Daniel ArapMoi, Sumbeiywo did not enjoy the same access tothe current president, Uhuru Kenyatta, orinfluence inside his administration. Thoughinitially content with his role as deputy to Seyoum,the two men had very different styles, and theformer general soon chafed at being second-in-command.Muhammad Ahmed al-Dabi’s appointment as a

third mediator was a puzzling decision for mostobservers, even when accounting for IGAD’sinternal politics. Al-Dabi had once served as thehead of Sudan’s powerful military intelligenceagency and was implicated in the darkest eras ofKhartoum’s domestic wars. Moreover, only a yearearlier al-Dabi had served as chair of an ArabLeague observer mission in Syria where he cameunder widespread criticism for his public

21 African Union Peace and Security Council, communiqué from 411th meeting, Banjul, the Gambia, AU Doc. PSC/AHG/COMM.1(CDXI), December 30, 2013; UNSecurity Council, “Security Council Press Statement on Situation in South Sudan,” UN Doc. SC/11227-AFR/2775, December 30, 2013.

22 Chapter VII of the UN Charter (Articles 52–54) outlines subsidiarity norms, noting that regional entities “shall make every effort to achieve pacific settlement oflocal disputes through such regional arrangements or by such regional agencies before referring them to the Security Council.”

23 United Nations, Guidance for Effective Mediation, p. 6.

Page 11: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

statements and handling of the mandate.24 WhenSudan began dabbling in South Sudan’s conflict,including occasionally supplying weapons andammunition to Riek Machar’s opposition forces,concerns about al-Dabi’s role resurfaced: Might hecompromise confidential information or attemptto influence the parties? In practice, al-Dabi provedto be the least engaged of the three mediators, andhis involvement proved mostly innocuous. But theoptics of his appointment would damage thecredibility of the mediation in the eyes of somestakeholders and observers.The Secretariat

During Phase I of the peace process, the mediatorswere supported by a very small team of Ethiopianand Kenyan aides. The chief mediator regularlysought counsel and informal support from a smallgroup of senior international diplomats, includingmembers of the so-called Troika (the US, the UK,and Norway), as well as the UN and EU. But hemaintained a tight circle when it came to strategy,relying on substantive inputs from his deputymediator, his Ethiopian chief of staff, and a fewtrusted American diplomats.Before Phase II of the peace process began, the

Americans, supported by European partners,presented Seyoum with a memo proposing a morerobust secretariat. An inclusive political dialogueaiming to remake South Sudan, they argued, wouldrequire a secretariat of technical specialists tailoredto the task. The memo thus recommended that themediators handpick advisers in process design andstrategy, economics, security, constitutionalmatters, and strategic communications. The USand European partners offered to foot the bill, butthe mediators could recruit whichever experts theysaw fit. Seyoum declined. Reluctant to cede control,the veteran diplomat opted not to widen his tightcircle. Seyoum was grateful for the private supporthe was getting from peace process supporters but

was also “under pressure from the region to ‘makeit local,’” one member of the IGAD secretariat laterexplained. In practice, this made for a shorthandedmediation team, as “there was not enough capacity,not enough resources.”25

Four months later, the chief mediator agreed tohire one principal outside adviser and slowlyexpanded the support staff. But concerned aboutoptics, the mediators remained reluctant to harnessoutside expertise. Peace process supporters lateroffered to fund the hiring of experts to help managespecific issue sets—some with country-specificknowledge, others with thematic expertise,including a team from the UN Mediation SupportUnit in New York.26 But again, despite capacitydeficits, the mediators made little use of outsideassistance.27


At the request of IGAD, Norway and otherEuropean partners financed the peace process,supplemented by contributions from the UnitedStates.28 Despite the readiness of partners toprovide funds and thereby take a major concern offthe mediators’ plate, coordination and expectationmanagement remained constant challenges. As theUS envoy later reflected, “When Western nationsfund peace processes they do not control, tensionsemerge as those funds are sometimes attached topolicy preferences, legal requirements, or politicalobligations back home.”29 IGAD, meanwhile, hadits own priorities, which did not necessarily reflectthose of the donors.Despite the challenges presented by this division

of labor, Seyoum and his deputies nonetheless hadstrong backing from a coterie of internationaldiplomats. Political support from IGAD memberstates, however, was lacking. Though regionalheads of state had tasked them with an officialmandate, in practice the mediators quickly foundthemselves operating on a very short leash. The


24 For example, see Kareem Fahim, “Chief of Arab League’s Mission in Syria Is Lightning Rod for Criticism,” New York Times, January 2, 2012. 25 Interview with member of IGAD secretariat, Addis Ababa, January 2018.26 The UN Mediation Support Unit sent experts in transitional governance, security, and constitutional processes. They were among the experts who delivered

presentations to South Sudanese stakeholders on a range of post-conflict transitional models during a symposium organized by IGAD and supporters in June2014.

27 Later in the process, when agreements being finalized had implications for UNMISS, the mission sent planners from Juba to Addis Ababa. UN officials, USdiplomats, and other peace process supporters had to make repeated entreaties to the mediation team to ensure these officials were party to discussions that madedemands of UNMISS resources or impacted its operations.

28 The full list of partners contributing resources to the peace process and supporting cease-fire mechanisms includes China, Denmark, the European Union, Italy,Japan, the Netherlands, Norway, Sweden, the United Kingdom, and the United States.

29 For a fuller discussion of financing dilemmas, see Booth, “South Sudan’s Peace Process.”

Page 12: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

heads of state retained ultimate control over thedirection of the process and its outcomes. Whensummit decisions ignored the mediators’ progressor simply overturned their decisions, their lack ofauthority was exposed, and the parties henceforthcalculated accordingly.CONSENT

“Mediation is a voluntary process that requires theconsent of the conflict parties,” notes the UNGuidance for Effective Mediation. Without it, thoseparties are unlikely to “negotiate in good faith or becommitted to the mediation process.”30 Consentmay be the most crucial ingredient of anymediation effort, and yet very often—as in SouthSudan—it is also the most elusive. The Colombianpeace process is one recent exception, though ittook nearly half a century of conflict before boththe government and the Revolutionary ArmedForces of Colombia (FARC) came to the negoti-ating table of their own volition.31

South Sudan’s warring parties, by contrast, cameunder extraordinary international pressure tonegotiate a cease-fire and were forced to the tablewithin weeks of the outbreak of conflict. Nohurting stalemate existed on the battlefield, nor dideither party express a desire to do anything butdefeat its opponent by military means.32 Thesecurity situation remained highly fluid, with fewclear battle lines and an impromptu rebellion stillin the making. In short, this was not a conflict“ripe” for settlement. But the human costs of theconflict’s first weeks, marked by brutal violenceand civilian targeting, meant outside actors wereunable to stand by and wait for the parties’ consent.The situation thus presented a familiar dilemma ofinternational peacemaking in the modern era: howto compel a mediated resolution of conflictbetween two parties bent on war.If the IGAD mediators and their international

partners had waited for the parties to requestmediation, they would likely still be waiting. Nopeace process would have materialized, and withcombatants free from any constraints, millions ofvoiceless civilians would be in grave danger. The

ethnically motivated massacres in Juba and revengeattacks that followed suggest that a cycle ofgenocidal violence might have ensued, with notelling how high the death toll might have climbed.Regional states and international partners thus

resolved that they must use whatever diplomatictools available to get the parties talking. Callspoured in from heads of state and foreign ministersaround the world; expressions of outrage andwarnings of sanctions and criminal accountabilitywere coupled with appeals to each leader’s ego,moral standing, and legacy. The arm-twistingrequired left no doubts as to the absence of consent,and thus to a decidedly unfavorable mediationcontext. Seyoum and his team would begin theprocess with the deck stacked against them.Kiir’s administration went along with the process

halfheartedly, keen to alleviate internationalpressure and interested only insofar as the processmight help it stamp out the rebellion. But whenIGAD and peace process supporters dismissedJuba’s narrative of an attempted coup d’état,already reluctant government negotiators seethed.When Seyoum and bilateral partners—specificallythe United States—later pressed for a politicaltransition and a multi-stakeholder format for thetalks, the government withdrew further, stymieingthe process at every turn while doubling down on amilitary solution. At moments of maximuminternational pressure, Juba made temporaryconcessions to avoid opprobrium, only to return toa policy of obfuscation when attention subsided.Machar’s coalition of anti-government forces

likewise began the process reluctantly and withmaximalist positions. They had more to gain frommediation, but the opposition rank and file wasbent on overthrowing Kiir and his government. Itwas not until mid-2015 that partial consent wasforthcoming from Machar, and even then it did notrepresent the majority of constituencies fightingunder his loose command. Consent to a mediationis often vulnerable when disputes within negoti-ating parties arise, creating new pressures on theprocess and presenting mediators with difficult

8 Zach Vertin

30 United Nations, Guidance for Effective Mediation, p. 8.31 For more on the Colombian peace process, see Renata Segura and Delphine Mechoulan, “Made in Havana: How Colombia and the FARC Decided to End the

War,” International Peace Institute, February 2017, available at www.ipinst.org/2017/02/how-colombia-and-the-farc-ended-the-war .32 A “hurting stalemate” exists when both (or all) warring parties are suffering and have reached the conclusion that they cannot achieve an outright victory. This is

often the moment when third-party mediation has the best chance of success.

