Humanitarian aid and development  »  History, milestones and reforms  »

Ruanda 1994 and the birth of modern humanitarianism

  • Page updated onMarch 29, 2026
Humanitarian reform after the 1994 Rwanda crisis
Text and photo: Bruno Abarca

The genocide and major humanitarian crisis in Rwanda in 1994 was a turning point in the history of humanitarian action. First, it exposed the failures of the international community to prevent and stop genocide. In addition, it also exposed weaknesses in the preparedness, coordination and accountability of the humanitarian system in the face of complex emergencies, in which violence overlaps with large forced population displacements.

The evaluation of the humanitarian response to this crisis, together with analyses of interventions in other complex humanitarian crises such as Bosnia and Somalia, was a key factor in prompting a reform process that was already incipient and had taken significant steps in multiple directions.

The 1994 Rwanda crisis and the evaluation of humanitarian response

Violence, genocide, forced population displacement and public health emergencies

Social and political tensions between Hutu and Tutsi, exacerbated during the Belgian colonial period (1916-1962) and after the declaration of independence in 1962, resulted in 1990 in a civil war that reached the most extreme levels of violence in 1994. On April 6 of that year, an attack (whose authorship has not yet been clarified) killed the president of Rwanda. This event served as a pretext for Hutu extremists to activate an already prepared extermination plan against the (minority) Tutsi population and against moderate Hutus, perceived as traitors. In just one hundred days, from April to June, the presidential guard, militias, local authorities and mobilized civilian population killed between 500,000 and 800,000 Tutsi in systematic massacres. There was no coherent action by the international community to stop the genocide.

At the beginning of July, the Rwandan Patriotic Front, after three months of military confrontations, managed to regain power in the country, establishing a unity government. Fearing reprisals, more than two million people (mostly Hutus, including extremists, militia members and perpetrators of the genocide) fled to neighboring countries: Burundi, Tanzania and Zaire (today, the Democratic Republic of Congo). Among them, 850,000 fled across the northern border to Goma, concentrating with extraordinary speed in and around the city.

In the weeks that followed, Goma was the scene of a public health emergency of enormous proportions. Approximately 50,000 people (between 6% and 10% of the population) died in just the first month, mostly due to epidemics of cholera and dysentery, aggravated by dehydration and violence. The volcanic rock terrain in which they were located made it difficult to drill wells, provide clean water and install basic sanitation infrastructure. This, together with the overcrowding caused by an immediate and massive migratory flow, favored feco-oral transmission of diseases. Humanitarian organizations were overwhelmed by the situation, which was also complicated by the extreme insecurity caused by the armed Hutu militias that controlled the improvised camps.

Humanitarian response to a complex emergency: scale and characteristics

Once the emergency and cholera outbreak was declared in Goma, and spurred on by extensive media coverage, the international community poured into a large-scale humanitarian response. In addition to the usual international NGOs and UN agencies in complex crises, a multitude of actors joined in, including civil protection and disaster response agencies from donor countries, military contingents, as well as many other less experienced NGOs. They soon succeeded in distributing food and water and setting up health care centers.

Within Rwanda, the extreme violence that began in April 1994 was a major obstacle to the humanitarian aid needed by thousands of internally displaced persons. Only a few actors were able to operate, with great difficulty and risk, in the areas controlled by the genocidaires. On the contrary, in the areas that the Rwandan Patriotic Front was controlling, there was better humanitarian access. The courage and determination of organizations such as the ICRC (International Committee of the Red Cross), MSF, the United Nations advance team, Caritas/CRS, and the World Food Program made it possible to avoid famine. In the southwest of the country, a French military force was able to create a safe zone. While this allowed the protection of civilians in this area, it also attracted humanitarian actors who took time to establish operational capacity and diverted their attention from the more concentrated and less protected migratory flows in the northwest.

