ELITE AMBITION, THE ‘REBEL LEADER’ AND ILLUSIONS OF ETHNIC REBELLION

At the elite level, the impetus to rebellion is easy to understand – rebellion pays. Most political and military leaders in South Sudan have joined an armed rebellion at some point in their careers and eventually won promotions as a result of peace negotiations facilitated by foreign governments. Some supporters of the ruling SPLM refer to themselves as loyalists but the SPLM itself began as a rebel movement. Led by officers who mutinied against their Arab Sudanese commanders, the SPLM only became a government after signing a series of power-sharing agreements with the Sudanese government in 2005 known as the Comprehensive Peace Agreement (CPA). Brokered by the East African member states of the Intergovernmental Authority for Development (IGAD), as well as the USA, the UK and Norway, the CPA preserved Sudanese President Omar Al-Bashir's grip on power in Khartoum. However it also gave Southern Sudanese a ‘golden chance’ for independence and turned SPLM chairman John Garang, a former colonel in the Sudanese army, into the ruler of an autonomous region and the Vice-President of Sudan (Deng Reference Deng2013: 26). Often held up as a success story, the CPA did not unify Sudan or end its civil wars. The SPLM began preparing for Southern secession even as they signed the CPA and Al-Bashir reignited the civil war with members of the SPLM in Blue Nile and South Kordofan States even before South Sudan's official independence. Ultimately, the greatest beneficiaries of the CPA were the elites who signed it.

The incredible rewards SPLM elites reaped from the CPA have inspired numerous officers or politicians, embittered by unfavourable election results or their inability to secure government appointments in South Sudan, to launch their own ‘liberation struggles’. Though aware that the CPA did not solve Sudan's problems, the SPLM and international peace brokers have repeated this ineffective power-sharing strategy whenever they fear mutinous officers have significant militia support. Using crude ethnic stereotypes about ‘hotheaded Nuer’ or ‘cowardly Equatorians’ to decide if they should combat or placate rebel leaders, the SPLM has papered over personal and political differences with promotions and cash and effectively subsidised ‘ethnic’ rebellions since before South Sudan's independence. Academics have long explained this problem by referencing John Garang's observation that ‘in the South it pays to rebel’ but this consensus about elites’ incentives does not provide a real alternative to ethnic stereotypes or explain why rural civilians decide to support, ignore or even execute officers who rebel (El-Batahani et al. Reference El-Batahani, Elbadawi, Ali, Sambanis and Collier2005: 193).

David Yau Yau has become a particularly famous example of how ambitious politicians appear to have ethnicised conflicts in the Greater Upper Nile region as they strategically transition from candidate to insurgent and back again. Yau Yau campaigned in the 2010 elections for a seat in the state assembly against an incumbent who trounced him at the polls. A former clergyman, Yau Yau had no military training before losing this election and decided to rebel, and was not particularly popular within Murle communities either. Footnote ⁴ Yet, he still managed to convince locals to shelter him from the government because members of his Murle-speaking community had long resented the Sudan People's Liberation Army (SPLA) and were especially bitter about a recent disarmament campaign. Unable to capture Yau Yau with force, the SPLM co-opted him and made him a Major General in the SPLA. Ever ambitious, Yau Yau revolted again the next year and won even greater concessions. In 2014, international mediators praised Salva Kiir for signing an accord that made Yau Yau governor of an autonomous Greater Pibor Administrative Area carved out of Jonglei State. Yau Yau's repeated successes have reinforced the perception among foreign observers that rebellions are fundamentally ethnic or ethnicised. In reality, ambitious officers of virtually every ethnicity have regularly rebelled, even those of the ruling Dinka ethnic group, and while their success has depended on civilian militias, ethnicity has not determined that support.

Yau Yau's rebellion, along with numerous others in 2010, show that ambition, rather than ethnicity, explains who launched revolts, and why. But they also show that military experience, and even political connections and patronage networks tend to matter less than popular militia support. For example, the late Major General, George Athor Deng (from the Dinka-speaking Luach community), also rebelled against the government after losing his electoral bid for state governor in 2010. George Athor had decades of experience as a guerrilla commander. He had also served as Deputy Chief of General Staff for the entire SPLA and had used his position to accrue exactly the kind of political patronage that, in theory, should have helped him win a new office. Despite his military experience and political capital, Athor could not replicate Yau Yau's successes, even after winning several battles against the SPLA. Unlike Yau Yau, Athor Deng could not find support or shelter among local communities. This was not because Yau Yau was significantly more popular among ‘his own people’ than Athor Deng, but rather because the SPLA was far more hated in Yau Yau's area than in Athor Deng's. Unlike the Murle who had come to Yau Yau's aid against the SPLA, the Luach-Dinka communities where Athor Deng sought refuge did not have a history of violent SPLA occupation, and thus did not find a common cause with Athor Deng even though he was technically one of their ethnic brethren. Unable to find support (or even refuge) in the cattle-camps, Athor Deng was shot crossing the Ugandan border, hundreds of miles from his home.