Page 13: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

choices. The tide eventually turned against SouthSudan’s anti-government forces, which weremeanwhile bending under the weight of internaldivisions. Machar knew it was time to deal andstruggled to balance his consent for a negotiatedsolution with more strident popular demands byhis supporters.33

As the increasingly polarized conflict wore on andnarratives in both camps hardened, the prospect ofa negotiated settlement became even harder foreither leader to countenance. Even if Kiir or Macharcalculated that a negotiated settlement was theirbest course of action, both men were politicallyvulnerable; hardliners in their respective campsmade threats that effectively raised the cost ofconsent. For example, on numerous occasionssenior military officers in Kiir’s government,including then army chief of staff Paul Malong,threatened to kill the president if he agreed to a deal.IMPARTIALITY

For any mediation effort to succeed, the partiesmust believe the designated mediator to be fair andbalanced. It goes without saying that any materialinterest in the outcome of a negotiation wouldcompromise a mediator’s credibility with theparties.34 Unfortunately, IGAD’s peace process wascorrupted by both perceptions of mediator bias andthe material interests of its member states.Of the interests of regional states that compli-

cated the process, Uganda’s military interventionin South Sudan was most damning.35 PresidentYoweri Museveni sent Ugandan troops and warplanes into South Sudan in the earliest days of theunfolding war, fighting alongside governmentforces and positioning himself as Juba’s principalally. Museveni’s army played a major role in haltingthe opposition’s January 2014 assault on the capital

and was widely credited with saving Kiir’s govern-ment and preventing further atrocities in Juba.Many in the region and the internationalcommunity welcomed this limited preventiveaction, as well as the safe evacuation it affordedforeign nationals. But they were simultaneouslyconcerned about Museveni’s rhetoric and the riskspresented by the possibility of more partisanUgandan involvement.36 When that initial contestwas over, the Ugandan army not only remained inSouth Sudan but also led coordinated air andground offensives against the opposition.37

As a result, the cessation of hostilities agreementsigned in January 2014 explicitly called for thewithdrawal of all “armed groups and allied forcesinvited by either side from the theatre ofoperation.”38 The language was a thinly veiledreference to Uganda’s army, but Musevenifrustrated IGAD partners and peace processsupporters by ignoring it. Later, as Phase II began,a Ugandan diplomat was invited to join the discus-sions alongside the mediators, but oppositionnegotiators refused to engage him. “You’rewelcome to participate,” one of them declared, “aslong as you sit with your colleagues on the govern-ment’s side of the table.”39

Uganda’s continuing presence bolstered thegovernment’s position and made it even lessreceptive to mediation. President Museveni’srefusals to withdraw or to use his leverage with Kiirwere undermining the mediation and taken asslights in Ethiopia, where chief mediator SeyoumMesfin and Prime Minister Hailemariam Desalegnfumed. They handled Museveni delicately inperson but were apoplectic behind closed doors,and after their own entreaties failed, they repeat-edly petitioned the Americans to force his hand.40


33 Just weeks before the ARCSS was signed, two senior opposition commanders announced a split from Machar, a declaration the government attempted to exploitat the negotiating table. Later, just days before the accord was inked, another senior member of the opposition announced his defection and the creation of a newpolitical entity.

34 United Nations, Guidance for Effective Mediation, p. 10.35 The interests of regional states and international partners are further detailed in the fourth section of this paper (p. 18).36 Contrary to widespread rumors at the time, Ugandan intervention was neither requested nor facilitated by the United States.37 Several UN reports detailed UPDF involvement in the conflict, including the alleged use of cluster munitions. See UNMISS, “Conflict in South Sudan: A Human

Rights Report,” May 8, 2014.38 “Agreement on Cessation of Hostilities between the Government of the Republic of South Sudan and the Sudan People’s Liberation Movement/Army in

Opposition (SPLM/A in Opposition),” Addis Ababa, January 23, 2014, available at https://peacemaker.un.org/southsudan-cessation-of-hostilities-SPLM/AOpposition .

39 Discussions with official present in the meeting, Addis Ababa, February 2014.40 Both Seyoum and Hailemariam traveled to Uganda on separate occasions to meet with Museveni, but their appeals yielded no change in his approach.

Washington made numerous private appeals to Museveni and publicly reiterated the cease-fire’s stipulation that foreign forces be withdrawn but chose not toconfront him publicly. For more on US engagement with Uganda, see the fourth section of this paper (p. 18).

Page 14: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

10 Zach Vertin

41 United Nations, Guidance for Effective Mediation, p. 11.42 Though the war was being fought primarily in territories dominated by Dinka and Nuer communities, mediators and peace process supporters spent considerable

time contemplating an appropriate role for the so-called “Equatorians.” The many ethnic communities that call the Equatoria region home could not be left out ofdiscussions on the future of South Sudan, yet many worried that introducing Equatorian representatives as an independent bloc could set a dangerous precedentand reinforce the ethno-regional divisions many South Sudanese sought to overcome.

43 IGAD’s communiqué from its March 2014 summit reaffirmed the need for an inclusive process and called on the parties to allow participation from otherpolitical parties, former detainees, and civil society organizations as deemed necessary by the mediators. The May 9th agreement, signed by Salva Kiir and RiekMachar in 2014, explicitly committed the parties to an inclusive process “in order to ensure broad ownership of the agreed outcomes” and named the sixstakeholder groups first identified by the mediators in February. “Agreement to Resolve the Crisis in South Sudan,” Addis Ababa, May 9, 2014, available athttps://peacemaker.un.org/southsudan-agreement-resolve-crisis2014 .

44 For example, a broad range of civil society actors were invited to Addis Ababa in March 2014 to determine their role in the process and choose a team of delegatesto represent them. When the mediators convened this group to elect representatives, Sumbeiywo allowed the government and opposition parties to weigh in,fatally corrupting an already fraught exercise.

45 “S Sudan Civil Society Chairman Has Theory Why He Was Shot,” Radio Tamazuj, September 18, 2014, available at https://radiotamazuj.org/en/news/article/s-sudan-civil-society-chairman-has-theory-why-he-was-shot .

Perceptions of mediator bias also undercutIGAD’s impartiality. True or not, allegations thatLazaro Sumbeiywo was aiding governmentnegotiators swirled about hotel corridors in theearly stages of Phase II, raising eyebrows amongopposition delegates, peace process supporters, andSeyoum himself. IGAD secretariat officialsprivately confirmed that Sumbeiywo had leakeddocuments to government negotiators. During andafter the process, unconfirmed allegationscirculated that Sumbeiywo had further collabo-rated with Kiir’s government to influence theprocess in its favor. Though it proved a compara-tively lesser concern, questions also lingered aboutal-Dabi’s perceived impartiality, given Sudan’stroubled history with South Sudan.INCLUSIVITY

Who gets a seat at the table, and who does not? Thisis a central question in any peace process, and thepolitics of participation are often heated. Again,South Sudan was no exception. Mediation bestpractice suggests inclusive peace processes aremore likely to address the root causes of conflictand increase the legitimacy of the process.41 Apower negotiation between armed combatants maybe enough to halt immediate violence, but forginga sustainable peace in South Sudan would requireboth wider participation in a political process(including defining its objectives) and ownershipof its outcomes.With this in mind, the Americans proposed a

