Institutionalising an Emergency Response: ‘Protection of Civilians’ Sites at UN Bases in South Sudan as a Way to Deal with Violence Against Communities

ABSTRACT

South Sudan’s civil war has since 2013 threatened the protection of civilians, which UN peacekeepers are mandated to ensure. One aspect represented a new challenge, due to its scale and rapid onset: more than 200,000 civilians fled to UN bases across the country, seeking protection. This amounted to more than a tenth of all internally placed people at one stage. The response by the UN helped to save many lives, but created further dilemmas. Real problems were also experienced, with attacks on these ‘sites’ and on peacekeepers. This article analyses the significance of the phenomenon and how it unfolded.

Introduction

South Sudan’s civil war has brought enormous suffering for its people in terms of direct attacks, ethnic and gender-based violence, and mass displacement. All of this amounts to a significant challenge for the UN Mission in South Sudan (UNMISS), which is mandated to protect civilians, like most peace support operations now. Sadly, none of this is unique. But what is unusual about this case is the way in which more than 200,000 people sought safety on or beside UN bases across the country. The resulting protection of civilian sites were initially an emergency response, but as war and insecurity became protracted, these temporary sites have developed into long-term settlements. This article looks at the way these camps arose, evolved, and the challenges they pose for civilian protection.

Although protection of civilians (PoC) has been a core function of peace operations for some 20 years now, South Sudan’s ‘PoC sites’ as they are called are different in several ways to the long tradition of people seeking shelter at UN bases. They are on a much larger scale, arose quickly and with limited prior planning and have required both the UN and humanitarian organisations to change their operations and to find new ways of working together. Furthermore, the sites have significant implications for how the UN mission uses its resources. Their existence has generated a range of additional UN policies, with preparations now made by missions in other countries for a similar influx. Specific problems arose, such as attacks on the sites which were sometimes met with an ineffective response by peacekeepers. They also highlight how the different functions of civilian protection can have an effect on each other, such as promoting political dialogue while also dealing with a hostile government which sometimes obstructs the UN and humanitarian agencies.

This article looks at ‘PoC sites’ at UN bases in South Sudan and asks what contribution they make to the overall aim of protecting civilians. There is in fact a long history of people spontaneously seeking refuge at UN sites, where they have often been accommodated. This was seen, for example, in South Lebanon over the years in response to cross-border shelling by Israel and in Srebrenica in Bosnia and Herzegovina. However, most previous instances had been only for brief periods.¹ What happened in South Sudan in 2013 was on a different scale and speed. While the outbreak of civil war was not entirely unforeseen, the scale and sudden onset of the fighting caused tens of thousands of people to turn up at UN bases within a few days in December 2013. Briggs sets out the unusual nature of the situation, saying that ‘what evolved over the next three years was unlike anything humanitarians had experienced previously’ (2017, p. 17). This required an immediate response and ultimately led to the development of policies and practices with wider implications for peacekeeping. By 2014, more than 200,000 people were living in the camps set up inside and around UN bases, amounting to approximately one-tenth of all internally displaced persons in South Sudan. All of this raised important questions for the mission and for protection in general, with implications which go beyond the immediate questions of safety and physical needs.

The need to protect civilians within the perimeter of UN bases was never something which the mission planned for or sought to do but was rather an organic response to a rapidly developing situation.² The need, and the response to it, highlight many of the difficulties and tensions for peacekeeping missions in complex conflict environments. Concerns existed within UNMISS from early on in the response about people being drawn to the ‘sites’ (the term preferred by the UN) because of services provided there, despite the difficult living conditions. Having wanted to divest themselves of the sites for some time, the mission ultimately started handing the sites over to the government of South Sudan in 2020. Other significant issues in PoC which are underlined by the sites include allocation of resources, training and preparation, risk aversion, and double lines of command for particular contingents.

How Civilian Protection Emerged

Peacekeeping has gone through radical changes since it first emerged in the 1950s. Some of those changes were conceived in policy documents and high-level reviews. Others were ‘field innovations’ which were driven by new challenges and then became institutionalised practice. Protection of civilians as a concept has developed incrementally since 1999 through both policy and practice and is now a core element in most peacekeeping mandates. But one unexpected development emerged from the choice made by civilians under threat literally voting with their feet, as they sought shelter in UN bases in South Sudan when civil war broke out in December 2013. Thus, the protection of civilian sites – or camps for people seeking protection directly by the United Nations – came into being (Arensen 2016, Center for Civilians in Conflict 2016b). Just as protection sheds light on peacekeeping as a whole, these PoC sites reveal a great deal about the challenges and consequences of how the UN deals with direct violence against civilians.

Protection of civilians (PoC) goes to the heart of what peacekeeping should involve in the popular imagination. It has become relevant as the UN increasingly operates in areas of ongoing conflict where civilians are directly targeted by armed actors, rather than separating two sides who have reached an agreement. Yet protection presents some of the greatest challenges for the UN, which can find itself damned if it acts and damned if it does not. Peacekeeping has in any case changed radically since the early missions, where use of force was a last resort and only when the UN mission itself was under attack. Protection is defined by the UN in terms of direct physical violence rather than structural violence or a denial of rights:

without prejudice to the primary responsibility of the host state, integrated and coordinated activities by all civilian and uniformed mission components to prevent, deter or respond to threats of physical violence against civilians, within the mission’s capabilities and areas of deployment, through the use of all necessary means, up to and including deadly force. (UN Department of Peace Operations 2020, p. 3)

This involves a much wider range of activities than military options, so all aspects of a multidimensional peace support operation have a role to play.³ The term is different from the ‘social protection’ referred to by humanitarian and development organisations, which relates to a much wider agenda of welfare and rights, rather than direct physical violence. The UN’s Report of the High-level Independent Panel on Peace Operations (HIPPO) said in 2015 that protection of civilians ‘is a core obligation of the United Nations, but expectations and capability must converge’ (United Nations 2015b, p. 11). This means that mandates should be ‘realistic and linked to a wider political approach’ (p. 11).

Making the protection of civilians an explicit responsibility of peacekeeping missions has happened incrementally, after some very tentative beginnings, but there is an identifiable moment when the commitment was made and a clear context leading up to that. It emerged at the end of the 1990s, which was a decade of hope, ambition and shocking setbacks for the idea of a more peaceful world. The Cold War had ended, leading to optimism about greater freedom of action for the UN to prevent conflicts. UN peacekeeping expanded drastically, both in terms of numbers of personnel and scope, with the Security Council less constrained by Cold War tensions. However, the wars in the former Yugoslavia and the genocide in Rwanda in 1994 undermined any post-cold war optimism. The failure to prevent atrocities as the world looked on raised difficult questions about who could – and should – respond, in what circumstances, and on what basis.

The idea of peacekeepers protecting civilians was given further support by the UN Secretary General in a 1999 Report to the Security Council. He said: ‘The plight of civilians is no longer something which can be neglected, or made secondary because it complicates political negotiations or interests. It is fundamental to the central mandate of the Organization’ (1999, p. 22, par. 68). The Council subsequently expressed its willingness – in very tentative terms – to consider how peacekeeping mandates might address the issue better. The first mandate to do so came later that year in relation to Sierra Leone. The United Nations Mission in Sierra Leone (UNAMSIL) was authorised – with caveats – to use force to protect civilians based on Chapter VII of the UN Charter. The resolution stated that the mission ‘may take the necessary action … within its capabilities and areas of deployment, to afford protection to civilians under imminent threat of physical violence, taking into account the responsibilities of the Government of Sierra Leone and ECOMOG [Economic Community of West African States Monitoring Group, a regional peace support operation]’ (UN Security Council 1999). The qualifications are significant and reiterate that primary responsibility for protection rests with the state, with a complementary role for the international community. The limitation regarding the mission’s capacity and geographical deployment carries real weight, especially where freedom of movement may be restricted or the areas covered are vast. The mandate also refers only to imminent physical violence, rather than wider definitions of violence or other forms of rights violation.