The case of ‘the Nuer rebel’ Gatluak Gai also illustrates the near impossibility of rebelling successfully without local support, even when the government is perfectly willing to reward rebellion because of their stereotypes about ‘Nuer rebels’. Gatluak Gai was a lieutenant in the prison service who declared his ‘rebellion’ from a hotel room in Khartoum in 2010. Eager to prevent a ‘Nuer rebellion’ Salva Kiir's government co-opted Gatluak Gai by making him a general in the SPLA. When the new general returned to his home community in Unity State, he proposed launching a second rebellion to emulate Yau Yau's success. Gatluak Gai's men were disgusted. His own second in command (another mutinous officer named Marko Chuol Ruei) helped lead the group of men who shot the self-styled rebel leader right in his hometown of Pakur. Taken together these cases from 2010 show that local support, not the extent of a general's patronage network nor the government's decision to combat or co-opt, determined rebel leaders’ success.

Analysts of political violence have long recognised the ‘banality’ of describing a conflict as ethnic and noted that instrumentally deployed discourses, such as ethnic patriotism, seldom explain why people choose to fight (Kaufmann Reference Kaufmann1996; Mueller Reference Mueller2000; Chandra Reference Chandra2006). Yet, academics who analyse conflicts that clearly possess some ethnic dimensions face a perpetual temptation to reduce ethnicity to a tool in the hands of political elites who ‘reach for the “ethnic” card’ (Hutchinson Reference Hutchinson2000: 6). Such tales of elites’ guile harbour pejorative undertones, implying that sophisticated urban politicians dupe simpleminded cattle-keepers into fighting wars contrary to herders’ own interests (see Tuttle Reference Tuttle2013). Rural militias who choose to fight do draw on various ethnic discourses as an organising principle for local, violent collective action. They also selectively claim or ascribe gendered stereotypes about ‘thuggish’ militiamen, and ‘peace-loving’ women and elders to serve their interests. Nevertheless, such discourses are both contested and flexible because community members invoke, counter, or ignore these narratives for their own strategic reasons.

When viewed from their own cattle-camps, rather than the capital cities, the histories of militias known as the Nuer White Army reveal that women and men who support rebellions have coherent motivations of their own, which shape their military conduct and objectives more than the politicians who claim to command them. These militias do form alliances of convenience with rebellious SPLA officers and politicians, but they also disregard, attack or even kill Nuer politicians whose positions they oppose. Time and again, militias’ simmering resentments over the huge concentrations of foreign aid and material wealth in South Sudan's urban areas have precipitated attacks that levelled these ‘islands of development’. Demonised in the press as mercenaries, hell-bent on rape, mayhem and ethnic cleansing, these militias are really a franchise of temporarily militarised cattle herders and cross-sections of local communities’ fighting-age males rather than consortiums of malevolent men.Footnote ⁵

Not especially eager to die in battle, most of these young men fight to preserve their prospects of acquiring cattle for bride-wealth and fathering as many legitimate children as possible so they can eventually retire as successful patriarchs.Footnote ⁶ Ballooning military payrolls have allowed a significant number of Nuer herders to transform into professional soldiers during recent wars. However, the contrast between civilian militiamen (whose patriarchal aspirations depend on serving local elders and accessing unmarried daughters in rural cattle-camps) and professional soldiers (who organise their lives around generals’ patronage) remains a useful incentive-based distinction that helps to explain different combatants’ choices. Governments and international bodies have often co-opted professionals with promotions and cash, but neither the UN, the SPLA, nor mutinous officers have convinced militiamen to undermine their own self-interests by opposing the locally powerful men and women who rule their futures. Militias’ violent actions may strike development agencies and the communities they pillage as wanton, but militias have consistently targeted local sources of mounting wealth inequalities, paying themselves in loot while destroying centres of perceived hoarding that sparked their jealousy and threatened their more communal mores.

A MILITIA'S MOTIVES: A SHORT HISTORY OF THE NUER WHITE ARMY (1991–2012)

• Ciëŋ da ni Thɔɔc ë /cade    Our home is Thoich [in Maiwut County] not one

• mak ε raan εni mi ε nyiith     person [can] invade except a mosquito,

• ε jεn de γä makɔ.          it alone can invade me.

