Twenty years on: the Rwandan genocide and the evaluation of the humanitarian response

Twenty years ago in the first week of April, an ODI colleague and I were in Brussels to launch a new network for relief workers (what later became the Humanitarian Practice Network). It was there we learnt that ten Belgian paratroopers forming part of the United Nations (UN) peacekeeping force in Rwanda had been killed trying to protect the Rwandan Prime Minister. The Prime Minister had also been killed. This came just one day after the presidents of Rwanda and Burundi had both been killed when their aircraft crashed on its approach to Kigali. It later emerged that the plane had been shot down. These shocking events marked the start of a genocide that claimed the lives of an estimated 800,000 people and displaced three million – two million as refugees in Tanzania and eastern Zaire (now the Democratic Republic of Congo) and one million inside Rwanda.By the end of 1994 an international evaluation process was about to start – the Joint Evaluation of Emergency Assistance to Rwanda (JEEAR), instigated and led by the Evaluation Department of the Danish Ministry of Foreign Affairs. ODI had been awarded the contract to undertake JEAAR Study 3 to evaluate the international humanitarian response, and I had been appointed as Team Leader.  The following 14 months were an intense rollercoaster of an experience as our 20-strong team carried out its work. Four separate studies together with a synthesis report of the JEEAR were published in March 1996.

The evaluation was highly critical of the international community’s political and military response to the situation in Rwanda: in the months preceding the genocide (when there were clear signals that mass killings were being planned); during the 100 days the genocide was allowed to continue; and in its immediate aftermath. In the absence of an effective political and military response, official funding of humanitarian action in effect became a substitute for political action. Humanitarian agencies were obliged to operate in situations made almost impossible by political and military failings.

A clear example was in the refugee camps that formed in Zaire after the genocide. UN member states were unwilling to provide troops for a proposed UN force to provide security in and around the refugee camps and to separate the killers (genocidaires) from the majority of refugees who had not participated in the genocide. Humanitarian agencies were forced to choose between providing assistance to all those in the camps (including to genocidaires), or pulling out – as Médecins Sans Frontières (MSF) did in November 1994.

Whilst the major failings clearly lay within the political/military domains, shortcomings in the humanitarian response were also apparent. Study 3 was particularly critical of: coordination arrangements within the UN; the lack of preparation by key agencies ahead of the massive refugee influx into Goma in July 1994; and the poor performance of some international NGOs. Whilst many international NGOs had performed impressively, the unprofessional and irresponsible behaviour of some had, most probably, caused unnecessary loss of life. It was our view that such international NGOs should not have been allowed into Zaire, let alone to work in the refugee camps. Without effective support from the Zairean authorities it was striking that the UN High Commissioner for Refugees (UNHCR) was unable to address the issue of unprofessional and irresponsible international NGOs. Even when UNHCR was able to close certain national NGO operations in the refugee camps, the agency was unable to prevent them from relocating their activities to just beyond the camp perimeters.

Aware that the JEEAR’s recommendations might result in an externally administered accreditation system for NGOs, members of the Steering Committee for Humanitarian Response (SCHR) had begun discussing self-regulation approaches. Just before the evaluation was published, SCHR launched a project to develop a Humanitarian Charter and Minimum Standards – what soon became known as the Sphere Project. Within a year this was followed by a British-based project to explore another of the JEEAR’s recommendations – that a Humanitarian Ombudsman mechanism should be established. The subsequent Humanitarian Ombudsman Project evolved into the Humanitarian Accountability Partnership (HAP), which established a voluntary process for certifying the quality of NGOs. In 1997, the Active Learning Network for Accountability and Performance in Humanitarian Action (ALNAP) was set up.

As well as being an appalling tragedy, shocking in its horror and scale, the Rwandan genocide and its aftermath was also a traumatic experience for the humanitarian community. Initiatives on accountability and performance would almost certainly have been developed even in the absence of the JEEAR, but there is general agreement that the evaluation gave impetus and focus to such efforts.

Looking back at these and other related efforts from 20 years on, there is much to admire in what has been achieved. Nevertheless it was striking how, during the response to the 2010 Haitian earthquake, it was still possible for unprofessional and irresponsible NGOs to set up ‘humanitarian’ operations.The ongoing efforts to develop Core Humanitarian Standards (by HAP and People in Aid) and a Certification Model for Humanitarian Organisations (by SCHR) are therefore commendable developments. But measures to strengthen legal frameworks will also be necessary if unprofessional and irresponsible international NGOs are to be prevented from accessing disaster-affected populations. The low-profile work of the International Federation of the Red Cross/Red Crescent (IFRC)’s Disaster Law Programme to help guide national legislators through its ‘Model Act’ and its proposed work to develop a ‘Model Emergency Decree’ is likely to form part of the necessary framework.