Tanzania received 200,000 people in April 1994 fleeing the genocide. There, the response was coordinated very effectively by UNHCR, which had the initial support of the Tanzanian government. However, the migration flow continued to increase in the months that followed and, when the Goma crisis occurred and international attention was focused there, the situation in Tanzania deteriorated. Water supply and infrastructure had not been planned for a response of this duration.

In Burundi there was a lower density of refugees from Rwanda. Despite encountering a highly violent and politically unstable context, UNHCR and some international NGOs were able to provide an effective humanitarian response.

The Joint Evaluation of Emergency Assistance to Rwanda and its findings

Due to the magnitude of the Rwanda crisis, an unprecedented multinational evaluation was launched in late 1994 to draw lessons that could be relevant to ongoing operations. One of the four studies that formed part of this evaluation, led by John Borton and published in 1996, focused on analyzing the performance of the humanitarian system.

The study highlighted a fundamental weakness of the humanitarian system: the absence of mechanisms to monitor and analyze information on the evolution of the situation and to issue alerts in areas with almost no coverage by humanitarian agencies. The few organizations with response capacity in the midst of the violence offered partial and fragmented data. There were data and estimates from some organizations on people in need of assistance that did not reach other key organizations. At the same time, military operations were altering the dynamics of the crisis and response as they went along. While many actors were concentrated on the Tanzanian border and in the «safe zone» of southwestern Rwanda, in Goma there was barely a small remnant of personnel left, with a contingency plan for some 50,000 people, far fewer than eventually arrived. It was hoped that, should a more intensive humanitarian response be necessary, air assets could be activated quickly, but no assets or equipment were pre-positioned.

The situation in Goma could have been much worse had there not been a new multi-agency mission to assess needs in the northwest of the country just after the victory of the Rwandan Patriotic Front and before the arrival of the large migratory flow. This allowed for an extraordinarily rapid mobilization, although initially insufficient for the scale of the crisis. The dependence on air transport was a logistical bottleneck, complicated also by the urgent shipment of some non-priority cargoes that did not correspond to what was requested. The need for heavy machinery, for example, was not met until September.

Similarly, it was not until March 1995 that the problem of insecurity and violence in the camps in Goma was addressed with the necessary priority, which made humanitarian action extremely difficult. There were also problems of coordination between important actors, creating duplication and leaving unmet needs.

Learning and recommendations for improved humanitarian action in complex crises

The joint assessment revealed key problems, but also proposed recommendations and solutions. Among the priority issues were:
  • The political vacuum of the international community to stop the extermination plans and genocide. The violence drove many humanitarian organizations out of Rwanda and persisted during the movements of refugees to camps in Goma and other border areas. Humanitarian organizations did what they could in an extremely high-risk scenario, but their staff were forced to leave the cholera treatment centers in the camps every night and it was impossible to ensure equitable food distribution, for example. All this showed that humanitarian action cannot be a substitute in complex emergencies for political, diplomatic and (in some cases) military action.
  • The lack of investment in crisis preparedness. Donor risk aversion limited the funding of contingency plans for emergencies that might not occur, regardless of the fact that, if they did occur, rapid response would be too late. It was not only a matter of prepositioning essential goods, but also of mobilizing key technical and logistics personnel and preparing communication and transportation equipment.
  • The importance of early warning mechanisms. In order to have useful contingency plans, it was also necessary to improve the flow of information and communication between agencies and with NGOs at all levels.
  • The need to improve coordination among actors. Between 250 and 300 humanitarian organizations and more than 20 donors participated in the response in Rwanda and neighboring countries. UNHCR took the lead in responding to the displacement of refugees, with financial and institutional support. However, within Rwanda, coordination of the response was more fragmented. There, a temporary ad hoc Rwanda response coordination structure (UNREO) was established in April 1994, supported from abroad by the UN Department of Humanitarian Assistance. But it arrived late, lacked sufficient resources and experience in humanitarian coordination and did not have a clear relationship with the UN operational agencies and NGOs.
  • The performance of NGOs and the accountability of the humanitarian system in general. The work and vital role of many NGOs was recognized, but the lack of professionalism, responsibility, technical coordination and rigor in accountability for impact and performance assessment of many others was also criticized. This underscored the need for a set of humanitarian standards (already under development by then) and their global dissemination. It also went so far as to suggest the creation of an accreditation system for NGOs and a humanitarian «ombudsman».