Within a few years, nearly all uniformed UN personnel would be operating under similar protection mandates (Hultman 2013, United Nations 2015a). Besides being quickly established as an emerging norm, the language has steadily become stronger, with fewer caveats. Starting in 2015, the term ‘imminent’ was dropped in relation to the threat of physical violence as mandates were renewed for operations in a number of missions: the UN Multidimensional Integrated stabilisation Mission in Mali (MINUSMA), the UN Multidimensional Integrated stabilisation Mission in the Central African Republic (MINUSCA), the UN Organization stabilisation Mission in the Democratic Republic of the Congo (MONUSCO) and the UN Mission in South Sudan (UNMISS). How these mandates are interpreted, prioritised and operationalised is of course the important question. UN missions are now multidimensional, covering a wide range of tasks from political dialogue to supporting humanitarian access. When it comes to protection, the rules of engagement, types of patrolling and ultimately the willingness to take risks can have a significant bearing on mission practices and mandate outcomes. Those risks are not just physical, but may involve compromises when it comes to dialogue between conflict parties and managing relations with the host government, who can always hinder the mission or even expel it from the country.

Given the risks and potential fallout associated with using force, it is understandable that some of the most intense debate on PoC involves the use of force. But it is important to stress that protection actually requires an integrated, holistic approach which operates on many levels and is much broader than military action. These other options are not just a preamble to ‘real’ action, but can be as effective and just as essential to longer term protection as the use of force. The official UN policy organises the work of protection into three tiers (UN Department of Peace Operations 2020). The first one is essentially political, involving negotiation, mediation and dialogue in order to steer the conflict away from violence, along with public engagement. This aligns with the ‘primacy of politics’ which was championed by the HIPPO report, stating that lasting peace is found through political solutions rather than military and technical engagements (United Nations 2015b). This issue is discussed in more detail by Day et al. (2020) and Russo (2022), among others. Tier one PoC recognises that troops can only do so much for as long as they are deployed, especially when thinly spread across a vast area, so a peace process must ultimately be owned by the conflict parties rather than by outsiders.

The second tier involves police and military elements of a UN mission, using a range of tactics. These can include presence, patrolling and deterring or responding to attacks – possibly with the threat or use of force. The third tier of protection is a broader programmatic one aiming to create a better environment, which might sometimes be categorised as medium or long-term peacebuilding objectives. That includes human rights monitoring, accountability and security sector reform. It is possible that progress in some of these tiers may seem to make work in another tier more difficult, or that they operate on different timescales. Such is the nature of attempting a holistic approach, which recognises the complexities of the situation and the realities of seeking to have ownership by ‘local’ actors who have different perspectives and interests.

Context of South Sudan

South Sudan had a long struggle for independence from Sudan dating from the early 1960s, complicated by the later discovery of oil and the ascendance of an authoritarian regime in Khartoum, which resisted calls for greater autonomy and a more inclusive political process. The long history of conflict in South Sudan, which helps to explain some of what has happened since independence, is explored in appropriate detail elsewhere (for example, Johnson 2016a). This is also explained along with more recent aspects of the politics and conflict by de Waal (2014), Johnson 2016b, Jok 2017 and Pinaud (2021), among others. A Comprehensive Peace Agreement signed in 2005 created a pathway out of decades of war, during which conflict within South Sudan was often backed by Khartoum with deadly consequences. The peace process included a referendum in the South on independence. The vote in 2011 was overwhelmingly in favour, and the new state of South Sudan came into being in July of that year. It was quickly recognised and received considerable international support in making the difficult transition from independence movement to state. After so many years of isolation, war and exploitation by a range of powers, key development and gender equality indicators were extremely poor, with minimal state services such as health or education. The UN mission which had been present since the war of independence stayed on under a new mandate and name, becoming the UN Mission in South Sudan (UNMISS).

The new country fell back into war just over two years after independence, when tensions within the power-sharing government led to its collapse in December 2013. The power struggle between various groups led to widespread ethnically targeted violence, spurring a rapid escalation with reprisals throughout the country. While there are many intersecting and localised conflicts often tracing back to conflicts that pre-dated independence, the new war increasingly saw the polarisation of politics and access to resources along ethnic lines, making its resolution all the more difficult. What became clearer as the war developed through 2014 and 2015 was that people were being killed or driven out of their homes on the basis of ethnicity, often by government forces or their proxies. Human rights abuses and conflict-related sexual violence were widely reported (UNMISS and Office of the High Commissioner for Human Rights 2017). While estimates of fatalities are always difficult, one study based on excess mortality put the number at close to 400,000 by 2018 (Checchi et al. 2018).

Over a third of the estimated population of 11 million fled their homes, with roughly half of these four million displaced people becoming refugees in neighbouring countries, the others being internally displaced. The disruption to agriculture and to the movement of goods meant that food insecurity followed, and some areas of Unity State were (as predicted) categorised as being in a state of famine in 2017. More than half the population was food insecure in 2020. The space for civil society continued to narrow, making it harder for those courageous voices going beyond a zero-sum game to be heard. Humanitarian organisations and the UN faced restrictions, obstruction or the threat of expulsion in the case of NGOs.⁴ A number of peace agreements and ceasefires were signed after mediation by regional bodies and neighbouring countries, with limited results. A more comprehensive agreement, amounting to an elite pact sharing out resources for the main parties, was adopted in 2018, though implementation of the plan for a return to power-sharing was often delayed and marred by a lack of trust. Violence continued nevertheless, some of it in localised conflicts, with little sign of a significant move by people to return to their homes.

The long history of civilians seeking safety at peacekeeping bases of their own accord, quite independently of any UN policy and whether or not their arrival was expected, has been seen in South Sudan since shortly after independence in 2011. They turned up at various bases during communal violence from time to time, to the extent that a policy on how to respond had to be developed by UNMISS (Lilly 2014, Foley 2017). However, the numbers involved changed the nature of the phenomenon when the more serious violence erupted in December 2013, and it was clear that ethnicity was the basis for attacks. As Foley notes, the outbreak of civil war ‘caught UNMISS by surprise, and the scale of the influx overwhelmed it’ (2017, p. 328). Shortly after it broke out, an estimated 30,000 people entered the UN bases in Juba, 5,000 in Bentiu, 12,000 in Bor and 20,000 people entered the UNMISS base at Malakal. By 24 December, more than 60,000 people were seeking refuge in UN bases across the country, presenting an enormous challenge to those in charge (Arensen 2016). The numbers grew steadily during 2014 and 2015, exceeding 200,000 within two years, declining only incrementally towards approximately 170,000 by 2020 (Gregory and Gorur 2020). This amounts to approximately a tenth of all those internally displaced in South Sudan. Van der Lijn says it is important to note that ‘the POC sites were not planned, but the result of an urgent humanitarian need – large numbers of civilians looking to the UN to keep them safe’ (2017, p. 196).

In response to the unprecedented influx of IDPs, UNMISS had to work out a division of labour with humanitarian organisations for provision of the physical protection, space and services needed by such large groups of people (Lilly 2014). There was reluctance, however, on the part of UNMISS to provide resources that would project the image that the camps were to become permanent, even though they were very basic and sometimes did not meet minimum humanitarian standards. The mission always referred to them as PoC sites, rather than camps, and ultimately handed most of them over to the government in 2020. As Arensen noted in a 2016 assessment of the PoC sites,

Former camp managers … argued that UNMISS was reluctant to allocate additional space for the PoC sites, as improvements would potentially increase the appeal of the camps. The IDPs were often located on very congested parts of land, adjacent to empty spaces UNMISS had set aside for developing their bases. Dozens of new staff residences were being built in UN House while IDPs had only 3.5 m2 per person in parts of the PoC site. (Arensen 2016, p. 59)

Relationship with Host Government

One of the many constraints and dilemmas for UNMISS was having to walk a fine line in dealing with an openly hostile government. The UN is not an occupying force and is only present in the country with the explicit consent of the government, and to a lesser degree other parties to the conflict. This creates a clear dilemma when there is firm evidence that the government is one of the armed actors attacking civilians.⁵ The question is how to execute the mandate to protect civilians, without being expelled. In the language of the United Nations, maintaining host state consent is one of the contradictory tasks given to the mission. Sebastián and Gorur (2018) and Labuda (2020) have set out the difficulty of negotiating and maintaining host state consent, and its implications for effective peacekeeping. Duursma (2019) also looks at how obstruction to the mission can arise.

