Citation
The International Response to Conflict and Genocide:
Lessons from the Rwanda Experience
March 1996
Published by:
Editor:
Cover illustrations:
Graphic design:
Prepress:
Printing:
Steering Committee of the Joint Evaluation of Emergency Assistance
to Rwanda
David Millwood
Kiure F. Msangi
Designgrafik, Copenhagen
Dansk Klich‚, Copenhagen
Strandberg Grafisk, Odense
ISBN:
ISBN:
ISBN:
ISBN:
ISBN:
87-7265-335-3 (Synthesis Report)
87-7265-331-0 (1. Historical Perspective: Some Explanatory Factors)
87-7265-332-9 (2. Early Warning and Conflict Management)
87-7265-333-7 (3. Humanitarian Aid and Effects)
87-7265-334-5 (4. Rebuilding Post-War Rwanda)
This publication may be reproduced for free distribution
and may be quoted provided the source
- Joint Evaluation of Emergency Assistance to Rwanda is mentioned.
The report is printed on G-print Matt, a wood-free, medium-coated paper.
G-print is manufactured without the use of chlorine and marked with the Nordic Swan,
licence-no. 304 022.
2
The International Response to Conflict and Genocide:
Lessons from the Rwanda Experience
Study 2
Early Warning
and Conflict Management
by
Howard Adelman
York University
Toronto, Canada
Astri Suhrke
Chr. Michelsen Institute
Bergen, Norway
with contributions by
Bruce Jones
London School of Economics, U.K.
Joint Evaluation of Emergency Assistance to Rwanda
3
Contents
Preface
5
Executive Summary
8
Acknowledgements
11
Introduction
12
Chapter 1: The Festering Refugee Problem
17
Chapter 2: Civil War, Civil Violence and International Response
(1 October 1990 - 4 August 1993)
Track I: Civil war and the response
Track II: Civil violence and human rights conditionality
20
Chapter 3: Implementing and Subverting the Peace Agreement
(4 August 1993 - 6 April 1994)
Implementing the peace agreement
Civil violence and the subversion of the peace agreement
34
Chapter 4: Crisis and Withdrawal
(6 April 1994 - 21 April 1994)
Crisis
Withdrawal
The media
40
Chapter 5: Retroactive Conflict Management
(29 April 1994 - August 1994)
Changing the mission: UNAMIR II
Deploying UNAMIR II
Opération Turquoise
48
Chapter 6: Refugees and Internally Displaced Persons:
The End and the Beginning of Conflict
(May 1994 - May 1995)
The refugee camps: humanitarian and security issues
Searching for solutions
Internally displaced: Kibeho
57
Chapter 7: Conclusion
Early warning
Conflict management
65
65
70
20
27
34
36
40
42
45
48
51
53
57
59
61
4
Chapter 8: Recommendations
General - humanitarian realism
Policy approaches to conflict management
Early warning
The parties
76
76
77
79
80
Endnotes
82
Appendix: References
Interviews
Bibliography
UN documents
Other official documents
Newspapers and periodicals
100
100
101
107
108
109
5
Preface
Within a period of three months in 1994, an estimated five to eight hundred thousand people
were killed as a result of civil war and genocide in Rwanda. Large numbers were physically and
psychologically afflicted for life through maiming, rape and other trauma; over two million fled
to neighbouring countries and maybe half as many became internally displaced within Rwanda.
This human suffering was and is incomprehensible. The agony and legacy of the violence create
continuing suffering, economic loss and tension both inside Rwanda and in the Great Lakes
Region.
For several years preceding the massive violence of 1994, the international community
contributed to efforts to find a peaceful solution to escalating conflict and provided substantial
assistance to alleviate the human suffering. During the nine months of the emergency in 1994,
April to December, international assistance for emergency relief to Rwandese refugees and
displaced persons is estimated to have cost in the order of US$ 1.4 billion, of which about onethird was spent in Rwanda and two-thirds in asylum countries. This accounted for over 20% of
all official emergency assistance, which in turn has accounted for an increasing share, reaching
over 10% in 1994, of overall international aid.