Effects and consequences of the joint evaluation of emergency assistance in Rwanda

The evaluation stressed that funding for humanitarian action was used as a substitute for political and military actions that could have curbed genocide. Despite this, reviews of compliance with these recommendations in 1995 and 2004 showed little progress in preventing and suppressing genocide and massive human rights abuses.

As far as humanitarian response is concerned, not all recommendations had the desired effect, nor did all subsequent changes in the humanitarian system coincide with the vision and proposals of the evaluation. Some proposals were rejected outright, such as the recommendation to create a humanitarian «ombudsman», to combine the humanitarian functions of the different UN agencies into one, to have all humanitarian funding controlled by a neutral entity, or to establish an international accreditation system for humanitarian organizations, so that only those that met certain requirements could register, receive work permits for their staff, or have import privileges in the countries of intervention.

However, many of the document's conclusions and proposals catalyzed important changes and advances in accountability, humanitarian standards and professionalization. They also helped to consolidate and sustain improvement initiatives that were already in progress, the importance of which was now beyond doubt.

The Code of Conduct for Disaster Relief

The origin of the Code and its coincidence in time with the Rwandan genocide

In 1991, the French Red Cross proposed to the International Federation of Red Cross and Red Crescent Societies and the International Committee of the Red Cross the development of a code of conduct in humanitarian action. The Cold War had just ended, and the humanitarian system did not seem capable of providing quality assistance at scale in complex humanitarian crises, with strong war and political undertones. Many independent organizations, coming from rural development and natural disaster relief work, or unprofessional, had no experience in conflict situations.

The proposal was brought before the Steering Committee for Humanitarian Response, which commissioned Peter Walker (IFRC) and Tony Vaux (Oxfam) to draft it. In 1994 they published the Code of Conduct for the International Red Cross and Red Crescent Movement and Non-Governmental Organizations (NGOs) in Disaster Relief, at the same time as the Rwandan genocide. This proximity in time helped it to be perceived as an important tool for addressing the problems and lack of standards seen in that and other complex emergencies of the time.

The contents of the Code of Conduct: 10 ethical principles of conduct and 3 annexes

The Code of Conduct was intended to establish standards of behavior for all humanitarian organizations, based on the humanitarian principles of the Red Cross and Red Crescent Movement. Moreover, its first four ethical principles of conduct are basically a reformulation of the humanitarian principles of humanity, impartiality, independence and neutrality (although the reformulation of this principle in Article 3 is particularly ambiguous). The following six principles of conduct, however, address other issues, such as respect for local culture, strengthening of local capacities, participation of beneficiaries, reduction of vulnerabilities to future disasters, accountability and respect for people's dignity in public information and communication actions.

In addition, the Code included three important advocacy annexes. Two of these directly called on governments to respect the independence and impartiality of non-governmental humanitarian agencies, facilitate access to affected people, assume their political responsibilities in planning and coordinating aid, and comply with international humanitarian law. A third annex urged intergovernmental organizations (such as the United Nations) to support local humanitarian coordination and to recognize nongovernmental humanitarian organizations as allies and partners, protect them and give them access to the same information that is guaranteed to their own agencies. For many, however, these three important annexes remained on the invisible back burner.

Impact and legacy of the 1994 Code of Conduct

The Code of Conduct, in reality, did not bring about much immediate change in the way the organizations that signed it behaved, as Walker himself acknowledged years later. In part, this was because it was voluntary and not legally binding. As such, there were no clear mechanisms for accountability or sanctioning violations of the Code. In addition, neither the grand statements nor the dissemination of the Code by organizations to their staff was transformed into consistent commitment and change. The Code of Conduct became an essential part of induction processes for new staff, and was often consulted when updating the organizations' strategic approaches, but then put away in a drawer during program implementation. Nor did NGOs take sufficient advantage of the advocacy potential of the Code's annexes.

