Roadblock ethnography: negotiating humanitarian access in Ituri, Eastern Dr Congo, 1999-2004.

ABSTRACT

Based on field research carried out in 2004, this article focuses on how Ituri-based Congolese relief workers negotiate humanitarian access with roadblock militias. Experiences and testimonies highlight the importance of sociocultural and political awareness during relief work induced by conflict. It is demonstrated that relief workers in conflict zones do not (and cannot) shed their ethnic identities; that instead they accept that a perceived ethnic identity brings strategic advantages as well as disadvantages. Further, a relief worker's bargaining power is shown to be influenced by militia perceptions of how his/her organization is positioned in the conflict. The overall argument responds to the renewed policy interest in debating the political context of humanitarian intervention.

RESUME

S'appuyant sur des travaux de terrain realises en 2004, cet article s'interesse a la maniere dont les travailleurs humanitaires congolais bases a Ituri negocient l'acces humanitaire avec les milices en charge des barrages routiers. Les experiences et les temoignages soulignent l'importance de la connaissance socioculturelle et politique dans les operations humanitaires suscitees par des conflits. Il est demontre que les travailleurs humanitaires en zones de conflit n'abandonnent pas leur identite ethnique (et ne le peuvent pas) et qu'ils acceptent au contraire qu'une identite ethnique percue presente des avantages en meme temps que des inconvenients strategiques. D'autre part, l'article montre que la puissance de negociation du travailleur humanitaire est influencee par la perception qu'ont les milices de la position de l'organisation de ce travailleur dans le conflit. L'argument general repond a l'interet accru a debattre du contexte politique de l'intervention humanitaire.

**********

This article focuses on the negotiation skills of Ituri-based Congolese relief workers in their dealings with roadblock militias. The testimonies I collected highlight the importance of socio-cultural and political awareness during relief work induced by conflict. It is hoped that their availability will stimulate discussion within the humanitarian community and thus help improve future access policies. The article demonstrates that relief workers in conflict zones do not (and cannot) shed their ethnic identities; that instead they accept that a perceived ethnic identity brings strategic advantages as well as disadvantages. Further, a relief worker's bargaining power is shown to be influenced by militia perceptions of how his/her organization is positioned in the conflict. The article is based on fieldwork in war-torn Ituri, carried out in April-May 2004, prior to which I interviewed expatriate relief workers and missionaries with Ituri experience.

NEW CHALLENGES IN HUMANITARIAN INTERVENTION

Despite the ferocity of the six-year conflict in Ituri, which caused the deaths of some 60,000 civilians, (1) only half a dozen international non-governmental organizations (NGOs) brought relief throughout or during most of the crisis. They included: Medair, the International Committee of the Red Cross (ICRC), Medecins Sans Frontieres (MSF), Oxfam, Caritas-Belgium, German Agro Action (Welthungershilfe) and the Italian NGO Cooperation Internationale (COOPI). My research was restricted to relief agencies and the United Nations Office for the Coordination of Humanitarian Affairs (OCHA); it did not cover humanitarians working for the UN Mission in the Democratic Republic of Congo (MONUC). By focusing on access negotiations, and especially on the role of local humanitarians, this article captures an aspect of political awareness that is rarely acknowledged. The testimonies contextualize Hugo Slim's argument that every relief worker, whether part of a UN force or a relief agency, needs to have 'a strong sense of his or her individual position in relation to the violence' in order to maintain morale (Slim 1997: 3). Individual positioning, I shall add, is also a political act that makes relief workers more effective in delivering their assistance.

The 'political box' that humanitarians work inside received attention at an Overseas Development Institute (ODI) seminar on Ituri in July 2003. The debate inspired me. At the seminar, Anneke van woudenberg, former Oxfam representative in the DRC, now with Human Rights Watch, addressed the subject of politics:

   To remove oneself from politics is exceedingly difficult in a
   context like Ituri. When you are there, the definitions of
   neutrality and impartiality seem to change; it can be very
   difficult to remove yourself from that situation and to say,
   'We are here and we are not with the Hema or with the Lendu.'
   It is a reality that we work inside a political box.

      I came to the conclusion ... that [humanitarians] have to
   work within the political box. It's the only way to do it. Within
   that box, you can set boundaries about what you will do and what
   you won't do, with whom you will negotiate and whom you will not
   negotiate with.... While we have to work within the 'box', we
   still need to find ways to set conditionality. (ODI 2003)

Van Woudenberg's perspective is grounded in the realization that humanitarians tend to treat political economy issues 'as background to the essentially technical business of delivering aid, rather than as a central and immediate aspect of humanitarian need and humanitarian response' (Collinson 2002: 3). Endorsing the position taken up by Duffield (1994), Collinson reiterates that relief agencies 'cannot afford to depoliticize policy by reducing it to a technical matter of organization or good practice' (Collinson 2002: 3). The present article focuses on the ethnic identity of local humanitarian workers, an issue central to any politicized policy approach. Through this focus I hope to contribute to an improved understanding of the complex political and economic contexts of relief operations, thus inviting agencies to consider applying 'an ethical framework to their work [so they can] manage their security more effectively in situations of conflict and political instability' (2002: 3-4, with reference to van Brabant 2000: 4). Applying an ethical framework includes considering the possibility that under certain circumstances, such as when militias massively loot humanitarian stock, relief aid may do more harm than good (Black 2003: 96).

