Peace Series y and Securit FES Alan Boswell Do local peace deals work? Evidence from South Sudan’s civil war

Alan Boswell Do local peace deals work? Evidence from South Sudan’s civil war

About the author Alan Boswell is the Senior Analyst for South Sudan at International Crisis Group. This paper was commissioned and researched in his previous independent capacity. During South Sudan’s civil war, Boswell has served as a formal advisor and researcher to a number of organisations, including the UN, USAID, US Institute of Peace, European Institute of Peace, the Centre for Humanitarian Dialogue, and the World Peace Foundation. Boswell is the author of several reports on South Sudan’s civil war for the Small Arms Survey institute in Geneva. Boswell covered South Sudan’s run-up to independence as a journalist in Juba. He is currently an associate of the Conflict Research Programme at the London School of Economics. Imprint Friedrich-Ebert-Stiftung Uganda and South Sudan Office 5B, John Babiiha Avenue P.O Box 3860 Kampala, Uganda. Tel.:+256(0)393 264565/(0)757 345535 E-Mail: fes@fes-uganda.org www.fes-uganda.org © Friedrich-Ebert-Stiftung 2019 Layout: Green Eyez Design SARL, www.greeneyezdesign.com ISBN: 978-2-490093-15-1 “Commercial use of all media published by the Friedrich-Ebert-Stiftung(FES) is not permitted without the written consent of the FES. The views expressed in this publication are not necessarily those of the Friedrich Ebert Foundation.”

Table of Content Key findings 4 Introduction 5 Analytical framework 6 Ample time and space are critical for strong processes 6 Broad participation, including by“godfathers” and spoilers 6 Guarantors to hold the peace 6 Strong feedback from local to national and national to local 6 Rituals and restitution 6 Case Studies 8 Pibor: Church mediation with unaligned militia 8 Yambio: Church mediation with unaligned militia 9 Lakes-Unity: Chief-to-chief peace 11 Mundri& Wondoroba: Church-mediated failure 12 Yei: Co-optation of rebel defectors 12 Concluding notes and takeaways 15 What makes local peace deals work? 15 Formulas for failure 16 “Uninclusive” 16 “Post-peace” mobilization 16 “Easy wins” 16 Divided civil society 16 “Bad faith“ 16 Appendix 17 Pibor peace agreement with David Yau Yau and the Murle community 17 Yambio peace agreement with SSNLM 33 Lakes-Unity peace agreement 40 Mundri peace agreement 42 Wonduruba peace agreement 46 Yei peace agreement 50

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Key findings Across five case studies, criteria for successful local peace agreements include … • Be inclusive of main conflict actors, including the top leaders. • Be a supplement, not substitute or alternative, to a national peace process. • Involve credible united civil society mediation with political backing. • Put emphasis on strong processes during mediation and monitoring mechanisms after. • Offer structural solutions to structural problems. However, the‘pockets of stability’ strategy has clear limitations, especially in areas affected by conflict between national armed actors, and may under certain circumstances pose risks of exacerbating conflict. 4