This growth reflects the worldwide proliferation in recent years of so-called complex
emergencies. These tend to have multiple causes, but are essentially political in nature and entail
violent conflict. They typically include a breakdown of legitimate institutions and governance,
widespread suffering and massive population displacements, and they often involve and require
a range of responses from the international community, including intense diplomacy and conflict
resolution efforts, UN policing actions, and the provision of multilateral and bilateral
humanitarian assistance by official and private agencies. A complex emergency tends to be very
dynamic, characterized by rapid changes that are difficult to predict. Thus complex issues are
raised regarding the timing, nature and scale of response. The Rwanda complex emergency
shares all these characteristics and more.
Although some evaluations of international assistance for complex emergencies have been carried
out, experience from the planning and execution of large-scale aid for relief, rehabilitation and
reconstruction has not been extensively documented and assessed. Recognizing both the
magnitude of the Rwanda emergency and the implications of complex disasters for constricted
aid budgets, the Danish Ministry of Foreign Affairs, through its development cooperation wing,
Danida, proposed a Joint Evaluation of Emergency Assistance to Rwanda.
This initiative resulted in the launching of an unprecedented multinational, multi-donor evaluation
effort, with the formation of a Steering Committee at a consultative meeting of international
agencies and NGOs held in Copenhagen in November 1994. This Committee1 is composed of
representatives from 19 OECD-member bilateral donor agencies, plus the European Union and
theDevelopmentAssistance Committee (DAC) of the OECD; nine multilateral agencies and UN
units; the two components of the International Red Cross and Red Crescent Movement (ICRC
and IFRC); and five international NGO organizations
1 Australia, Austria, Belgium, Canada, Denmark, Finland, Germany, Ireland, Italy, Japan, Luxembourg, Netherlands, New
Zealand, Norway, Spain, Sweden, Swizerland, United Kingdom, United States of America, Commission of the EU,
OECD/DAC, IOM, UN/DHA, UNDP, UNHCR, UNICEF, WFP, WHO,IBRD, ICRC, IFRC, ICVA, Doctors of the World,
INTERACTION, Steering Committee for Humanitarian Response, VOICE. Several other countries supported the evaluation,
but did not participate actively. France suspended its participation in the Steering Committee in December 1995. The cost
of the evaluation has been met by voluntary contributions from members of the Steering Committee.
6
Objective of the Evaluation2
The main objective of the evaluation is to draw lessons from the Rwanda experience relevant for
future complex emergencies as well as for current operations in Rwanda and the region, such as
early warning and conflict management, preparation for and provision of emergency assistance,
and the transition from relief to rehabilitation and development.
In view of the diversity of the issues to be evaluated, four separate evaluation studies were
contracted to institutions and individuals with requisite qualifications in the fields of (i)
emergency assistance planning and management; (ii) repatriation and rehabilitation of refugees;
(iii) history and political economy of Rwanda and the surrounding region; (iv) institution and
capacity building in development; (v) conflict and political analysis; and/or (vi) socio-cultural and
gender aspects. Institutions and individuals were also selected for their proven ability to perform
high-quality, analytical and objective evaluative research.
The institutions and principal individuals responsible for the four reports are listed below. Space
precludes listing all team members for each study, which ranged from four persons for Study I
to 21 for Study III; in all, 52 consultants and researchers participated. Complete identification of
the study teams may be found in each study report. Several of the studies commissioned substudies that are also identified in the respective study report.
Study I:
Historical Perspective: Some Explanatory Factors
The Nordic Africa Institute (Uppsala, Sweden)
Tor Sellström and Lennart Wohlgemuth.
Study II: Early Warning and Conflict Management
Chr. Michelsen Institute (Bergen, Norway)
York University (Toronto, Canada)
Howard Adelman and Astri Suhrke.
Study III: Humanitarian Aid and Effects
Overseas Development Institute (London, United Kingdom)
John Borton, Emery Brusset and Alistair Hallam.
Study IV: Rebuilding Post-Genocide Rwanda
Center for Development Information and Evaluation,
US Agency for International Development; Development Alternatives, Inc.;
Refugee Policy Group (Washington, DC, USA)
Krishna Kumar and David Tardif-Douglin.