Another important criticism of the Code comes from its Western roots, from the vision of the headquarters of international organizations, with little consideration of the ideas and voices of local organizations. This, which also affected many other initiatives of the time, clashed with the emerging vision of a decolonized and localized humanitarian action.

In any case, the Code did play a key role in disseminating the humanitarian principles of the Red Cross and Red Crescent Movement and solidifying a common normative basis for virtually all the heterogeneous actors in the humanitarian system, from NGOs to UN agencies. In addition, it also served to pave the way for the success of the Sphere Project, which years later would integrate the Code of Conduct into its essential chapters. Many of its elements also served as the basis for the subsequent development of the Core Humanitarian Standard.

The Sphere Project: Effectiveness and Accountability in Humanitarian Action

What was the origin of the Sphere Project?

If the Code of Conduct talked about how to provide humanitarian assistance, it suddenly became necessary to agree on what it should consist of. In fact, there had already been some attempts in this direction. For example, the U.S. NGO alliance InterAction had begun work in the early 1990s, under the coordination of Lisa Mullins, on a set of standards for private voluntary organizations. The «Humanitarianism and War» project, led by Larry Minear and Thomas G. Weiss, also played a key role at that time in promoting debate around humanitarian principles, operational dilemmas, and the professionalization of aid. These and other initiatives served as precursors to one more, which would end up having a much greater impact.

In 1995, Peter Walker of the IFRC and Nicholas Stockton of Oxfam sketched the first outlines of a new project in Geneva, after many discussions about the internal and operational challenges of humanitarian response. Out of that concern came the initiative, which would take off with the support and enthusiasm of the Steering Committee for Humanitarian Response organizations. Other alliances of organizations soon joined the project, such as ICVA, the U.S.-based InterAction and the European VOICE, uniting their efforts for a common global goal that would end up being called the Sphere Project.

The Sphere project sought to answer a key question: What should humanitarian organizations guarantee to people affected by crisis? Moreover, this question was posed from the perspective of these people, their rights and the minimum standards they would expect to receive to help them preserve their lives with dignity. To find answers and develop such an ambitious project, they tried to involve as many actors as possible, seeking also the support of more than ten humanitarian donors. Sphere did not have the logo of any organization on it. It belonged to everyone who felt the need to improve the effectiveness and accountability of humanitarian action.

Launching Sphere shortly after the Rwanda tragedy was a key factor in the project receiving near-global support. It was a collective effort of self-regulation by humanitarian organizations, in part also to anticipate a possible recommendation by the Joint Evaluation of Emergency Assistance to Rwanda to create an external regulatory mechanism. And it worked. The Rwanda evaluation team itself recognized the value of the initiative, rather than promoting more interventionist alternatives.

The 1998 Sphere Handbook was received with great enthusiasm, but also with criticism

The first pilot version of the Sphere Handbook was released in 1998. It was very well received by the general public. This was reflected in the rapid multiplication of translations of the handbook into many languages and the high demand for Sphere training workshops in many contexts. There was a thirst for common technical standards.

However, there was also criticism and even distancing from the project by several humanitarian organizations, mainly French. What if such a strong emphasis on technical standards ended up being used as a rigid prescription, understood as an end in itself, prioritized ahead of accountability to people? Were the standards applicable to any context? How were the technical criteria linked to the Humanitarian Charter? Where was the importance of protection addressed? How could humanitarian principles be effectively applied if there was no hierarchy among them? Finally, and above all, there was a fear that the Project would end up shifting political responsibility for humanitarian crises and the burden of accountability from governments to humanitarian agencies, and that the standards would become a mechanism for donor control over implementers.