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Introduction In December 2013, violent conflict broke out in South Sudan, the continent’s youngest country which had gained independence only two years earlier. Originally triggered by a political conflict in the ruling Sudan People’s Liberation Movement(SPLM) elite, ethnic divisions soon became a defining feature of the new civil war. To date, the conflict has resulted in close to 400,000 excess deaths, 1 nearly two and a half million South Sudanese refugees in the region 2 and nearly two million internally displaced persons. 3 The first attempted peace accord, the Agreement for the Resolution of Conflict in South Sudan(ARCSS), signed in August 2015 and mediated by the regional Intergovernmental Authority on Development(IGAD), collapsed when renewed fighting broke out in the capital Juba in July 2016. This led to the further proliferation of armed groups and political fragmentation. The Revitalized Agreement for the Resolution of Conflict in South Sudan (R-ARCSS), signed in September 2018 4 , has significantly reduced fighting on the ground. Yet the success of the peace process remains in doubt, as implementation has faced repeated delays, and several armed actors have rejected the agreement. As of October 2019, the parties had yet to form the unity power-sharing government envisaged in the agreement. From the beginning, local dynamics, too, shaped the agendas of armed actors. The link between local and national conflict dynamics is usually complex. For example, the creation of new administrative boundaries further politicized and exacerbated traditional boundary conflicts. Local conflict revolving around cattle grazing is often linked to elites in Juba. The design of the regional peace process poorly reflects these local-national conflict links. In fact, attempts at conflict resolution at the national level may also fuel new violence, such as through ‘cantonment’ policies incentivise more armed mobilisation by local actors. The localised conflict dynamics and the difficulties in reaching a national political settlement frequently raise the question of whether and how conflicts can be resolved or mitigated at the local level. Historically, people-to-people processes, particularly the 1999 Wunlit conference, played a critical role in reducing fighting among South Sudanese. In the current conflict, too, non-state actors have forged intra- and inter-communal peace agreements at the local level. The results of such local peace initiatives in the ongoing South Sudan war vary considerably. Systematic analysis of these local peace initiatives – varying widely in context, scope, inclusiveness, and so forth – has also been wanting. This paper maps five cases of recent local level peace deals in South Sudan between 2014 and 2018 to identify key criteria that strengthen the prospects of successful implementation and sustainability of local peace agreements. Critically, there are no generic formulas to achieve effective sustainable local peace deals. What works is highly contingent on the specific local context. Nonetheless, the set of criteria identified in this paper offers a useful guide for external actors who engage in peace-making efforts at the grassroots. 1 Checchi, F., Testa, A., Warsame., A. Quach, L.,& Burns, R.(2018): Estimates of crisis-attributable mortality in South Sudan, December 2013- April 2018: A statistical analysis. London School of Hygiene& Tropical Medicine, September 2018. 2 UNHCR, South Sudan data, September 30, 2018. 3 UNOCHA, Humanitarian Snapshot September 2018, October 10, 2018. 4 For a full text signed copy, see JMEC: IGAD- HLRF Agreements: Revitalised Agreement on the Resolution of the Conflict in the Republic of South Sudan, September 2018. 5

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Analytical framework The office of the Friedrich-Ebert-Stiftung(FES) in South Sudan convened a Regional Experts’ Reflection Group on South Sudan in June 2018. The experts, some of them involved as mediators in past local peace initiatives, developed a number of criteria that are necessary (but not sufficient) for effective and sustainable peace processes at local level. Ample time and space are critical for strong processes Extensive quiet and informal diplomacy, facilitated by mediators and/or informal influencers, is required to build trust between parties. This requires sufficient time and space to allow engaging especially spoilers shielded from the public eye and its expectations. Informal processes are critical in trust-building. External actors must display courage to engage in flexible and open-ended processes rather than pre-determined, limited projects. Broad participation, including by “godfathers” and spoilers Peace-makers and spoilers- including ‘godfathers’ and violent youth –, alike must participate in local agreements to make them sustainable. Active agency is important rather than mere representation. Formal and informal authority often overlaps and both must be involved. The capacity, deep understanding and(moral) authority of mediators are instructive for trust-building and changing of the narrative. Churches and leaders of faith, followed by local elders, are regarded as informal influencers unrivalled in capacity to engage‘godfathers’ in quiet diplomacy. Guarantors to hold the peace Guarantors must have credibility, capacity and legitimacy. They may include a variety of different actors with different capacities, such as control over immaterial and material resources that enables them to offer incentives and impose sanctions. Guarantors may include institutions of authority or specifically set up follow-up mechanisms, as well as external actors, be they states, INGOs or NGOs. Strong feedback from local to national and national to local Effective feedback mechanisms are required to link the local, national and regional level, empower communities to hold their leaders accountable and create local ownership of current peace processes. The failure to explain proceedings and agreements resultant from ongoing national and regional processes leaves space for misunderstandings, mobilization and incitement. At the same time, community voices must be heard by national leaders. Rituals and restitution People-to-people peace processes must address both the emotional and material needs of the communities in conflict through symbolic and material means in order to be sustainable. Both rituals and restitution contribute to a changing of the narrative from conflict to harmony by addressing psycho-social needs and delivering tangible justice and material benefits from peace agreements. To further this inquiry, this paper takes as a case study local peace deals in South Sudan from the period of 2014 to 2018. This period 6