Evaluation oversight was performed by the Steering Committee (which held four meetings
between December 1994 and December 1995), and by a Management Group, comprised of one
lead bilateral agency for each study: Study I: Claes Bennedich, Sida, Sweden; Study II: Jarle
Hårstad, Ministry of Foreign Affairs, Norway; Study III: Johnny Morris, ODA, United
Kingdom; and Study IV: Krishna Kumar, USAID/CDIE, USA; and Niels Dabelstein, Danida,
Denmark, as chair. The evaluation teams were responsible to the Management Group and the
Steering
Committee
for
guidance
regarding
2 See Appendix I of the Synthesis report for the full Terms of Reference.
7
such issues as terms of reference and operational matters, including time frames and budget
constraints, and they were obliged to give full and fair consideration to substantive comments
from both groups. The responsibility for the content of final reports is solely that of the teams.
The approach taken to this evaluation has reflected two concerns:
•
•
to try, through involving experienced outsiders, to examine as objectively and critically as
possible an experience about which it is impossible for any person with humane values not to
be deeply affected;
to engage leading Africans in a critical review of the analysis, findings and recommendations
while they were still in draft.
For this last reason, a panel of distinguished experts from Africa has provided a critique of the
report through participation in two panel discussions with the authors of the reports and selected
resource persons. The panel comprised: Reverend Jos‚ Chipenda, General Secretary, All-Africa
Conference of Churches, Kenya; Dr. Adama Djeng, President, International Commission of
Jurists, Switzerland; Professor Joseph Ki-zerbo, Member of Parliament, Republic of Burkina
Faso; and Dr. Salim A. Salim, Secretary General, Organization of African Unity, Ethiopia. Also,
Mr. Gideon Kayinamura, Ambassador of Rwanda to the UK; Ms. Julie Ngiriye, Ambassador of
Burundi to Denmark; and Ms. Victoria Mwakasege, Counsellor, Embassy of Tanzania,
Stockholm, made significant contributions through their participation in the December 1995
Steering Committee Meeting.
While the Steering Committee is particularly grateful to these African participants for contributing
their wisdom and keen insights at one stage of the evaluation process, it is also acutely aware of
the fact that African researchers and institutions were not, with the exception of selected substudies, involved in its execution. However, the Steering Committee is committed to disseminate
the evaluation widely among African leaders and organizations and anxious that they participate
fully in discussions about the evaluation's recommendations.
The following resource persons have commented on drafts at various stages and/or participated
in panels or workshops: Mary B. Anderson, Consultant, USA; Hanne Christensen, Independent
Bureau for Humanitarian Issues, France; John Eriksson, Consultant, USA; Professor Andr‚
Guichaoua, Universit‚ des Sciences at Technologies de Lille, France; Sven Hamrell, Dag
Hammarskjöld Foundation, Sweden; Larry Minear, Humanitarianism and War Project, Brown
University, USA; Professor Rodolfo Stavenhagen, Colegio de Mexico, Mexico; and Stein
Villumstad, Norwegian Church Aid, Norway.
The Synthesis Report was prepared by John Eriksson, with contributions from the authors of the
four study reports and assistance from Hanne Christensen and Stein Villumstad in the
preparation of findings and recommendations.
This evaluation was initiated on the premise that in spite of the complexity and chaos that
characterize Rwanda's experience, it would be possible to identify applicable lessons to be
learned by the international community in attempting to respond to future complex emergencies
and in its continuing attempt to help Rwanda rebuild its society. The international teams who
have produced this evaluation believe they have identified such lessons. It will be up to the
governmental and non-governmental leaders of the international community for whom this
evaluation has been prepared to apply the lessons.
Niels Dabelstein
Chairman of the Steering Committee for
Joint Evaluation of Emergency Assistance to Rwanda
8
Executive Summary
Using the techniques of critical policy analysis, this report examines the effectiveness of early
warning and conflict management of the Rwanda conflict. The study begins with the refugee
problem prior to 1990 and ends with the genocide of 500-800,000 persons, mainly belonging to
the minority Tutsi community, in the second quarter of 1994. Security issues related to the
subsequent formation of refugee camps are also examined.
Once the civil war broke out in 1990, increasingly there were warnings, supported by evidence,
that large-scale civilian massacres might occur. Nevertheless, virtually no-one anticipated genocide
on the scale that took place. Preparations to deal with the contingency of massive violence that
targeted civilians were inadequate.
