Secret UN briefing on 1994 Rwanda atrocities – full text of the “Gersony Report”
HIGH COMMISSIONER FOR REFUGEES
11 October 1994
Dear Mrs. Molina-Abram,
We refer to UNHCR’s briefing to the Commission of Experts on Monday, 11 October 1994.
As requested by the Commission, we are forwarding herewith a written summary of Mr. Gersony’s oral presentation and copies of some field reports sent to UNHCR Headquarters by UNHCR Field Offices.
We are confident that, as agreed by the President of the Commission of Experts, these documents will be treated as confidential and only be made available to the members of the Commission.
Chef de Cabinet
Mrs. B. Molina-Abram Secretary to the Commission
of Experts on Rwanda Room 141.1 1 Palais des Nations
SUMMARY OF UNHCR PRESENTATION
BEFORE COMMISSION OF EXPERTS
10 OCTOBER 1994
PROSPECTS FOR EARLY REPATRIATION OF RWANDAN REFUGEES CURRENTLY IN BURUNDI, TANZANIA AND ZAIRE
Between April and August 1994, over one million Rwandans fled their country to refuge in Burundi, Tanzania and Zaire. Some 100,000 reached Burundi and Tanzania in July and August alone.
On July 27, the High Commissioner appointed an Emergency Repatriation Team to conduct a thorough, systematic and comprehensive field assessment of the prospects for repatriation and to devise mechanisms which could accelerate the safe return to Rwanda of the refugee population. In conducting its assignment, the team received information which suggested that conditions were not yet conducive for such return, its findings are briefly summarized in this document.
During a five-week period in August/early September, the team visited 41 (28%) of Rwanda’s 145 communes, and collected detailed information about ten additional communes. In nine UNHCR refugee camps in Burundi, Tanzania, and Zaire, and in 91 locations inside Rwanda, the team conducted more than 200 individual, private interviews of about one hour each with local residents, former displaced persons, spontaneous refugee returnees and current refugees.
More than 100 additional persons were interviewed in small groups in a less private manner. Their first-hand eyewitness accounts — which comprise the main basis of this report – were detailed and credible, and they responded spontaneously to many follow-up questions, clearly describing events which they had seen. The most important elements and patterns of the Information they provided, particularly for the south/south-eastern regions, were corroborated independently by the team’s interviewees in several countries outside Rwanda and by those in different locations inside Rwanda. [These reports are also mutually corroborative with information gathered independently by UNHCR offices in neighboring countries and through contacts established in Rwanda. According to the more recent of these reports, the types of activities described later herein in the south/south-eastern region have continued during the first half of September.] These accounts, if accurate, compel certain unavoidable findings which are set forth later herein.
The Government of Rwanda permitted the team full, independent freedom of movement, without accompaniment and control, through the issuance at the outset of the mission of an official letter of authorization which, with few exceptions, was honored throughout the country.
Genocide of Tutsis
The Government of Rwanda, the UNHCR and the international community have placed considerable emphasis on the early return of all refugees, and particularly those who fled Rwanda during the past year. Of these, the refugees currently remaining in neighboring countries are in the main Hutu people. It was not within the team’s mandate, nor was it able to conduct a systematic inquiry into the killing in April 1994 of hundreds of thousands of Tutsi people and many Hutu people by soldiers, militia and surrogate forces of the former Government of Rwanda. The killings of the Tutsi people have been described” by the Special Rapporteur of the Commission on Human Rights in his June 28, 1994 report, as genocidal in nature.
Due to the nature of its mission, the team’s findings concern principally the conduct of the Rwandan Patriotic Army (RPA), the military forces of the current Government, with respect to the Hutu residents of Rwanda. [Previously known as the Rwandan Patriotic Front (RPF), this force has since been redesignated as the Rwandan Patriotic Army (RPA) and is referred to throughout this document as such.] However grave the team’s findings with respect to RPA conduct, they do not mitigate, nor should they be permitted to obscure, the genocidal violence unleashed against the Tutsi people in April 1994 by forces of and associated with the previous Government.