• /cade kewε ca mak dä yac …   None [can] insult or invade our [home] …

• γɔk ε niεεn kε cak ε ruk      Cattle sleep with milk [still] inside.

• män kε yaat jal /ca niεεn i …   Mothers tie skirts to walk [but] not to sleep…

• mɔkapä makɔ          I mobilise [and] invade [a cattle-camp, killing even]

• taa kε bat nyaaŋ         those who were not my adversaries.Footnote ⁷

Nuer-speakers have had a bellicose reputation among outsiders for roughly two centuries. In fact the word ‘Nuer’ actually means ‘killers’ and the fact that they are known by this epithet and not by their own name for themselves (nei ti naath, literally ‘people of the people’) is a testament to outsiders’ bigotry (Johnson Reference Johnson1981). One reason for this stereotype is that Nuer men, such as those who composed the praise song above, seldom pass up a chance to boast of the fighting prowess which makes otherwise expendable males valuable within their polygynous communities. Many of their neighbours also promoted this image of the Nuer as aggressive savages to goad impressionable Arab and British commanders into military operations that allowed them to loot Nuer settlements (Johnson Reference Johnson1994). Historically Nuer-speaking herdsmen have also proven tenacious defenders of their own areas – always armed, and easily mobilised when necessary for community defence or preemptive raids – even defeating various rifle-wielding invaders from Ethiopia and northern Sudan with only spears. Rather than concentrating on taking land, these herders often used their 19th century offensive operations to protect their communities, and also benefit from looting, by obliterating the bases of merchant slavers and their local allies (Johnson Reference Johnson1981). This historic context has contemporary resonances since merchant slavers founded virtually all of South Sudan's cities and rural communities have often understood their relationships with urban centres through this historical lens of exploitation and inequality (Leonardi Reference Leonardi2011).

Despite this long history of cyclical, community militarisation, people first started referring to Nuer-speaking militias as the White Army, or dεc in bor (sometimes rendered dei in bor or jiech mabor, literally ‘white soldiers’) in October 1991 when describing groups of armed civilians who were supposedly loyal to the Nuer-speaking prophet Wutnyang Gatakek. Since then the term has come to refer to all Nuer-speaking militias who do not belong to a salaried, uniformed army. This ‘first’ White Army did form after various elites initiated a series of national conflagrations, but it was never controlled by any elite actors. In August of 1991, the SPLA split into factions when Riek Machar, and other leaders from various ethnic groups, made a failed bid to replace John Garang as the SPLA's supreme commander. After allies of Riek Machar began attacking Garang's forces in Equatoria, Garang decided to eliminate his opponents’ bases in Greater Upper Nile. Sensitive to ‘ethnic politics’, Garang intentionally selected his second in command (a Nuer general named William Nyuon) to lead this assault into these Nuer areas. However, such elite-centred logic mattered little on the ground. Not especially concerned with the native tongue of the generals involved, local men took exception to the ill-supplied and predominantly foreign SPLA units marching through their areas and supporting themselves on plunder. Mobilised against SPLA intruders, militiamen routed an attack in September, and a second two-pronged attack in October, before deciding in November to launch their own offensive against the SPLA, in the hope that they could dissuade further incursions (Johnson Reference Johnson2006: 97–9).

Pursuing their assailants back to Garang's home area near Bor town, the White Army participated in the now infamous Bor Massacre, killing thousands of innocent civilians and rustling innumerable cattle over a two-week period. Just as many observers gloss recent violence as targeting specific ethnic groups, most reporting on the Bor Massacre conflated militias’ motives with their ethnic discourse, and interpreted these punitive attacks simplistically as vengeful violence targeting Garang's relatives. At the time of the Bor massacre, the prophet Wutnyang Gatakek, the White Army's purported leader, encouraged this view when he commented that, ‘People are saying the relatives of John Garang have received a lot of food’ (Scroggins Reference Scroggins2002: 258). Yet militias vented most of their wrath on the Dinka-speaking Bor and the mini-metropolis of Bor itself, not Garang's natal Twic East community who lived farther north. What explains the unexpected contours of this militia violence?

White Army militias had come to view the town of Bor as a centre for the corrupt hoarding of foreign aid. By 1991 the United Nations’ Operation Lifeline had sent huge quantities of aid down the White Nile to Bor town but the government of Sudan in Khartoum, not John Garang or the SPLA, had prevented virtually all of this aid from reaching Nuer-speaking communities farther downstream. Many Nuer-speakers interpreted this apparent hoarding as a Bor conspiracy. One popular rumour among the Nuer was that aid-filled shipping containers marked with the numbers 13 0 12 actually stood for B O R, a suspicion only strengthened when John Garang's forces seized a long idle Red Cross barge that had delivered this aid to Bor and used it in their assault on Nuer areas (Scroggins Reference Scroggins2002: 258–9).