By failing to deal with the festering refugee problem prior to 1990, both the Rwandese and the
Ugandan governments set the stage for future conflict. The build-up of tension leading to the 1990
invasion by the Rwandese Patriotic Front (RPF) was inadequately monitored. Once the civil war
started, however, it triggered an intense and effective diplomatic process that eventually secured a
peace agreement. The process was initiated and led by the Organization of African Unity (OAU)
and the regional states, particularly Tanzania, and received considerable international attention and
support. The United Nations assumed formal responsibility for overseeing the implementation of
the consequent Arusha Accords, but failed to make adequate use of the OAU and local African
states in this regard. As a consequence, there was a disjuncture between the mediation and
implementation phase, which contributed to undermining the Accords. Throughout, insufficient
attention was paid to the warnings from human rights organizations and other sources that
extremist forces linked to the regime were progressively organizing and arming themselves to
derail the peace process and massacre the minority group.
The UN force mandated to oversee the implementation of the peace agreement (UNAMIR) was
structured and financed to satisfy a cost-conscious United States increasingly unwilling to support
UN peacekeeping. The force was inadequately supported and slowly deployed, despite warnings
that speed was essential to maintain the momentum of the peace process. The operation had no
flexibility to respond to changing circumstances such as those caused by the crisis in neighboring
Burundi in October 1993. Unequivocal warnings reached the United Nations in January 1994
regarding a planned coup, an assault on the UN forces to drive them out, provocations to resume
the civil war, and even detailed plans for genocide. The cable was placed in a separate Black File,
designed to draw attention to its content, and circulated to several departments in the
UNSecretariat. However, senior officials in the Secretariat questioned the validity of the
information and made no contingency plans for worce-case scenarios. Similar intelligence failures
were evident on the state level, particularly in France and Belgium, both of which had a
considerable capacity for overt and covert information gathering in Rwanda at the time. Generally,
the UN Secretariat interpreted UNAMIR's mandate and terms of engagement narrowly, and on
several occasions denied the Force Commander permission to search and seize arms caches. When
developments in early 1994 further eroded the peace accords, the Secretary-General and the
Security Council threatened to withdraw the UN force, hence strengthening the hands of the
extremists.
When the crisis came to a head on 6 April, there was an absence of leadership at UN headquarters
in New York. The Secretary-General misread the nature of the conflict. The killing of 10 Belgians
created a political surge in the Security Council to withdraw, although this was not recommended
by UNAMIR's Force Commander or African countries contributing troops.
9
Once the direction and magnitude of the genocide became undeniable, the UN reversed itself and
accepted an obligation to protect civilians. However, the realization of this peacekeeping mission
(UNAMIR II) was hampered by the unwillingness of key Council members to pay for or provide
troops, and to match African troops with equipment in an expeditious manner. The force was
deployed only after the genocide and the civil war had ended. The French Op‚ration Turquoise
was deployed with great speed and efficiency, but the political decision to intervene was only
taken two and a half months after the genocide commenced and when the civil war was almost
over. The intervention stemmed the mass outflows of refugees in the south-west, but in some
respects was compromised by France's close relations with the former regime.
When massive numbers of refugees, retreating government forces, and the assumed perpetrators
of the genocide crossed into Zaire and Tanzania in April-July 1994, the UN High Commissioner
for Refugees (UNHCR) warned the UN in New York about the attendant security problems. The
Secretariat took the unprecedented step of examining the issue in a peacekeeping context, but the
Security Council provided no support. The problem fell back into the hands of UNHCR, which
resorted to a novel and reasonably effective solution to police the refugee camps. The arrangement
did not and could not deal with the broader security threats posed by the existence of militarized
communities in exile, and these were left to fester. The large concentrations of internally displaced
persons in south-west Rwanda presented a domestic version of similar problems. These came to a
head in the operation when the Kibeho camp was closed in April 1995 with a large loss of life.
Nevertheless, the efforts and planning that went into the coordination of UN agencies, NGOs and
the Rwandese government were clearly steps in the right direction.
The report attempts to explain why the signals that were sent were not attended to or not translated
into effective conflict management.
Failures of early warning are attributable to many factors. To briefly summarize:
•
•
•
•
•
•
•
As a universal membership organization, the UN is poorly suited to collect and flag
information about human rights violations and certainly genocide in member states;
Failure in both the UN system and the NGO community to link human rights reports to
dynamic analyses of social conflict;
An internal predisposition on the part of a number of the key actors to deny the possibility of
genocide because facing the consequences might have required them to alter their courses of
action;
The mesmerization with the success of Arusha and the failure of Somalia, which together cast
long shadows and distorted an objective analysis of Rwanda;
The vast quantity of noise from other crises that preoccupied world leaders;
The confusion between genocide as a legal term referring primarily to an intent, and the
popular association of genocide with massive murder in the order of hundreds of thousands;
Finally, the general desensitization that has emerged to mass slaughters, and the incredulity
that a massive genocide might actually occur.
