Camille Perreand, Paris, MSF-Crash

This study started out looking at the shift to a post-conflict situation in Katanga; the intention was to describe how MSF’s sections negotiated the transition from war to post-war. This raised the question of where to draw the line that distinguishes a conflict from a post-conflict situation. This was no easy task in North Katanga between 2000 and 2006. The situation at the time could be described neither as war nor peace; armed groups were committing widespread atrocities against civilians and this in turn led to extensive internal population displacement and generalised poverty. Our study does not therefore examine a post-conflict situation as such, but an unstable situation that is hard to classify within the dominant descriptions as being conflict/post-conflict, post-war or in transition. Having examined MSF’s archives, held interviews and discussions with MSF actors and studied academic papers, I will now attempt to recount the difficulties encountered, and the solutions offered, in Katanga from 2000 to 2008 by the MSF Belgium and MSF France teams.

1. DEMOCRATIC REPUBLIC OF CONGO: THE CHALLENGES OF TRANSITION

This first chapter examines the situation in the DRC during the years 1998 to 2008. The idea is to provide a brief sketch characterising the period: the shift from armed conflict to transition, upsurges of conflict during the transition, successive UN and donor policies, the organisation and state of the Congolese healthcare system, international healthcare cooperation, epidemiological alerts and high mortality rate caused by the conflict as well as the violence inflicted by armed groups. These factors combined to ensure that the DRC constituted a major international governance challenge during the war and into the transition period, which began in 2002. A number of studies have been produced on international governance in the DRC as well as on the armed groups and rebel movements, Congolese political culture, the electoral process, the pursuit of conflicts in the Kivus and Province Orientale, the pillage of mineral wealth, etc. This is far from being an unknown war. The information presented here is drawn from these studies, but is necessarily only a very small part of the facts and figures they contain.

1.1. A BRIEF HISTORY OF THE TRANSITION

In 1996, in East Zaire, the rebels of the Alliance of Democratic Forces for the Liberation of Congo (AFDL) and Rwandan forces launched what was termed a war of liberation. This war brought about the downfall of Mobutu’s regime in May 1997, with AFDL chief Laurent Désiré Kabila and his Rwandan allies taking up the reins of power.

In 1998, Kabila renounced his alliance and Rwandan meddling; war erupted on 2 August. A rebel movement was formed under the direction of former members of Zaire’s armed forces and Rwandophone Congolese Tutsis. This movement was backed by Rwanda and Uganda, whose armies were also engaged. In the space of a few days, these forces seized the main towns of the east. Buoyed by this success, the rebels created the Rally for Congolese Democracy (RCD) in August 1998. Kinshasa forged alliances with Zimbabwe and Angola, and the presence of these countries’ forces in the field led to a stabilisation of the front-line.

The main aim of the agreement reached during July and August of 1999 in Lusaka, under the guidance of the UN and the OAU, was the progressive establishment of an inter-Congolese dialogue intended to prepare the ground for a new political and institutional framework. The agreement stipulated that foreign troops would withdraw at a later stage, following the establishment of new institutions and the deployment of an international peacekeeping force. However, Laurent Kabila tried to replace the terms of the agreement with a national debate between equals. Initially viewed by the international community as the key obstacle to peace in the region, the DRC gained diplomatic leverage thanks to the outbreak of fighting for control of Kisangani, in August 1999 and again in May and June of 2000, between former allies Uganda and Rwanda. The UN Security Council demanded that Ugandan and Rwandan forces withdraw from the DRC (resolution 1304, 16 June 2000). It also demanded the creation of an expert panel to examine “the illegal exploitation of natural resources and other forms of wealth of the DRC”. The panel was formed in June 2000 and in April 2001 it published its first report, highlighting the “pillage” occurring in the east of the DRC at the hands of Ugandan and Rwandan groups, as well as other military, political and commercial networks.

Laurent Désiré Kabila was assassinated on 16 January 2001. He was immediately replaced by his son, Joseph Kabila, in circumstances that remain shadowy. Joseph Kabila denounced the previous agreement on ending the crisis, making the early departure of foreign troops a condition for continued Congolese acceptance of its undertakings.G. de Villers, République démocratique du Congo. De la guerre aux élections. L’ascension de Joseph Kabila et la naissance de la Troisième République (janvier 2001-août 2008), Royal Museum for Central Africa (Africa Tervuren), L’Harmattan, 2009, p. 38-41.The international community then undertook to put an end to the “unresolved turmoil”, described by some actors as the First African World War.Term used by Madeleine Albright in her opening remarks to the UN Security Council meeting on the DRC, New York, 24 January 2000, cf. T. Trefon, Réformes au Congo : attentes et désillusions, L’Harmattan, 2009, p. 19.The idea of sending a UN mission, approved in November 1999 under the terms of Security Council resolution 1279, was confirmed with the resumption of talks. Once Joseph Kabila had greenlighted the United Nations Organization Mission in the DRC (MONUC) deployment plan, it was put into effect as of March 2001. The mission’s medium-term objective was to obtain the “full and definitive withdrawal of all foreign troops from the DRC” (UN Security Council, resolution 1341, 22 February 2001).

On 17 December 2002, the parties to the inter-Congolese negotiations in Pretoria signed a “global and inclusive agreement” concerning the transition period. Crucially, the agreement heralded the adoption of a power-sharing formula known as 1+4: the head of State was required to work with 4 vice-presidents. The agreement also demanded “the organisation of free and transparent elections”. The elections on 6 December 2006 signalled the end of the transition period, and Joseph Kabila was subsequently elected president after the second round of voting, garnering 58% of votes compared to 42% for Jean-Pierre Bemba. There were 25.4 million registered voters, with a turnout of 65.4% for the second round (29 October 2006).

1.2. THE INTERNATIONAL COMMUNITY IN THE DRC

1.2.1. CIRCUMSCRIBED SOVEREIGNTY AND PARTIAL TRUSTEESHIP

Created in November 1999, initially as an observer mission and subsequently entrusted with a peacekeeping mission, MONUC had troops on the ground as of March 2001; it was operating in a country beset by both internal and regional fighting. Following the official withdrawal of Rwandan and Ugandan troops in 2002, stabilisation efforts were concentrated on the east of the country (in particular Kivu, Ituri, Maniema, and the city of Kisangani) and on setting up new political and socio-economic systems. MONUC was granted more extensive prerogatives as the transition progressed. The conditions for the use of force and contingents’ rules of engagement were redefined at a time when the mission was tasked with significant new roles covering the reorganisation of the army and police, political structures, the organisation of elections and support to humanitarian actions. In October 2004 (Security Council resolution 1565), its mandate was altered and MONUC became an integrated mission: headed by a Special Representative of the Secretary-General, all its actions were coordinated with those of other specialist UN bodies (UNDP, UNHCR, UNICEF, OCHA, etc.). In this way, the UN played a part in establishing a “regime of partial trusteeship, with limited and controlled sovereignty” G. de Villers, République démocratique du Congo. De la guerre aux élections, 2009, p. 227. in the DRC.

Another component of this regime based on partial trusteeship was the International Committee in Support of the Transition (CIAT), headed by the Special Representative of the Secretary-General and whose members comprised foreign ambassadors, representatives from MONUC, the EU and the OAU. Under the terms of the global and inclusive agreement (Pretoria, December 2002), the CIAT was defined as one of the guarantors of the transition, and it was to “arbitrate and make a decision in any disagreement that may arise between the Parties” to the Pretoria agreement. This power placed the committee at the heart of the political transition mechanism.

1.2.2. “PUTTING THE STATE BACK IN THE DRIVING SEAT”

This expression is used by Théodore Trefon as well as in Congolese government documents, such as Programme minimum de partenariat pour la transition et la relance en DRC, 17 May 2004.