Impact and relevance of the successive updates of the Sphere Handbook to date

Dissemination of the Sphere Handbook was not just a matter of making itself known. On the contrary, and always with the intention of making the project global and of the entire humanitarian system, exchanges of experiences were held on its applicability in numerous contexts and countries.

Over time, new editions of the handbook (in 2000, 2004, 2011, and 2018) incorporated changes based on these discussions, improvements, and expansion of its contents, and even standards from other initiatives consistent with the project's vocation, such as the Core Humanitarian Standard. This evolution helped Sphere to overcome some of the initial limitations and to find its place in the humanitarian system, more as a reference than as a fixed standard. It also consolidated its role as a technical and normative foundation for advocacy with governments, responsible for creating a humanitarian space in which compliance with Sphere standards is possible.

The Sphere Project has introduced a common language to the humanitarian sector, becoming over the years a fundamental element of the identity and culture of many of its organizations. Sphere remains today a benchmark and a cornerstone for driving accountability and quality in humanitarian action, although there is still a long way to go in its operationalization and implementation.

The birth of new organizations: ALNAP, People in Aid, HAP, and OCHA

In the years following the Rwanda crisis, there were also strong institutional responses to the failures of the humanitarian system that had been identified in this and other complex emergencies: they had to do mainly with an interest in overcoming problems related to the lack of systematic learning, lack of accountability to populations in need of humanitarian assistance, and weaknesses in humanitarian coordination and leadership. 

ALNAP: a commitment to systematic learning in the humanitarian system

Some of the people who participated in the evaluation of the humanitarian response in Rwanda, including John Borton, the director of the third study, decided to continue the initiative. To do so, they created a forum to talk about accountability and learning in the humanitarian system. This is how the Active Learning Network for Accountability and Performance in humanitarian action (ALNAP) was born in London in 1997.

Today ALNAP continues to be one of the most influential evidence and learning networks in the humanitarian system. Its reports and studies contribute to generating evidence on what works, how and why in humanitarian action, to improving evaluations of humanitarian projects and the overall performance of the sector, and to addressing new challenges that shake the foundations of the humanitarian system.

The birth of People In Aid and the Humanitarian Accountability Partnership

In 1995, a research project focused on the risks arising from the expansion of the number of NGOs and the need for professionalization of their staff. The study showed that humanitarian work and international development cooperation required a particular combination of skills and training, and that this was a determining factor in the quality and effectiveness of aid. As a result, the organizations that developed the study, a year later, published a code of good practice in the management and support of humanitarian workers, along with tools for signatory organizations to assess their progress against. This initiative resulted in the creation of People in Aid, an organization established to support humanitarian organizations wishing to improve the management of their staff.

While People In Aid was consolidating its mandate, in 1997 a group of British organizations tried to move forward with the idea of a humanitarian ombudsman proposed in the joint evaluation of the international response to the Rwandan genocide. This mechanism was envisioned as an impartial and independent entity that could investigate complaints and abuses reported by aid recipients and regulate the activities of NGOs that failed to comply with their legal obligations and established codes of practice. This already complex proposal was met with skepticism by many humanitarian actors who preferred self-regulation (along the lines of the Sphere Project).

🧠 Let's pause and reflect

Why did the humanitarian system opt for self-regulation instead of an external accountability mechanism, and what consequences has this decision had for the populations it is supposed to protect?

  • 1 Think and write your answer.
  • 2 Click on «Copy and open».
  • 3 Paste to receive feedback.

📚 This is the NotebookLM for this topic. It uses only carefully selected references. | What is NotebookLM?

Although the idea of the humanitarian «ombudsman» did not prosper in its initially proposed form, it eventually evolved into the creation of HAP International (Humanitarian Accountability Partnership) in 2003. It was a self-regulatory initiative launched by humanitarian, development and donor organizations to improve the accountability of humanitarian action to people affected by crises, thereby protecting their dignity and rights. In 2010 they published their standards on quality management and accountability and in 2014, the first edition of the Core Humanitarian Standard. Just one year later, People In Aid and HAP International merged into what is now the Essential Humanitarian Standard Alliance.