The media, with some exceptions, played an irresponsible role in their reporting on Rwanda. The
overall failure of the media to report accurately and adequately on a crime against humanity
significantly contributed to international disinterest in the genocide, and hence to the inadequate
response.
Conflict management, however, is a function of capacity, interest and commitment as well as
information. In the Rwanda case, the relevant actors knew at a critical stage that the situation was
unstable and dangerous. Yet the sustained and careful attention so necessary to successful conflict
management was lacking. Admittedly, some individual or collective actors did the most with the
least under difficult or adverse circumstances. Human rights NGOs monitored the situation
10
throughout. Tanzania struggled to turn the Arusha process into effective preventive diplomacy.
UNAMIR I tried to function proactively despite tight reins prior to 6 April, and many remaining
units - along with ICRC - bravely sought to save civilians once the killings started. This could not
compensate, however, for the overall failure of the international community to prevent, mitigate or
stop the genocide. When the crisis struck, and it later became clear that a massive genocide was
unfolding, there was still no effective international action.
In one sense, this inaction can be seen as a propensity of states to be guided by narrow selfinterest rather than moral obligations to uphold international norms of justice. However, this
propensity has historically varied over time and place. In this case, it was allowed significant play
because of a structural mismatch between responsibilities of institutions and interests in the
contemporary world.
Revitalized by the end of the Cold War, the United Nations in the 1990s rapidly expanded its
peacekeeping operations. Rwanda was added to the list in October 1993. The framework for
peacekeeping, however, was set by the distribution of power in the Security Council, which
represented the world as it was half a century ago. Except for France, the major powers on the
Council were uninterested in a small Central African country that was marginal to their economic
or political concerns, and peripheral to international strategic rivalries. By their power of veto and
finances, the Permanent Five controlled the peacekeeping and enforcement operations of the UN.
The only state with a demonstrated ability to energize the Council in a crisis – the United States –
was haunted by the memories of Somalia and determined not to get involved in another African
conflict. Washington was also preoccupied with crises elsewhere, especially in Bosnia and Haiti.
Hence, a principal lesson from the Rwanda conflict is that in a world of multiple crises, even
major disasters in a seemingly peripheral state may fall victim to neglect.
As the major Western states were uninterested in Rwanda, and the regional states were unable or
unwilling to maintain a sustained presence, France was left to define a large part of the policy
field. The result was to magnify the consequences – negative as well as positive – of unilateralism.
The consequence of these cumulative fault lines in the international system was a genocide of
immense proportions.
The report ends with a number of suggested recommendations that follow from the analysis and
are consistent with the principles of humanitarian realism.
11
Acknowledgements
We would like to acknowledge a number of individuals without whose assistance this report could
not have been completed. First, the members of the inter-governmental committees supervising the
study were always supportive and understanding without trying to influence the outcome. Second,
we wish to thank the other teams involved in the study. It was recognized from the beginning that
the teams had overlapping areas to cover and we found that the teams openly and willingly shared
information and materials. We are particularly grateful to the leaders of Study III, with whom we
had the greatest overlap. Third, this study has depended on the contributions of numerous persons
and agencies. Several academics and scholars in effect became members of an expanded team:
their research is listed in the bibliography and acknowledged in the text. A much larger group of
academics and other observers gave valuable comments and some read parts of the draft report. A
large number of officials in Africa, Europe and North America gave of their valuable time, taken
from already overburdened schedules, to provide information, analysis, interpretations and
explanations, and patiently subjected themselves to being questioned. Many shared documentation
to ensure factual accuracy. Their agencies and organizations are listed in the Appendix.
Workshops to test out ideas and receive critical feedback were held at the Centre for Refugee
Studies, York University in Toronto, and the New School for Social Research in New York. An
international conference on Conflict Resolution in Africa organized by the Chr. Michelsen Institute
in Bergen helped to crystallize some of the conclusions.