“multi-stakeholder” format for the second phase oftalks. They worked closely with Seyoum andSouthern Sudanese constituents to ensure a place atthe table not just for the warring factions but alsofor other political parties, the former detainees,civil society groups, women’s organizations,churches, and elders.42 But the government refused,

reluctant to invite any talk of political transition.For weeks, an increasingly frustrated Seyoumattempted to secure agreement simply on thestructure and rules for the proposed dialogue, butgovernment negotiators fought tooth and nail tocurtail the scope of the process and the list ofparticipants.Machar’s opposition had more to gain from an

inclusive process, as most of the additional voiceslikewise sought to constrain the power of thegovernment. But Machar and his delegation failedto appreciate this reality, believing they too wouldbe better off dividing the cake in two. As a result,government and opposition delegates ultimatelyconspired to restrict participation.The mediators, with strong backing from peace

process supporters, pushed ahead with the multi-stakeholder format nonetheless. They secured anexplicit mandate and subsequent affirmations fromIGAD leaders,43 but critical process mistakes madean already uphill battle even more difficult.44Wheninvitations for inclusive talks were finally issued,the government first refused to appear and laterintimidated or co-opted representatives from otherstakeholder groups. On one occasion, the govern-ment blocked invited political party leaders fromparticipating by preventing them from boarding aplane bound for Ethiopia. On another, aftercriticizing government and opposition negotiatorsfor ignoring widespread suffering, the leader of thecivil society delegation was shot by an unidentifiedassailant and never returned to the negotiations.45

Peace process supporters pleaded with themediators not to ignore such blatant processviolations. But in both instances, the mediatorschose to proceed with talks as scheduled. Toodesperate to keep the process alive, they seemed notto appreciate that their repeated concessions meant

Page 15: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


they were forfeiting any chance of success.Frustrated by the parties’ intransigence, the task ofmanaging a multi-stakeholder process, and the lackof political reinforcement from their bosses, themediators capitulated. “This inclusive process isjust too difficult,” one of them admitted privately.By the end of 2014, the peace process would bereduced to a three-faction affair. It seemed theremaking of South Sudan would belong to threeSPLM-dominated groups—the government, theopposition, and, in a minor role, the formerdetainees.Mediators face a difficult task in finding a sweet

spot between inclusivity and efficiency, as the valueof wider consultation can easily be outweighed bythe costs of a slow and unwieldy process. The UNGuidance for Effective Mediation emphasizes thatparticipation need not equate to a seat at the tableand recommends that mediators develop alterna-tive mechanisms to broaden participation.46 To thisend, US diplomats penned a proposal outlining aseries of measures the mediators might use toexpand participation, including town hall meetingsin South Sudan, a formal feedback loop to funnelinput from South Sudanese communities back tothe negotiating table, and dedicated radioprogramming to broadcast plenary sessions andupdates from the mediators. These low-cost activi-ties might help increase awareness, combatconcerns about dialogue happening on foreign soil,and build national ownership, but the mediatorscould also use these wider inputs to expand thepolitical agenda in Addis Ababa. But unaccus-tomed to such transparency, and doubting theutility of an active communications strategy, themediators again passed on the opportunity.STRATEGY AND SEQUENCING

Strategy and sequencing are foundational elementsof any mediation effort and can position a mediatorto both drive the process and respond effectively tochallenges that inevitably arise. Strategies for eachphase of a mediation can be built and adjustedbased on clear identification of objectives, acomprehensive analysis of the conflict, mapping ofstakeholders (including their positions, interests,and bargaining power), and an appreciation of the

operating environment (including the availabilityof leverage).47

Conflict narratives often differ among competingparties, but as South Sudan’s peace process enteredits second phase, there remained no consensusamong stakeholders (or the wider communitysupporting the peace process) as to the nature andorigins of the conflict or what was required to settleit. This lack of consensus made the first-ordertasks—analyzing the conflict and identifyingappropriate objectives for the peace talks—all themore difficult for the mediators.Also critical for any credible negotiation is a

mediator’s acceptance by the parties, which mustsee the designated individuals as capable, judicious,and authoritative.48 IGAD’s appointed mediatorsfaced an extremely difficult mediation context, asevidenced in the preceding sections. But the lack ofa coherent strategy was made worse by regularprocess mistakes, delays, and occasional confusion,which in turn weakened the mediators’ credibilityin the eyes of the parties and contributed tomounting frustration among peace processsupporters.Time and again the mediators shifted approaches

or made unwarranted concessions, signaling to theparties that intransigence could be rewarded. Thechief mediator came under criticism for being tooflexible with hardened South Sudanese negotiatorsand too willing to accommodate. South Sudanesestakeholders and observers complained that he andhis colleagues needed to show more backbone. Onone occasion when the mediators did take anunusually firm line and stuck to it, even SouthSudanese delegates whose interests were hurtapplauded their resolve.On other occasions, the mediators seemed not to

make use of the tools available to them. Amonitoring mechanism was established during thefirst phase of the process to verify compliance withan agreed cease-fire. The monitoring body wouldissue public reports of cease-fire violations, whichcould be harnessed to hold the parties accountableand even create some leverage during the secondphase of talks. But once established, Seyoumconfused stakeholders and peace process

46 United Nations, Guidance for Effective Mediation, p. 11.47 Ibid., p. 6. 48 Ibid.

Page 16: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

12 Zach Vertin

49 Peace process supporters eventually convinced the mediators to release the reports publicly, but not until 2015.50 Ibid., p. 23.51 Synthesized analysis and some text/quotations included in the following sections are drawn from a forthcoming book by the author, A Rope from the Sky: The

Making and Unmaking of the World’s Newest State (2018).52 United Nations, “Secretary-General, Welcoming Cessation of Hostilities in South Sudan, Calls for Accord’s Immediate Implementation,” UN Doc. SG/SM/15610-

AFR/2801, January 23, 2014.

supporters when he decided not to release thereports. Absent a single, credible authority toestablish the facts, the parties repeatedly violatedthe cease-fire and pointed fingers withoutconsequence. Whatever the reasons for hisreluctance, the mediator effectively took a valuablenegotiation tool off the table.49

Seasoned mediators appreciate the fact thatprogress toward peace is often made not duringofficial talks but in the days, weeks, or monthsbetween formal discussions. However, giventensions among the members of the mediationteam and their respective IGAD countries, Seyoumand his deputies regularly went their separate wayswhen one round of negotiations finished. Theysometimes reconvened just hours before talks werescheduled to resume and began consultingstakeholders anew without any game plan for howto advance the process. Peace process supportersmade several offers to host “strategy sessions” inthe intervening periods, but no such workingretreats ever materialized.

The Negotiations

Each of the principles examined above can shapethe mediation context, and clearer adherence tothem—by IGAD, its designated mediators, or theirdiplomatic supporters—may have altered thecharacter of the South Sudanese process. Yet inreviewing the peace process and the approach ofthe mediators, it is imperative to reiterate thecentrality of political will. “The success or failure ofa mediation process,” notes the UN guidance,“ultimately depends on whether the conflict partiesaccept mediation and are committed to reaching anagreement.”50 South Sudan’s principal combatantsnot only lacked the political will to make peace butwere also often hostile to the very idea of a negoti-ated settlement. Against this backdrop, this sectionoffers further insight into the first and secondphases of the negotiations and considers sevenfactors that shaped the environment and outcomesof the peace process.51


As the parties arrived in Addis Ababa to begintalks, one African diplomat opined that if third-party intervention could not arrest the violence “inthe first two weeks,” it would likely “last for monthsor years. There is usually no middle ground.” Withthis tipping point in mind, and the battle for SouthSudan’s capital city intensifying, the mediatorsresolved that an immediate cessation of hostilitieswas the first order of business. The national armyhad fractured in two, battles raged north of Jubaand in three state capitals, and more and moreSouth Sudanese were being drawn into the fight.Seyoum understood that no such cessation wouldlast unless underlying political grievances wereaddressed, but he and the IGAD heads of stateresolved that a reduction in violence was necessaryto create the space for that discussion. Thissequencing decision was supported by diplomatspresent from Africa and the wider internationalcommunity.Nineteen days later, the parties signed a cessation

of hostilities agreement—the product of intensivenegotiations, focused international attention, andclose cooperation between mediators and peaceprocess supporters. The deal committed govern-ment and opposition forces to cease offensiveoperations, freeze their forces in place, and refrainfrom attacks against civilians. It also established amonitoring and verification mechanism andoutlined its composition, operations, and reportingmandate.UN Secretary-General Ban Ki-moon was among

the world leaders who welcomed the deal. Echoinga wider sense of urgency, he also underscored “thenecessity to continue without delay” toward a“national political dialogue to reach a comprehen-sive peace agreement.”52 Though intended to belean and mobile, the monitoring mechanism wouldtake weeks to set up. In the interim, with the ink onthe deal barely dry, new clashes erupted, and thetwo sides began pointing fingers.