Based on information gathered through the procedures described earlier, following are the principal findings of the Emergency Repatriation Team which affect the prospects for an early large-scale repatriation of Hutu refugees currently residing in neighboring countries:
WITH RESPECT TO BUTARE. KIBUNGO. AND PARTS OF KIGALI PREFECTURES…
1. Significant areas of Butare Prefecture, Kibungo Prefecture, and the . southern and eastern areas of Kigali Prefecture have been — and in some cases were reported to remain as of early September – the scene of systematic and sustained killing and persecution of the civilian Hutu populations by the RPA.
These activities are reported to have begun, depending on location, between April and July 1994, immediately following the expulsion from each area of former Government military, militia and surrogate forces. These RPA actions were consistently reported to be conducted in areas where opposition forces of any kind – armed or unarmed – or resistance of any kind – other than attempts by the victims of these actions lo escape — were absent. Large-scale indiscriminate killings of men, women, children, including the sick and the elderly, were consistently reported. The reported violence includes:
– Mass killings at meetings. Local residents, including entire families, were called to community meetings, invited to receive information about “peace,” “security” or “food distribution” issues. Once a crowd had assembled, it was assaulted through sudden sustained gunfire; or locked in buildings into which hand-grenades were thrown; systematically killed with manual instruments; or killed in large numbers by other means. Large-scale killings which did not involve such “meetings” were also reported.
House-to-house killings, and attacks on villages and displaced populations. — Pursuit of hidden populations. In response to the above actions, significant segments of the peculation flee into hiding in swamps, bush areas, banana plantations and other areas of difficult access to RPA soldiers. In many cases, they remain in hiding for extended periods of time. RPA soldiers’, in a few more recent cases accompanied by civilian Tutsi surrogates armed with spears and other manual weapons, were reported to actively pursue the hidden population through:
–Sudden, apparently well-coordinated attacks with gunfire;
–Silent attacks in which hidden groups are killed with manual weapons;
–Burning of swamp areas to prompt movement by civilians who are then killed;
–Periods in which operations are suspended, followed by invitations to the hidden families to return home in peace. Shortly thereafter the villages are attacked and returnees are killed.
- Killing of asylum seekers. Asylum seekers – particularly those fleeing from Rwanda in a southerly direction towards Burundi are reportedly systematically intercepted, ambushed and killed in significant numbers. These actions are reported with particular frequency in the commonly-used routes via road and through the countryside which pass through Matongo and Musenyi in southern Birenga Commune, and in areas close to the river which defines the Rwanda/Burundi border.
- Killing of returnees. Hundreds of Rwandan refugees in Burundi have heard Government of Rwanda radio broadcasts inviting and calling on the refugees to return home, and some of the refugees tried to do so. Some attempted to return to Rwanda with their entire families; others sent one or two individuals as “advance scouts” to confirm that conditions are peaceful. Still others traveled to their home areas to find and/or bring their relatives back to Burundi, or to harvest food to bring back to Burundi for their own consumption or for sale. Regardless of the motivation for their return, those who attempted to return and who were able to flee back to Burundi report consistently eye-witnessing killings and seeing the relatively fresh bodies of their Hutu neighbors and relatives.
–Killing of sick and elderly. Some individuals who are too old or too sick to flee to a neighboring country or to hide in the swamps, remain at their homes. Reports indicate that at least some of these are killed by RPA soldiers.
The foregoing activities have not been reported to be carried out throughout the geographical areas described above, nor carried out simultaneously in the areas which are affected. Nonetheless, an unmistakable pattern of systematic RPA conduct of such actions is the unavoidable conclusion of the team’s interviews.
More than 80% of the interviewees in this region — both inside Rwanda and in six refugee camps in neighboring countries provided detailed, credible reports of the conduct summarized above.
As the team traveled extensively within Rwanda through some areas close to the borders with Burundi and Tanzania, no RPA military personnel or checkpoints reported or alerted me team to security danger in these areas which it might encounter, or made reference to the fact or possibility of armed conflict between the RPA and former Government soldiers or militia conducting incursions into these areas. During its journey, the team often took care to check with RPA military personnel and was always assured that it faced no security problems in the conduct of its travels.
[Two incidents of the killing of Tutsi returnees by Hutu men armed with machetes were reported by Tutsi interviewees in two locations. The interviewees characterized the events as isolated incidents. One incident involved the killing of a Tutsi returnee from Burundi who appropriated the house of the Hutu owner who was still hiding in a nearby area. The second incident involved the killing of a Tutsi family by six Hutu men armed with manual weapons said to have returned to their home area from Burundi.]