UNMISS faced frequent obstructions by the government to resupplying its northern bases with fuel in 2014, leading to critical shortages, and the Humanitarian Coordinator and Deputy Head of UNMISS, Toby Lanzer, was expelled from South Sudan in 2015.⁶ The government also strenuously resisted and frustrated the deployment of a Security Council mandated Regional Protection Force of about 4,000 troops which was intended to respond to violent outbreaks in Juba in 2016 (Spink 2016).

It is important to understand the sensitivities around showing respect for South Sudan’s sovereignty after its long and painful struggle for self-determination. But managing relations with the government can seriously limit action to protect civilians. Day et al. describe how ‘systematic obstructions to the freedom of movement of UNMISS and its partners – by the Government and other parties – has created serious impediments to the Mission’s ability to deliver on its protection and humanitarian mandates’ (2019, p. 13). A parallel can be seen in the difficulties faced by peacekeepers in Darfur in Sudan, which was initially an African Union mission from 2004 and later became a joint operation with the UN.

UNMISS’s original mandate after South Sudan’s independence in 2011 had focused on state-building through UN support to various government institutions. Disarmament and security sector reform were key mission components, alongside UN police support for the development of the South Sudan National Police Service. To the extent that the mission looked to play an active role on PoC, much of the focus was on the conflict between the government and David Yau Yau’s ‘Cobra Faction’ armed group in Jonglei in 2012 and 2013.⁷ This focus on Jonglei, and the support of the government, contributed to the comparatively limited attention the mission gave to the rising tensions in Juba during the escalating political crisis in 2013. Political analysis in the mission recognised the rising tensions within the government in the November 2013 report of the mission to the Security Council but significantly underplayed the risk of violence.⁸

The breakdown of the government and the massacres in Juba in December 2013 therefore created a degree of dissonance and paralysis among some of the senior mission leadership as they struggled with the implications of the radically changed mission setting and the threat that the government presented to large sections of the civilian population.⁹ Such were the levels of confusion within the mission that UNMISS officials in Bentiu passed over confiscated weapons to commanders of opposition forces because of previous working relationships with them. Similarly, some of the senior leadership in Juba argued for the turning over to the government of weapons confiscated from people seeking protection in UNMISS PoC sites. This is despite the escalating violence and the contradictions of providing weapons to armed actors during the course of a conflict in which civilians were being targeted. Only an intervention from UN headquarters in New York (at the request of some junior UNMISS personnel) halted the transfer (Craze and Jérôme 2016, p. 45). South Sudan was eventually placed under a Security Council arms embargo in 2018 though, notably, UNMISS has no role in monitoring its implementation.

Complexity as the Theoretical Approach

The analytical lens used in this article is complexity theory. This opens the way to seeing more nuanced, non-linear causal connections between the numerous actors and variables in the politics and conflict dynamics of South Sudan. Besides recognising a larger number of variables, it allows for the interaction between these variables, rather than regarding the effect as merely an additive one. The series of interactions and feedback loops can amount to the actors beginning to behave like a system (such as an ant colony or an ecosystem) rather than piece of clockwork with fixed connections and relationships between its parts. The theory has been developed and applied in a number disciplines in both the natural and social sciences. More recently, it has been brought to the development and humanitarian fields (Ramalingam 2013, Andrews et al. 2017), with the promotion of adaptive practices which take account of unexpected results or lack of predicted outcomes.

More specifically, complexity theory has been brought to bear on peacebuilding in the last decade or so (Brusset et al. 2016, De Coning 2016, 2018). Day (2022) uses it as the basis to explain failures in the state-building projects in the DRC and South Sudan. The lens of complexity theory helps to make informal patterns, power structures, practices and relationship visible and reveal the way they interact in sometimes unexpected ways. Tipping points are less surprising when the dynamics are understood through complexity, along with a system’s ability to resist and adapt to change, making inputs by international actors appear to be less effective. Closer again to the topic of this article, complexity is more recently being used to understand peace support operations (Hunt 2015, 2020, De Coning 2020). Day and Hunt (2022) point out that interventions like peacekeeping cannot effect linear change when engaging with a complex system, that unintended consequences can be understood through complexity theory, and that peacekeeping itself shows fundamental characteristics of a complex adaptive system.

This approach is important in attempting to understand the complex dynamics in South Sudan, with its long history of intersecting conflicts, multiple ethnic groups and changing alliances, not to mention the wide range of outside actors which have been involved. Complexity theory exposes the importance that relationality plays – often unacknowledged – in both peacebuilding and UN operations (Hunt 2017, Day and Hunt 2022). Based on extensive fieldwork in South Sudan, Gray (2022) uses feminist relational theory to look at civilian protection, saying this illuminates a broader range of actors and indeed processes: ‘By understanding connections between actors, and broadening understanding beyond narrow conceptions of protection, we can better identify how and by whom power is used and push for transformation of unjust structures’ (p. 166). Based on this emerging literature, we feel that complexity offers a way to see and understand processes, co-evolution and relationships which would otherwise be hidden.

Methodology

This research is based on a review of UN documents and fieldwork in South Sudan involving semi-structured interviews with a wide range of actors including PoC site residents, and those working for local civil society organisations and international agencies, and UN officials. It draws on insights gained through interviews with uniformed peacekeepers, officials and organisations familiar with protection issues elsewhere, including fieldwork on PoC in Mali. It is also informed by one of the authors’ personal experiences of working for UNMISS when protection was a particularly pressing issue.¹⁰ Publicly available UN documents and the policy and academic literature have also been consulted.

Literature Review

Civilian protection has been explored in both policy and academic literature since the early 2000s, though with increasing frequency in recent years as its importance and challenges have become clearer (for example, Hultman 2013, Mamiya and Willmot 2013, Willmot and Sheeran 2014). Most of the discussion takes place within the peacekeeping literature in general. Particular case studies are explored such as Côte d’Ivoire, or the Democratic Republic of Congo, where the use of force or ‘robust peacekeeping’ has become an issue (Bellamy and Williams 2011, Boutellis 2013, Bellamy and Hunt 2015, 2021, Müller 2015, Murphy 2016). The use of force has raised particular concerns about unintended consequences and risks which arise, as debated within the UN and beyond (Peter 2015, Andersen 2018). Legal and policy aspects are also considered (Willmot et al. 2016) and issues such as training and the contribution of troops (Curran 2017).

The ways in which both humanitarian and peacekeeping actors have a role in protection and how they interact with each other in attempting to carry out their tasks are discussed by Metcalfe (2012). This study helps to explain the way in which roles and responsibilities are taken up while preserving the non-military character of humanitarian organisations. These dilemmas parallel some of those seen in the running of PoC sites in South Sudan, as well as the process through which informal accommodations and arrangements which can emerge.

The importance of early warning systems within UN peacekeeping in relation to civilian protection is examined by Mamiya and Willmot (2013), who highlight the need to engage effectively with affected communities, with South Sudan as one of the case studies. They set out an initial set of principles for such a system, such as a link to actual response, and effective dissemination of information. These ideas proved sadly to be relevant when it came to the later violence in South Sudan, especially the situation in the capital, Juba, in July 2016, after which the UN faced very stark criticism from its own internal review.

Williams (2013) reviews UN mandates for civilian protection, asking how these can be reconfigured to take account of local communities and how they can be empowered and allowed a greater say in national governance. In some ways, this points to differences between the three pillars in civilian protection which were later developed, covering a wide range of activities from physical intervention to political processes. Action in one area alone is less likely to bring sustainable change.