Ever since the second to last decade of the Mobutu regime, observers had been vying to outdo each other in their descriptions of the dysfunctional Congolese State. Theodore Trefon takes issue with this view: he maintains that such atti- tudes speak of a lack of understanding of the “journey” taken by the post-colonial State, which does not correspond to the model of the western State ruled by law.T. Trefon, Parcours administratifs dans un Etat en faillite. Récits populaires de Lubumbashi (DRC), Africa Tervuren, L’Harmattan (coll. Cahiers africains), 2007.It is true that the various reports and institutional papers usually only point to a causal link between the emergence of the conflict and the collapse of the State,K. Ballentine and H. Nitzschke, Profiting from Peace: Managing the Resource Dimension of Civil War, Boulder, Lynne Rienner, 2005.or cite the final decades of the Mobutu era in order to describe a journey wherein they recognise the characteristics of a failing State, with the war acting as an “accelerator”.European Union, Cooperation Strategy 9th EDF 2003-2007, 2003, p.10.This perception, using terms such as chaos, breakup, bankruptcy and collapse, fails to acknowledge that the Congolese State in fact showed a remarkable capacity for resilience. Years of internal conflicts, international wars and foreign occupation of part of the country never called into doubt the existence of a unified and legitimate State.T. Raeymaekers and K. Vlassenroot, Reshaping Congolese Statehood in the midst of Crisis and Transition, in U. Engel and P. Nugent (ed.), Respacing Africa, Boston, Leiden, Brill, 2009, p. 137.

Timothy Raeymaekers and Koen VlassenrootT. Raeymaekers and K. Vlassenroot, op. cit.believe that the strategy of political power-sharingThis strategy governed the quest for a “global and inclusive agreement” embracing all actors in Congolese political life.enabled the emergence of a new State model: during the transition period the State seemed to have lost its Leviathan role for all time, ceding its place to alliances and regulatory systems agreed by economic actors, parties to the conflict, local authorities and public agencies. Yet this seeming displacement of State sovereignty into the private sphere was not a linear process. The emergence of new parallel economic and political terrains within which the State is but one actor amongst others served to weaken the peace process. However, in a context of neither war nor peace, these governable zones made it possible to keep negotiations open between the State and the many other actors present on the ground. Most importantly, they played a part in sustaining the sense of being members of a nation-State within an environment marked by widespread cronyism.The authors cite the example of the “coercive governance structure” that emerged during the transition period in the North Kivu’s Walikale territory, with adoption of a system of institutionalised political and economic regulations governing exploitation of resources. In this specific governance context, the military, territorial government and economic actors exploited minerals for their own ends and produced a complex governance system by redistributing resources using an institutionalised taxation system.

The international community’s goal for the DRC consisted, for some, in extricating it from a “conflict trap”,World Bank researcher Paul Collier, specialist in the study of strategies used in post-conflict situations, coined the term “conflict trap”. Aid must focus on the real causes of conflict if it is not to break out again. whilst others wanted it to break free of a “vicious circle” that maintained the country in a state of chronic conflict and poverty. Political changes and the prospect that they could lead to a way out from the crisis were thought of as being a “window of opportunity”Many of the reports produced by international organisations working in DRC used the notion of a “window of opportunity”.that needed seizing: this is the reason why the political priority for international action at the time was holding “free and transparent” elections, along with maintaining peace and the withdrawal of foreign forces.

1.3. THE CONGOLESE HEALTH SYSTEM: AT THE CROSSROADS BETWEEN WAR AND PEACE

1.3.1. THE CONGOLESE HEALTH SYSTEM

Efforts were made to plan and structure a health system in Zaire from the 1960s. The main inspirations behind this effort were the community medicine projects run in the districts of Bwamanda (Equateur), Kisantu (Bas-Congo) and Kasango (Maniema) as well as theoretical work undertaken by Christian networks (protestant and catholic) under the patronage of the Ministry of Health. These multiple influences played a part in the organisation of a health system based on decentralised territorial entities where private actors and the population at large both contributed to making the system work: health zones.

The Ministry of Health had been setting national health strategies since the late 1980s. The head doctors for each zone, backed by their administrative staff, were in charge of implementing it in each zone. Despite enjoying considerable autonomy, the zone managers were subject to two-tier oversight from the district and provincial medical inspectors.

The zones comprised primary health centres run by professional paramedicals who provided the first level of care via consultations and primary care services. In cases where hospitalisation was required, patients were cared for in local health centres that offered better levels of staffing and equipment. Where there were complications or special requirements, patients were referred to the zone’s hospital, which should be able to offer a wider range of treatments.

The presence of health providers at health centres and at the zone, district and provincial levels meant that the state of open warfare (1996-1997, 1998-2002) and internal conflict after 2002 (especially in Katanga) did not lead to a generalised breakdown of care provision. Indeed, the cause of the gradual decline in public health services can be traced to disengagement by the State that began well before the 1996-2002 period. The system of health zones doubtless contributed to preserving a minimum level of provision as it made it easier for external actors to intervene in a context of highly decentralised funding. The health zones are now at the heart of the health strategy adopted by donor bodies and the national government.

Actors involved in the transition determined and implemented a policy for the reconstruction of the health system: the policy was founded on the government’s adoption in 2005 of a national strategy for reinforcing the health system, whose implementation was in turn supported by funding support programmes run by the main financial donors. Spending was concentrated on health zones with populations in excess of 100,000, judged to offer good prospects and that retained sufficiently operational infrastructures. The “spillover” policy in these zones focused on supporting health centres that remained functional, with the idea that they served as models for others.

The stated priority of the national strategy was reconstruction of the health system, with the risk of not having any immediate impact on health indicators. The emphasis was on extending coverage and a considerable portion of the funds were spent on setting up a system for coordination at the central and provincial levels. Dovetailing with indicators set by the Millennium Development Goals was not prioritised for the first few years.

As well as rehabilitating structures and supporting their functioning, aid actors undertook to deliver a minimum package of activities (MPA) at health centres and a complementary package of activities (CPA) at hospitals. MPAs comprised curative minor surgery, consultations, care for chronic STDs, preventive activities such as pre- and post-natal consultations and immunisations, as well as programmes to promote good health practices. The package concept aimed over time to make it possible to eliminate vertical approaches that offer funding for specific pathologies (TB, HIV, etc.). Donor contributions resulted in patients paying around 15% less than they otherwise would. The EU sought to establish a fixed price service based on the average income of the population concerned. Donors active in this governmental policy were involved in healthcare at the provincial level. NGO partners and those qualified to set up projects were responsible for implementing the strategy within health zones.

Support for Katanga, for instance, came from a number of sources: the World Bank’s Health Sector Rehabilitation Support Project (HSRSP) and Emergency Multisector Rehabilitation and Reconstruction Project (EMRRP) provide guaranteed support up until 2011. NGOs running HSRSPs have a contract with the Bank and undertake to deliver the minimum health services package in the selected districts and zones, support malaria-fighting efforts and help build monitoring and management capacity at all levels of the health system. Katanga province was also a USAID priority up until 2009. As part of its AXxes (Integrated Health Services Project) programme, USAID contracts partner NGOs and asks them to establish elements pre-selected as part of the World Bank project within a given zone.

1.3.2. PUBLIC HEALTH AND THE INTERNATIONAL COMMUNITY

Public health was already a major component of international policies in the 1990s, well before the return of international community cooperation in 2002. When aid ceased being provided to the Mobuto regime, the European Commission made public health the core component of its decentralised cooperation during the 1992-2002 period. A transition support programme (PATS 1 and PATS 2) worth €54 million was designed to provide financial and material support to health zones.The support programme was, however, subject to repeated interruptions linked to the war and changes in geographic coverage. Cf. European Commission, Cooperation Strategy 9th EDF, 2003-2007.The return of major donors to the country announced significant new expenditure: the World Bank became involved in 2002 with a 44 million dollar payment made via its EMRRP programme.The Emergency Multisector Rehabilitation and Reconstruction Project (EMRRP), set up in 2002 by the World Bank, covered a number of sectors. Its primary objectives were infrastructure rehabilitation (74%), support for social services (23%) and fostering “community development”. The project’s stated aims were to provide health services, rebuild infrastructure and support sub-projects aiming to “strengthen the capacity of local communities”. The programme was wound up in March 2010, having provided project funding equivalent to €700 million from an initial target of €1.7 billion. AS far as health services were concerned, the HSRSP and EMRPP programmes focused their efforts on administrative and logistical reorganisation and rehabilitation of the health zones and intermediary decision centres (provincial administration).This funding was supplemented in 1995 by a HSRSP project worth a further 150 million dollars. This sector is also one of the three focal sectors for European Commission development funding covering the transition and post-transition periods.The focal sectors are the areas prioritised for the Commission’s rehabilitation, reconstruction and development programme.