2. Manner of killings. According to the interviewee reports, a common manner of effecting these killings is through the use of hoes to the skulls of the victims, or of machete blows to their heads and necks. The burning of victims in rural houses which have thatch roofs was widely reported. Gunfire is reportedly used in large-scale massacres and attacks. Otherwise, according to the interviewee reports, it seems to be used judiciously, and principally against fleeing males, who appear to be the priority — but by no means exclusive -targets of these actions.
3. Disposal of bodies. When local residents remain in an area where such killings have taken place, particularly when they know the victims and the number of bodies is manageable for them, and after RPA forces withdraw from the area, the remaining local residents themselves undertake some burial or covering of bodies. Eyewitness reports indicated that RPA forces sometimes burn or otherwise dispose of large numbers of bodies of those killed in public buildings. Large numbers of bodies are reported to remain above the ground, where some are partially consumed by dogs. Some formerly displaced persons and others have reportedly been ordered or obliged by RPA forces to bury some of the remaining bodies.
4. Culpability of victims. It appeared that the vast majority of men, women and children killed in these actions were targeted through the pure chance of being caught by RPA. No vetting process or attempt to establish the complicity of the victims in the April 1994 massacres of the Tutsis was reported.
5. Number of deaths. It is estimated that from late April/May through July, more than 5,000, and perhaps as many as 10,000 persons per month may have been killed in the manner described above. During August, while some 60,000 new refugees were arriving in Burundi and Tanzania, the number killed may have been somewhat less than in previous months, but probably at least 5,000. By that time, the people in many cases no longer responded to calls to attend RPA meetings and fled their homes when RPA soldiers approached. Also by then, many were dead or in exile in Burundi and Tanzania.
Through extensive travel on secondary and tertiary roads, the team observed that significant geographical areas of southern Kibungo and southern and eastern Butare prefectures were virtually deserted, while others were partially populated with varying degrees of apparent stability and security. In order to not compromise the security of the interviewees, the team declined the frequent offers of local residents to visit locations which they asserted were within walking distance to observe what they described as the relatively fresh bodies of Hutu people killed in the incidents which they had reported. Nonetheless, the team directly observed 150 bodies which appeared to have been dead for about 7 -10 days and which included men, women and children. The strong smell of deteriorating bodies was often present in the areas visited by the team.
During the last few days of August and the first days of September, new refugee arrivals to Tanzania diminished somewhat, but remained substantial. The lower arrival rates could be attributable to the factors described in the preceding paragraph. The systematic interdiction and killing of asylum seekers, described earlier, may be discouraging those who feel compelled to flee from attempting to do so. The communal origins of the latest arrivals suggest that reported RPA actions may, at that time, have been concentrated in areas more distant from the border, for example inTeastern Kigali Prefecture. It could also be that such actions were, at that time, at least temporarily diminished. With the exception of the returnee cases reported within paragraph one, above, no significant return to Rwanda of Hutu people from Burundi or Tanzania was reported or observed during this period.
WITH RESPECT TO RUHENGERI PREFECTURE…
6. Ruhengeri Prefecture.- In striking contrast to the southern and southeastern regions described above, refugee returnees from Zaire (of whom there were more than 100,000) and local residents in parts of western Ruhengeri visited by the team (including many who had been temporarily internally displaced in July and had since returned to their homes) reported overwhelmingly that conditions in their area were secure, stable and peaceful at the time of the team’s visit in early August. A liberal border-crossing policy by local authorities permitted families to walk home for brief exploratory visits, confirm for themselves the security situation, and return to Zaire to bring back their families.
7. Killing of returnee group. Based on credible reports gathered by the UNHCR office in Zaire from two seriously wounded survivors, it appears that on August 3 a group of up to 150 refugee returnees, including entire families of men, women and children who were walking from Zaire to northern Ruhengeri, were intercepted, questioned and killed by RPA soldiers just at the Zaire/Rwanda border north of Kinigi. To date it appears that, if the report is accurate, this mayihavo, been an isolated incident in this prefecture.
WITH RESPECT TO GISENYI PREFECTURE…
8. Unclear situation. In several Gisenyi communes visited by the team, less than 10% of the normal population appeared to be present. The proportion of population present in other communes was considerably higher, though in general lower than Ruhengeri Prefecture at that time.