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war corresponded with South Sudan’s civil war and a national peace process primarily based in Addis Ababa, hosted by the regional IGAD bloc of countries. This paper does not look at strictly inter-communal conflicts isolated from rebellion against the government. The paper only examines local peace efforts in the context of South Sudan’s current national war. One peace deal, in Pibor, was negotiated prior to the outbreak of the war but remains of interest given its implementation and challenges in the course of the civil war. 7

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Case Studies This study comprises five recent case studies of local peace agreements signed between 2013 and 2018 in different parts of the country. The five case studies also differ greatly with regard to the actors involved, the nature and scope of the agreements, and implementation of the respective provisions. Given the small number of cases and the wide variation across numerous variables, the validity of the findings remains limited. Ultimately, the criteria for successful mediation and implementation of peace agreements at local level are highly context-specific. Nonetheless, this comparative study offers valuable insights and points to important questions external actors must consider before getting involved in local peace making. The full texts of the peace agreements are included at the end of this paper. The case studies are updated on events up to November 2018. Pibor: Church mediation with unaligned militia Relative success. Background: Nearly half of all violent deaths and displacements in South Sudan in 2009, the year of the worst violence since the signing of the Comprehensive Peace Agreement, occurred in the Jonglei region that witnessed a series of inter-tribal clashes, including between the Murle and the Lou Nuer. David Yau Yau, an ethnic Murle local official in Pibor county, was one of several candidates who contested and lost in the 2010 parliamentary elections to seek and receive arms from Khartoum for rebellion against the government. Following mediation by the South Sudan Council of Churches, Yau Yau accepted government amnesty in 2011 and integrated his troops into the SPLA. This peace was short lived: In April 2012, the rebellion re-emerged when Yau Yau defected to Khartoum. By 2013 the Yau Yau-led insurgency among the Murle – now known as the Cobra faction of the Khartoum-aligned South Sudan Democratic Movement/ Army(SSDM/A) – had gained significant strength and support in the context of Murle opposition to the SPLA civilian disarmament exercise. 5 However, several other conflict dynamics were at play, including long-standing competition over water resources and cattle between different ethnic groups and contests between local political leaders, which were aggravated by an influx of small arms in the region. Mediation: Several attempts at governmentinitiated attempts at mediation failed. The Murle objected to mediation by South Sudan Council of Churches Archbishop Daniel Deng, of the Murle’s rival neighbors, the Dinka Bor. The government also hired a British defense contractor to negotiate directly with Yau Yau, which also failed, as did outreach attempts by local Murle elites and interlocutors close to the Juba government. Finally, a mediation team of three nationally respected church officials not from the Jonglei area(all three were Equatorians) succeeded in bringing Yau Yau to the table, establishing trust, and holding the peace talks together even as South Sudan fell into civil war in December 2013. In fact, the emergence of a national level rebellion, the SPLM-IO, likely strengthened the government’s commitment to this local peace deal. The peace process involved wider support, and was brought into the IGAD-led process in Addis Ababa in January 2014. 5 For more background, see‘My neighbour, my enemy: Inter-tribal violence in Jonglei’, Small Arms Survey, 2012;‘David Yau Yau’s rebellion’, Small Arms Survey, June 2013. 8