Published literature on post-conflict management describes the political aims of investment in the health sector: these reflect donors’ desire to coordinate emergency actions with development measures. The move from crisis to reconstruction also offers the assurance of making the benefits of the interventions rapidly visible, thanks to the Quick Impact Projects (QIP) funding mechanism. The effect, and thus the value, of health sector funding is to consolidate the autonomy of the State in the medium term and, in the short term, to enhance its popular legitimacy by delivering immediately available services. Furthermore, donors view health as being a neutral public good that plays a role in bringing peace to communities previously in conflict: according to an ECHO representative in the DRC, even if these programmes did not improve the population’s health indicators during the war, they did make it possible to get into the war zones and, by kickstarting collective structures, played a part in bringing fighting to an end.Conference given by an ECHO representative in DRC during the transition period, as part of the MSF-organised symposium on rebuilding the health system in DRC.

The technical justifications proffered in post-conflict literature regularly mention the motives that lead international actors to invest in public health. According to Béatrice Pouligny, beyond the move from war to peace or increased security, what is at stake is redefining a social contract. Spending on social programmes is a way of resolving the political contradictions inherent in modern State-building policies. Whilst playing a role in importing a “turnkey State model”, UN leaders also attempt to use specific policies (especially the holding of elections) to manufacture a popular legitimacy and to “encourage a measure of renewed commitment to living together.”R. Caplan, B. Pouligny, “Histoire et contradictions du state building”, Critique internationale, 28, July-September 2005.

1.3.3. THE “HUMANITARIAN CRISIS”: MORTALITY AND CONFLICT

At the end of the 1990s, faced with a lack of quantified data and difficulties in measuring the extent of the “humanitarian crisis”, a number of bodies conducted retrospective mortality surveys in various provinces of the DRC. In 2000, International Rescue Committee (IRC), an NGO operating public health missions in South Kivu since 1996, carried out epidemiological surveys of five sites: Kisangani (Orientale), Moba (Katanga) and three sites in South Kivu. Based on these surveys, and assuming them to be representative of the 5 provinces in eastern Congo (South Kivu, North Kivu, Maniema, Katanga, and Orientale), IRC estimated that mortality caused by the conflict in these provinces could be estimated at 1.7 million people over the 22 months preceding the survey (Mortality in Eastern DRC. Results from Five Mortality Surveys, IRC, May 2000).

In 2004, IRC decided to carry out a fourth retrospective mortality survey so as to be able to extrapolate the data gathered to the nationwide level, polling 19,500 households across the country (Mortality in the DRC: Results from a Nationwide Survey. Conducted April-July 2004, IRC). IRC concluded that the net mortality attributable to the conflict (for the period August 1998-2004) was 2.1 per 1000, i.e. 3.9 million deaths. However, the survey also concluded that most were not the direct consequence of acts of violence, which accounted for less than 2% of deaths. A fifth IRC survey from May to July running in 11 DRC provinces made it clear that the “humanitarian crisis” was ongoing despite the official end of hostilities in 2002: IRC estimated that 2.1 million deaths related to the conflict occurred during the years 2002 to 2007. From August 1998 to 2007, according to IRC, a total of 5.4 million deaths can be attributed to situations provoked by the war and its aftermath as well as to internal conflicts (Mortality in the Democratic Republic of Congo. An ongoing Crisis, IRC, 2007).

The conclusions of these epidemiological surveys led IRC to label the situation in DRC as “the world’s most deadly [conflict] since the end of World War.”B. Coghlan, R. J. Brennan et al., Mortality in the Democratic Republic of Congo: a nationwide survey, The Lancet, 367 (9504), 7 January 2006.

This position was used to justify the organisation’s operational choices: IRC retargeted its emergency programmes towards providing support for the public health system and for reconstruction. Backed by the estimate that 98% of conflictrelated deaths were caused by the destructive and pillaging acts of armed groups, forced displacements and precarious living conditions, in 2004 it adopted a strategy of providing primary health support.Statement by Loïc Aubry, a public health specialist with IRC, MDM symposium on rebuilding the DRC public health system, June 2006.IRC also lobbied for MONUC to provide civilians with better protection.

In 2001 and 2005, MSF-Belgium also carried out retrospective surveys that produced data similar to the IRC: these highlighted the excess mortality in zones afflicted by conflict (MSF, Violence in the Congo and access to care (DRC). Results of five epidemiological surveys, Brussels, December 2001). Acting on these observations, the Belgian section refocused on frontline zones and transformed the nature of its mission: from programmes mainly intended to support the health system, it switched to emergency projects to reach out to populations impacted by the conflict. MSF-Belgium did, however, feel that it was not possible to extrapolate its results to include the whole country. Moreover, war was not the sole factor accounting for the state of the public health system, excessive mortality rate and difficulties in accessing care. For example, mortality rates in 2005 in provinces far away from any fighting remained extremely high. The 2005 report published by MSF-Belgium compares mortality rates for 2001 in the health zones examined with those in conflict-impacted health zones where MSG was present: mortality rates rose markedly, from 1.1 in 2001 to 1.8 in the case of Kilwa (Katanga), and they soared in health zones far away from the conflict, as was the case in Inongo, where the rate rose from 0.4 in 2001 to 2.2 in 2005.MSF, Access to care, mortality and violence in DRC. Results of 5 epidemiological surveys: Kilwa, Inongo, Basankusu, Lubutu and Bunkeya. March to May 2005, Brussels, October 2005, p. 21.

The IRC surveys led to optimistic claims concerning the potential role of epidemiology in conflict prevention. A paper published in The Lancet (2006), for example, claimed that “We can no longer claim ignorance about this [situation […]. In this sense, mortality studies could play a pre-emptive role to provide further justification for peace initiatives when conflicts threaten to break out.”E. Depoortere, F. Checchi, “Pre-emptive war epidemiology: lessons from the Democratic Republic of Congo”, The Lancet, 367 (9504), 7 January 2006.The fact remains, however, that the methodology employed in the IRC surveys and the resulting estimated mortality rates remain controversial. CriticsCf. HNTS (Health and Nutrition Tracking Service), Re-examining mortality from the conflict in the Democratic Republic of Congo, 1998-2006, WHO, 2009.contest the robustness of the assumptions underpinning the extrapolations made by IRC epidemiologists in 2001, point to the fragility of data surrounding the cause of death and the number of deaths per household, and stress the difficulty of assessing high mortality caused by the conflict itself in the absence of any credible demographic data describing the population. Yet there can be no doubt that the international attention focused on the IRC’s excessive mortality rate data did serve in a very real way to raise the alarm and call attention to the situation.

This first section has looked at the DRC background, initiatives and mechanisms from a national perspective. The following section homes in on local actions in Katanga. At the local level, it becomes clear that health system support policies had little impact in many sites where MSF operated. However, certain generalised characteristics can be applied to situations specific to Katanga: the dilapidated state of hospitals and health centres, the impoverishment of the population, the scale of the exactions and their continuation since 2002. Another point to highlight is the convergence between the analyses of governance specific to the DRC, as presented by Timothy Raeymaekers and Koen Vlassenroot,Cf. T. Raeymaekers and K. Vlassenroot, “Reshaping Congolese Statehood in the midst of Crisis and Transition”, in U. Engel and P. Nugent (ed.), Respacing Africa, Brill, Leiden, Boston, 2009. See also K. Vlassenroot, “Négocier et contester l’ordre public dans l’Est de la République démocratique du Congo”, in Politique africaine, 111, October 2008, p. 44-67.and the local-level governance that MSF missions engage with. Locally, including in the remotest villages of North Katanga, there are public actors (administrators, police officers, internal security agents, members of the armed forces) and public and private health system actors. Even in the most remote zones, emergency actions never take place in an institutional vacuum; in reality, it is necessary to deal with a number of power centres in order to be able to make decisions and take action. This point is borne out by the descriptions of MSF action in North Katanga.