Page 17: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


53 United Nations, Guidance for Effective Mediation, p. 6.54 IGAD, “Communiqué of the 24th Extra-ordinary IGAD Summit on South Sudan,” Addis Ababa, January 31, 2014.55 Centre for Humanitarian Dialogue, “A Guide to Mediation: Enabling Peace Processes in Violent Conflicts,” 2007, p. 3.56 Each negotiating team was led by a high-profile figure, but neither had been given the autonomy required to negotiate. Nearly every decision required phone calls

back to Juba or Gadiang, where Machar’s impromptu rebellion had established its bush command post.57 The May 9th negotiation was the first time both men participated in the process in Addis Ababa, but despite initial plans, they did not meet face-to-face or shake

hands upon signature of the interim accord.58 That agreement came just days after US Secretary of State John Kerry visited Juba and pressured Kiir to commit to negotiate the terms of a transitional govern-

ment for South Sudan. “Agreement to Resolve the Crisis in South Sudan,” Addis Ababa, May 9, 2014, available at https://peacemaker.un.org/southsudan-agreement-resolve-crisis2014 .


Despite its limitations, the cessation of hostilitiesnominally provided a basis on which to begin talkson the underlying causes of the conflict. The speedwith which South Sudan had come undone hadprompted hard questions. Was South Sudan inneed of a course correction? Was it fast becoming afailed state, or had it failed before ever becoming astate? What exactly should Phase II of the talks aimto achieve?As the process entered this more complex phase

of dialogue, one of the challenges Seyoum faced (aswould any mediator) was to “maintain a sense ofurgency while avoiding quick-fix solutions.”53Immediate political and military realities could notbe ignored, but a quick-fix accommodation amongSouth Sudan’s ruling elite might simply set thestage for a repeat disaster. The initial reflex ofIGAD leaders and the mediators was to engage thepolitical elites they were used to dealing with. Butpeace process supporters urged the mediators toinclude a much broader swathe of South Sudanesesociety in the process. Many in South Sudan andabroad wanted to see Phase II yield more than justa division of spoils between warring factions orSPLM power brokers. They sought justice andreconciliation, economic and institutional reforms,and political accountability. Could Phase II be anentry point for more transformational change?IGAD mandated the mediators to “develop a

framework” for an inclusive dialogue, which was toinclude “specific modalities on structure, represen-tation, and timeframe.”54 In light of the ongoingwar, the mediation context, and demands for acomprehensive political agenda, Seyoum and hisdeputies also had to determine what kind ofmediators they were going to be. Third-partymediators can adopt roles ranging from “facili-tator” (creates conditions for dialogue) to “problemsolver” (drives the process and proposes solutions)

to “power mediator” (uses leverage to coerceparties to reach a settlement).55 Given the parties’lack of consent and many stakeholder groups’limited capacity, there was no doubt that the IGADmediators—with firm backing from peace processsupporters—would have to actively shape anagenda and then drive the process forward.56

Following consultations with a wider set of SouthSudanese voices and peace process supporters, themediators began shaping modalities for the talksand an agenda that included transitionalgovernance, security arrangements, economic andfinancial reform, reconciliation, constitutionalreform, and, ultimately, implementationmechanisms. Process battles chewed up severalmonths of negotiations, and the scope, format, andobjectives of the talks were only resolved when Kiirand Machar convened in Addis Ababa and signedthe “May 9th agreement.”57 Though again forgedunder international pressure, that preliminaryagreement marked a breakthrough in the process,nominally committing the parties to “engage insubstantive discussions…on the Agenda as consti-tuted by the mediation process.” Most concretely, itcommitted the parties to negotiate a politicaltransition in South Sudan.58

More than a year of acrimonious negotiationsfollowed, yielding halting progress on a range ofissues. The most divisive issues, however, wereroutinely bracketed and left for Kiir, Machar, andthe IGAD heads of state. The “multi-stakeholder”format for dialogue was gradually undermined bythe two dominant parties, a highly personalizedconflict, and a disproportionate focus on two menand their control of political power at the center.Despite this de facto reduction to elite bargaining,

South Sudanese individuals, the mediators, andpeace process supporters nonetheless sought toembed the kind of transitional mechanisms andreforms that could begin to diffuse power andreshape political dynamics in the country.

Page 18: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

14 Zach Vertin

59 The proposed accord included protocols for: (1) transitional governance, (2) security arrangements and security sector transformation, (3) humanitarianassistance and reconstruction, (4) economic and financial management, (5) transitional justice, (6) a permanent constitutional process, and (7) an implementationand monitoring mechanism.

60 Four others also signed the accord that day as “stakeholders,” including representatives of civil society and women’s groups, eminent persons, and faith-basedleaders. A representative of “other political parties” was not able to sign, as a dispute persisted over which political parties were legitimate opposition actors andwhich had been co-opted by the government. See “Agreement on the Resolution of the Conflict in South Sudan,” Addis Ababa, August 17, 2015, available athttps://peacemaker.un.org/node/2676 .

61 Booth, “South Sudan’s Peace Process.”

Box 1. Concluding the Agreement on the Resolution of the Conflict in South Sudan (July-August 2015) After eighteen months of negotiations, the parties had logged substantive progress but remained far from anegotiated settlement. Hoping to unlock the stalled peace process and confront the regional divisions thatwere undermining the mediation effort, US President Barack Obama invited the leaders of IGAD and theAU chairperson to a July 2015 summit in Addis Ababa. After a lengthy meeting, the group announced thatSouth Sudan’s warring parties should finalize a negotiated settlement by August 17, 2015. Meanwhile,drawing on a year and a half of consultations, position papers, and face-to-face dialogue, the mediators—with input from select peace process supporters and technical experts—shared a draft agreement with theSouth Sudanese parties.59 They stipulated that amendments could be made to the draft through jointagreement by the two warring factions—the government and the opposition.On August 10th, just one week before the deadline, President Museveni invited his Kenyan and Ethiopiancounterparts to Uganda for yet another summit. Confusing the process and angering the mediators, theSouth Sudanese opposition, and other stakeholders, President Museveni proposed substantial changes tothe draft agreement already under review by the parties. These included alterations to the proposed power-sharing formulas and, more consequentially, to its prescribed security arrangements. This includedsoftening or deferring provisions on the integration of forces, the demilitarization of Juba, and the introduc-tion of a third-party force to provide security for transitional institutions in the capital city. (Many of theseissues were deferred to two subsequent and deeply flawed workshops on security, nominally “technical”exercises that occurred after the accord was finalized and wider attention had subsided).On the deadline day of August 17th, after several weeks of focused international attention, Riek Macharsigned the accord on behalf of the opposition, together with a representative of the SPLM former detainees.60But with Salva Kiir unready to commit, the Ethiopian and Kenyan leaders decided to afford his governmenttwo additional weeks to return to Juba and secure support for the deal. In the ensuing days, the United Statesintroduced a draft resolution in the UN Security Council threatening an arms embargo and targetedsanctions against South Sudan if the government failed to finalize the deal. Nine days later, at a ceremonyin Juba, Kiir—citing a series of “reservations”—signed the agreement in the presence of IGAD leaders,ending the second phase of the process.Despite notable flaws, including eleventh-hour changes to the security protocol, the Agreement on theResolution of the Conflict outlined a “plan to end the fighting, frame a post-conflict transition, and beginthe tasks of reconciliation and reform.”61As such, it reflected the objectives that guided the mediators duringPhase II of the peace process. Nonetheless, the deal would come undone a year later, in July 2016.

Seven critical dynamics shaped this second phaseof the mediation effort: government denial, battle-field stasis, forum shopping, summit diplomacyand empty threats, a fractious opposition, a souringmediation, and elite compromise. While notexhaustive, this list offers important insight into theprocess, the challenges confronted, and theenvironment in which an ill-fated peace agreementwas forged.