There is a wealth of research on peacekeeping operations in general showing that they have a positive effect on the reduction of conflict, as measured in a number of different ways (for example, Hultman et al. 2019, Walter et al. 2020). More recently, the effectiveness of PoC operations in particular has been demonstrated. Carnegie and Mikulaschek (2020) show a greater reduction in civilian casualties correlated to areas where more blue helmets are deployed. Two other studies use recently available geocoded data to look at deployments and violence at a sub-national level (Fjelde et al. 2019, Phayal and Prins 2020). They found that greater presence of peacekeepers locally is correlated with a reduction in abuses or one-sided violence against civilians, especially when perpetrated by rebel groups rather than governments.

More specifically, PoC in South Sudan has generated a wide range of policy literature (Spink 2016, Gregory and Gorur 2020). The progress of the mission as a whole is assessed as part of the Effectiveness of Peace Operations Network (Day et al. 2019). There is also a limited number of academic publications (Foley 2017, Zambakari et al. 2019, Sundberg 2020). Since the start of the most recent conflict in 2013, protection has been mentioned regularly in scheduled reports by the UN Secretary-General to the Security Council on the mission, the situation in South Sudan, and on PoC as an issue in its own right (for example, UN Secretary General 2019). Specific UN operations and agencies also deal with it, such as UNMISS and the Office of the High Commissioner for Human Rights (2017), and the regular updates from the UN Office for the Coordination of Humanitarian Affairs (UNOCHA) (2017).

The literature on PoC and gender is, however, comparatively limited, despite the fact that gender-based violence is one of the most serious threats facing civilians. While all civilians can, of course, suffer violence, women and girls experience wars and peace processes in different ways to men and boys (Ní Aoláin et al. 2011). There are also different roles, expectations, stigma and power dynamics for each gender. However, the discourses between PoC and women, peace and security do not overlap in the way they should (Hultman and Muvumba Sellström 2019, Cuzzuol and Wels 2021). Moreover, the number of female peacekeepers in uniform remains far below the UN’s own targets.

When it comes to PoC sites themselves, less has been published than on civilian protection in general, since these are a recent phenomenon, and naturally most of it appears in the policy literature. Lilly (2014) draws on his experience of working on protection for UNMISS, to give an early account of how the UN system adapted to working with other actors in managing the sites. Arensen (2016) provides important testimony from the perspective of those surviving in the camps as well as humanitarians and UN staff involved in the response and details the attack on the PoC site at Malakal. The Center for Civilians in Conflict also documents these attacks (2016a and 2016c) and other issues such as the planned handover of the sites by the UN. Stern (2015) considers establishing security at the sites, while Briggs and Monaghan write for the Norwegian Refugee Council (2017) on the lessons to be learned from the experience. Kilroy (2018) makes an early assessment of the significance of the sites and the issues they raise. More recently, discussion about the UN exit from the sites has prompted analysis and advocacy (Pendle 2019, Spink and Levine-Spound 2021, Amnesty International 2021).

Murphy (2017) gives one of the most detailed accounts of the PoC sites in the academic literature, including the way in which they came about. He describes them as an ‘untested initiative’ (p. 381) which was, however, the correct decision, although the scale of the problem was not anticipated. He says it ‘was not a pre-planned or a well thought out response and was initially intended to be a short term means of providing refuge for those fleeing imminent violence’ (p. 377). This article helps us to understand the way in which decisions came about, and the failings from which lessons can be learned. He highlights that resources used for the sites can take away from the capacity to provide protection more broadly, but says that they remain ‘an essential component’ of the UN’s protection strategy (p. 392).

Foley (2017) writes from the perspective of an international legal scholar who has worked extensively for peace support operations. He charts the development of policy on protection, its basis in the UN charter and international law and varying conceptions of it by different actors. He then looks at it in practice in the DRC, Darfur and South Sudan. He provides a detailed account of the speed at which the situation developed in South Sudan, the challenges faced and the ways in which the sites themselves came under attack, leading to loss of life. The choices facing the mission are also highlighted: protecting and facilitating the PoC sites meant that fewer personnel and other resources were available to provide protection outside of them, where the vast majority of people lived. He also explores the difficulties and legal basis for detention of camp residents accused of violence or other crimes within the sites.

Berdal and Shearer (2021) also write from the perspective of both academic and practitioner.¹¹ They point out that in 2017, half the mission’s resources were estimated to be devoted to protecting the six sites, which accommodated about a tenth of the IDPs.

Munive (2021) uses the idea of ‘the politics of resilience’ (p. 1880) to understand the tensions over longer term existence of the sites. UNMISS was keen to keep services to a minimum and encourage people to leave the sites by providing supports elsewhere, but humanitarians wanted to provide better services and help build the capacity of residents to withstand future shocks. While the focus of this article is on how ‘resilience’ is conceptualised, it provides useful detail on concerns within UNMISS that the sites were attracting new arrivals and taking up resources which should have been used for protection in the community at large. The tensions with humanitarians are clear.

Paddon Rhoads and Sutton (2020) look at how community justice systems within the PoC sites can be seen as a form of self-protection by civilians. This is based on an extensive range of interviews and is one of the few academic articles to investigate the challenges faced by those living in the sites, whose safety was far from guaranteed, including from fellow IDPs. It also highlights the gendered nature of insecurity, and how it is experienced differently by men and women. The lens of community self-protection practices is a useful one for South Sudan as a whole, as it forms part of the mosaic of practices, options and actors – both military and non-military. It also recognises the agency which civilians and communities have, whether or not this is realised by international actors.

Lilly (2021) looks at the difficulty task of protection, while UN missions are going through transitions to a new phase at a time when threats to civilians persist. This is especially relevant when it comes to UNMISS handing over responsibility for PoC sites to the government of South Sudan. He points to the need to foreground civilian-led protection, a shift towards peacebuilding, and to foster national ownership of the project.

Donais and Solomon (2022) contrast the resources and attention focussed on PoC sites in comparison with other ways of dealing with the conflict. They call for a broader and more locally based peacebuilding effort to be prioritised, as part of the protection agenda, rather than relying on the sites as a primary means of protection. This is of course consistent with the UN’s own protection policy, which encompasses a wide range of actions. It also mirrors the rationale put forward by UNMISS for handing over the sites to other actors, so that they could focus on protection in other spheres.

Keen (2017) includes PoC sites in his analysis of UN-designated ‘safe havens’ which were seen in other conflicts such as Bosnia, Iraq and Rwanda. He recognises that the sites in South Sudan emerged much more spontaneously than the planned safe zones in other contexts. While acknowledging that the sites made a real contribution to civilian safety, there were problems with fatal attacks on people living in the sites, arms being brought into them and poor living conditions. The fact that little protection was available to those living outside the sites is also highlighted by Keen, with mission resources tied up in the sites rather than elsewhere.

The relevant analyses provide an excellent background to civilian protection, and a more recent part of the academic literature now deals with the specific question of PoC sites, which is the subject of this article. The sites are still significantly under-researched, however, and gaps remain in the literature which can be addressed here. These include the use of complexity theory to understand the dynamics affecting protection and the sites and the way in which seemingly spontaneous events and top-down initiatives interact with each other.

Analysis

Response by UN and Lack of Safety within the Camps

The UN Secretary General reported to the Security Council that he was proud of the quick action and courage of the mission staff to open the gates in the emergency, and that this had saved tens of thousands of lives. He described the action as ‘correct, unprecedented and not without considerable risk – to United Nations staff, to our relations with communities and to those we are trying to shelter’. Arensen (2016) and Briggs (2017) agree that thousands of lives were saved, and that humanitarian organisations saw the camps as meeting an urgent need. A comprehensive review by Day et al. (2019) says the sites ‘unequivocally saved lives and helped prevent a far worse trajectory for the conflict in South Sudan’ (p. 3).