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Results: A Cessation of Hostilities Agreement was signed between the Government of South Sudan and the SSDM/A – Cobra faction in January 2014 in Addis Ababa, and followed by a comprehensive peace deal in May 2014. The core plank of the peace agreement was the devolution of significant local power through the creation of the Greater Pibor Administrative Area(GPAA), which effectively gave the Murle a self-governing administrative unit with statelike powers. 6 The agreement further provided for increased political representation of Murle at national level and a special government development fund for the GPAA.(However, the relevance of the GPAA was greatly diminished by the creation of 28 states by presidential decree in October 2015, in place of the formerly ten states spread over three historic provinces.) Lead mediator Bishop Paride Taban personally lobbied government in 2015 to prevent Yau Yau’s forces from being deployed en masse against the SPLA-IO. Takeaways: The mediation of the Pibor conflict began prior to the outbreak of South Sudan’s civil war. The Pibor case is instructive in part because it was the last major local peace deal of its kind due to the eruption of national conflict. The government made major concessions towards the national power structure, namely the devolution of significant powers. This clearly underlaid the sustained success of the accord. When local civil society and elites failed to bridge the gap between the parties, national church officials succeeded in establishing credibility with both sides. The mediation was lengthy and engaged at the senior levels on both sides. Juba negotiated directly with the insurgents in recognition of the national roots of the conflict. South Sudan’s sudden descent into wider civil war likely had a counterintuitive stabilizing effect on the peace process. The raging war with the SPLA-IO strongly incentivized the government to prevent a wider, united insurgency. Key Variables: • Strong sustained lengthy process. • Inclusive, including the top commanders. • Structural solutions involving devolution of power. • Strong support and troubleshooting afterwards. • Strong incentive for government to contain local insurgency to prevent wider united insurgency. Yambio: Church mediation with unaligned militia Limited success. Background: In August 2015, South Sudan President Salva Kiir sacked and detained the Governor of Western Equatoria state, Joseph Bangasi Bakosoro. Two armed groups with Bakosoro’s patronage subsequently declared a rebellion against the government: one, the “Arrow Boys”, a previously existent loose network of community defence forces in Western Equatoria, declared their allegiance to the main rebel group, the SPLA-IO. The other, the newly formed South Sudan National Liberation Movement(SSNLM), declared an independent rebellion. Sporadic clashes with government forces broke out from the end of 2015 and into 2016. 7 6 See also‘Real but fragile: The Greater Pibor Administrative Area’, Small Arms Survey, October 2015 7 For more background, see‘Spreading Fallout: The collapse of the ARCSS and new conflict along the Equatorias-DRC border’, Small Arms Survey, May 2017; Schomerus, Mareike,‘Non-state security providers and political formation in South Sudan: the case of Western Equatoria’s Arrow Boys’, ODI, April 2016.. 9

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Mediation: Peace talks with both groups were mediated by an“interfaith” group of crossdenominational church leaders in Yambio, the Western Equatoria capital. Despite the church leaders’ efforts, however, the local SPLA-IO leadership refused to negotiate formally with the state government on the grounds that such negotiations needed to take place through SPLA-IO at a national level. The SSNLM, unaligned to SPLA-IO, did respond to church-led overtures, leading to direct talks with the state government. The state government in Yambio and the National Security Service pushed the SSNLM peace deal, but it received opposition in Juba, especially from the SPLA under thenSPLA chief of staff Paul Malong. Surprise SPLA offensives on the SSNLM positions killed the SSNLM leader and marred and delayed the peace process. Results: A peace deal was signed in 2016 between the SSNLM and the state government, backed by Juba. The agreement made explicit reference to ARCSS and was presented as complementary to the national process, and provided for legislative action of Gbudue and Maridi states to initiate a reconciliation process. At the core of the deal though was the integration of SSNLM troops into national security services. Informal dialogue proceeded with the local SPLA-IO, resulting in periods of effective truce and real positive gains for human security and freedom of movement. However, the success of the accord was also limited, primarily on two counts. First, this did not end the local insecurity since the SPLA-IO units, a larger force than the SSNLM, continued to insist peace must be negotiated at a national level. Heavy levels of insecurity in 2018 highlight this point. Secondly, the SSNLM peace deal produced a significant pocket of risk prior to the integration of the SSNLM force. Local elites used the SSNLM peace deal for“post-peace mobilization”, as a force of no more than a few hundred claimed 3000 members and soon swelled to over a thousand. This post-peace mobilization served three clear purposes: i) for the governor, to justify the peace deal against Juba opposition, ii) for local elites, to increase the number of locals capturing gains of the process, iii) for those mobilized, to capture the gains from this process. Takeaways: Overall, the SSNLM peace deal shows the value of church-led mediation. However, it also shows the limits of local peace talks in the context of the national civil war. Post-peace mobilization temporarily increased militarization in order to bring peace. This created a risk for greater instability, amid fears the now-larger SSNLM would defect again. Further, the sharp rise in insecurity due to SPLA-IO insurgency in 2018 shows the gains of this approach was limited to the non-SPLA-IO group. Since most armed opposition factions are now part of the national peace process, the SSNLM itself was a contingent outlier in the current war, rendering its local resolution of limited relevance to other areas. Key variables: • Limited inclusivity. All parties were engaged, but the largest of the rebel factions did not participate in the accord. • The peace therefore served to supplement, not substitute for, the national peace process by resolving conflict with the rebel group not aligned to SPLA-IO. • The peace deal was pushed through by a strong, credible united civil society in the form of a cross-denominational local interfaith group. • Strong sustained peace processes, and very strong monitoring and support by the mediators after the signing of the accord. 10