Government Denial

Despite the existential crisis facing his country andwidespread civilian suffering, Salva Kiir’s govern-ment sought from the outset to project a narrativeof “business as usual.” It was a fiction that infuri-ated both South Sudanese opponents and the manyinternational supporters that had fought to helpsecure the country’s independence. Juba’s negoti-

Page 19: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

ating team nonetheless made clear that it need notnegotiate anything other than a cease-fire; itavoided substantive discussions by effectivelytrapping Phase II in months of circuitous discus-sions on the terms of the process.62

Calculating that any political negotiation underIGAD auspices could only result in checks on itsown power, the government argued that anypolitical discussions, should they be necessary,could occur back in Juba once traitorousopponents surrendered their arms. Not surpris-ingly, this was a nonstarter for all other SouthSudanese stakeholders. Meanwhile, Juba sentdiplomatic teams on a global propaganda blitz,seeking to present itself as a legitimate governmentsimply defending itself against a baseless rebellion.Still peddling narratives of an attempted coupd’état, its spokesmen went to African capitals,regional fora, and the embassies of UN memberstates in New York.63

But the reality was that Kiir’s government wasneither stable nor in control of an explosivesecurity environment. Its dogged and ill-fatedpursuit of a military solution would divide SouthSudanese society, run the economy into theground, and prompt a famine. Its legitimacy,already diminished by the atrocities its forces hadcommitted in December 2013, would continue towane as the war dragged on. But governmentofficials continued to pretend otherwise.Battlefield Stasis

Round-the-clock negotiations in January 2014 hadhelped secure a temporary cease-fire in less thanthree weeks—a remarkable pace in comparison tomany contemporary wars. But as Phase II began,this momentum diminished. Offensives andcounteroffensives continued in breach of cease-fireagreements, but by and large, battle lines werestabilizing and a de facto partition was emerging.64Delegations and mediators alike settled into more

established routines in Addis Ababa, much to thechagrin of South Sudanese who were suffering andwhose well-being urgently depended on results. Asweeks of fighting turned to months, it appeared theparties would remain unreceptive to negotiatedcompromise until a mutually hurting stalemate wasreached or the relative balance of power shifteddramatically.Forum Shopping

A cardinal rule of international peacemaking is toavoid “forum shopping,” whereby competing foraemerge, and conflict parties either seek out theforum most favorable to their interests or play twoprocesses against one another.65 In the spring of2014, an alternative mediation initiative material-ized, designed to pursue a settlement by reconcilingSouth Sudan’s ruling party, the SPLM. But the“Arusha process,” as it came to be known, wouldprove ill-conceived in terms of both substance andprocess.The Arusha process was an initiative of

Tanzania’s leading political party (Chama ChaMapinduzi) and a small European nongovern-mental organization (the Crisis ManagementInitiative).66 Its creation reflected the lack of clearconsensus about both the conflict and its idealremedy, as well as emerging frustrations with theIGAD process.67 But there was more at play behindthe scenes, and its organizers’ appreciation ofconflict dynamics and party politics in South Sudanseemed dangerously superficial. Despite forcefuland repeated warnings from international actors—about a party-only process and the dangers offorum shopping—its organizers proceeded inbreaking the cardinal rule. If one or more partiesdid not like what they were getting in one forum(IGAD), they could simply take their chances inanother (Arusha). Moreover, they could play oneforum against the other, in an attempt toundermine both. And that is exactly what they did.


62 After failing to secure agreement on a set of principles to frame the process, Seyoum spent weeks attempting to negotiate a substantive agenda for Phase II. Bothsides proved difficult, but the government was the primary obstacle to progress.

63 Several offices within the government also hired foreign lobbyists to advance the government’s narrative and interests in Washington and other capitals.64 The January 2014 cessation of hostilities agreement was reaffirmed in the May 9th agreement.65 United Nations, Guidance for Effective Mediation, p. 19.66 Members of Tanzania’s ruling party, Chama Cha Mapinduzi, worked with representatives from the Crisis Management Initiative, a Finnish conflict resolution

organization founded by Martti Ahtisaari. Ahtisaari was not personally involved in the Arusha effort.67 Many also ascribe the Arusha process’s beginnings to wider competition for preeminence among African states, including South Africa, which actively backed the

Arusha process.

Page 20: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

16 Zach Vertin

“We thought it was the way to go,” one seniorAfrican diplomat later reflected, having activelybacked the parallel track. “I’m admitting now, thatwas a mistake.”68 In theory, a truly complementaryprocess that helped the SPLM devise a path to amore democratic ethos could have been valuable.But Arusha’s elite group of invitees haddemonstrated little interest in party reform. It wasevident from the outset that all three SPLM factionswere approaching the Arusha process either inpursuit of narrow self-interest or with perniciousintent.69 Increasingly wary of the IGAD process,and angry that the opposition had been put onequal footing, President Kiir had personallyrequested that the Tanzanians establish the alterna-tive party process, though he did not advertise hishope that it would sabotage the IGAD talks.70 Inpredictable fashion, once elements were agreed inArusha, the government began using them toupend the IGAD agenda, marginalize participants,and walk back items already agreed upon. Summit Diplomacy and Empty Threats

Each time negotiations reached an impasse andshuttle diplomacy failed, the mediators were forcedto ask the IGAD heads of state to convene. Thissometimes yielded damaging delays, as coordi-nating the schedules of three or more heads of stateon short notice was not easy.71 Each time it alsoreinforced what the parties already knew—that themediators themselves possessed little authority.Seyoum and his team prepared written and oral

briefs to frame the issues, but the unscriptedsummits often took on a life of their own.Sometimes the obstacles articulated by themediators were addressed; other times decisionswere taken that ignored the mediators’ input or

further complicated their task. Even more frustrat-ingly, despite the issuance of official communiqués,participants sometimes emerged from the summitswith different understandings of what had beenagreed. This put the mediators in a difficultposition, forcing them to improvise or split thedifference.Seven extraordinary IGAD summits were

convened during the peace process. On five suchoccasions, the resulting communiqués threatenedpunitive measures, including specific mentions oftravel bans, asset freezes, and even militaryintervention.72 But time and again the rhetoricalflourishes proved to be empty threats, and theparties quickly surmised that inaction spoke louderthan words. “Everything professed at thosesummits was half-hearted,” one frustrated memberof the IGAD secretariat later reflected. “Theywanted to pretend they were supporting sanctionswhile servicing their own interests via the backdoor.”73 This further undercut the mediation teamand meant they operated without any real leverage.The empty threats were not lost on the parties,

which continued to wage war, flout their commit-ments, and stonewall the peace process withoutconsequence. In an unusually frank memo in thespring of 2015, Seyoum argued that the parties didnot take “the repeated threats of IGAD action andintervention seriously” and had effectively “calledIGAD’s bluff.”74

UN Security Council members, led by the US andUK, several times sought to move forward onmultilateral sanctions against South Sudaneseindividuals who were waging war or obstructingthe peace process.75 Political support for punitivemeasures arose periodically in African capitals as

68 Interview with senior diplomat, Addis Ababa, January 2018.69 Invitees included senior party members from the government, members of the opposition, and a delegation of former detainees. In addition to the government

motivations outlined above, Machar hoped to recapture the party hierarchy and the state resources that came with it, participating despite strenuous objectionsfrom his deputies, most of whom despised the SPLM. The SPLM former detainees—all of them first-tier party members—had been sidelined from the IGADprocess by mid-2014 and saw Arusha as their ticket back to relevance.

70 Tanzanian Directorate of Presidential Communications, “Statement on SPLM Agreement in Arusha,” January 20, 2015.71 The heads of state of Ethiopia, Kenya, and Uganda were essential for a summit to proceed, together with a high-level representative from Sudan. Additional IGAD

leaders (from Somalia and Djibouti) sometimes participated but were not deemed essential. The occasional participation of Salva Kiir—himself an IGAD head ofstate—in such meetings infuriated the opposition.

72 In response to continuing violations of the cessation of hostilities, the IGAD heads of state—led by Ethiopia—mandated the deployment of a “protection anddeterrent force” during their third summit on March 13, 2014. They called on the AU and UN for support, but because they did not first consult with thesepartners on troops, resources, or concept, support for the proposed force was not forthcoming. IGAD, “Communiqué of the 25th Extra-ordinary Session of theIGAD Assembly of the Heads of State and Government on the Situation in South Sudan,” Addis Ababa, March 13, 2014.

73 Interview with member of IGAD secretariat, January 2018.74 IGAD, “Developments in the IGAD-Led South Sudan Peace Process: Challenges and Recommendations for the Way Forward,” memo from the mediators to the

IGAD chairman and IGAD heads of state, March 5, 2015.75 In April 2014, the United States established its own sanctions regime for South Sudan, targeting any individual undermining the peace process, but Washington is

conscious of the limited reach of bilateral sanctions.