In strategic terms, the Bor massacre backfired. Dinka-speakers in the SPLA and international aid workers who had formerly concurred with criticisms of John Garang's dictatorial style could not condone the bloodbath and ceased supporting everyone who claimed to control the White Army. For the White Army, the looting of Bor was also a hollow victory against inequality because, after they returned home with Bor cattle infected with rinderpest, most of their own herds also died. The attack also failed to eliminate the true sources of wealth disparities that foreign aid agencies and the government of Sudan continued to manufacture. In the end, Riek Machar proved at least as adept at hoarding aid as the politicians whom he had criticised for their perceived corruption. In his quest to supplant John Garang, Riek Machar had forged an alliance with the very government of Sudan against which he had struggled for independence. This new alliance inspired the government of Sudan to allow international aid agencies to ship aid to Riek Machar's headquarters in Nasir, where he quickly squirreled away these resources, literally burying stockpiles in the ground. This new hoarding soon produced new inequalities that turned Lou-Nuer militiamen against their Jikany-Nuer neighbours around Nasir, though until that moment the two groups had been comrades in arms for two centuries. In 1993 Lou chapters of the White Army responded by invading Jikany areas and burning Nasir to the ground.

Looking closely at how the White Army mobilised young men after the Bor massacre reveals that militias who invoked ethno-linguistic nationalism as a discourse took up arms for far more concrete reasons, and that male and female elders, as well as their unmarried daughters, had considerable control over male combatants. At the time of the Bor massacre, both Lou and Jikany communities were experiencing the most severe famine in over a century and were profoundly vulnerable.Footnote ⁸ When civilians were weak from hunger, local militias might have shrivelled and ceased to exist had their young male members been motivated merely by abstract, ethnic animosity. Moreover, elders interested in recruiting young men into militias also had to compete with the Sudanese government and the SPLA – both of which offered handsome rewards to any youth willing to fight – as well as with the refugee camps that offered refuge and significant aid for those less eager for battle. By 1993 the United Nations High Commission for Refugees (UNHCR) had established a camp across the Ethiopian border at Fugnido, where young men could receive free food, clothing, housing and quality schooling. However unromantic an existence, the UNHCR offered refugees food, safety, education and prospects for economic gain. The fact that many young men continued to fight in civilian militias despite the countervailing incentives testifies to the continued power of elders, and their marriageable daughters, to compel young men to fight on their behalf.

During these times of scarcity, most militia conflicts revolved around control over critical local resources. After four Lou men died in a scuffle over access to a seasonal fishing pool with some Gaajiok Jikany men, both Lou and Jikany elders ended up recruiting men to launch retaliatory raids by creatively leveraging prospects for legitimate marriages – a resource controlled through local patronage networks, which NGOs could not provide and warlords could not monopolise. As Pinaud has detailed, generals had the coercive power to challenge local patriarchs’ traditional ‘ownership’ of cattle and marriageable daughters (the Nuer word guan means both ‘father’ and ‘owner’).Footnote ⁹ Yet women, who were never passive chattel in their own marriages, still exerted considerable control over the courting process and exercised the ability to shame men, as Sharon Hutchinson and Jok Madut Jok have already described in detail (Hutchinson & Jok Reference Hutchinson, Jok and Werbner2002: 103). Playing to this strength, Lou elders assembled marriageable daughters into brigades that marched a few hundred metres behind militiamen and, at least according to Jikany men and women who survived these skirmishes, sang this ‘song of the Lou daughter’ (diit nyiet Loukä):

• /Cu rͻdu bar kε mac yaŋ.    You yourself don't run with the gun of a cow.

• /Cu rͻdu bar kε matu yaŋ εε.  You yourself don't run with the Kalashnikov of a cow!

• Donε εε tipbε luͻc naath.    The Yellow maize nourishes the hearts of the people.

• Donε Jikäny tipbε luͻc naath.    Jikany yellow maize nourishes the hearts of the people.

• Donε εε tipbε luͻc naath.    The yellow maize nourishes the hearts of the people.

• (Riem & Gatchai 2012 int.)