We would also like to thank the Chr. Michelsen Institute and the Center for Refugee Studies at
York University for providing institutional and intellectual support. The graduate and postgraduate students at the Center for Refugee Studies made a singularly valuable contribution by
assisting with data collection and index preparations.
Finally, and not least of all, we would like to thank our spouses and children, who had to put up
with an intense travel schedule and our absence even when physically present as we struggled
dispassionately to dissect a human catastrophe.
Bergen and Toronto
December 1995
A.S., H.A.
12
Introduction
Analytical Framework
Subject matter
This report examines the effectiveness of international monitoring and management of the Rwanda
conflict, which culminated in genocide of half a million to one million persons in the second
quarter of 1994. The report has two distinct yet closely related foci: (i) the existence and
effectiveness of early warning mechanisms anticipating escalations in violent conflict and mass
murder; (ii) the nature and appropriateness of the responses of significant international actors
towards the evolving conflict.1 Since the 1990 invasion marked the opening of the conflict cycle
that culminated in genocide, the study begins with the factors leading up to the invasion of October
1990 and ends in early 1995. Within this time frame, different phases of the conflict are identified
and studied in varying degrees of depth. In the period after August 1994, consideration is given
only to conflict management issues concerning internally displaced persons and refugees. Given
the constraints of this study, early warning relating to providing humanitarian assistance is a
peripheral rather than central focus.
Process
Data collection for the study commenced at the beginning of 1995 and has been carried out by
Howard Adelman (York University, Toronto) and Astri Suhrke (Chr.Michelsen Institute,
Bergen), with the assistance of research associates working on assigned topics. These included:
Agnes Callamard, Shally B. Gachuruzi, Peik Johansson, Bruce Jones, Kate Halvorsen, Steven
Livingston and Todd Eachus, Ogenga Otunnu, Gérard Prunier, Turid Lægreid and Amare Tekle.
The research is based on secondary resources (books and articles by academics and journalists,
media studies, reports, etc.) as well as considerable primary data collected through interviews and
document searches in the UN system (New York and Geneva), the NGO community, and visits to
relevant national capitals in Europe and North America (Paris, Rome, Brussels, London,
Washington and Ottawa), and in Africa (Nairobi, Kigali, Kampala, and Dar-es-Salaam). Some of
the most valuable documentation was made available for exclusive use of this study. Nevertheless,
documents have limitations: a great deal of communication and key parts of the decision-making in
this case were not recorded. As for interviews, personal recollections of participants always pose
difficulties with respect even to basic facts; there is a need for caution, given the propensity to
rewrite history in the reconstruction of memory. This difficulty was compounded by the traumatic
nature of events in Rwanda and controversies about international responses and responsibilities.
With these constraints, we have tried to provide coherence in reconciling the various perspectives
on the crisis.
If the report does not adequately reflect the position of some parties, one reason is that
documentation and personnel were not available. Most of the persons interviewed have been
identified only by their institution (see Appendix), but some insisted on complete anonymity.
Although the research responsibility was initially divided according to issues – Adelman taking
primary responsibility for early warning and Suhrke for conflict management – the final product is
the responsibility of both. The dialectic of reaching a consensus on virtually all issues was a
discipline in itself.
13
Premises
This study falls outside the bounds of evaluation in a conventional sense. To address the questions
posed by the terms of reference required two tasks: sorting out the cause-and-effect relationships
of complex historical developments and assessing their consequences, not only in relation to
formal stated goals but equally in relation to higher standards of conflict management and
international norms. In these respects, the study follows in the tradition of critical policy analysis.
There is a recognition of competing goals, of unintended, unforeseen or indirect consequences, yet
a presumption of some choice and consequent responsibility by the actors concerned.
The report is based on some general premises concerning the interpretation of a complex social
conflict. The first premise is that historical actors, in comparison to the critical analysts of those
acts, interpret events and their significance from very different perspectives. Though both base
their assessments on inadequate information and communication, analysis of a conflict “from
within”, while it is evolving, differs significantly from ex post facto analysis. The analyst is in
some respects better positioned to understand and interpret what occurred. In retrospect, the
equivalent of early warning data may be interpreted differently and “more correctly” in light of
what actually happened; signals that at the time were ambiguous may now seem compelling, and
additional information may be available. The analyst may be constrained by lack of access to
information sources but can view the issues from several perspectives. Herein also lies the danger
of misrepresenting the ability of historical actors to interpret signals correctly within their own
situational context. A related distinction lies in the set of filters that screen incoming information;
the range of action that policy-makers consider realistic will influence the attention paid to
information, how they interpret it, and to whom it is communicated. In a feed-back mechanism,
this in turn influences the formulation of policy options.