2. MSF MISSIONS IN KATANGA. WHY STAY? HOW TO LEAVE?

2.1. WHY KATANGA?

2.1.1. MSF-FRANCE: JUSTIFICATION FOR ITS PRESENCE IN KATANGA

During the 2001-2002 period, MSF-France made efforts to move its operations closer to the areas of North Katanga impacted by violence. This was a choice designed to bring its actions in line with MSF’s guiding principle: priority must go to providing medical assistance to people living in critical conditions caused by conflicts. “Before going to Katanga, we were running projects by default as we weren’t managing to get into provinces impacted by the conflict, and these were the provinces where our working methods were most suited.”Interview with a former head of the DRC programme, Paris, January 2010.More than any other type of situation, open warfare provides a legitimacy for the way that the organisation works. And, in July 2007 when MSF judged that levels of violence by armed groups were under control, the decision was taken to close the mission in Katanga.

For MSF as well as the international community in general, the notions of war and post-war, although somewhat hazy, remain vital categories as they serve as justification for creating a “zone of exception” that provides legitimacy for international rescue missions and makes it possible to delimit them in space and time.

However, as we will see, the choice to enter and remain in Katanga was the result of ad hoc decision-making and compromises whose roots did not lie in the conflict alone. Throughout the mission, the context in Katanga placed the organisation at the heart of competing interests caused in part by the conflict, and in part by the long history of Mobuto’s regime and its effects on the health system. This mission forced MSF to negotiate a context for its work that lay at the crossroads of emergency, reconstruction and public health.

2.1.2. THE QUEST FOR THE FRONTLINE

References to the conflict are central to the decision-making process, and explain why MSF decided to intervene in Katanga in 2001, as attested to by internal documents of the time. Activity and annual plans justify MSF’s presence because of “the dichotomy between the policy of pacification and transition and the continued existence of extreme levels of violence in the east.”MSF-F, 2004 Annual Plan, p.152.The Annual Plan for 2005 similarly cites “the violence that persists in Kivu and North Katanga.”MSF-F, 2005 Annual Plan, p.119.

From the very first year of its engagement, MSF set in place a mechanism with a dual purpose: partly to offer care to the sick and those displaced by the violence, and partly to combat epidemics.

The mission in Katanga marked the return of MSF-France to areas of DRC affected by violence; MSF had been on the ground in 1996-1997 in Zaire when Rwandan refugees were being pursued.Cf. Laurence Binet, Traque et massacres des réfugiés rwandais au Zaïre-Congo 1996-1997, Paris, MSF, 2004.The decision is also viewed, with hindsight, as signalling a return to MSF’s founding mandate: priority to medical assistance close to zones of conflict. The fact is that the Congolese experience is a perfect illustration of the antagonisms that existed within MSF concerning its operational policy: some referred to the utter absence of health provision in order to make access to care the central problem to be faced in the country, whilst others disapproved of any strategies that involved playing a part in projects that did not relate directly to the conflict. For MSF France, the decision to enter North Katanga can also be explained by an operational desire to no longer have to deal with the difficulties encountered in setting up a health system support project in the hospital at Bandundu.

MSF-France and the Bandundu hospital

In early 2000, MSF was present in Bas Congo running an assistance programme for Angolan refugees in Kimpese (operational since 1998) and a centre for displaced Congolese at Sicotra, close to Kinshasa. The demands in terms of access to care throughout the region and the difficulties in reaching conflict zones led MSF to commit to a medium-term programme. In August 2000, the decision was taken to support the Bandundu health zones by setting up within the main hospital and providing assistance to three health centres. For MSF, opening a programme in a stable region was the sign of a new ambition: to work on structural pathologies (malaria, respiratory infections, diarrhoea) and to forge ties with actors in the public health system. For a time, MSF was considering using this programme as leverage to wage a campaign against cost recovery.Isabelle Mouniaman-Nara, MSF à Bandundu, janvier 2000-juillet 2002, internal report, MSF (13/10/2002).
The intervention came under severe pressure almost from the start. Negotiations on cost recovery failed; MSF’s salary offers were rejected and relationships with the hospital managers became embittered. The programme review cites as a major difficulty MSF’s inability to “come up with a suitable intervention project and positioning, its poor understanding of the health system and how it operates, as well as poor management of agreement and salary protocols.” With the parties unable to reach agreement, managers in Paris decided to pull out of the programme in 2002.

The failure of the Bandundu operation confirmed the belief, held by many, that this type of programme was not what MSF is about. Katanga, on the other hand, did meet with their approval, as it demanded traditional emergency response-style working methods.

The first exploratory missions to North Katanga in 2001 revealed a situation far removed from the initial concerns. The frontline was stable, and was actually more of a demarcation line between areas administered by Kinshasa and those to the east under the control of RCD-Goma (a party whose armed forces were backed by Rwanda).“From the Lusaka ceasefire agreement in mid-1999 until the end of 2002, frontlines between government forces and the Goma rebels were more or less stable. […], in total, over the entire period following the Lusaka agreement, military clashes between the main protagonists were rare”. G. de Villers, De la guerre aux élections, op. cit., p. 132. The RCD-Goma continued to control large sections of North Katanga after 2002.Although there remained pockets of displaced people, the overall context was closer to chronic crisis than generalised conflict. The first reports MSF issued spoke of a post-conflict situation,MSF-F exploratory missions, Malemba Nkulu and Kitenge health zones, Haut Lomami administrative district, Katanga province, August 2001.a forgotten crisis where the true causes of mortality were endemic in nature. This is underlined by the fact that the third mission was forced to suspend its explorations in order to provide support to an epidemic response programme. The desire to intervene was also driven by the state of the public health system in North Katanga and its relative isolation from international aid mechanisms. In the various reports, the theme of the forgotten crisis was more effective than references to a conflict in arguing for a programme to be opened.Describing the Kitenge and Malemba zones, the report concludes with the following operational recommendations: “both are contexts of extreme precariousness in socio-economic and health terms alike. In both instances, the situation is ‘post-conflict’ or ‘chronic post-emergency’ with massive displacement of civilian populations and health structures in disarray. In humanitarian aid terms, all of Kitenge and the Mai-Mai sub-zone in Malemba are forgotten and/or actor-free areas, remembering also that ACF has no medical capabilities in the FAC sub-zone and that no other actors have come forward to provide medical aid in Malemba.” August 2001, p.6.The situation changed, however, with the upsurge in fighting between the armed forces of the DRC (FARDC) and militias who had not disarmed after the end of the fighting with RCD-Goma forces and Rwanda. “In North Katanga, the Mai Mai phenomenon arose from the creation, at the instigation of the government in Kinshasa, of popular self-defence forces intended to resist the advance of the RCD and its Rwandan allies. After the end of the war in 1998, the Mai Mai movement in North Katanga […] remained one of the country’s primary sources of disturbance and insecurity.”G. de Villers, op. cit., p. 193.

2.1.3. MSF-BELGIUM: BACKGROUND AND DILEMMAS SURROUNDING THE DRC INTERVENTION

This paragraph draws on an MSF-Belgium policy paper from 2006 about its activities in the DRC.

The origins of MSF-Belgium’s presence in the then Zaire can be traced back to the arrival of 35,000 Angolan refugees in Katanga province (formerly Shaba) in 1985. In 1986, the original emergency response project to provide care to the refugees from Angola was expanded to embrace long-term projects for providing structural support to over twenty health zones, the creation of a local NGO, Horizon Santé (1994),This project was handed over to Fometro in 2001; MSF remains on the governing board.to ensure the project’s sustainability, and other vertical projects to tackle AIDS (1993) and trypanosomiasis (1996).

The number of emergency situations continued to grow as the health indicators continued to decline. To cite only the largest operations, 1994 saw the flight of one million Rwandan refugees in Kivu and the exodus of persecuted Kasaiens from Shaba (the former name of Katanga province), followed by the ebola epidemic in Kikwit (Bandundu) in 1995. These emergencies led to the creation of an emergency team in 1995, the PUC (Pool d’Urgence Congo); initially based in Kinshasa, it extended its reach to cover a large area of the national territory with regional satellites in Lubumbashi, Mbandaka and Kisangani. Other emergencies dealt with were violence against Rwandan refugees moving within the DRC in 1996-1997, cholera epidemics and open warfare (Kinshasa, Kisangani, Mbandaka and Katanga).