The idea that civilians should have to move to a camp in order to be free from violent attacks is, ideally, an interim emergency measure, an option only taken when other forms of protection are unavailable or unsuccessful. Guarding a fixed site does, however, provide some advantages. Having the people in need of protection in a clearly designated, accessible area is an easier option for peacekeepers than trying to cover vast geographic areas with limited uniformed personnel, particularly when freedom of movement is often restricted and in the face of serious logistical challenges during the rainy season.

However, protection of the sites has not been straightforward, with several attacks on the camps leading to dozens of fatalities. The failure of the peacekeepers to protect residents mirrors the wider challenges for protection in general: risk aversion, unclear lines of command, lack of preparedness and a slow response when an attack is reported. One such incident occurred in Bor in April 2014, when peacekeepers were reported to have retreated into the site and left unarmed NGO staff to deal with the attackers (Foley 2017). Dozens were killed before some peacekeepers fired shots, leading the attackers to leave the base. Briggs says that since they were established ‘four POC sites have been overrun or shelled and over 180 IDPs have been killed during attacks on the sites’ (2017, p. 19). These were Bor (discussed above), Akobo, Malakal and Juba.

Another deadly attack on a PoC site occurred at Malakal in February 2016, with similar failings on the part of the peacekeepers (Murphy 2017). This site was a mixed one, with Nuer, Dinka and Shilluk populations in separate sections. The attack followed disturbances within the camp between these groups, involving firearms and outside forces. It lasted almost a full day and was clearly targeted at specific ethnicities in the camp. Nearly a third of the camp was burned down, including all of the Nuer area and much of the Shilluk parts but leaving most of the Dinka housing intact (Arensen 2016). The Sudan People’s Liberation Army (SPLA), the Dinka-dominated state forces, was seen to take part in the assault, and more than 30 civilians died (Center for Civilians in Conflict 2016a). This highlights the reality of the state being one of the groups attacking civilians, despite having primary responsibility for its own citizens’ protection. It also underlines the difficulty for the UN in how it confronts a state threatening its civilians, when the peacekeepers also require permission from this host state in order to operate or even be in the country. In this case, the UNMISS response on the day was criticised for being slow, limited and excessively risk averse. The use of force is always problematic, with significant risks and possible negative consequences, but the fact that peacekeepers were present yet did not act effectively creates real doubts among the population that they will be protected inside the PoC sites.

An even more grave crisis erupted later the same year as an uncertain peace agreement fell apart and fighting broke out in Juba between the two main factions to the power-sharing agreement. One of the leaders, Riek Machar, was forced to flee the city, eventually escaping the country, and again civilians were attacked on the basis of ethnicity. It also created a difficult situation for UNMISS, whose base and PoC sites came under attack, and not just from small arms fire. Helicopter gunships were operating in close proximity to the UN House base, and UN humanitarian stores were looted. Two Chinese peacekeepers lost their lives when their armoured personnel carrier was struck by a rocket propelled grenade. The response to attacks on the PoC sites was varied and poorly coordinated, with contingents from different troop contributing countries sometimes taking their own approach. Some peacekeepers actively helped civilians, while others abandoned their posts or even used tear gas on civilians who fled the violence and entered the main UN base from the neighbouring PoC site (Center for Civilians in Conflict 2016c).

There were also major failings amid the confusion when the peacekeepers did not respond to sexual violence against civilians taking place within their sight immediately outside the PoC site. They did not react to repeated calls for help during a prolonged attack on humanitarian staff involving murder and rape. A Quick Reaction Force which was supposed to be mobilised did not even leave the base (Center for Civilians in Conflict 2016c).

The UN’s own report on the failures is damning and highlights many important issues and lessons regarding civilian protection (United Nations, 2016). It says that there were adequate warning signs that hostilities would recur, but that the mission did not prepare properly for scenarios that were in fact foreseeable. Its criticism is direct, adding that ‘Lack of leadership on the part of key senior Mission personnel culminated in a chaotic and ineffective response to the violence’ (p. 2). It adds that a ‘culture of reporting and acting in silos inhibited effective action [when] swift, joint action was essential’ and that the ‘force did not operate under a unified command, resulting in multiple and sometimes conflicting orders’ (p. 3). The force commander was effectively dismissed shortly after in response to these failures, leading to a strong protest by the country he came from, Kenya. The experiences in Juba are seen as some of the starkest examples of the UN coming under physical attack, and of failing to protect civilians, since PoC has become a significant part of the mandate.

Resistance from within UN and Handover of Sites

The attitude towards PoC sites within UN leadership revealed a certain ambivalence. Many lives were saved by swift and innovative action in the field – experiences that were drawn on to develop new UN PoC policies (Lilly 2014). Other UN missions drew up detailed plans on how to deal with large numbers seeking protection at each of their bases, so that they would be more prepared in case this was needed. However, there was also resistance and a concern that the existence of the camps would attract a population of long-term residents because of services provided there.

While UNMISS sought to accommodate and protect the people within its bases, and indeed frequently pointed to the sites as evidence that the mission was fulfilling its PoC mandate, it also sought, even from the earliest days of the conflict, to divest itself of responsibility for the sites. The sites, it was argued, were a drain on UN resources and reduced UNMISS capacity to provide protection to the majority of IDPs in South Sudan who were outside of its bases.¹² Moreover, the mission leadership was concerned that the PoC sites, in providing humanitarian services, would attract yet more IDPs and others seeking support. As noted in a 2016 report by the International Organization for Migration (IOM):

As UNMISS leadership did not have an interest in increasing the pull factor to the PoC sites, it allegedly sought to limit the services offered by INGOs. For example, one organization, reportedly, infuriated UNMISS leadership by creating a child friendly space in UN House as it was perceived to create an additional draw factor. (Arensen 2016)

Van der Lijn (2017, 197) notes that ‘UNMISS policy has been to keep living conditions in the POC sites basic to avoid providing incentives for civilians to stay long term or attracting additional civilians to the bases’. There was therefore a Janus-like ambiguity to UNMISS’s presentation of PoC sites. On the one hand, they were held up as evidence of UNMISS’s achievements in protecting civilians in South Sudan while simultaneously being pointed to as limiting the mission’s achievement of its PoC mandate. This contradiction is a key tension within UNMISS’s execution of its mandate where PoC sites are presented as both a success and an impediment.

This proved to be a significant tension between mission and humanitarian partners (and by extension the mission and IDPs) as the different UN pillars attempted to coordinate and design a long-term response, with the mission emphasising the temporary nature of the PoC sites while not having a clear sense of when the protection needs of those in the PoC sites would facilitate voluntary return (see, for example, Munive 2021).

The PoC sites also presented a serious problem for the mission in terms of law and order. As the population of the sites increased, and the duration of the stay became drawn out, dealing with crime became an increasingly urgent and complex issue for the mission. Violent criminality presented a protection problem for residents of the PoC sites, but UNMISS had no mandate or facilities to detain or process offenders. The mission was forced to deploy UNPOL and prison advisors to develop an ad hoc detention facility and informal justice system. By 2017, an estimated 3,800 detainees had been held (Briggs and Monaghan 2017).