As this rousing chorus made clear, any Lou militiamen who retreated would have to literally run past all the unmarried women of his community, broadcasting his cowardliness and thus his inability to provide for a household and the community in general.Footnote ¹⁰

Thus organised and motivated, Lou militiamen depopulated Ulang County, burned Nasir town to the ground, and continued raiding farther eastward until female elders among their Jikany opponents devised their own recruiting strategy, making intensive use of the same premise of monopolising marriages. The Gaajak-Jikany to the east were essentially unaffected by these raids but the Gaajiok-Jikany who bordered the Lou were soon scrambling to recruit militiamen (Bup 2013 int.). Unable to meet their needs with existing customs, they created a new political office, the ‘chief of the daughters’ (kuäär nyiet) and, at the height of the conflict, these women became some of the most powerful figures in the various payams (or cluster of villages) along the Sobat River (Hothnyang & Mach 2013: int.). These matriarchs ensured that no able-bodied men even conversed with unmarried girls without first fulfilling their guard duties (Bun 2013 int.; Deng Reference Deng2013 int.). They also retained militia manpower by levying cattle fines against men who left the area to conduct private business. By 1996, Jikany militiamen, and non-combatant women and elders who policed them, had confined Lou raids to the original Lou-Jikany border area and local ‘chiefs of the daughter’ surrendered their authority as better rains reduced competition for local resources and the conflict deescalated.

Existing histories of both the Bor Massacre and the subsequent Lou-Jikany conflict have tended to over-emphasise the agency of the Sudanese Armed Forces, political elites like Riek Machar, and prophets like Wutnyang Gatakek, without ever mentioning locally powerful matriarchs like the kuäär nyiet or the Lou ‘daughter brigades’. Scholars who have neither learned local languages nor conducted research in the payams have tended to over-inflate rebel leaders’ abilities to control combatants because they cannot analyse local actors whom they have not even heard of. Portraying women as victims rather than leaders who recruited reluctant young men by leveraging their patriarchal ambitions, elite-centred narratives have also treated Nuer discourses of male machismo and feminine pacifism as literal truths rather than cultural ideals which people deploy instrumentally.

On the ground, Lou and Jikany men and women have generally agreed that neither Wutnyang Gatakek nor Riek Machar really controlled the White Army. They have not agreed on who to blame for the violence of the 1990s but their debates do reveal the importance of local politics and competitions for power. Women have blamed men for the destruction by turning men's own boasts of military prowess against them. Men have countered by invoking another cultural presumption – that adult children listen to their mothers more their fathers – and blamed mothers for asking their sons to go raiding (Hutchinson Reference Hutchinson, Schlee and Watson2009). Pastors have blamed ‘pagan’ elders for blessing military campaigns, while elders have countered that educated Christian elites have presided over more destructive wars than the spear feuds of the past. These competing explanations co-existed because, however much outsiders may see them as a monolithic group, individuals within Nuer communities have multiple identities and competing agendas, and conduct most of the politics of mobilising and governing militias at that local level on the basis of more parochial motives.

Refusing to believe that these militias were truly community governed, both the SPLM and the United Nations continued to erroneously believe in Wutnyang Gatakek and Riek Machar's authority decades later. In 2006, the SPLA co-opted Wutnyang Gatakek and sought to use his imagined authority to ease tensions after Lou-Nuer communities began complaining that some SPLA soldiers had commandeered their cattle during a government disarmament campaign. The SPLA brought the prophet to the disarmament site but Lou militiamen, who generally blamed him for offering bad council during the civil war, shot him on the spot.

The best publicised example of this failed elite-centred policy was the Nuer-Murle conflict of 2011–2012. In June 2011, White Army ‘captains’ (bunomni in Nuer) from the Lou, Jikany and Gawaar confederations assembled for their semi-annual regional militia conference and discussed taking collective action to end a pattern of midnight raids which they blamed on Murle-speakers. Hoping to repeat a 1998 invasion, which had prompted Murle elders to return kidnapped children, roughly a hundred captains agreed (purportedly by unanimous consent) to launch a massive invasion at the next semi-annual conference in December unless the situation improved (Hutchinson & Jok Reference Hutchinson, Jok and Werbner2002: 105; Nyang 2013 int.). The UN and the leadership of SPLM were aware they had not managed to prevent particularly gruesome assaults in Mareng and Nyandit payams in 2009 or Uror County in 2011 but still hoped to stop a retaliatory strike that so blatantly challenged the state's supposed monopoly on legitimate force. When thousands of militiamen began marching toward Murle areas in December, the UN immediately airlifted then Vice President Riek Machar to the scene in the misguided hope that he could control his ‘ethnic brethren’. The captains who were leading militiamen from each participating payam agreed to meet the vice president, but shouted him down and continued their march when they realised he had no new plan to prevent raids on their home communities. As these episodes demonstrate, policymakers’ strategy of co-opting commanders from a particular ethno-linguistic group, and then requiring them to oppose militia members’ senses of self-interest, have consistently failed.