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Lakes-Unity: Chief-to-chief peace Possible limited success. Background: When war broke out in late 2013, the border between the southern Unity state, dominated by ethnic Nuer, and Lakes state, inhabited mainly by ethnic Dinka, formed a natural front line in the ethnicized national conflict at local level. The national-level conflict disrupted longstanding traditional kinship, trade, and neighborly ties for the bordering communities, specifically the Nyuon Nuer in Unity’s Panyijar and the Dinka of Amongpiny and Dinka of Yirol East, and exacerbated preexistent violent inter-communal conflict over cattle raids, grazing rights and water points. In addition to the formal SPLA-IO and SPLA fighters, both communities had their own informal community militias, known as the“gojam” in Nuer and the“galweng” for the Dinka. 8 Mediation: The mediation and process, formally initiated in 2015, was complex. Since the national level armed parties on both sides were still at war, the administrators, most importantly the county commissioners and governors, could not officially declare peace with each other. The first thaw in the relations was on the initiative of the Amongpiny commissioner, a South Sudan government official, who approached the Panyijar commissioner, an SPLA-IO rebel official. This created space for a chiefs-to-chiefs led process, mediated and supported by a national NGO, Assistance Mission for Africa. The talks between chiefs produced provisions for the freedom of movement for civilians and commitments not to raid each other. This in turn created confidence for a local truce to be negotiated and extended to the PanyijarYirol border. Results: A peace proclamation was signed between the paramount chiefs of Ganyiel, Amongpiny, and Yirol in 2018. The commissioners, representing warring parties, lent political support to the process but did not sign. The mediation has reported a limited, communityto-community success, although the impact of the recent accord is difficult to verify due to lack of visibility on the ground. The impact of the agreement was limited by design. It stipulated renewed community to community relations but did not address the official armed parties to the war, the SPLA-IO and the SPLA. The local peace therefore produces some positive but very frail gains. Additionally, the commissioners on both sides navigated difficult, shifting internal politics within their own warring parties. Takeaways: Traditional mediation between neighboring communities achieved some critical local human security gains. However, since these did not and could not include the national armed parties to the war, the process was fraught and politically contested. Gains were limited and could prove frail. Key variables: • Strong process, clear mechanisms taking recourse to traditional structures and practices. • Supplement, not substitute, national peace process. Stayed within its limits. • Traditional leaders and civil society took lead roles. • Did not bypass the official local political figures in the SPLA and SPLA-IO, thus trying to prevent emergence of immediate and clear spoilers. 8 For more background, see‘Peace is the name of our cattle camp: Local responses to conflict in Eastern Lakes state’, Rift Valley Institute, September 2018;‘Community peace resolutions agreed between Payinjiar, Yirol and Amongpiny’, PAX Protection of Civilians, May 2018. 11