The Belgian section was divided in its goals between attempting to make up for the structural shortcomings of the health system and dealing with emergencies created by political and social factors. It based its work on the following premise: “there are two crises in Congo: the crisis of violence, which overlays the other crisis related to the country’s public health structures.”Interview with a manager from the Analysis and Advocacy Unit, MSF-B, January 2010.As Françoise Wuillaume wrote: “in the Congo, MSF (Belgium) is confronted with a timeless existential question: faced with the enormity of the country and its needs, how can accessibility to its projects best be ensured?”Discussion sur les interventions de MSF dans les Zones de Santé du Congo Démocratique – Rapport de visite, 08-28 July 2002, p. 5.

From the late 1980s onward, questions were asked within the Belgian section about the relevance of its emergency programmes for displaced populations, as its ad hoc actions had no impact on the stubbornly high mortality rates. Having identified the system’s structural deficit, it then started to work closely with the Ministry of Health and to provide support to isolated populations by setting up in referral hospitals and some local health centres. This first foray into cooperation focused on distributing medicines and providing in-patient and out-patient hospital care by a mixed team of MSF international personnel and ministry staff.This approach was initially described as a spillover strategy.

The withdrawal of international donors from the country in the 1990s and the suspension of international aid served to increase the pace of this new way of working: MSF-Belgium formalised this strategy as the “operational district” concept and extended its operational scope to include a large number of provinces. This extension went hand in hand with a reduction in the intensity of its programmes and an increase in indirect aid to districts: international teams ceased to provide care and restricted themselves to supervision. Cooperation occurred at every level: indirect support for the operation of health centres and hospitals, administrative support for zone and district offices and staff training. In Katanga, MSF-Belgium committed itself to involvement in the Kilwa, Pweto and Kasenga zones, as well as three further zones in the inner suburbs of Lubumbashi, the provincial capital.

Changes in the institutions and general context led to a gradual dilution of the district-based approach: the outbreak of warfare in 1998 forced MSF to respond to new emergencies and to diversify its programmes. The Ministry of Health gradually increased its control over districts managed by MSF by reclaiming its rights to appoint medical staff. Furthermore, the La Mancha process that the MSF movement as a whole committed to encouraged sections to focus their activities on providing treatments and medical interventions that centred on patient care.

The more deep-seated desire to “act to drive down mortality rates” in the immediate term was backed by mortality surveys run by the Belgian section from 1999 onwards: mortality rates were higher in areas where there was fighting and it became harder to identify the benefits of the district-based approach. From 2005 onwards, MSF-Belgium altered its initial strategy, turning instead to emergency actions in areas impacted by violence caused by armed groups. District support programmes were progressively wound down.

2.2. CONTINUING EMERGENCY: RELEVANCE AND LIMITATIONS OF MSF-FRANCE ACTIONS IN KATANGA

2.2.1. CHRONOLOGY OF MSF-FRANCE OPERATIONS IN KATANGA

Chronology

A feature of 2000 was the determination to enter Katanga province and move closer to the frontlines in eastern Congo. Acting on a request in Kinshasa from the Commissariat General for Professional Reintegration (CGR), the DRC’s official body for displaced persons, and in the light of its contacts with other international bodies active in the area, MSF-France decided to undertake an exploratory mission in North Katanga.
An initial mission in late December 2000 failed as the national security agency (ANR) cancelled all permits for foreigners to enter Katanga during the holiday period. In mid-January, a second attempt was halted following the assassination of President Kabila on 16 January.
A third exploratory mission was conducted by a doctor and a logistics specialist from MSF-France between 19 May and 9 June 2001. After further discussions with the CGR and international organisations present in Kinshasa (MSF-B, ACF USA, CICR, UNOCHA), MSF decided to immediately open a mission in the Kabalo health zone, as a cholera epidemic had broken out in the zone (Mulimi health area). The exploratory mission thus came to an abrupt halt before all the towns and villages had been evaluated. From January to June 2001, a series of six reports and missions were used to define MSF-France’s position.
Despite the greater vulnerability of the Malemba N’Kulu health zone to attacks by the Mai Mai, the focus was placed on the Kitenge zone, also plagued by militia violence. The aim was to have the programme operational before the November rainy season. However, the number of epidemic crises in the Malemba health zone forced MSF-France to come and go between the two health zones, as no decision had been made to commit extra resources. The development of its activities in Kitenge was delayed, and the intervention was not operational until January 2002. MSF-France’s operations were to continue up to 2007, moving from the north of the province toward the south as the seat of the violence moved.

2002-2005

Malemba N’Kulu health zone: MSF-France focused on providing care to people fleeing fighting between the DRC armed forces and the Mai Mai, seeking refuge from the exactions of both sides. The intervention provided outpatient services. The primary activity was fighting epidemics: in all, 6,000 cases of cholera were recorded by the four cholera centres; measles was also a significant problem, with 6,373 cases treated and 53,976 children inoculated. Kitenge health zone: MSF-France concentrated its activities on providing structural support to the referral health centre at Annuarite in Kitenge, as well as to village health centres at Kaloko and Kiléo. The main activity was outpatient consultations (120,000 consultations in three and a half years). Treatment was also provided for outbreaks of measles, malnutrition, typhoid and TB.

2003-2007

Mukanga health zone: following fighting between the Mai Mai and DRC armed forces, MSF-France offered structural support to the health centres at Mukubu, Mukanga, Kyolo and Kasenga: in all, 110,000 consultations were made and 2,000 malnourished children were cared for. Non-Food Items were distributed to 1,992 families.

2004-2007

Ankoro health zone: in 2002, fighting between the DRC armed forces and the Mai Mai in Ankoro forced MSF-France to intervene. The operation started during the final months of the year, with a mobile clinic set up and treating displaced people. The injured were cared for at the hospital in Ankoro. Faced with an epidemic of measles, vaccinations were administered to 24,761 children; 1,562 infected patients were treated.
Starting in 2004, work on setting up a secondary care project aimed at covering the injured and major complications seen by all projects led to MSF taking over management of the hospital at Ankoro, which was fully rehabilitated. Medical and surgical activities were restarted, and care was progressively offered for specific pathologies (TB, HIV, vesicovaginal fistula, malnutrition). Epidemics of typhoid in Kitanda (237 cases) then cholera (166 cases between December 2005 and March 2006) were also treated.

2006

Nyonga health area, lake Upemba: operations by the DRC armed forces to wipe out armed groups led to major population movements. After rehabilitating several health centres, MSF-France handled 20,937 consultations, 582 hospitalisations and 160 births. Obstetric cases were seen at Kikondja, and surgical cases at Ankoro. Non-Food Items were distributed to 473 families. Considerable anti-epidemic work was also done owing to the high endemicity of vibrio along the river banks.
From January to October 2006, the emergency desk dealt with an outbreak of cholera in the Kikondja health zone. Following the detection of cases of measles in the Nyonga region (Butumba and Mukanga), a care plan was devised to prevent it spreading and reduce its lethalness.For details on the emergencies in Katanga in 2006, see Fabrice Resongles, Revue critique des urgences 2006, Paris, MSF, p. 43-46.

2007

Kabondo Dianda health zone: MSF-France intervened in response to a measles epidemic, with 909 cases treated and 73,641 children inoculated. MSF-France also dealt with a cholera outbreak, treating a total of 636 cases.
MSF-France withdrew from Katanga during the first half of 2007. 2008
In early 2008, MSF-France acted in support of the Belgian section during a cholera outbreak in Likassi, in southern Katanga.
In June 2008, MSF-France was active in Tanganyika district in north-east Katanga, acting on a request from MSF-B and the Congo Emergency Team for it to lead a mass vaccination drive in a number of health zones. MSF-France limited its campaign to a single zone. The campaign got underway two months behind schedule, in September 2008.For more information about this intervention, see Delphine Chedorge, Revue critique des urgences 2007-2008, Paris, MSF, June 2009, p. 40.

The various exploratory missions recommended that an emergency response mechanism be put in place. Their conclusions were only partially in agreement with the initial analyses behind deployment of the missions, but the rapid rise in cholera epidemics and the extreme fragility of displaced people constituted solid reasons for intervening in a province where international assistance had very little impact in comparison to Kivu and Orientale provinces, at least between 2001 and 2006, and particularly North Katanga.