Page 21: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


76 Seyoum Mesfin, interview, Addis Ababa, January 2018.77 The dynamics around the pursuit of a UN arms embargo are further detailed in the fourth section of this paper (p. 18).78 Intra-party wrangling flew largely below the radar; even the mediators and peace process supporters in Addis Ababa did not appreciate the depth of opposition

rancor until late in the peace process.79 IGAD, “Developments in the IGAD-Led South Sudan Peace Process,” March 5, 2015.

well. But IGAD states were never committed topressure tactics, and subsidiarity norms meant thatneither the AU nor the UN Security Council wouldhave the votes to act unless punitive measures werefirst endorsed by IGAD. In practice, this meantregional opponents—most notably Kenya andUganda—had a veto on international action inSouth Sudan. “A stronger and more united region,”the chief mediator later lamented, “would havepaved the way for AU and UN action.”76 (Despitethis pursuit of punitive measures, the United Statesdeclined to press for an arms embargo early in theprocess, which many believed could also havechanged early calculations in Juba.)77

Fractious Opposition

Divisions plagued both the government and theopposition, but the fractious character of Machar’santi-government coalition—and its competingobjectives—would color the peace process frombeginning to end. In addition to seeking justice forthe massacre of Nuer civilians in Juba, mostopposition fighters retained more hardlinedemands than their de facto leader. Many soughtmilitary victory, while those who supported apolitical deal sought one rather different than whatMachar had in mind. It was a marriage of con -venience; he needed them to project force, and theyneeded his political heft.While talks continued, Machar and his deputy

kept allies in line in part by providing them armsand ammunition. Though they cobbled togetherenough supply from external sources to sustain thefight, it would never be enough to defeat thegovernment. Later, Machar twice convenedopposition conferences in an attempt to alleviateinternal pressure. On both occasions he retainedtacit support for his leadership, but resentmentcontinued to fester.78

When these cleavages came to the fore in mid-2015, the mediators faced another difficult choice:whether or not to engage competing voices fromwithin Machar’s camp. Though dissenting generalsprivately requested an audience with Seyoum, hedeclined. The chief mediator wanted negotiations

to reflect realities on the ground but calculated thatengaging Machar’s disgruntled affiliates wouldlikely encourage further splintering. He also knewthat they were far less receptive to (and capable of)a negotiated compromise.The IGAD mediators later noted that both Kiir

and Machar appeared ill-positioned to make thecompromises necessary for peace, as intra-groupdynamics had rendered both men “prisoners oftheir own constituencies.”79

A Soured Mediation

Seyoum and Sumbeiywo had been given anextremely difficult task and faced a decidedlyunfavorable mediation context. Still, many of thechallenges they faced were made worse by personalfriction between the two men, their two “frenemy”countries, and the poisoned environment createdby the competing national interests of IGAD states.As a frustrating process wore on with few results,the IGAD mediation team soured. Tensionsbetween Seyoum and Sumbeiywo were no secret inthe negotiating corridors, and the dynamic wassoon mirrored by their support staff, whosometimes worked effectively together, andsometimes at cross-purposes. This dynamiccomplicated information sharing, undermined ashorthanded secretariat, and confused a processalready wanting of a strategy.By late 2014, an embittered Sumbeiywo had also

developed increasingly antagonistic relations withpeace process supporters, hoping to shut them outof the mediation effort. Until then, these diplomatshad taken their cues from the mediation team—coordinating messaging, undertaking specifictasks, and offering technical and strategic advicewhere requested. But Sumbeiywo’s effortssucceeded in closing this consultation channel,weakening a mediation effort already sufferingfrom capacity gaps.Individual differences aside, by early 2015 the

larger divisions among IGAD states could nolonger be ignored. “Each of our capitals…aresending different, and sometimes contradictory,messages to the parties,” Seyoum reported to his

Page 22: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

18 Zach Vertin

collective bosses. National interests had “prevailedover IGAD’s regional common interest and leftIGAD in paralysis.” As a result, the parties felt noneed to negotiate the compromises required forpeace. They had come, he explained, to see thatdifferences among regional states were “there to beexploited. [The] parties had mastered the game ofplaying one member-state off against the other.”80

Elite Compromise

At the onset of the peace process, the naturalinclination of the IGAD heads of state and theirappointed mediators was to engage the SPLM partyelites who controlled the nascent government andwhom they were accustomed to dealing with. Butthis narrow focus failed in some ways to appreciatethe nature of the war, the character and interests ofthose fighting on the ground, and larger questionsabout the source of the SPLM’s legitimacy.At the onset of Phase II, the mediators were

warned about the risks of a stitch-up between thevery actors whose power struggle had ignited thecivil war. Stability was important, but a return tothe status quo would not only fail to address thecorruption, mismanagement, and poor leadershipthat had helped erode fragile state institutions butwould also risk setting the country up for anotherbreakdown. At the urging of other South Sudaneseconstituencies and peace process supporters fromthe West, the mediators came to embrace thenotion that a comprehensive and sustainable peacewould require a more diverse and inclusivedialogue.IGAD’s January 2014 communiqué thus called

for a “truly inclusive” dialogue involving a “broadrange of government, political, and civil societyactors,” and its subsequent resolutions affirmed theplan for a “multi-stakeholder” process.81 But thepush for inclusivity and transformational changecame from powerless constituencies, the mediators,and other outside actors. Absent political resolvefrom IGAD, and amid active opposition from thewarring parties, it could not be sustained. And soby late 2014, the peace process was reduced to a

three-faction affair, each dominated by members ofthe fractured SPLM.

A Theater for RegionalCompetition

The participation of immediate neighbors inresolving a conflict can be a wild card, as interestedstates can variously support, shape, or spoil a peaceprocess, depending on the context. Their compara-tive advantages must be weighed against the risksof partisan interventions that may complicate orprolong a conflict. Would a disinterested outsider,with no history or direct ties to the region, make abetter mediator? The UN Guidance for EffectiveMediation notes that “proximity to the partiesshould be neither dismissed nor taken for grantedas an automatic advantage.”82

In time, the values brought to the South Sudanesemediation effort by IGAD’s frontline states—unique knowledge, relationships, direct interest instability—were outweighed by their competingnational interests and stakes in the outcome.Uganda’s army doubled down in support of thegovernment, provoking strong reactions fromSudan. Ethiopia was frustrated by its inability tounite the region or deliver a settlement. Weaponsand ammunition flowed into the country fromKenya, Sudan, and Uganda. Wider regionalrivalries meant Egypt and Eritrea also paid closeattention and were suspected of partisan involve-ment. It was a dizzying mix of competing interestsand egos, all playing out on a South Sudanesecanvas, while those suffering as result of the warsometimes got lost in the mix.The rest of the African continent, and the world,

initially shuddered at the thought of a proxyconflict unfolding inside South Sudan or, worse, aregional war. Later, frustrated by a process trappedin regional divisions and a mediation in disarray,peace process supporters asserted their ownnotions of the way forward, ultimately proposing astructural change to the mediation architecture.

80 Ibid.81 See, for example, the IGAD summit communiqués of March 13, 2014, and June 10, 2014.82 United Nations, Guidance for Effective Mediation, p. 19.

Page 23: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


83 “S. Sudan Defence Minister Admits Government Is Paying Ugandan Army,” Sudan Tribune, February 14, 2014, available atwww.sudantribune.com/spip.php?article49963 .

84 Reports suggest Museveni altered his advice to Salva Kiir only following the summit convened by US President Barack Obama in July 2015.85 Sudan also had a substantial economic interest in South Sudan. Per the terms of 2012 agreements between Sudan and South Sudan, which outlined their coopera-

tion in the wake of separation, Khartoum received payments for the transit of South Sudanese oil exports across its territory, as well as processing and otherrelated fees.