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Mundri& Wondoroba: Churchmediated failure ‘Good faith’ failure. Background: In late 2015, violence escalated against the government in the Mundri and Wonduruba areas west of Juba in Central and Western Equatoria. In both places, insurgencies began in 2014 and then erupted into larger scale violence following the August 2015 national peace accord in Addis Ababa. The Mundri rebellion was led by Wesley Welebe, and the Wonduruba rebellion was led by John Kenyi Loburon, both of whom had joined the SPLA-IO. In both clashes, attacks on the SPLA by insurgents led to heavy retaliatory abuses by the SPLA against the civilian population, including extensive destruction and looting, resulting in the flight of civilians into the bush. 9 Mediation: Following their success in brokering peace with Yau Yau’s group in Pibor in early 2014, the same three bishops sought to mediate the conflicts in both Wonduruba and Mundri in late 2015. However, this latter mediation differed substantially from the Pibor mediation effort. Rather than negotiate between Juba and the insurgent leaders, in both cases they mediated between local SPLA commanders and“community” representatives. The mediation was conducted quickly, with little time spent on the ground by the bishops, and never received any official blessing from the Juba government or the SPLA-IO leadership. Local clergy were also involved in the mediation but did not play the leading role. Results: The Wondoroba and Mundri peace agreements signed after only two days of meetings set out a series of pledges of good will on behalf of the community and the SPLA but apart from the establishment of committees of inquiry did not provide for mechanisms for implementation and monitoring. As a result, the“peace deals” quickly fell apart. Civilians heeded calls to return to the towns, despite the absence of a ceasefire between the opposing belligerents. In both cases, renewed clashes sparked additional reprisals against civilians by the SPLA, further alienating the community and plausibly strengthening the insurgencies. Takeaways: The mediation failed to acknowledge or end the SPLA-IO rebellion, resulting in the quick collapse of the accords. The peace deals were uninclusive, vague, and did not address structural issues. Rather than a sustained peace process, the church mediators opted for a brief visit. Key variables: • Not inclusive: Did not include the principle actors on either side of the conflict. • Did not supplement but rather sought to bypass the two warring parties. • Short and shallow mediation process without broad civil society engagement. Yei: Co-optation of rebel defectors ‘Bad faith’ failure. Background: South Sudan’s southern Central Equatoria“Yei state” descended into heavy ethnic political violence in 2016 following the collapse of a national peace deal in July 2016 and the escape of SPLM-IO leader Riek Machar through the Equatoria region into the DR Congo, with government forces in hot pursuit. Local groups mobilized heavily into the SPLA-IO. 9 For more background, see Small Arms Survey,‘Conflict in Western Equatoria’, July 2016. See also UNMISS,‘Peace agreement signed in Mundri West’, May 2015. 12