The situation in Katanga is worth examining as it offers an illustration of the gap between reality and the situation typologies applied by most international agencies. Three elements, sometimes converging and sometimes conflicting, shaped the complexity and uniqueness of the MSF-France teams’ working environment over the years. The internal conflict between the DRC armed forces and various local militias, latent during the early months, worsened as of 2001. Aside from sporadic outbreaks of extreme violence, teams noted the population’s poor general health as a result of failures in the public health system, which had deteriorated since the Mobutu era. Finally, the geographical position of the site created logistical problems that remain vivid in many peoples’ mindsEverybody interviewed emphasised the scale and difficulties of the logistics efforts required by the North Katanga missions.and served to forge the mission’s reputation.

The unpredictable bouts of pillaging and violence against civilians, difficulties in reaching the populations concerned, demands from people under care for the provision of “structural” care are all realities that have to be taken onboard when characterising the context. It cannot simply be defined as a zone of exception, a conflict zone within which the legitimacy of MSF-France’s interventions cannot be challenged. The reality is that MSF-F was in a process of constant negotiation with itself and its environment, seeking to preserve a space for its interventions at the crossroads of emergency, development and public health provision.

2.2.2. KATANGA: HALFWAY BETWEEN AN EXCEPTIONAL AND AN ORDINARY SITUATION

Emergency organisations generally take the view that the time to pull out is once warfare has ended. This is often a difficult decision to take, and it usually arises once traditional definitions of a zone of exception no longer apply: population movements decrease, violence abates, the State recovers its prerogatives. Needs and context evolve. Emergency aid organisations, which cite international humanitarian law when intervening during periods of conflict, feel that the post-war situation does not provide their interventions with sufficient operational or legal legitimacy. However, as Rony Brauman points out, “the line between an ordinary and an exceptional situation is not one that can easily be drawn. The end of a conflict does not signal an end to its consequences […].”R. Brauman, La médecine humanitaire, Paris, PUF, coll. Que sais-je ?, 2009, p.107.

The experience in Katanga shows the problematic nature of the notion of “exceptional”, the keystone of MSF’s operational methods and legitimacy. Before MSF-France entered Katanga, there was fighting in the north of the province, but the conflict that pitted the DRC against Rwanda and its RCD-Goma allies was fairly stable by 2002. Regular exactions against civilians started occurring in 2001; these were the work of militia as well as DRC armed forces. However events such as these were not widespread throughout North Katanga, and they did not appear to be coordinated within this part of the province. Population movement in response to attacks and the actions of the armies did not lead to the creation of major displaced person camps or of any specific social structures: “displaced people living in neighbouring villages are often mixed in with the general population.”Interview with a former medical manager for the Katanga mission. The 2003 medical report also mentions the following difficulty: “in this zone as elsewhere in North Katanga, the status as a displaced or needy person is fairly common, as a major section of the population has been on the move for a varying length of time”, MSF-France, 2003 Medical Activity Report, p.13.

In addition, the war had not destroyed the public health and administrative structure in the area, although they were structurally weakened by a fall-off in funding since the second half of the Mobutu era. Its meagre resources were not the result of a state of exception related to the conflict, but were a historical given. The State had not disappeared during the conflict. Its structures remained in place in the form of localised governance that was restrictive and repressive. An MSF-France mission manager puts it this way: “we never considered acting outside the provincial health structures as the bureaucracy was so stifling.”Interview with an MSF manager, Paris, January 2010.

This was a paradoxical form of authority, providing administrative continuity that was only partially in contact with central government, and bereft of any of the resources needed to provide a meaningful public health service: “the situation isn’t unique to Katanga. You could apply it to the entire Congo, but it’s not so extreme in other countries where we’re working. Wherever you go in Congo you’ll find health centres open, staffed with public servants and a still-functioning organisation, but with absolutely no drugs available.”Interview with an MSF manager, Paris, December 2009.

Thus, the violence in 2002 was all too real, but did not fit the expected pattern of a state of exception. In the face of sporadic localised violence, as well as the catastrophic health system, MSF-F progressively began to set up actions that suited the realities on the ground, negotiating an operational space at the crossroads of the exceptional and the ordinary.

The first concession to the exceptional situation was that MSF-F treated the public health authorities as a partner of first choice, acting as far as possible to place its actions within the context of the public health structures. This policy, which was not much debated at the time, seems all the more surprising given that the Congolese State was party to the conflict via its armed forces, which were responsible for major exactions committed against the civilian populations in Katanga.

In reality, relations between MSF-F and the public health authorities were far from smooth. They swung between a desire to make a partial break from bureaucratic restrictions in order to deliver optimal assistance, and the need to adapt in order to work with the administrative apparatus that continued to organise health actions in the areas where MSF-F worked. “To be honest, I don’t think that we ever thought about setting up a structure outside those operated by the health ministry.”Interview with a former MSF manager for the DRC, January 2010.

This ambiguous relationship led MSF-F into negotiations to preserve its objectives and working space; a space that was forever questioned and readjusted, and maintained at the cost of inconsistencies and concessions that at times ran counter to how MSF-F perceived missions in a conflict zone. Hence, building an autonomous working space to meet ad hoc emergencies, in an exceptional situation, required negotiating, and on occasion making concessions over the usual operating framework. A number of recurring elements illustrate this: the policy of consistently offering 100% free care in conflict zones was put under

- severe strain; MSF eventually conceded in certain zones, such as Kitenge,In Kitenge, the health centre where MSF operated belonged to the diocesan works office, a religious body firmly opposed to free care. Negotiations about cost recovery and wages for the centre’s staff spread over several months and hindered activities within the health zone. Interview with a former doctor to the mission, Paris, January 2010.

Mukubu and Ankoro, allowing the maintenance of a cost recovery system limiting access to care.Studies carried out by a field manager in Ankoro, based on questionnaires returned by the local population, show that disappearance of currency meant that even households that were not at all destitute were unable to access care. The 2003 annual medical report notes increased activity at the health centre in Kitenge, once cost recovery had been abandoned. After lengthy negotiations, MSF managed to abolish this system in Kitenge and Ankoro;

- aid practices also had to live side-by-side with public health requirements and protocols established by public health authorities: attitudes to vaccination are a good reflection of these tensions. On a number of occasions, MSF-F decided to set up a policy of systematic vaccination of displaced populations, in accordance with its own health protocols.According to MSF’s guide to refugee health in emergency situations, immunisation against measles is the second highest priority, immediately after assessing needs and the general situation. Vaccinations can be selective (with prior assessment of individual status) or non-selective (no assessment of individual status). Even where public health authorities are involved in the immunisation programme, MSF recommends non-selective vaccination (cf. Médecins Sans Frontières, Refugee Health, an approach to emergency situations, 2008, p. 59).But this policy ran contrary to that of the provincial public health authorities, which attacked MSF-F for acting outside the health resources plan by limiting vaccination to areas where refugees were living. By making ad hoc adjustments to vaccination rates in limited areas, the MSF-F teams were criticised for upsetting national arrangements. In fact, the risk was that raising vaccination rates in certain zones would lead to the halting of WHO funding for Expanded Immunisation Programmes (EIP).This question was raised in interviews on a number of occasions by actors from MSF (Paris, 2009, 2010).This question is all the more critical as it is often difficult to tell the difference between refugee and local populations. In order to safeguard its working space, MSF-F was on a number of occasions obliged to alter the way it worked to suit choices and directives coming from the public health authorities. In this way, MSF-F obtained permission to vaccinate, but on condition that it expanded its immunisation policy to include all antigens recommended under the terms of the national protocols. During another emergency situation, it could only obtain authorisation to proceed by expanding its vaccination campaign to include zones unaffected by armed groups.Interview with medial and non-medical managers in the DRC, Paris, January 2010.