In terms of impact, Uganda’s role in the SouthSudanese civil war was second to none. Its militaryintervention on behalf of Kiir’s government andunilateral decisions were a constant strain on thepeace process, an x-factor that made an alreadycomplicated knot harder to untie. In addition toundermining IGAD’s impartiality, the tensionbetween Museveni and what he saw as “junior”partners in Ethiopia and Kenya prevented regionalconsensus at critical junctures.The Ugandan People’s Defence Force (UPDF)

had a long history of regional deployments gearedtoward political and financial gain, and thisappeared to explain at least part of its presence inSouth Sudan. Though never expressly articulated,observers speculated on Museveni’s economic andstrategic objectives. First, South Sudan wasUganda’s largest export market; everything fromfuel to produce was trucked in from the south byUgandan traders. It was also widely rumored thatKiir’s government had paid Kampala substantialsums for the UPDF’s reinforcement—whether tothe Ugandan treasury or its president’s campaignchest.83

Second, Museveni’s record of regional interven-tions reflected his desire to dictate events in hisbackyard. A longtime ally of Southern Sudan’sliberation army, Museveni had backed its guerrillasin their fight against Khartoum. He retained deepreach inside the SPLM, and in the lead up to theconflict he had kept close tabs on the intra-partypower struggle. In this context, Museveni’s personalcontempt for Riek Machar was well known.Despite Uganda’s outsize influence on the

ground, among IGAD’s major players it was theleast engaged in the peace process. Repeated callsfor Ugandan troops to withdraw from South Sudanwent ignored, as did appeals for Museveni to usehis leverage to move Kiir toward a negotiatedsettlement.84


Given its historical role as host to the process thatyielded Sudan’s 2005 Comprehensive PeaceAgreement, Kenya’s government had likewisehoped to host the new peace talks. When theymaterialized instead in Ethiopia, Nairobi chafed.The relationship between these two neighbors—sometimes cooperative, sometimes competitive—was defined by latent tension, as was their sharedstewardship of the peace process. PresidentKenyatta tried on several occasions to negotiate hisown deal, secretly convening the factions in anattempt to short-circuit the process and score amajor diplomatic victory on Kenyan soil. Suchattempts were not only unlikely to succeed butundermined the credibility of the officialmediation.Financial interests—some state, some

individual—also shaded Kenyan engagement. Notonly did Kenyan elites own businesses in Juba, butit was widely believed that South Sudanese officialshad parked millions of dollars of stolen cash inKenyan banks and real estate, much of it with thehelp of local facilitators. When the internationalcommunity attempted to create leverage by threat-ening economic sanctions against South Sudan,Kenyan diplomats publicly supported the callswhile privately campaigning against them.Sudan

Khartoum had a long history of playing SouthernSudanese groups against one another in service ofits own interests. And so Sudanese leaders kept afoot in both of South Sudan’s warring camps,publicly supporting the government in Juba whileprivately supplying enough ammunition to keepthe opposition afloat.85 But Sudanese attention wasalso piqued by Uganda’s provocative interventionand the presence of UPDF fighter jets near Sudan’ssouthern border. The two countries had long beenideological adversaries and occasionally engagedproxy groups to destabilize the other.

Page 24: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

20 Zach Vertin


The IGAD chair proved, in comparative terms, tobe the most responsible actor in the region, guidedboth by its desire for stability in a volatile neighbor-hood and aspirations of regional (and continental)leadership. Its management of the peace processwas frustrated, however, by competition withneighbors. Ethiopian officials resented Kenya’sparallel initiatives and were deeply frustrated by theadventurism of Ugandan President Museveni andhis army. But Ethiopia’s mediation effort was alsosometimes clouded by preoccupations with itsprestige as chair of the regional body and desire tomaintain a veneer of IGAD unity despite profounddivisions over South Sudan. Occasional frictionamong senior government personnel also hinderedEthiopia’s leadership, as communication gaps andlack of coordination led to mistakes. But anyassertions of Ethiopian interests in South Sudan,whether personal or institutional, were negligiblein comparison to the other frontline states.PEACE PROCESS SUPPORTERS

Interested states and institutions can play anindispensable role in supporting conflict partiesand mediators in their pursuit of a negotiatedsettlement, including through the provision ofdiplomatic, financial, technical, and other forms ofsupport. Beyond the immediate region, this thirdring of actors can constructively shape a process,including by bringing leverage to bear on theparties (first ring) or on regional actors (secondring).86

In South Sudan, the peace process supporterswere generally united in support of IGAD’s leader-ship, and many offered critical support to themediation effort. While each had ideas of what asettlement should look like, few had strategicinterests in the country, so they refrained frommaking undue impositions on the process. Even inthis generally collaborative environment, onesenior diplomat later reflected that “balancingsometimes different approaches and differentpriorities presents constant challenges—which, if

not carefully managed, can complicate peaceprocesses or compromise outcomes.”87 Peaceprocess supporters expressed concerns at regularintervals, sometimes reflecting larger philosophicaldifferences over the nature of the process, itsdirection, the role of the mediators, and the desiredend state. But unless these third-ring actorsasserted a far greater role in negotiating andenforcing a peace deal or in pressuring second-ringstates, there was little they could do to alter thecourse of a process owned by the region.United States

Washington had long been South Sudan’s principalforeign benefactor, having worked with the regionto negotiate an end to Sudan’s civil war in 2005 andprotected the South’s right to self-determination in2011. But the war quickly upended that relation-ship. American pressure on the government overits conduct, as well as US support for a balancedand inclusive IGAD peace process, began a markeddeterioration in relations with Juba. Of the peaceprocess supporters, US Special Envoy Booth andhis advisers were most active, both in shuttlediplomacy and direct support to the mediators.Their advice and technical support, provided on aninformal and voluntary basis, was sometimesemployed by the mediators and sometimesdiscarded.88 Personal interventions by Secretary ofState John Kerry and President Barack Obamahelped breathe life into the stalled process, and theUS sought to influence the process through its ownbilateral sanctions and actions at the UN SecurityCouncil. But while Washington occasionallysought to exert greater influence, it was reluctant to“own” the process. US policy sometimes appearedinconsistent as a result, and important opportuni-ties were missed.First, early in the conflict, the United States had

an opportunity to push for a UN arms embargo,but it demurred. The proposed weapons ban washotly debated inside the US government, andEuropean allies lobbied Washington to support it.As weeks turned to months, and the window of

86 The “three rings” metaphor is drawn from the former UN mediator in Syria, Lakhdar Brahimi. See I. William Zartman and Raymond Hinnebusch, “UNMediation in the Syrian Crisis: From Kofi Annan to Lakhdar Brahimi,” International Peace Institute, March 24, 2016, available at www.ipinst.org/2016/03/un-mediation-syrian-crisis .

87 Booth, “South Sudan’s Peace Process.”88 Some observers mistakenly likened American involvement in the peace process to that of IGAD in Sudan in the 1990s and early 2000s. The dynamics and

structure of the mediation were fundamentally different, however, and the United States did not have, or exert, the same influence over the process.

Page 25: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


opportunity seemed to close, proponents werefrustrated by the lack of a clear rationale from theadministration. Those reluctant to impose anembargo, most notably National Security AdviserSusan Rice, cited varying reservations, includingover the potential inefficacy of an embargo and thepossibility that it might unduly aid the opposition.89Some also posited that the threat of an embargooffered the international community the mostleverage over the government, while its impositionmight push Juba beyond a point of no return.Second, despite repeated appeals from Ethiopia

and South Sudanese constituents, Washington wasalso reluctant to lean too heavily on PresidentMuseveni over his increasingly unhelpful posturein South Sudan.90 Senior US diplomats madenumerous private demarches, including severalvisits to engage Museveni, but these appeals wereultimately tempered by the perceived importanceof wider regional security cooperation withUganda.Troika and European Union

A small group of designated special envoys fromNorway, the United Kingdom, and the EuropeanUnion regularly worked in lockstep with theUnited States, providing diplomatic support andessential funding to the mediation effort. TheEuropean Union also imposed sanctions againstselect individuals for obstructing the peace processor committing atrocities and coordinated itsdesignations with the United States.91


Beijing’s Africa envoy was dispatched to the AddisAbaba talks on several occasions, and thoughmediators and other peace process supporterswelcomed China’s presence, its engagement wassparing and inconsistent by comparison.92 InJanuary 2015, Chinese Foreign Minister Wang Yimade a visit to Khartoum, during which he invitedSouth Sudan’s warring parties to a “special consul-tation” and reiterated Chinese support for theIGAD process and its aims.93

United Nations

UN Special Envoy for Sudan and South SudanHaile Menkerios was based in Addis Ababa andserved as another informal but important adviserto Seyoum Mesfin. His interventions were oftenclosely coordinated with other peace processsupporters, though other UN obligations meant hecould not dedicate his full energies to South Sudan.Back in South Sudan, the head of the UNpeacekeeping mission, Ellen Løj, was understand-ably frustrated by her distance from the process.The mission had an enormous task in hosting morethan 200,000 internally displaced persons on itsbases and facilitating humanitarian access. The UNwas not positioned to play a larger political roleafter relations with the government soured inDecember 2013. Nonetheless, Løj sought a strongerconnection with the IGAD process, as decisionsbeing made in Addis Ababa would inevitablyimpact the mission’s mandate, posture, andresources going forward.African Union

Having endorsed IGAD’s leadership of themediation, the African Union’s Peace and SecurityCouncil largely deferred to the decisions taken byIGAD heads of state. In December 2013, however,it did mandate the body’s first-ever commission ofinquiry to investigate human rights violations andother abuses committed during the initial phase ofthe conflict and to make recommendations ontransitional justice.94 Though AU officials latershared wider frustration with the IGAD process,Peace and Security Council member states werereluctant to assert higher authority or exertleverage on the South Sudanese parties—partiallyas a result of sustained lobbying efforts by thegovernment of South Sudan.When the commission of inquiry completed its

investigations in 2014, aggrieved South Sudanesecitizens and peace process supporters eagerlyawaited their report. Many hoped it could be agame changer, introducing accountability into the

89 Colum Lynch, “Inside the White House Fight over the Slaughter in South Sudan,” Foreign Policy, January 26, 2015, available athttp://foreignpolicy.com/2015/01/26/exclusive-inside-the-white-house-fight-southsudan-obama-conflict-susanrice-unitednations/ .