The detention facilities were constructed quickly in response to an emergency situation and without the mission being specifically budgeted for their development. The conditions were therefore extremely basic, and concerns were raised that they were not only inadequate but violated prisoners’ rights (Briggs and Monaghan 2017). But perhaps the deeper dilemma for UNMISS was that there was no legal standing to detain South Sudanese citizens, a mandate which was only held by the national authorities. However, given that IDPs feared they would be harmed if turned over to the national security services, and UNMISS was unable to guarantee the safety of persons detained by national authorities, detainees were in a legal and protection limbo: held to protect IDP communities, but also held to potentially protect them from the South Sudanese government. This was a no-win situation for the mission and a dilemma that could be presented by the government as UNMISS exceeding its mandate while also violating detainees’ rights. Ultimately, UN peacekeeping wanted to get out of the business of protecting long-term sites, and started redesignating them as Internally Displaced People (IDP) camps in preparation to passing on responsibility for security in them (Mold 2020). By early 2021, numbers had fallen to about 125,000 people, and five of the six sites had been handed over by UNMISS to the government. This was the same government whose forces had carried out many of the organised attacks on civilians, including assaults on the very sites now coming under their control. This caused very real concerns among some residents, with fears for their safety after the UN’s departure (Mednick 2021, Spink and Levine-Spound 2021).¹³ One of the arguments put forward by UNMISS was that exiting the camps would free up UN troops to carry out PoC work in the rest of the country. However, the Special Representative of the Secretary General (SRSG) in South Sudan said that the transfer of PoC sites to the government in fact allows for the troop and police numbers to be reduced by about 1,400, as well as others being redeployed within South Sudan (UNMISS 2021).

Findings

This analysis is based on documents, interviews and reports of the events in South Sudan, as outlined in the Methodology. It is informed by complexity theory in order to take account of the interactions between factors and emergence of systems behaviour in this context. The argument is made here that PoC sites are one of a number of significant factors in the wider spectrum of protection activities. Also, that they highlight issues which help us to understand some of the questions and critical tensions involved in UN missions fulfilling their PoC mandate in certain conflict conditions. While they do have a clear and specific role, especially in the short term, that is a particular part of a wider holistic response. It is important to understand protection in this way, making use of complexity theory in order to see the causal dynamics in a non-mechanistic way. This approach is particularly relevant in a complex conflict such as that in South Sudan, with deep historical and sociological roots, and multiple actors. It allows us to ‘see’ a wider range of variables, and how they interact with each other, rather than looking for simplistic causal connections. One analogy for PoC sites might be that of an accident and emergency department within a wider health service: it can play a vital immediate function but does not replace the rest of the system, and its effective input may be limited to a specific time period. A peacekeeping mission can only have so much influence on the overall conflict dynamics, rather being able to control it entirely. However, it is also important to avoid false dichotomies which reduce success and failure to all-or-nothing binaries: just because the sites are not a long-term solution for PoC on their own does not mean they have no role to play in making a contribution to protection in important ways. That argument could be used against all aspects of PoC by anyone using a more simplistic causal model. Protection is in fact a multifaceted and complex set of inter-related tasks. To turn it the other way around, failing to have a plan for the arrival of people seeking protection, or just focussing on a simpler challenge, is not a solution either. Our analysis aims to recognise that power takes many forms, and actors are influenced in any number of ways besides physical force. Sometimes the most lasting changes can be created slowly and below the radar, at the microlevel and without even being noticed.

The sites were the most visible manifestation of UN protection in South Sudan after 2013, in the context where a third of the population was displaced and nearly 400,000 people died during the four first years of the war (Gregory and Gorur 2020; Checchi et al. 2018). The sites were also the most clearly measurable PoC result, with metrics showing how many people were directly under UN protection (for example, UN Office for the Coordination of Humanitarian Affairs, [UNOCHA] 2017). This is valuable when assessing the impact of peacekeeping operations as other metrics have often proven to be less clear to the point of being opaque.¹⁴ Independent assessments cited earlier agree that many thousands of lives were saved by allowing civilians to enter UN bases and setting up the sites (see also Foley 2017). This indicates that PoC sites are a valuable response for UN missions under extreme conditions and, therefore, should feature in mission planning where PKOs face similar conflict conditions.

The design of UNMISS’s mandate in 2011 contributed to difficulties in implementing a PoC mandate under the conditions of the expanding conflict in 2013. The mission had been built with capacity support and partnership of the government as central to its understanding. Now that these aims were not only no longer viable but also contrary to the aim of establishing impartial and effective practices during the conflict, the mission became caught in a bind. The government was also not interested in seeing the UN act with impartiality in the conflict and sought to pressure the mission to maintain close alignment with the government’s goals. PoC sites increasingly became a source of tension as the mission sought to meet the escalating needs of IDPs while being attacked, both verbally and physically, by the government.

The government maintained a consistently negative rhetoric against the PoC sites, claiming that the UN was using them to justify expanding the mission, while also claiming that the UN was supporting the rebellion by hosting hostile elements in the sites.¹⁵ The PoC sites raised uncomfortable questions about the government’s legitimacy as it was not fulfilling its obligations to provide protection, while also demonstrably being a significant human rights abuser. The sites – with Bentiu providing shelter for more than 100,000 people – were a visible example of the scale of displacement in the context of killings, ethnic cleansing and gender-based violence (Amnesty International 2016, Arensen 2016, UNMISS and Office of the High Commissioner for Human Rights 2017). The PoC sites were therefore emblematic of these failings, a factor that provides a significant explanation for the desire of the government to remove this high-profile illustration of its failures.

Initial Response

In December 2013 UNMISS had approximately 7,000 force personnel in the country. These were widely distributed throughout South Sudan, with concentrations of various size and capabilities in Juba, Bor, Bentiu, Malakal, Akobo and Pariang, among other sites. This wide geographical distribution had three key impacts on the mission’s response to the civil war. Firstly, the comparatively few UN troops in Juba when the fighting broke out limited the UN’s potential response in the capital.¹⁶ Second, due to most bases only being accessible by aircraft and with communications out in many locations, bases became quickly isolated from the broader UN management structure of the mission. This resulted in staff in different locations operating largely autonomously, particularly in the early days and weeks of the conflict. For example, UNMISS Bentiu, with UN force, police and civilian elements working together, had established a makeshift but distinct PoC site in an area adjacent to the base almost immediately as the fighting began, whereas UNMISS Malakal struggled to organise a response as IDPs occupied the entire base.¹⁷

As UNMISS headquarters in Juba struggled to understand the emerging conflict dynamics and effects on the mission, the third impact of the distribution of bases began to take shape. IDPs that could not receive protection in their current location were in many instances able to access protection by coming to the mission. In the opening weeks of the conflict, IDPs from most of South Sudan’s ethnic groups, along with foreign nationals, were present in many of the PoC sites.

The response and performance of UN forces and UN police contingents continued to vary across the South Sudan as contingents from different countries interpreted their responsibilities differently and often sought direction from their home headquarters in preference to direct orders from UN force headquarters in Juba. Some forces were extremely proactive in both developing the emergent PoC sites and establishing defensive structures, while others were reluctant to enter the PoCs or to actively patrol beyond the perimeter of their bases.¹⁸

As large areas of UNMISS’s mandate were, in the wake of December 2013, no longer operable (such as support to South Sudan’s police service), the mandate restructured organically while it awaited a new Security Council resolution, which came in May 2014. Resolution 2155 expanded the mission’s troop ceiling to 12,500 and UN police numbers to 1,323. Significantly, however, the mandate renewal was for only six months (12 months had been the previous standard). There was, therefore, a perception among some UNMISS leadership that the new mandate was temporary, and that the mission’s ‘real’ mandate would be restored once the fighting had subsided. This implicitly placed the mission in something of a holding pattern: while dealing with the immediate needs for planning and responding to the crisis, a return to ‘business as usual’ was anticipated. The PoC sites sat uncomfortably with this perception since, by May 2014 and as the numbers of displaced persons continued to rise into the hundreds of thousands, the PoC sites were clearly a more long-term proposition that the mission would have to reckon with for the foreseeable future.