RURAL RESENTMENTS AND INTERNAL POLITICS OF THE WHITE ARMY (2012–2013)

As with the elite-centred histories of past conflicts, most accounts of how the current war began focus on persons and events in Juba without describing the war's rural antecedents or the politics of civilian combatants. In 2012 and early 2013, rural Jikany men who were (and are) members of White Army chapters commonly criticised Salva Kiir, not for ethnic favouritism per se, but for being far too easily manipulated through ethnic rhetoric. As one former official put it, if you want an appointed position, ‘you go to Salva while he is drinking, if anyone refuses you say [with pretend anger] “It is because I am Nuer” and he signs’. Reacting to the suggestion that corruption might be preferable to war, the same former official responded that ‘Where there is corruption, you must fight against it or it will grow’ (Nhial 2013 int.). Nuer communities certainly noticed that the SPLA hunted down Dinka rebels like George Athor Deng and bought off minor Nuer figures like Gatluak Gai, but they felt contempt, not gratitude, for a government easily manipulated by its own ethnic stereotypes.

Rural herdsmen also resented officials from other areas who sought to enforce national laws that conflicted with local mores. In 2012 and 2013, these herdsmen loosely referred to any South Sudanese official from outside their immediate area as a ‘Dinka’ (jaa ŋ), and often deployed the term as a euphemism for an unethical outsider. In everyday expressions of antipathy toward the government, armed Jikany men repeatedly offered to harbour the author if a ‘Dinka’ official in Nasir (who was actually of Shilluk background) found fault with his passport. Thus militiamen invoked ‘Dinkocracy’ as an idiom for ‘corruption’ among officials who might possibly set the government's authority against their personal commitments to hospitality.

Perhaps most importantly, rural Nuer communities resented wealth inequalities that had only continued to intensify since the CPA had created a flood of oil revenue and post-conflict development projects. Following the CPA, both the government and NGOs had focused the majority of their spending in Juba and other major cities, rather than in the rural areas where most South Sudanese actually lived. By 2013 service delivery among Nuer communities in the Upper Nile State was extremely uneven, with most of the resources being devoted to the capital of Malakal or to Nasir town, rather than Nasir County's outlying payams or the state's other Nuer counties. Nasir town also received a significant amount of attention from aid agencies, which warped the local economy. Benefiting from inflated NGO salaries, many men in Nasir had become quite wealthy, by local standards, but little of this wealth trickled into the countryside. NGO employees spent much of their earnings in a parallel economy, run almost exclusively by merchants from Ethiopia, Darfur and Shilluk areas around Malakal, which inspired bitter locals to complain of merchants relaying profits to relatives abroad.

NGOs’ out-sized resources also encouraged local governments to serve foreigners who provided services for them, rather than investing in local service delivery. This unsustainable trend resulted in a paradoxical scenario in which some towns with less NGO activity actually had noticeably better public services. For example, Mathiang town could boast a serviceable roadway (constructed by the state government), wireless internet in the county's medical centre, and several hours of free public electricity each day. Maiwut County used its meagre budget to employ a licensed physician, born in the county, who travelled to remote cattle-camps to evenly distribute medical services. In contrast, Nasir possessed none of this, but it was the only Jikany town to maintain an airstrip for NGOs’ chartered flights and a fire department (with 50 salaried staff but not one hose, pump or bucket). By the spring of 2013, men in the payams surrounding places like Nasir often spoke jealously of poorly utilised (or mismanaged) stockpiles of urban resources as property they might justly redistribute.

By 2012, the Nuer White Army was also a different entity from the old militias of the 1990s and included many members who had received a formal education while living as refugees, especially in Ethiopia's Gambella Region. Still an unsalaried civilian force, the militias blended bureaucratic strategies they had honed in other countries with local traditions and religious beliefs. While local chapters remained autonomous, they adhered to a common governing structure that allowed local commanders to coordinate with counterparts across the Upper Nile region even as they partnered with local elders and women who could recruit manpower. This modern militia bureaucracy did run on patronage, but local businessmen and herders supplied more of the resources, not elites in Juba, Malakal and Bor or their salaried surrogates.