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war The state government relied on primarily ethnic Dinka SPLA soldiers, which committed widespread atrocities and torched numerous villages, depopulating and alienating the community, hundreds of thousands of whom fled to Uganda and Congo as refugees. 10 Mediation: In early 2017, a Yei charismatic bishop, backed by American evangelicals, the Yei government, and Juba’s National Security Service(NSS) began talks in Kampala with a faction of SPLA-IO self-proclaimed local commanders led by Hilary Yakani, a former political operative, civil society, and NGO worker. By Yakani’s own private admissions, he had arrived in the Yei battleground area just months earlier, the end of 2016. SPLA-IO officially denies Yakani’s role as an officer, but privately acknowledged Yakani played a role in non-military functions. Yakani publicly described himself, falsely, as overall commander of the SPLA-IO forces in Yei state. Results: A“peace deal” was signed in Kampala in April 2017 after four days of consultations. 11 The“peace deal” committed the parties to a continuous grassroots process for conflict resolution, provided for presidential amnesty to those engaging in dialogue, encouraged the voluntary return of those who fled to Uganda and DR Congo, and so forth. At its core though was the formation of a Joint Military Committee that would oversee the assembly and training of SPLM-IO fighters for the formation of a pro-government militia. However, Yakani was only able to bring back a handful of defectors from the thousands he claimed to command. This prompted a heavy “post-peace” recruitment process by Yakani and the Yei state government to create a pro-government local militia. The local bishop who had been instrumental in the peace talks opened up a training camp in Morobo for Yakani’s force. Some who escaped say the trainees were held involuntarily. Ugandan border and security officials have accused Yakani’s group of recruiting refugees inside Uganda to join the group, with accusations of forced abduction and threats against refugees. Several Yei civil society leaders claim that a group of youth was deceptively recruited in Yei and also involuntarily sent to the Morobo training. The bishop received financial backing for the implementation for the peace deal from a small group of American evangelical organizations. Numerous allegations have been leveled from Yei civil society and SPLA-IO officials that the Bishop used the resources for nefarious purposes. An interfaith council in Yei, which includes the Catholic and the ECS, have denounced Bishop Elias’s activity and distanced itself fully from his peace process, which the government continues to strongly support. Takeaways: The Yei process is an extreme example of a“bad faith” local peace initiative. The local peace process thinly covered for a politicized attempt to splinter the opposition and mobilize a new pro-government militia. Key variables: • The peace deal did not include the rebel leadership but a defecting rebel official who overstated his credentials. • The peace deal was not supported by the Yei civil society. The bishop was denounced by fellow clergymen. • The peace process attempted to under mine and replace the national peace 10 For more background, see‘Spreading Fallout: The collapse of the ARCSS and new conflict along the Equatorias-DRC border’, Small Arms Survey, May 2017. 11 South Sudan warring factions sign peace pact’, The East African, June 6, 2017;‘S.Sudan rebels distance themselves from Kampala-based faction’, Sudan Tribune, March 25, 2017. 13

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war process without buy-in from the local SPLA-IO. • The mediation and implementation appears to be primarily be a ruse designed to attract and maintain external private funding from well-wishers. • The Yei peace process has not led to any tangible structural concessions. 14

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Concluding notes and takeaways The‘pockets of stability’ strategy has had a strong allure for donors and diplomats when national and regional level processes are stalling or lacking credibility. However, if these alternative peacebuilding efforts are not well thought through, they can indeed exacerbate conflicts – particularly if the external actors are perceived to be lacking impartiality. What makes local peace deals work? This paper identifies five key variables in the five case studies examined of recent local peace processes in South Sudan. • Inclusive of main conflict actors. • Success as supplement but not substitute or alternative to national peace process. • Civil Society mediation with political backing and strong, active monitoring and engagement. • Strong processes and follow-up. • Structural solutions to structural problems. The five key variables identified by the paper as relevant to the success or failure of local peace deals – inclusivity, complementarity, civil society participation, strong processes and structural solutions – heavily overlap with the ones identified by the FES Reflection Group on South Sudan. Crucially, these criteria appear critical but not sufficient to make local peace deals work. A key take-away is that local peace deals are not substitutes for a national process. Two of the three(relatively) successful cases involved(rare) non-aligned opposition groups not linked to a national rebel group such as the SPLA-IO. The third successful case represented neighboring communities across the SPLA-IO and SPLA front line, but did not claim to strike peace between the warring parties themselves, and was thus limited as a result. Among the cases examined, no local peace deal has succeeded in ending a localized conflict with the SPLA-IO. In sum, local peace deals in the context of a national conflict must acknowledge and seek to incorporate the national conflict while still providing direct benefits to communities. Of the two substantial failures, one involved a matter of‘good faith’ but shallow mediation attempt which quickly collapsed. The other failure can be classified as a‘bad faith’ effort to be avoided. It is noteworthy that the one clear successful local peace deal – Pibor – included a substantive structural solution, the creation of a new administrative unit and the devolution of powers. Importantly, in Pibor, the government had strong military incentives to pursue a settlement and make concessions in the form of structural solutions in order to prevent the emergence of a wider, united armed uprising. This study pointed to the difficulties of forging local peace in areas contested by national armed groups. At the same time, if external actors in consequence shift their attention to areas not(or less overtly) affected by national conflict dynamics, this may nurture perceptions that external actors favor communities in government-held territory – and by extension display a bias towards government. 15

