Joint Evaluation of Emergency Assistance to Rwanda
by Laura Jackson June 2003

In this, the penultimate issue of the RRN Newsletter before we embark on a new three-year phase (October 1996 – October 1999) supported by new donors, it is very encouraging to report on the positive responses we have received to the members’ questionnaire sent with the February 1996 mailing (see Feedback section). Your comments have reinforced our enthusiasm and conviction that the sharing of experience and lessons learned between personnel engaged in the provision of relief and rehabilitation assistance is of critical importance in raising the standards and effectiveness of humanitarian aid programmes. The cornerstone of RRN publications, Good Practice Reviews, have been particularly well received. Good Practice Review 3 on General Food Distribution in Emergencies was mailed in February 1996, offering a comprehensive review of agency practice in the field and documenting the extensive personal experience of the two authors in this key area of humanitarian assistance. The forthcoming Good Practice Review 4, to be mailed at the beginning of July, [actually GPR no. 5] looks at the different methods of estimating and counting populations used by humanitarian agencies, including the often fraught question of registration, with a view to equipping agency personnel with basic principles to be applied in the field.

The Rwanda Report

This RRN mailing follows shortly after the publication of the Joint Evaluation of Emergency Assistance to Rwanda, widely recognised as a unique and ground-breaking report. Network Paper 16 outlines the main findings and recommendations of Study III of the Evaluation which focused on the humanitarian assistance to Rwanda and its effects between April 1994 and December 1995 (Studies I, II and IV focused respectively on the historical background, the build-up to the genocide and war and the repatriation and rehabilitation phases). The Joint Evaluation represents an unprecedented collaborative study of a major humanitarian operation supported by a significant number and variety of organisations and agencies. Some of the most important issues highlighted by the Rwanda experience were: lack of policy coherence, donor funding and preparedness measures, earlywarning and contingency planning, coordination, the role of military forces in humanitarian operations, improving NGO performance and accountability.

Study III was undertaken by a 17-member team, based at the ODI in London, and led by John Borton, founder and former Coordinator of the RRN.


Here, we offer a brief résumé of recent initiatives resulting from the Evaluation. At this stage, it is difficult to gauge the impact of the Evaluation’s recommendations on the international relief system as a whole, but in the two months since its publication in March 1996, it has been discussed in numerous fora and some of its findings are being acted upon at a number of levels. Discussions have ranged from specially convened meetings of national NGOs in various donor countries and academic conferences to the High Level Council of the Development Assistance Committee (DAC) and the UN Inter-Agency Standing Committee. At a special meeting of Emergency Aid services of eight donor organisations – from Denmark, Finland, Germany, Ireland, the Netherlands, Norway, Sweden and the UK – in Copenhagen on 7-8 May, issues of accountability, standards, monitoring and reporting; early-warning and preparedness; and coordination, raised by Study III, were discussed. In order to monitor and report on actions resulting from such meetings, a ten person Follow-Up Network has been formed, comprising representatives from the Evaluation Studies themselves, and members of the Management Group and Steering Committee that were set up to oversee the Evaluation. It is proposed that this group will report to a specially convened meeting of the Joint Evaluation Steering Committee to be held towards the end of 1996.

Accountability, Standards, Monitoring and Reporting

Debates on standards and codes of conduct in the provision of humanitarian assistance are not new. The International Federation of Red Cross and Red Crescent Societies and NGOs in Disaster Relief Code of Conduct, published in 1994 (Network Paper 7), and today supported by 77 signatory agencies, marked an important step in the setting, attainment and maintenance of standards. Further initiatives are underway within the Steering Committee for Humanitarian Response (SCHR), ECHO and in other fora to develop codes and standards. There is both a need for coordination of such initiatives at the international level and for more detailed technically-based codes to be developed, before agencies can judge their performance in relation to established standards and donors can begin to use them in screening funding proposals and selecting high quality implementing agencies. One such initiative, People in Aid, reported in the News section of this issue, is currently drawing up a draft Code of Best Practice for the recruitment, training and support of aid workers, based on one of the principal recommendations of the recent report, Room for Improvement (Network Paper 10). Although originally conceived as a UK and Ireland initiative, recent discussions with US and UN agencies have revealed their interest in following developments in the Code and one UN agency has indicated its intention to use the Code, once endorsed, as a basis for selecting UK/Irish partner agencies.

The Joint Evaluation of Emergency Assistance to Rwanda also highlighted fundamental issues of NGO accountability, to both donors and beneficiaries. Regarding the former, steps are being taken, as part of the follow-up process, to develop a ‘core’ set of indicators and questions to be included in the monitoring and reporting requirements made of implementing partners. The lack of standardisation in reporting during the Rwanda crisis added significantly to the difficulties of evaluating performance (see Network Paper 15 on Cost Effectiveness Analysis).

Network Paper 14, a Critique of Psychosocial Trauma Projects, together with the article on Refugees in South Kivu featured in this Newsletter, offer another perspective on humanitarian agency accountability – that of consultation with, and responsibility to, beneficiaries. Derek Summerfield, a psychiatrist at the Medical Foundation for the Care of Victims of Torture, questions the usefulness of ‘western’ models of psychosocial care for those who witness and experience conflict and atrocities in developing countries. He advocates an approach based on the strength of relationships forged with those needing assistance and projects drawn up according to the priorities of users, which do not challenge their own cultural frameworks and interpretations. Danielle de Lame, one of the anthropologists contributing to Study III of the Rwanda Evaluation, writing in the Articles section of this issue on her experience in the Rwandan refugee camps in eastern Zaire, offers a reminder of the sophisticated social and power structures within camp populations. Failure on the part of agencies distributing food and healthcare to look beyond what may appear to be a homogenous group and recognise differences in need and access to information – not just between adults and children, or men and women, but more subtle distinctions between educated and peasant women for example – can lead to the delivery of inappropriate assistance. Both contributors argue that understanding the historical and social contexts giving rise to conflict and complexity of relationships within refugee populations are an essential and intrinsic part of any aid programme.