In the public domain, however, agency discourses continue to uphold the image that humanitarians work outside the political box. In a report on its activities in 'neglected conflicts', Oxfam, for example, states: '[H]umanitarian agencies and the United Nations must often negotiate access agreements with all the warring parties. To be successful in such negotiations, it is essential that humanitarians can assert their independence and impartiality from politics' (Oxfam 2003:16-17). Do agencies work inside or outside the political box? Hoping for some answers, I visited Ituri to explore access negotiations.

SOCIETY AND HISTORY IN ITURI

For Jean-Pierre Lobho Lwa Djugudjugu (1971 a, 1971b, 2002), a Hema anthropologist trained in British structural-functionalism, the violence between Hema and Lendu, the main protagonists in the conflict, (2) is endemic; it stems from historically evolved socio-cultural and economic differences. But this notion of endemic conflict needs qualifying. While Ituri has a history of intermittent yet contained violence, the escalation in 1999 would not have occurred had there been a semblance of central authority, justice and policing. From July 1999 onwards, Hema landowners paid Ugandan soldiers to protect and further their interests. A Ugandan military unit commanded by Captain Kyakabale entered Djugu Territory 'to kill Lendu and wipe out entire villages' (ASADHO 1999). (3) Ituri's violence is a transnational affair, initially on account of the Ugandan occupation, more recently because of the involvement of Rwanda. Before Uganda became attracted by 'Congo Gold Fever' (Lobho 2002: 43), most instances of Hema-Lendu violence had erupted following local administrative decisions--some colonial, some post-colonial--that could be construed as favouring one ethnic group over the other. Clashes were violent, but always contained.

Where did it begin? According to Aldan Southall, who studied Alur society in the 1950s, the 'Sudanic' Lendu may have been first (or among the first) to migrate into the area west of Lake Albert some three centuries ago (Southall 1954a: 142). Southall refers to Lendu and Okebo as 'non-Alur tribes whom the Alur had been continuously absorbing as their subjects' (Southall 1954b: 485). He makes the same point about Hema-Lendu relations, situating the earliest Hema migrations in the late seventeenth century. Hema who live in Ituri came from Bunyoro, Uganda, where they were known as Hima, nowadays as Batoro. (4)

   The earliest Hima chiefs to cross Lake Albert were the Gegere
   who settled among the Lendu south-west of Mount Aboro. They
   were recognized as overlords by subsequent Hima groups who
   joined them. Presumably Bantu speakers on arrival, they
   gradually became entirely Lendu in speech. (Southall 1954a: 151)

While Lobho accepts Southall's reconstruction of pre-colonial Hema-Lendu relations, he stresses that the initial 'integration' of Lendu was not only gradual and peaceful, but also much needed. Lendu, Lobho argues, lived in dispersed clans that clashed frequently and violently. Lendu accepted the authority of Hema chiefs, and of Hema generally, because 'the Muhema used diplomatic skill to maintain law and order' (Lobho 1971b: 90). With the passage of time, 'every Hema family head gathered around him Walendu clients, whom he called "ma bale", my Walendu. The Hema political role became so enormous that the Hema imposed upon all Walendu an entire political organization imported from Bunyoro' (1971b: 90-1). (5)

Lobho's reading of Ituri history reminds us of the functionalist anthropology of Alexis Kagame and Jacques Maquet in Rwanda in the 1950s, an antiquated perspective which the Rwandese Patriotic Front has tried to resuscitate (Pottier 2002). Whatever one's opinion on this, Lobho's reconstruction of harmonious pre-colonial Hema-Lendu relations is valid only up to the late nineteenth century, at which point firearms were introduced along with extreme forms of violence. Southall recalls how the PaNduru (Alur) chief Ujuru 'subdued the Lendu of Anzhou south of Juganda, then, returning south at a call from Nblukba, the Hema chief of the Gegere, he massacred the Lendu as far as Blukwa and Mambisa' (Southall 1954a: 163).

When the Belgian colonialists arrived, 'Alurland had entered upon a period of destructive turmoil which was still on the increase' (1954a: 163). Regarding Hema-Lendu relations, the problem was two-fold: not only had Hema ruling groups imposed domestic serfdom on (many) Lendu, but they had also begun to displace Lendu from their land. Lobho details the displacement:

   While the Bahema were happy just to graze their cattle, they
   gradually encroached upon Lendu territory, and ended up
   disposessing the Walendu of a part of their domain. The
   abundance of land at the time, combined with Walendu hospitality,
   explains why the latter did not put up much resistance and
   why they readily accepted the new political order. (Lobho,
   1971a: 568-9)

Land dispossession occurred predominantly in Lendu-Gegere areas of Djugu Territory and much less in the Lendu-Bindi area south of Bunia (Irumu Territory), where land was more plentiful and access more even. (6)

To end Lendu subjugation, the Belgian colonial authorities created separate Hema and Lendu villages, gave Lendu their own chiefs, and stopped Hema from grabbing Lendu land. But the plan proved unpopular in the early years. Many Lendu rejected the authority of the newly appointed chiefs (Southall 1956: 320), while other Lendu groupings like the Lendu-Bindi (also known as Ngiti) retained their independence and continued to elude the authorities. Among these independent Lendu, political action was organized at the level of strongly autonomous sub-clans (1956: 161).

Lobho and Southall agree that the initial European encounter was disastrous for the Hema political elite. By the early 1950s, just two decades after villages were 'fixed' as mono-ethnic entities, Lendu had started 'to pretend that no other state of affairs had ever existed' (Southall 1956:153). Hema supremacy continued to decline during the first decade of independence when Mobutu's Zaireanization campaign gave Hema once again the upper hand in matters of administration, education, artisanal fishing and commerce (Lobho …