In the 1990s each payam had referred to its own local chapter of the White Army as their bunom, a term originally borrowed from their Anywaa (or Anyuak) neighbours once they began purchasing breech-loading rifles at the turn of the 20th century. By 2011 young men who remained active in White Army chapters had innovated more specialized terminology and used the once generalized word bunom to refer to one particular leadership position, the captain of a local White Army chapter.Footnote ¹¹ These captains (bunomni plural) maintained their prestige in peacetime by hosting security meetings at least once a month. Captains convened these gatherings by securing a steer to be sacrificed, finding a revered elder (ran mi pal ‘person of prayer’) to bless the sacrifice, recruiting a bevy of female cooks (thäät) to prepare the feast, and dispatching an official messenger (guan lääri or ‘father of sayings’) to tell members the time and location (Hashimoto Reference Hashimoto2013: 165). Militia captains did not need the blessing of every reputable elder, priest, or prophet in the area to legitimate their proceedings, nor the labour of every local woman to produce these feasts, but they did depend on significant support of both kinds.Footnote ¹²

During their meetings, captains also relied on literate bureaucrats to manage communal resources. These functionaries held official positions that the White Army had not developed back in the 1990s, including a secretary who ensured members paid their dues (including rotating responsibilities to supply animals sacrificed for these gatherings), a treasurer who kept the books, and a deputy (ran mi guru bunom) who managed relations with local government officials. Local men attended these security meetings voluntarily, but in practice every household had an interest in protecting their herds and few reasons to pass up a free feast. Most captains were fairly wealthy men with plenty of cattle. However they usually owed their fortunes to relatives and shrewd business dealings, not salaries or gifts from higher-ranking government officials. Regional warlords like Gathoth Gatkuoth did have a following in towns like Nasir and in their natal payams but these elites rarely had direct connection with militia captains. Some Nuer ethnic patriots in Nasir were upset when President Salva Kiir fired Riek Machar in July 2013 but the militias did not mobilise for war.

CONTEMPORARY VIOLENCE AND EVOLUTION OF THE WHITE ARMY (2013–2014)

When reports reached Upper Nile State in December 2013 that the Presidents’ supporters in Juba had massacred Nuer civilians, White Army members viewed this catastrophe through the lens of their own experiences with inequality and mobilised through their existing organisations. Messengers quickly summoned thousands of militiamen to meetings in each county where the assemblies agreed to march on the state capital of Malakal, in the opposite direction to Juba, and overthrow the state government. While militiamen used revenge against a theoretically Dinka SPLA as a discourse, they directed their military campaign at Governor Simon Kun Pouch, a Nuer-speaking native of Nasir County. The Governor was loyal to President Salva Kiir, but not especially implicated in the SPLM infighting 500 kilometres away from Juba, much less genocide against his own ethnic group. Moreover, when the conflict displaced Dinka civilians living between Nasir and Malakal in Baliet County, militiamen welcomed thousands of Dinka refugees who took shelter with their Nuer relatives in Dome payam (White Army member from Yomding 2014 int.). For many White Army militiamen calling for revenge against the Dinka, discourses of conflict between Nuer and Dinka or Riek Machar and Salva Kiir did not need to line up perfectly with reified ethno-linguistic categories because they used ‘Dinka’ as a euphemism for government. In fact, rural herders’ relationship-based concepts of being ‘people of the people’ were (and are) more behavioural than primordial. The placement of Nuer-speaking officials like Governor Simon Kun and the SPLA's army chief James Hoth Mai did nothing to appease militiamen because they deployed the derogatory concept of ‘Dinkocracy’ more as a critique of class privilege than as a simplistic expression of ethnic hatred. As proof of this, White Army members in Nasir decided to attack their own kinsman's government because, under his supervision, the booming mini-metropolis of Malakal, like Bor in 1991, had become an island of capitalist prosperity built on hoarding rather than distributing wealth.

Locals’ resentments toward governing elites ran so deep that even civilians who opposed the decision to attack Malakal chose to make their case by criticising rebel elites instead of defending the government. One man (who happened to be the county government's director of communication) made his case for opposing rebellion by asking ‘Should we fight for leaders [like Riek Machar] whose children have gone to other countries? Will they work for South Sudan? All their money will go to those places’ (Nasir County Government Worker 2014 int.). Such appeals skilfully blended herders’ traditional reasoning about familial wealth with nationalist and class-based discourses, but most locals believed that such clever rhetoric came from government appointees who were looking out for their own salaries.

Militiamen calling for rebellion argued instead that the war started because of the hoarding of resources by corrupt politicians. Elders, women and youths outside the militias generally concurred. Hoth Guandong, who has been the most respected judge in Nasir County's traditional courts since the early 1980s, sided with the militias and turned his back on the county commissioner whom he had worked alongside for several years. His counterparts in Longechuk County publicised their views by putting Salva Kiir on trial (in absentia of course) for massacres of Nuer civilians in Juba. Mothers helped beat the war drums by circulating fantastic stories about ‘the Dinka’ (White Army sympathiser 2014 int.). Such enthusiasm for rebellion did not eliminate tensions within local communities and sometimes aggravated them. Fifteen-year-old schoolboys who chafed at parental authority in Gambella crossed the border in Ethiopia to join the fighting, though, at least at the beginning of the war, the White Army typically refused to enlist many youths (fifteen-year-old combatant 2014 int.; child combatant's father 2014 int.).