The health resources plan and its ramifications were also the topic of negotiations. MSF-F relied on the existing system’s human and structural resources, but it made its investments according to objectives that it defined and pursued alone. These objectives occasionally ran counter to the authority’s demands in terms of health coverage. Sometimes an ad hoc intervention by MSF-F occurred outside the established referral system, and its support for certain centres instead of others meant that the health resources plan had to be altered. But it was a strategy that survived thanks to the provision of intermittent aid to structures that were disadvantaged by MSF-F’s activities. The organisation offered redress in a range of ways, such as support for national vaccination campaigns and or the construction of buildings for the disadvantaged structures.Interview with a former member of the board of governors (Paris): in 2002, MSF operated at a health centre run by the office of central diocesan works in Kitenge. As a way of recompensing the public sector, which was disadvantaged by this choice, shortly before leaving Kitenge the organisation financed construction of buildings to be used in the creation of a public referral health centre.These negotiations did not always prove sufficient to free MSF-F from the problems of dealing with the province’s administrative and health arrangements: setting up a referral mechanism alongside the ordinary mechanism for providing care for medical complications was not welcomed by managers where they felt this would disadvantage their health zones.

ANKORO

MSF’s intervention in Ankoro dates back almost to the beginning of the mission to the province. In December 2002, MSF decided to support the referral hospital and a number of health structures in outlying villages over a two-month period.
In 2003, MSF managers decided to take charge of managing Ankoro hospital for the treatment of severe illness and injury. This outlay was justified by the hospital’s “ideal” location, “equidistant from the two zones most effected by the violence and accessible from the RCD zone.”MSF project files, 2003, p. 5. Part of North Katanga was under the control of the government appointed by the RCD-Goma; the other part, where MSF-F operated, answered to the Kinshasa government. As Ankoro lies on the river that served as a demarcation line, the hospital was accessible to people from both sides.The objectives were ambitious: the hospital was to be totally rehabilitated (electricity, roof, water supply, equipment, etc), with MSF assisting with administration, staff training and setting up a system for charging a flat rate for healthcare. The project had twin goals: to provide local care to the people of Ankoro, and to act as a referral hospital for medical-surgical emergencies from neighbouring health zones.
The results were mixed by the time the programme closed in 2007. The hospital had kick-started quality and accessible primary care, but it was not possible to systematically achieve the goal of making it a referral structure for secondary care. Logistical difficulties coupled with administrative reticence prevented the referrals system from succeeding.
In a context of neither war nor peace, but of chronic violence against civilians, the Ankoro was an operational choice, signifying MSF’s desire to develop and strengthen secondary care in a context where such a choice was neither usual or easy.

The project at the Ankoro hospital is very revealing in that it illustrates MSF’s difficulties in reaching agreements with the numerous different local authorities, both medical and governmental, whose approval it needed: these are not one-off negotiations, but an on-going dialogue that has to be permanently nurtured. Intended as the secondary care referral structure for all zones surrounding Ankoro, the hospital never truly fulfilled this mission owing to logistical difficulties of accessing it, patient reluctance to be transferred when required to travel along unmade roads, and a lack of enthusiasm by Head doctors to refer patients to a location outside the area they administered. MSF-F’s presence in a given area represented a coveted financial and symbolic resource. Its action upset political and social balances and created jealousy between administrations, which rebounded on the organisation’s activities.Interview with MSF managers formerly in charge of the DRC, Paris, January 2010.

2.2.3. AN EMERGENCY AFTER ALL?

It is generally difficult to decide exactly where the line lies between exceptional situations and ordinary situations, and this applies equally to the situation in North Katanga. MSF dealt with the exceptional and the ordinary in parallel, emergencies relating to armed conflict, destruction, exactions and the chronic inadequacies of the health system. This situation generated two types of difficulty for the teams in the field:

- the reinvigorated structures and new prospects made possible by MSF-F’s interventions in the health system led to a rapid upsurge in activity at health centres where MSF-F treated both refugee and local populations alike. The high incidence of infectious pathologies required that the associated comorbidities also be treated, enlarging the scope of the medical care offered. TB treatments are a good illustration of the difficulties encountered: medical actors in the field want to respond as the numbers of cases rise. But the nature of the treatment, the major risk of resistance if treatments are not followed, led some managers to follow WHO recommendations and national protocols, refusing to enlarge the scope of intervention. Attitudes to TB, which was not at the time a reason for emergency intervention by MSF-F, varied amongst practitioners, but this did not stop treatment being offered at Kitenge and other projects, such as Ankoro;

- in operational terms, emergency response requires a complex mechanism to be put in place. Emergencies arose when villages were attacked then subject to endless exactions, all within zones that were themselves enclaves; MSF-F was therefore required to commit considerable mobile logistics resources. At the same time, the weakness of the health system justified the considerable sums that MSF-F committed to the referral structures it managed. This twin-track approach cost a lot of money and led to questions being asked within the organisation about the direction of its action and objectives. Dealing with emergencies is a process that “requires considerable resources but all too often only involves reaching small pockets of the population, sometimes several days after the crisis has ended.”Interview with a former field manager, Paris, January 2010. During the course of 2004, MSF-F attempted to reach displaced populations in the Ankoro health zone. Logistical obstacles meant that MSF-F only reached them once the situation had stabilised. The intended NFI distribution then became utterly pointless.Some field workers found these situations frustrating: “we go in and support the population, we help them through the crisis period until the time comes when they are able to resume the same miserable lives that they had before being displaced. I feel it’s like we say to them, well, you’re in the shit, and when the shit rises above your head, we’ll help you breath. When you manage to get you head above the surface we’ll leave... but you’ll still be in the shit! That sums up the chronic emergencies of the situation in Katanga!”Internal MSF correspondence between managers, MSF, 2005.These ad hoc responses contrasted with the major workload going on in the health centres. They were meeting fast-growing demands for treatment, yet this local healthcare mission brought into focus the competing antagonisms. Secondary for some, “chronic emergency”2003 Medical Report.for others, the mission was felt to be legitimate only because the violence, and hence the emergencies, persisted. The emergency situation explains the stopping and starting of medical support at health centres. However, the notion of emergency is not easy to grasp, different interpretations and uses of it are possible, and it can therefore justify various operational choices; how to decide between them when faced with a situation where armed violence could break out at any moment, dissipate, then settle in a particular area without directly impacting the neighbouring areas, but leading to a movement of refugees into them, for instance, into the urban centres at Ankoro, Kitenge and Malemba N’kulu? And then there were new outbreaks of violence, new emergencies, and the same process began again, and again responses had to be found.

The unstable nature of the line between an exceptional situation and an ordinary one makes taking decisions more complicated, particularly the decision to close a programme. On a number of occasions, plans to pull out were postponed because of the reappearance of sporadic fighting and the reservations expressed by teams in the field. For instance, the proposed closure of the Kitenge centre in 2004 was hotly disputed: the on-site team felt that the increase in activity since 2003, the local population’s precarious health situation and the agreement, after long negotiations, to provide free care all justified maintaining the medical activities and opposing the position of the operational managers in Paris. “The announcement of the closure of the Kitenge centre went down badly with the field teams. They accused us of being out of touch with reality, of having an overly-theoretical conception of what constituted an emergency, one that they did not share. They felt the Kitenge emergency to be on going. I was called a murderer.”Interview with a medical manager at the time, Paris, January 2010.Head office felt that the priority at the time was to concentrate operations on areas where the populace was directly impacted by attacks from armed groups and soldiers.

2.2.4. HOW MSF-BELGIUM MANAGED THE ORDINARY AND THE EXCEPTIONAL: THE CASE OF THE MITWABA PROJECT

Starting in 2005, MSF-B recentred its activities on conflict zones, giving rise to new emergency response programmes. Following the upsurge in violence in the Mitwaba health zone (North Katanga) in 2004, MSF-B decided to open a care programme for refugees, returnees and residents. The project consisted of “making contact with isolated groups thanks to a significant mobile and referral capability” through use of “integrated action if necessary.”MSF-Belgium, Mitwaba project document, November 2006.Drawing on the lessons of the district-based approach and its limitations during the early years of the conflict, the Belgian section was worried to see its programmes being extended and gradually meeting ordinary health needs.