The transformation of DHA into OCHA, to coordinate and lead the system

The Rwanda crisis exposed the limitations of the DHA (Department of Humanitarian Affairs) created by the UN Secretary General in 1992. It highlighted the need for stronger leadership and greater inter-agency capacity. A UN reform program was launched in 1998 that, among other things, addressed this priority. DHA became OCHA, which was given a clearer and more significant mandate around coordination, policy development and advocacy in humanitarian crises.

What about the principle of neutrality and its application in Rwanda?

The joint evaluation did not address the issue of humanitarian principles and their application in the response to the genocide and refugee crisis in Goma. Its focus was on other things: effectiveness, coordination, logistics, information and political accountability. However, this complex crisis once again brought to the table important dilemmas about neutrality and its operationalization.

The genocide and extreme violence in Rwanda pushed the operational interpretation of the principle of neutrality to its limits

Some authors criticize that, between April and July 1994, humanitarian organizations advocated the need for a cease-fire and the intervention of international troops from the United Nations, without giving support to the Rwandan Patriotic Front, when doing so was the only viable option to stop the genocide. They also criticize that many organizations did not want to intervene inside Rwanda in areas taken over by the rebel army trying to stop the genocide, but then hurried to respond to the refugee crisis fleeing to Tanzania or Zaire. Some reports estimated that among the 850,000 refugees were 20,000 well-armed military personnel who had been involved in killings, and who now controlled the camps, from which they could regroup and reactivate attacks. In Goma, while humanitarian workers were doing their best to save lives in the midst of a terrible public health emergency and great insecurity, the perpetrators of the massacres were taking advantage of humanitarian assistance to obtain resources and security and to consolidate their power in exile.

Eventually, organizations such as MSF France, for example, decided to denounce the threats and the control and diversion of humanitarian aid by local military leaders and announced their withdrawal, frustrated by the inaction of the international community to prosecute the genocide and to avoid being complicit in the vacuum of political accountability. This decision, however, was not well received by other sections of the organization, which did not consider it part of MSF's mandate to establish the guilt of alleged genocidaires, opening an important debate.

The experience of other organizations was different. The ICRC was already in Rwanda when the genocide began and had a unique mandate under International Humanitarian Law. That mandate allowed it to stay in the country when almost all other organizations evacuated their teams. They maintained, as required by International Humanitarian Law, a strict interpretation of neutrality and constant dialogue with all parties (including the genocidaires). This was essential to be able to operate in an environment of great violence, preserving access to the population that had been trapped and protecting the safety of their personnel. Circumstances did, however, compel the ICRC to publicly denounce the violence in a targeted manner on several occasions.

The debate, which is still open today, contributed to a more mature understanding of the principle of neutrality in the humanitarian system

The experience in Goma and the examples mentioned above highlighted the huge operational dilemmas and normative tensions that humanitarian organizations face in armed conflict. They also demonstrated that different organizations, making a coherent reading of their traditions, mandates, responsibilities and principles, could end up applying neutrality in different ways.

The dilemmas of 1994 did not resolve the debate, but rather took it further and deeper. Today the reflections on humanitarian principles and neutrality continue, the result of more conflicts, more humanitarian interventions, more mistakes, more successes, more self-criticism and more willingness to learn and be accountable to the populations affected by violence.

History, milestones and reforms

NotebookLM

You can review my bibliographical references on humanitarian reform after the 1994 Rwanda crisis with this NotebookLM, an artificial intelligence-based research assistant. Do you want to know more?

Learning pathway

This page is part of: The heart of humanitarianism: foundations, ethics and current dilemmas.

How to cite this page

Abarca, B. (March 29, 2026). Rwanda 1994 and the birth of modern humanitarianism. Salud Everywhere. https://saludeverywhere.com/en/humanitarian-aid-and-international-development/humanitarian-reform-after-the-1994-rwanda-crisis/

External links

Search
Scroll to Top