Consistent reports in Gisenyi Prefecture described a systematic pattern of arbitrary arrests and disappearances of adult males, all of whom were alleged to be suspected of being militia elements associated with the former Government. Of an estimated 400 or more arrests, it appeared that half may have been refugee-returnees from Zaire, some of whom were intercepted as they attempted to cross the first checkpoints in the prefecture. In some cases, an informal (though unevenly administered) vetting process to determine if those arrested were, in fact, militia or had participated in the April 1994 massacres, was reported. The execution of at least dozens of those arrested was credibly reported.
Both arrests and physical abuse of local residents were credibly reported in some areas of Gisenyi prefecture to which refugees had returned from Zaire and in some sectors from which the population had never fled. Local residents stated that they were considering fleeing to Zaire. In fact, a small trickle of refugees was noted to have arrived in Zaire from southern Gisenyi during the team’s visit to the Goma area during the following days. Finally, the liberal border-crossing policy which permitted refugees to visit their homes to assess the situation and then return for their families was tightened up. An attitude of impatience with the fefugees who had thus far failed to return to Gisenyi was manifested by some local civilian surrogates of the RPA. Such attitudes had been entirely absent in Ruhengeri during the team’s visit.
WITH RESPECT TO REASONS FOR FLIGHT TO ZAIRE…
9. Reason for flight. Contrary to early working assumptions, Rwandan refugees interviewed in Zaire did not refer to calls by Radio Mille Collines for their departure from Rwanda as determining factors in their decisions to flee. Roughly half of the interviewees in Zaire indicated that they had been urged to flee by former Government civilian and military officials in their home areas, or that on their own volition they had decided to flee with or immediately ahead of former army soldiers because of a general fear of the RPA not necessarily linked to direct negative experience. Many simply cited panic as their motivation. Those reporting these motivations tended to come from western Ruhengeri Prefecture and from parts of Gisenyi Prefecture.
However, half of the interviewees in Zaire reported eyewitnessing at the time’of their flight or in earlier contact with the RPA acts of violence by the RPA against Hutu civilians which were similar in type — though far less in frequency — to those reported in Kibungo, Butare and southern and eastern Kigali Prefectures. They included meetings which evolved into massacres and attacks on civilian villages. Those reporting such violence tended to-come from eastern Ruhengeri and some parts of Byumba Prefecture.
WITH RESPECT TO REASONS FOR RETURN FROM ZAIRE TO NQRTHERN RWANDA…
10. Reason for return. Returnees to Ruhengeri Prefecture from Zaire cited their perception of its relative security, and the messages they had received from [illegible] effect, as their principal motivations for return. A survey on one day [illegible] August of eighty returnee families who were walking along the Gissnyi/Ruhengeri road bound for different destinationsln Rwanda revealed that in these families, one out of five of the Immediate family members with whom they had arrived In Zaire had died there. The mortality rate reported by the families still living In the refugee camps In Zaire was reported by them to- be significantly less. This appeared to substantiate assertions by refugees still in Zaire that those who were In the worst physical condition and who had the least food were a high proportion of those who had overcome their fear and decided to return to Rwanda.
The confiscation by force from refugee families In Zaire by soldiers and militia of the former Government of the food, blankets and plastic sheeting provided (In some cases by airlift) by the intemational community through UNHCR and Its implementing partners was reported as a widespread problem which was affecting the relief operations. The deprivation caused by such actions appeared to be contributing to the hunger and sickness which seemed to be driving at least some of the refugees to return home at that time.
11. The overwhelming majority of refugees in Burundi, Tanzania and Zaire expressed a wish to return as quickly as possible to their homes. If they were in good physical condition, most could probably return home by foot in a journey of a day or so. They stated that their main –.and in most cases their only constraint to doing so was the fear of being killed in their home areas, either upon their return or at a later time. Many asked that the United Nations intercede with the current Government of Rwanda (and particularly with the RPA) to cease the killing so that they could return home promptly.
12. In order for both spontaneous and organized repatriation to proceed, the Government of Rwanda must create conditions conducive for the return in safety of the refugees. Progress will have to be closely monitored by the UN system both in towns along the main road and through deployment in a permanent and sustained manner particularly in off-the-road rural areas (such as specific commune capitals) to insure that the actions reported herein are promptly and durably stopped.
UNHCR Emergency Repatriation Team