With this in mind, it attempted to define intervention objectives that could be measured by a set of indicators. The special attention paid to the criteria for entering and leaving the programme can be explained “by the fact that the health situation in the country is generally poor, and that it is therefore hard to distinguish between structural problems at the origin of the worrying indicators for healthcare, and an exceptional situation that increases the vulnerability of a given population.”MSF-Belgium, Mitwaba project document, November 2006.In order to justify its decision to pull out, MSF-B turned to evaluation tools that tracked the evolving context. The mechanism consisted of “analysing the persistence of the consequences of displacement on the people receiving assistance”. It also distinguished between populations: the resident population was aided for so long as it suffered the consequences of the presence of refugees; aid to returnees continued “for so long as the population has not regained its autonomy, i.e. for so long as it remains dependent on WFP aid”. Officially, the MSF-B project was to provide medical and healthcare assistance to displaced people and the local population in the Mitwaba health zones via “a minimum surgical and obstetric capacity.”Interview with a former desk manager, Brussels, January 2010.MSF-B made efforts to break with the mindset of its previous action: “we will make every effort to act outside the health resources plan and only intervene to meet specific needs. However, it is hard to change professional habits. This is not helped by MSF-B’s traditional relationship with the Congolese authorities.”Interview with a former desk manager, Brussels, January 2010.

MSF-B’s worries lay partly in the context in Katanga, and partly in the section’s history in the DRC. Beyond the goals set out in writing, there existed operational practices that tended to hybridise the Mitwaba project’s objectives: it was hard for the teams to restrict their actions to just the target populations, and the section continued to ponder the issue of access to care. Such attitudes and habits blurred the programme objectives.

The project to disengage from Mitwaba was evidence of these conflicting currents. In the face of a situation characterised by the great distress and vulnerability of the population, the disengagement strategy again referred to context indicators to justify the departure: “now that the rebels have been disarmed and the zone stabilised, and in the absence of WFP food distribution during the months of June and July, displaced populations have started to return to their villages and their numbers have fallen sharply.”Dr Gbané, proposal to disengage from the Mitwaba project, MSF-Belgium.

Furthermore, the disengagement project involved handing over several centres to the health zone authorities: identification of the human resources needed for the zone to function, appointing MSF staff and the construction of a maternity building were all part of the exit strategy agenda.Dr Gbané, proposal to disengage from the Mitwaba project, MSF-Belgium.

The historical ambivalence of MSF-B’s position in the DRC led the organisation to seek the optimal operational compromise. Ever since its establishment in the former Zaire, and acting on the basis of mortality studies, the Belgian section had been seeking a healthcare model and was looking for an operational strategy that would have the greatest impact on the population’s health.

As part of its search for a new healthcare model, in 2007 MSF-B decided to invest in a referral hospital within a so-called post-conflict zone, but one with high mortality rates, where the populations suffered from limited medical provision, which many could not access because they were unable to pay. MSF-B’s objective in fitting out a hospital with the technical equipment to provide a variety of basic services was to quickly reduce mortality and morbidity rates, then build an efficient primary care and referral system, and finally set up a viable funding system so that it might be possible to hand the facility over to the Congolese authorities. A combination of two differing, even contradictory, strategies, the hospital at Lubutu was classed as an experimental project that might be adopted more widely if it succeeded, a model for specific post-conflict intervention.MSF-Belgium, Lubutu project document, pp.1-3, 2009.

2.3. PREPARING FOR DEPARTURE

It is never easy to close a mission. It is harder still when the crisis persists and the lines that divide an ordinary situation from an exceptional one are regularly discussed, re-evaluated and redrawn.

This uncertainty, and the fear of being trapped in a situation where providing assistance would no longer meet MSF’s principles for taking action, led managers to study changes in the environment to try to determine whether or not they felt these changes justified continuing operations. In the case of Katanga, the notion of conflict was the essential factor. It explained the organisation’s arrival in the province, and justified opening up a humanitarian working space that the organisation altered as it refined its ideas about the context for its intervention.

However, analysis of the missions in Katanga raises the question of whether the conflicts were the sole motivation for withdrawing or maintaining missions. MSF-F’s working space has never been a given. It is the fruit of constant negotiation within multiple parameters, among which variations in the intensity and mobility of fighting are not always decisive.

Various factors and institutional interests played a part in the decision to withdraw. This was the justification for quitting the province offered by MSF-F in July 2007: “the context showed signs of stabilising and there was a marked downturn in violence, so we decided to reposition ourselves starting in 2006: either we changed the objectives for our work, or we transferred operations and provided support to teams working in zones where the violence remained very high. We took our decision. The presence of other MSF sections in Katanga as well as budgetary and operational considerations all pointed to the need to strengthen and expand the programmes we were running in Kivu.”Interview with an MSF-F operations manager, Paris, December 2009.

The withdrawal from Katanga was often felt by actors within the organisation to be the immediate consequence of the progressive return to normal of the context in the province. However, the fact remains that other parameters were involved in making the decision: the presence of several MSF sections in Katanga, the influence of its budgetary position, and the logistical and operational constraints under which the Congolese missions operated. The decision to pull out was less the fruit of any one event than it was the result of a slowly maturing process. The questions that appeared gradually in the reports illustrated people’s doubts as to the legitimacy of the programmes in Katanga.

The organisation’s use in internal documents of certain terms rather than others makes it possible to track the growing attraction of a withdrawal over all other options.

The notion of post-conflict, referred to in the early reports made by the exploratory missions,MSF, exploratory missions summary report.disappeared in step with MSF-F’s decision to enter Katanga to offer emergency health care in response to the violence of the armed groups.Refer to the various annual plans.

The interviews reveal a more nuanced reality: the descriptions of the context varied as a function of the moment and location. The term post-conflict crops up orally on a number of occasions when describing the environment that teams were working in. It is a term subverted, however, by the institutional texts: the 2006 country data sheet refers to Ankoro and Mukubu as “zones of violence”MSF, DRC 2006 country sheet, p. 2.with a population in need of medical care; previously they were termed “terrains of conflict”. From 2007 on, the presence of MSF-F in these two health zones was motivated by the necessity of providing care to “indirect victims of the conflict.”MSF, DRC 2007 country sheet, p. 3.As the decision to withdraw began to gather speed within the internal debates, so the term post-conflict made a reappearance in the texts: the 2007 Mukubu project sheet shows teams wondering whether “they are not overly wedded to an operational pattern of providing assistance to victims of conflict in a situation of post-conflict”.MSF, internal document, Mukubu objectives 2007, p. 4. On this occasion, the presence of MSF-F is explained by the “fragility of the health situation of returning populations” and the “nascent post-conflict situation” which “offers an uncertain period of reconciliation.”MSF, internal document, Mukubu objectives 2007, p. 5.

Were the notions of conflict and post-conflict genuinely of any assistance in understanding the context in Katanga? They served in fact more as political labels that were used as justifications for operational decisions: labelling a situation as “conflict” legitimises the commencement and pursuit of medical assistance, but from the moment that such a label becomes uncertain, operations underway start to be called into question, and their interruption may be debated or decided by the operational centre, resulting in tensions between operational managers and actors in the field. In the case of North Katanga, on a number of occasions differences of opinion arose concerning MSF’s actions and the way that it used its resources. We have attempted to describe a few of these moments when the two mindsets clashed: the mandate versus the need. These opposing currents become evident when, for example, actors in the field judged that the decision to withdraw made by the operational centre was inappropriate, because the needs on the ground justified continuing medical assistance. They therefore opposed the mandate-driven approach: the mandate refers to a scale of values made explicit in the institutional messages; it concerns prioritising “medical assistance to populations whose living conditions are in a critical state as the result of conflict”. It is this type of situation where the organisation’s action is the most legitimate, the least open to criticism. The more that the situation diverges from this definition, the more that the actions undertaken come in for criticism by those who favour the mandate-based approach, and it is at these times that regular disagreements between the two different approaches arise. However, our study also shows that this situation of opposing viewpoints does not always result in clear-cut decisions. It is not, in fact, incompatible with simultaneously maintaining emergency operations in response to violence committed by armed groups, as well as providing support to hospitals and health centres, support that endures beyond spikes in violence against civilians. Everybody agrees that there is a difference between post-conflict and conflict, but the decision of where to draw the line fluctuates, unless a stable definition can be proposed (and imposed?), something that was a regular source of internal divisions from 2001 to 2006 in North Katanga.