90 Like the proposed arms embargo, a firmer line toward Uganda was also vigorously debated inside the Obama administration.91 The European Union also sustained an existing arms embargo that dated back to Sudan’s civil war period.92 Chinese interest spiked during the opposition’s 2014 assault on Chinese-operated oil fields in Upper Nile State.93 Chinese Ministry of Foreign Affairs, “Special Consultation in Support of the IGAD-Led South Sudan Peace Process,” January 13, 2015.94 African Union Peace and Security Council, communiqué from 411th meeting, Banjul, the Gambia, AU Doc. PSC/AHG/COMM.1(CDXI), December 30, 2013.

Page 26: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

22 Zach Vertin

peacemaking effort and possibly serving as a meansto exclude responsible parties from futuregovernance arrangements. Though the report wascompleted in 2014, the Peace and Security Councilopted several times to delay its public release. Thesedelays were often requested, or backed, by theIGAD heads of state and others who worried thatthe report could upend a fragile peace process. Thereport was ultimately released to the public oneyear later, in October 2015, after the peaceagreement was signed.Former Malian president Alpha Oumar Konaré

was ultimately appointed AU envoy for SouthSudan in June 2015 as part of an attempt todemonstrate enhanced AU support for the peaceprocess. But his appointment came late in theprocess and had little impact on the mediationeffort or institutional dynamics.“IGAD PLUS”

In June 2015, at the urging of the United States andother peace process supporters, Ethiopiaannounced the reconfiguration of the mediation as“IGAD Plus.” Supporters agreed that the processhad been poisoned and needed an antidote.95 Theexpanded mediation format thus added five AUmember states, the United Nations, the Troika, theEuropean Union, and China as official partners.96IGAD Plus was designed, first and foremost, as away to mitigate troublesome regional dynamics bywidening the circle of participants. It was alsointended to reinvigorate flagging internationalsupport for the process, invest a wider constituencyof African states in its success, and provide much-needed technical and strategic support to themediation team. In practice, however, littlechanged. Participation by AU member states wasminimal, and IGAD member states did not facili-tate the kind of structural changes necessary tomake IGAD Plus a reality.


The IGAD-led peace process for South Sudan mayhave prevented the country from plunging into awar even worse than the one it has endured. It may

have presented South Sudan’s leaders with aplatform for dialogue: an opportunity to beginreconciling their people and remaking their nation.But less than a year after the mediation processyielded a wide-ranging peace deal, war returned toSouth Sudan.The country’s leaders bear principal responsi-

bility for the conflict, the troubled nature of peacenegotiations, and the devastating suffering inflictedon millions of their fellow South Sudanese. As thefighting stretched from weeks to months to years,and negotiating teams failed to respond withrequisite urgency, those citizens most affected wereunable to impose a political cost on the warringfactions.Against that backdrop, the peace process

sometimes advanced important objectives andsometimes complicated matters. As the principalentry point for international actors, the processthus merits critical review. Fundamental issues ofconsent, preparedness, impartiality, inclusivity,strategy, and coherence are clearly identifiable andshould offer first-order lessons for futuremediation efforts, whether in South Sudan or inother conflict situations. More analysis is requiredto appreciate the second layer of dynamics thatcomplicated the peace process and weakened itsoutcome, from intra-group tensions to untappedleverage, and from elite bargaining to regionalcompetition.Two other important dynamics merit reiteration

in any assessment of the IGAD peace process, eachof which complicated the job of the mediators. Thefirst is the lack of consensus (among the parties, themediating institution, and the wider community ofsupporters) as to the nature of the conflict itself andthus to the scope and depth of its solution. And thesecond is the political and moral dilemmaconfronted by outside actors when a conflict is not“ripe” for settlement—when tradeoffs are madebetween ideal solutions and the imperative to stopthe violence.Some believe the 2015 peace accord would have

provided a sufficient blueprint for a meaningful

95 IGAD would remain at the center of the wider format because, as US Special Envoy Booth observed regularly, “If the region is not part of the solution, it will bepart of the problem.” Peace process supporters knew that even if the mediation was taken out of IGAD’s hands, the competing interests of IGAD member stateswould have remained a critical component of the conflict.

96 Chosen by AU members, the designated IGAD Plus members from Africa were Algeria, Chad, Nigeria, Rwanda, and South Africa.

Page 27: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,


post-conflict transition—if it had been accompa-nied by stronger oversight authority and sustainedinternational engagement. They argue that despiteits flaws, it could have nudged the parties down anirreversible path toward peace and stability,however slow or incomplete. They also emphasizethe fact that once the mediation architecture wasestablished and the process set in motion,scrapping a tainted effort provided no guaranteethat any alternative process would have taken itsplace, much less a better one.Others believe unrealistic objectives and fatal

flaws in the accord meant it was unlikely tosucceed—some of whom have come to this conclu-sion with the benefit of hindsight. “This thing wasad hoc. It began without a proper analysis of thereal state of affairs in South Sudan,” one seniorAfrican peace process supporter explained inretrospect. “We should have designed an objectivebased on what was really happening, rather thanbased on our best wishes.”97

As the IGAD region now attempts to “revitalize”the peace process, others continue to focus onmismatched incentives and deficiencies in themediation architecture. “Definitely, there was an

inherent problem in the structure of themediation,” chief mediator Seyoum Mesfinreflected in 2018. “I would not involve IGAD in thesame way if I did it all over again. I would changethe structure completely.” Indeed the need for adifferent mediation formula may be one takeawayfrom the process. Such a formula might couplecritically important regional players with bothoutside mediation expertise and a formal role for“third-ring” international actors. In theory, such astructure could better facilitate the use of pressureand incentives and draw on comparativeadvantages while preventing regional interestsfrom hijacking the process. Such coordination isalways more challenging in practice.The unsatisfying reality is that singular conclu-

sions are hard to draw. Those close to the processare sometimes limited in thinking creatively aboutalternatives, while those on the sidelines whoreadily dismiss the process fail to appreciate bothits merits and the political obstacles to fashioning aconstructive alternative. The best we can do,perhaps, is attempt to learn and build upon itslessons.

97 Interview with senior peace process supporter, Addis Ababa, January 2018.

Page 28: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,
Page 29: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,
Page 30: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,
Page 31: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,
Page 32: A Poisoned Well: Lessons in Mediation from South … · overcome generations of war and neglect to declare ... enterprise, its leaders dragged a war-weary people back into conflict,

777 United Nations Plaza

New York, NY 10017–3521


TEL +1-212 687-4300

FAX +1-212 983-8246

Freyung 3, 1010

Vienna, Austria

TEL +43-1-533-8881

FAX +43-1-533-8881-11


The INTERNATIONAL PEACE INSTITUTE (IPI) is an independent,international not-for-profit think tank with a staff representing more than twenty nationalities, with offices in New York, facing United Nations headquarters, and in Vienna. IPI is dedicated to promoting the prevention and settlement of conflicts between and within states by strengthening international peace and security institutions. To achieve its purpose, IPI employs a mix of policy research, convening, publishing, and outreach.


international not-for-profit think tank dedicated to managing risk

and building resilience to promote peace, security, and sustainable

development. To achieve its purpose, IPI employs a mix of policy

research, strategic analysis, publishing, and convening. With staff

from around the world and a broad range of academic fields, IPI has

offices facing United Nations headquarters in New York and offices

in Vienna and Manama.

www.ipinst.org www.theglobalobservatory.org

777 United Nations Plaza Freyung 3 51-52 Harbour House

New York, NY 10017-3521 1010 Vienna Bahrain Financial Harbour

USA Austria P.O. Box 1467

TEL +1-212-687-4300 TEL +43-1-533-8881 Manama, Bahrain

FAX +1-212-983-8246 FAX +43-1-533-881-11 TEL +973-1721-1344