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Formulas for failure The case studies also highlight“red flags” to watch out for: “Uninclusive” Successful peace deals either include the main actors to the conflict or else are naturally constrained to a more limited scope. Local peace processes have not found success in ending localized rebellions connected to the national conflict. Specifically, local peace processes have not found success in ending localized SPLA-IO rebellions outside the national peace process, though not due to a lack of trying. “Post-peace” mobilization The integration provisions of peace deals often create a sudden inflation in the size of the nonstate armed group. If the peace deal is unstable, this can pose risks of further militarization. Further, this can take a less benign form, as local elites instrumentalize local peace process for the goal of blessing the creation of a new local militia. “Easy wins” The Mundri and Wonduruba examples show the dangers of attempting quick peace deals without investing in the process and political capital necessary to create space for inclusive dialogue and resolving the conflict between the warring parties. Some community members believe these peace processes effected more harm than good. Divided civil society Local peace deals should at least achieve a base level of local civil society support. The Yei example is an extreme case that highlights the pitfalls of pursuing a deal not backed by the local civil society. ‘Bad faith’ Local peace deals are frequently wielded as a divide-and-rule strategy rather than a good faith effort at conflict resolution. These types of deals aim to further splinter forces or siphon off from the“bottom” rather than the“top” through a war of attrition. This strategy poses a number of issues: i. a reliance on partial cooptation and attrition rather than addressing root causes; ii. partisan mediators; iii. splinter groups, potentially rendering the conflict more long-term, fractious, and intractable. 16

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Appendix Pibor peace agreement with David Yau Yau and the Murle community 17

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 18

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 19

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 20

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 21

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 22

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 23

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 24

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 25

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 26

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 27

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 28

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 29

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 30

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 31

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 32

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Yambio peace agreement with SSNLM 33

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 34

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 35

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 36

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 37

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 38

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 39

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Lakes-Unity peace agreement 40

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 41

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Mundri peace agreement 42

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 43

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 44

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 45

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Wonduruba peace agreement 46

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 47

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 48

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 49

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war Yei peace agreement 50

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 51

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 52

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 53

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 54

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 55

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 56

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 57

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 58

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 59

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 60

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 61

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 62

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 63

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 64

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 65

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 66

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 67

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 68

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 69

Alan Boswell| Do local peace deals work? Evidence from South Sudan’s civil war 70

FES Peace and Security Series No. 36 About the FES Africa Peace and Security Series The lack of security is one of the key impediments to development and democracy in Africa. The existence of protracted violent conflicts as well as a lack of accountability of the security sector in many countries are challenging cooperation in the field of security policy. The emerging African Peace and Security Architecture provides the institutional framework to promote peace and security. As a political foundation committed to the values of social democracy, Friedrich-Ebert-Stiftung(FES) aims at strengthening the interface between democracy and security policy. FES therefore facilitates political dialogue on security threats and their national, regional and continental responses.The FES Africa Peace and Security Series aims to contribute to this dialogue by making relevant analysis widely accessible. The series is being published by the FES Africa Security Policy Network. About the publication The link between local and national conflict dynamics is usually complex. From the beginning of the outbreak of violent conflict in South Sudan, local dynamics, too, shaped the agendas of armed actors. However, the design of the regional peace process poorly reflects these local-national conflict links. This raises the question of whether and how conflicts can be resolved or mitigated at the local level. Historically, people-to-people processes played a critical role in reducing fighting among South Sudanese. In the current conflict, too, non-state actors have forged intra- and inter-communal peace agreements at the local level. The results of such local peace initiatives in the ongoing South Sudan war vary considerably. This paper maps five cases of recent local level peace deals in South Sudan between 2014 and 2018 to identify key criteria that strengthen the prospects of successful implementation and sustainability of local peace agreements.