Early Warning and Preparedness

The Rwanda Evaluation revealed that, in the period leading up to the Goma influx, there was no integrated mechanism for gathering and analysing information which might provide advance warning of large population displacement. Without such a mechanism, donors will continue to be reluctant to invest in preparedness measures. The Rwanda follow-up process identified two critical factors in ensuring that such a breakdown in the flow of information does not occur again: credibility of information and preparation of reliable probability estimates. Such gathering and collation of information on situations prevailing in different areas during emergency operations are an integral part of the coordination process as well as providing warning of population displacements.

Two articles appearing in this Newsletter offer practical information and details on how agencies can improve the quality and accessibility of such early warning and preparedness information. Alistair Hallam, who has recently been assisting the RRN editorial team, explores the adoption of technical processes in site planning and the management of information in emergency operations. In the first article, he considers the advantages of satellite imaging as a tool for locating sites which meet the basic access, water and strategic requirements of camp planners. In the second, drawing on the experience of Ian Attfield at the Integrated Operations Centre in Rwanda, he looks at the possibilities of standardising and sharing agency information via geographical information systems (GIS), which can be distributed via the Internet or email. Recognising that there are still a number of obstacles to be overcome in the distribution of agency information in this way – not least the need for agencies to become more open with their data – the fact that an agency can prepare a report or dataset and post it on the system, so that it can be immediately accessed by others, may help to avoid major duplication of effort and improve the timeliness and appropriateness of response.


The Joint Evaluation found major problems associated with the number of NGOs present and the scope of UN agency mandates. Issues of coordination have not only plagued agencies operating in Rwanda and neighbouring countries; donors are looking at their own role in ensuring that responses address the entire cycle of conflict and peace and do not set up systems for the distribution of humanitarian aid which leave an administrative vacuum on departure, in turn aggravating the rehabilitation process. In the News section of this issue, we report on an EU initiative taking place in the context of the wider debate amongst donors, UN agencies and NGOs, on the so-called “relief-development continuum”. (See RRN Newsletters of September 1994 and April 1995 for articles on the ‘continuum’ and for a related discussion on aid policy in transition). For its part, the European Commission is looking at the current split between humanitarian (ECHO) and development (DGVIII) departments, procedures and budgets. The news item on page 16 looks at the Commission’s recent communication which makes a number of recommendations for establishing a more holistic approach to linking relief and development, but suggests that without practical NGO involvement, such good intentions will not result in a realistic programme of action.

A potentially very valuable tool for the coordination and rapid dissemination of agency information, the Internet, and in particular the World Wide Web, offers relief agencies and individuals the opportunity to improve their awareness and knowledge of natural disasters and complex emergencies around the world. Network Paper 13, published and sent to members in February 1996, provides both a guide and directory to an extensive range of existing ‘sites’, offering up to date, local information on conflict, natural disasters and refugee situations, and agency responses to them. The directory now forms part of the RRN’s ‘Home Page’, located on the WorldWide Web at: [link no longer functioning]

This site will enable anyone with access to the Web to link to over 300 sites, including news updates, sitreps, agency home pages, educational and research organizations, and southern-based providers of Internet services. There is less and less excuse for pleas of ignorance.

The Role of Military Forces in Humanitarian Operations

The Rwanda Evaluation found military contingents to have played a significant role in humanitarian operations inside Rwanda and in eastern Zaire in the provision of relief assistance. However, the experience has raised questions about the predictability, effectiveness, high cost and ability of these contingents to participate in such operations. The question of military involvement demands more in-depth research, but the issue of Service Packages is considered in this Newsletter. Recognising the urgent need for considerable additional management and implementation capacity in Goma in particular, UNHCR requested donor governments to provide “selfcontained service packages”. Under the terms of these ‘packages’, donors were to take full responsibility for specific activities such as airport services, road servicing, sanitation, etc through the deployment of national civil defence teams, military contingents or civil disaster response teams. This Newsletter article, drawing on the experience of the Rwanda Evaluation and discussions with some of those involved in developing this ‘Service Package’ concept in UNHCR and its partner agencies, finds a number of problems in such deployment, not least in terms of coordination for UNHCR and from which budget such interventions would be funded.

A somewhat different account of the role of the military is provided by Koenraad van Brabant, an RRN member based in Sri Lanka, who describes recent first-hand experience of working with Sri Lankan government forces. The article (see Feedback Section) looks at the difficulties facing humanitarian agencies caught in the midst of a civil war where relief operations, and hence agency personnel, are regarded with suspicion and their neutrality repeatedly called into question. In this case, the military makes abundant use of its powers to restrict the supply of certain goods to rebel-held areas on the basis that they may be used by Tamil Tigers. Koenraad does not pretend to offer a solution to the problem, but gives examples of the ways in which he and colleagues seek to get around the restrictions. It is an experience that will be familiar to many relief agency personnel who have worked in on-going conflict situations.

This issue is also available in French: Échange Humanitaire No. 5