The short mandates for UNMISS during the first stage of the war, combined with the messaging from some senior management – particularly SRSG Ellen Løj who was appointed in September 2014 – on the temporary nature of the mandate activities, reinforced the perception that PoC sites were short-term solutions.¹⁹ This tended to impede contingency planning as there was an inherent tension between this short-term perspective and the long-term reality of the sites. This also encouraged the desire to seek ways to divest UNMISS of responsibility for the sites as discussions about handover to humanitarians or the government took place even as the war continued. This desire to withdraw from the sites was encouraged by mission leadership and UN force asserting that PoC was best conducted outside of the bases.²⁰

This dynamic created the perception within the mission that the PoC sites were a cause of UNMISS’s inability to provide more effective protection beyond the bases, partly due to the resources needed to host the IDPs seeking protection. However, this view is predicated on the assumption (by those who saw it this way) that UNMISS would and could conduct more effective PoC beyond its bases if only UNMISS forces could be released from the task of securing the fixed PoC sites. But this view misses a key point: the formation of the sites came about precisely because UNMISS was unable to provide protection beyond its own perimeters, so people seeking shelter moved to the peacekeepers rather than the other way around. It is therefore a false dichotomy between PoC in the bases and outside; under the conditions in South Sudan, and with the constraints on UNMISS force, PoC sites were for some the best and largely only form of protection available.²¹

Use of Force

Some of the debates on the use of force by peacekeepers have been referred to in the literature review. The risks associated with using force, in the context of responding to direct attacks on civilians, have been seen particularly in the DRC, where the Force Intervention Brigade used its additional authorisation effectively to carry out robust operations against the M23 militia which had preyed on the civilian population. But dealing with M23 (for a time before their resurgence later on) is not simply something to be ‘scaled up’ until the attacks have stopped, given the large number of armed actors. How many groups might be taken on, without significantly adding to the risks, costs and unintended consequences? These include civilian and peacekeeping casualties, changing the nature of the mission and its relationship with the population, armed actors and the state. In the case of South Sudan, would the regular use of force by the UN put its entire mission in jeopardy, given that the consent of the host state could not be taken for granted? This also has to be balanced with the very real risks and costs of not acting in a robust manner, when this is the only or best remaining option to prevent direct violence against civilians.

What is clear is that in situations where missions face complex conflict environments with threats to civilians from armed groups including the government, there needs to be a holistic approach from the UN system both to support the implementation of PoC and also develop sufficient pressure on the host government to refrain from obstructing the mission. This holistic approach aligns with the adaptive practices suggested by complexity theory. However, it has often been absent in South Sudan as divisions in the Security Council over issues such as targeted sanctions and an arms embargo have contributed to a divided UN response. The mission has therefore often sent mixed messages on some political issues.²²

Policy Implementation

This analysis of PoC sites aims to take account of the different timescales involved for various actions, and the fears that a short-term solution would become an open-ended commitment to hosting displaced people indefinitely. The context is one of growing urbanisation in South Sudan, and the appeal of services within the sites such as mobile network coverage which are not available to everyone outside of the sites. Close to large cities such as Juba, residents often leave the site for work or study, returning there in the evening. Any strategies for facilitating the return of residents to their place of origin are bound by the requirement that this is freely chosen, based on accurate information, and is supported. Returns are complicated by questions around recovery of land, housing and property, especially where these have been occupied or looted. Future research on PoC sites could include longitudinal studies, attitudes and motivations surveys, and analysis of displaced persons that establishes a clear factual basis for policy, rather than basing findings on assumptions about these communities’ priorities and needs. It would also be instructive to compare these with of the views of refugees or people living in other IDP camps.

One of the questions raised by the experience of PoC sites in South Sudan is the way policy is made, internalised and institutionalised by various parts of the UN system, after the initial response has been developed by those on the ground. The urgent and pressing needs saw what is usually termed ‘field-led innovation’ in its more extreme forms, where personnel on the ground were forced to innovate when faced with the arrival of so many people desperately seeking safety. Some of this was later codified and applied to other missions, in case they face the same sudden demands. Policies have been developed and refined since. UNMISS’s handing over of PoC sites to the government raises once again issues of the relationship, responsibilities, and sometimes divergent aims of the mission and the host state. The experience of large PoC sites sheltering tens of thousands of people at UN bases (as opposed to regular IDP camps) may turn out to be a one-off phenomenon which is not repeated outside South Sudan. If this is the case, how will peacekeepers respond if they are faced with mass movement to their bases in the future?

One implication of the experience of UNMISS with PoC sites is the need for independent oversight of a mission throughout the development and implementation of such sites. The Department of Peacekeeping Operations (part of the integrated Department of Political and Peacebuilding Affairs at UN headquarters in New York) nominally performs this role. But given the complexity of PoC sites and their intersection with the humanitarian pillar of the UN as well as non-UN agencies helping to run the sites, there is potential benefit to regular review of the sites from an entity external to the mission. This would aim to ensure that the analysis and planning for the sites is removed, to the degree possible, from the political pressures of the mission where coping with the conflict environment and a belligerent government can influence the mission’s disposition towards this key protection response.

Constraints on Protecting Civilians in General

Besides the use of force and host state consent, PoC sites highlight many of the existing constraints faced by the UN in tackling protection of civilians. These include the comprehensive nature of multidimensional mandates, which call for a wide range of actions which can interfere with each other, and require many different skill sets. How these mandates are operationalised and prioritised is where important choices are sometimes made. The issue of having enough, and appropriately motivated and led, uniformed personnel made available to the UN also arises, especially since troop and police contributing countries will have an understandable aversion to placing their people at risk. The capacity of the missions, their training levels, gender representation, dual lines of command and level of turnover are also relevant to both PoC sites and protection as a whole. The idea of protection involves creating and disseminating a new norm, which is rarely simple or straightforward, especially when it challenges other norms such as sovereignty, force or national ownership.

In this context, it is important that UN peacekeeping reflects on the successes and failures demonstrated in South Sudan. Despite the strain they placed on the mission, PoC sites were a demonstrable success against a backdrop of failures in other forms of PoC.

Conclusions

This article set out to explore the ways PoC sites in South Sudan evolved and the challenges that resulted for the UN in implementing its mandate to protect civilians in a complex conflict environment. The article is based on fieldwork and documents, and informed by first had experience. The sites were seen to have been a significant part of the protection ecosystem, rather than a stand-alone project or an alternative to PoC in general. While providing protection to thousands, many important lessons can be seen. Real difficulties, failings and dilemmas also arose, some of which reflect the complex nature of protection as a whole.

One of the key lessons from analysis of UNMISS operations during the civil war is that the PoC sites were the result of the mission’s inability to conduct PoC outside of its bases, contrary to views that argue that the PoC sites impeded PoC. This illustrates the gap between the ideal and the achievable. The PoC vision that informed the initial mission mandate was to support the government in providing protection throughout the country. When this proved impossible, people seeking protection created a solution by coming to the bases. A pragmatic assessment of this reality would guide mission planning in similar contexts. However, the dilemma remains deep as missions do not want to ‘plan to fail’. Accepting that PoC sites would form a predictable fall back for missions in the future would, for some, be tantamount to conceding before even beginning that other PoC options (such as robust patrolling) are unlikely to yield similarly demonstrable results as PoC sites.

Any useful analysis has to be a holistic one, which recognises the three tiers of protection and the role of many actors beyond those in uniform. It must also take account of systems thinking and understanding both obvious and subtle causal processes at work – and how these interact. PoC sites have been a significant innovation, saving many lives in the short term, but which brought important questions to the surface. Even if the South Sudan experience turned out be one which peacekeepers were reluctant to repeat, the decision is ultimately made by civilians deciding whether or not to seek protection at UN bases. Ultimately, outside actors may have vital roles, but the long-term protection of civilians is primarily the responsibility of the state. This means an inclusive, nationally owned, and sustainable peace process involving all parties and including women is part of any lasting solution.

PoC sites represent a very specific experience which has no parallel in peacekeeping, due to their scale and the way in which they arose. They are worth studying for those reasons alone. The UN response was unlike other situations where civilians have sought refuge, and some of which have been set out in new or amended policies and practices. They are seen to have provided immediate protection to many thousands of people who would otherwise have lost their lives. However, the sites can only be part of a holistic solution, which includes a wide range of actions somehow carried out in harmony. These range from physical protection by uniformed peacekeepers to the promotion of political dialogue and human rights monitoring. Berdal and Shearer remind us of the HIPPO report’s emphasis on the ‘primacy of politics’ and that ‘there is no technical fix to the challenge of civilian protection in civil-war situations, and durable solutions require addressing the political dynamics that drive violence (2021, p 76).