Civilians’ support for, or opposition to, the rebellion had more to do with how they felt about the kinds of inequalities the government had created than with their age or gender. Ultimately, civilians’ arguments for and against rebellion both linked the conflict to South Sudan's cash economy and invoked common criticisms of greedy individuals who have, as the Nuer say, a ‘big heart’ (diit lɔc). In this discourse, ‘corruption’ was less about clientelism, and more about a lack of ethical reciprocity between political elites and rural communities.

In contrast to this perceived injustice, White Army militias strove to uphold principles of reciprocity as they elected leaders and redistributed wealth. At White Army assemblies in December 2013, White Army chapters in each county decided to elect a single captain to serve as primus inter pares in the coming war and ultimately chose men with reputations for distributing wealth equitably. For example, the Nasir assembly elected a captain from Jikmir payam who had built a small fortune by shipping cheap Ethiopian maize to Nasir's busy markets and transporting travellers, including one of the authors, on top of his merchandise for a fee (White Army member from Korenge 2014 int.), (White Army Secretary from Ulang 2015 int.). Rather than building a compound in Malakal, this captain had invested his wealth in cattle, which he used to help local men marry. At least some of the militiamen who voted for this captain were reciprocating his patronage, but they were also supporting a man whose track record of redistributing rather than hoarding wealth suggested they could trust him to work for their common good.

Captains who were elected ‘generals’ resembled warlord-politicians like Riek Machar and George Athor Deng in some respects but they did not enjoy the same monopolies on wealth or attempt to demand the same kind of obedience. White Army chapters mobilised for war more by coordinating than commanding. Each chapter generally supplied their own war material and, at least among the Jikany, most did so by having each chapter's secretary take up a collection of cattle from their members. Select militiamen then drove these herds across the Ethiopian border where White Army treasurers sold the cattle and used the proceeds to purchase shoes, water jugs and ammunition. Upon their return, captains redistributed these goods among militiamen for the coming campaign (White Army Secretary from Jikmir 2014 int.; White Army Treasurer from Korenge 2014 int.). Militiamen who had not owned any shoes clearly benefited from this ‘patronage’ but, as the entire community had contributed to the herd, no one felt especially indebted to any general. A similar communal logic was also evident when the White Army captured, and then recaptured, Malakal in early 2014. Journalists’ reports of White Army commanders who seemed unable or unwilling to control or constrain combatants are disturbing but the observation that ‘everyone was carrying something’ out of the city also shows how this violence offered chances for opportunism and for the broad redistribution of wealth (Vice News 12.6.2014).

Since the moment this war began, White Army militias have consistently refused to honour the numerous ceasefires and agreements signed by Riek Machar (and even field commanders like Gathoth Gatkuoth). Finally, at the end of 2014, international negotiators agreed to a pause so that the SPLM and the forces Riek Machar has christened the ‘SPLM in Opposition’ could hold separate ‘popular consultations’ with civilians in the parts of Greater Upper Nile that each controls. At a December gathering in the Jikany town of Pagak, career generals like Peter Gadet, James Koang and Gathoth Gatkuoth sat down with White Army representatives to produce a declaration which, along with various other demands, insisted that any future peace agreement must allow White Army commanders to serve as brigadier generals in an entirely separate army (Radio Tamazuj 13.12.2014.; White Army Secretary from Ulang 2015 int.; White Army Secretary from Korenge 2015 int.).

One way of interpreting this development is to suggest that White Army ‘generals’ are attempting to follow the tried and true career path of earlier warlords. Whether or not these ‘generals’ hope to receive government salaries that tie them to the state, or plan to make themselves accountable to the kind of elections that first brought them into power, remain open questions. As the war drags on, and armed groups compete ever more intensely for remaining manpower, it is also not yet clear if ‘chiefs of the daughters’ (kuäär nyiet) will regain their former prominence.

What is undeniable is that the ostensibly successful peace treaties of the past such as the CPA (or the Addis Ababa Agreement of 1972 that ‘ended’ Sudan's first civil war) ‘worked’ for a few years but have not created lasting peace. Disputes in Juba can certainly encourage violence but the profound wealth inequalities between the rural majority and certain urban centres (Bor in 1991, Nasir in 1993, and in Malakal, Bor and Bentiu in 2013) are also an important part of the history of South Sudan's civil wars and a central reason why parochial civilian militias have continued to conclude that it pays to rebel.