Complexity theory helps to expose some of the non-linear causal relationships between the various aspects of PoC, and the ways in which factors can interact to produce sometimes unexpected results. It can make us more capable of seeing how various functions and actions – including PoC sites – can sometimes form part of a general ecosystem of protection and insecurity, which is of course nested within the wider conflict system.

There are tensions with the government, and ambivalence on the part of the mission, which did not want to become stuck with the sites. Protecting the sites from external attack and the authority and mandate around policing them raise further dilemmas, with very significant failings by some peacekeepers to respond to attacks on the residents. Some parts of the mission were highly committed to protection, while others failed.

One of the adaptations which had to be made was how long the sites have lasted: what initially thought of as a temporary emergency response quickly became established and ‘acquired a degree of permanence’ (Berdal and Shearer 2021, p 75). Handing sites over to the government also raises important questions about the safety of the residents, who naturally have fears about facing further violence.

Allocation of resources is also an issue, since UNMISS personnel and logistics became tied up in support the PoC sites, which were never part of its original plan on that scale. However, this is not the only constraint on UN effectiveness. Even if these resources were freed up, risk aversion, leadership, impassable roads in the rainy season, and government obstruction mean it cannot be presumed that these resources would automatically be used effectively to provide protection outside the sites.

All these make sites worth studying in their own right, and as part of the wider protection system.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Field work for this research project was supported by an Irish Research Council New Foundations grant, and by Dublin City University’s Faculty of Humanities and Social Sciences.

Notes on contributors

Walt Kilroy

Walt Kilroy is an Assistant Professor at Dublin City University’s School of Law and Government. He is also Associate Director of the University’s Institute for International Conflict Resolution and Reconstruction. His teaching and research include post-war recovery, peacebuilding, the UN, and peacekeeping. His PhD research looked at the reintegration of ex-combatants in Sierra Leone and Liberia. He previously worked in development and conflict in East Africa.

Klem Ryan

Klem Ryan served with UNMISS from 2013 to 2015 and was subsequently Coordinator of the UN Security Council Panel of Experts on South Sudan. As Head of Operations for the UNMISS Relief, Reintegration and Protection section, he worked extensively in the establishment of the PoC sites and was involved in the development of policy and coordination mechanisms as the Mission responded to the first years of the civil war. He holds a Doctorate in Political Theory from Brasenose College, Oxford. He currently consults on protection of civilians, peacebuilding, and mediation from his home in Stockholm, Sweden.

Notes

1. For example, in the Democratic Republic of Congo in 2006, and in Darfur between 2008 and 2014. See Protection of Civilian Sites, NRC (2017) available at: https://www.nrc.no/resources/reports/southsudan/. It was also seen in 2023-24 in Gaza, where large numbers of people sought shelter at approximately 150 UN facilities. Some 70 of these facilities were directly hit by gunfire, artillery, or tank shells, resulting in the deaths of almost 400 IDPs sheltering inside them. Some UN sites were forcibly emptied of IDPs by Israeli security forces (UNWRA 2024).

2. Though previous experience from Jonglei in 2012 had provided a recent precedent for UNMISS. Ibid.

3. UN peace support operations started to become ‘multidimensional’ around the end of the Cold War with the advent of second-generation peacekeeping, as they took on additional civilian tasks such as political dialogue, disarmament, humanitarian concerns, and policing. See, for example, Diehl and Druckman (2015).

4. https://theglobalobservatory.org/2014/07/agony-of-unmiss/

5. As extensively documented in various Security Council Panel of Experts on South Sudan reports. See particularly the final report of the panel for 2017 (S/2017/326).

6. As personally observed by the author.

7. See UNMISS mandate renewal on 11 July 2013 (SC/11058): https://press.un.org/en/2013/sc11058.doc.htm

8. The second paragraph of the UN Secretary General’s report to the Security Council (S/2013/651) states: ‘South Sudan celebrated its second anniversary of independence on 9 July 2013 amid concerns about potentially destabilizing divisions within the ruling party, the Sudan People’s Liberation Movement (SPLM), a serious security and human rights situation in parts of Jonglei State and uncertainty about the resumption of oil exports. Since then, the country has witnessed encouraging developments and positive steps in key areas, including the appointment of a leaner Cabinet on 31 July and the resumption of oil flow. While there is stability in many parts of the country, the security situation in parts of Jonglei State remains serious’ (2013, p. 1).

9. The senior leadership of UNMISS is comprised of the Special Representative of the Secretary General (SRSG), two deputy SRSGs, the Force Commander, UNPOL Commissioner, the Director of Mission Support, and the Chief of Staff. One of the authors worked closely this group in his various roles with UNMISS from 2013 to 2015.

10. Klem Ryan served with UNMISS between 2013 and 2015. He was Head of Operations for the Relief, Reintegration, and Protection Section from January 2014 and directly oversaw operations in the PoC sites in Juba.

11. David Shearer headed UNMISS 2017–2021.

12. The redeployment of UN troops to ‘hotspots’ for PoC was provided as one justification during the withdrawal of troops from sites in Wau and Bor in 2020: https://unmiss.unmissions.org/un-protection-civilians-sites-begin-transitioning-conventional-displacement-camps

13. Similar concerns have in fact been seen in Darfur in Sudan around the same time, although in this case it involved regular IDP camps rather than PoC sites at UN bases. The ending of the joint UN-African Union peacekeeping mission in Darfur, UNAMID, meant that responsibility for the camps came under the government of Sudan, some of whose forces had created the displacement of a large section of the population and who were not trusted by some camp residents (Kleinfeld 2021).

14. Discussions on how to assess the effectiveness of PKOs have been of increasing interest in academic literature. See Hultman and Tidblad-Lundholm 2020.

15. For example, see comments from Michael Makuei Lueth, South Sudan’s Minister of Information: ‘if there are [UN] people who are living on these people [IDPs], if you remove them, then you have deprived them of their livelihood. This is how I understand it. This is not correct. We want all these people to be taken to their home areas’, he said. If U.N. representatives ‘are unable to provide the protection, should they, will they assemble all the people of South Sudan in the POCs so they will protect them?’ (quoted in Craig 2016). Makuei was placed under US treasury sanctions in 2017 for, inter alia, attacks on the UN mission.

16. The then SRSG, Hilde Johnson, makes a point of the limited UN forces available in her book discussing the events of December 2013 citing this as, in her view, a significant factor in the UN’s failure to more actively limit the violence and protect civilians. However, she does not discuss in detail the failure of some of the forces present in Juba to effectively respond to the crisis even within the limited capacities they had available. It is not, therefore, clear that additional forces would have significantly altered UNMISS’s response in the opening days of the war.

17. As observed by the author in the company of Deputy SRSG Toby Lanzer on 24th and 25th of December 2013.

18. As directly observed by one of the authors.

19. SRSG Ellen Løj during an UNMISS press conference in Juba on 11 December 2014 strongly emphasised that she viewed the PoC sites as temporary and that the Mission’s focus was on getting people out of the sites.

20. The mission received significant criticism from NGOs, such as MSF, and members of the Security Council who shared this assessment of the need for UNMISS to ‘get out of the bases’.

21. See, for example, SRSG David Shearer’s speech to IDPs announcing the withdrawal of peacekeepers from some PoC sites in 2020: ‘Reducing the number of peacekeepers on static duties at these sites will enable us to do what we came here to do in the first place – protect all civilians who are in imminent danger to the best of our ability … . We can do more patrols in areas where there is real need, be more agile and effective, and build capacity among local law enforcement’. UNMISS 2020.

22. In one glaring example, in 2019, UNMISS hosted a South Sudanese army general under UN sanctions at the official UN day celebration: https://twitter.com/AmaralPhilip/status/1133809572693594113