Early warning is the collection, analysis and communication of the relevant evidence and
conclusions to policy-makers to enable them to make strategic choices.2 Unlike traditional
intelligence, which also collects and analyzes information and communicates the results, the object
of early warning is not primarily security for one’s self or one’s country, but the security of
another; in early warning, the security is not self-directed. The other party or parties are not
presumed to be adversaries, as is the case with intelligence analysis.
Human rights monitoring also involves the collection of information, analysis of that information
and communicating the results. But the prime audience is the public, even though human rights
reports are addressed to leaders. For human rights reports do not facilitate the decision-making of
those leaders, but rather serve as a superego, a haunting conscience for people in power. Like
early warning, it is other- rather than self-directed.
Early warning can be concerned with military conflicts, military coups, impending humanitarian
disasters, such as starvation and flows of refugees, slaughters, and, at the extreme, genocide. In
the Rwanda case, all of these were involved at different stages.
The spontaneous slaughter of one group by another, usually as a result of inter-ethnic strife, is not
generally referred to as genocide. Genocide must be the result of deliberate policies of the political
elite intent on wiping out the other group. According to the 1948 Convention, the “intent to
destroy” is an essential requirement for identifying genocide.3 Thus, understanding the political
process that leads to the genocide is critical. The decision of a political elite to foster or permit
genocide to occur may be the result of their implicit consent rather than explicit orders. In the
Rwandese case, no such subtle distinctions between implicit and explicit policies are necessary.
The genocide was actively promoted by political and military leaders, and there were apparently
concrete plans to kill at least all the Tutsi in Kigali. Once the killing commenced, there is a debate
on whether the leaders continued to manage the genocide until the end, or once the frenzy of
killing started and – given the organization in place – it acquired an additional momentum of its
own.
However, there need be no debate on whether the use of the term “genocide” is appropriate.
14
According to the 1948 Convention, intent to destroy a group “in whole or in part” constitutes
genocide, and the term was applied to describe even relatively small slaughters (300–2,000) in the
1990–93 period. In the public mind, by contrast, the term “genocide” is used only when
substantial proportions of a group are slaughtered. What happened after 6 April 1994 meets both
the legal and the popular meanings of the term.
According to what standards do we assess the anticipation of violent conflict? Were there relevant
structures for collecting and analyzing signals, and, if so, did these function in an efficient,
accurate and timely manner? What were the relevant structures? Who has – or should have –
responsibility for collecting and analyzing information and translating that analysis into options for
action? The assumption is that the responsibility for collecting and analyzing information flows
from the responsibility for action. That is, actors who played a role in the Rwandese conflict had
an obligation commensurate with their responsibilities to obtain and assess data about the evolving
violence. In each case, the overriding consideration must be whether the system(s) caught and
processed available information that at the time reasonably could be said to indicate that an event –
an invasion, an escalation of the conflict, genocide – was a possibility, likely, or already under
way.
In the ideal state, early warning serves prevention. In reality, preventive measures are among the
most difficult of all policies to formulate and carry out. The next best, as it were, is to focus on
conflict management. Here, the general criteria identified by the terms of reference are
“containment and de-escalation”. We see these as being derived from an ethical standard which
holds that the international community should have attempted to reduce violence in Rwanda in
general, and prevent genocide in particular. Given this premise, a specific response has to be
assessed not only in relation to its formal goals (e.g. of UNAMIR I); the question must also be
asked whether the formal mandate was appropriate in relation to a higher standard of conflict
reduction extant in the international community.
A number of generally recognized international norms for assessing conduct appropriate to conflict
reduction are relevant in the Rwanda case. The obligation exists for the parties to the conflict to
attempt to settle their differences by peaceful means, utilizing when ever possible the appropriate
international organization to which they belong. In refugee matters, countries of origin are obliged
to allow their nationals the right to be repatriated.4 States are not to permit invasions of other states
from their territories.5 There is an obligation not to intervene in the internal affairs of sovereign
states except in cases of genocide. These norms are not always compatible, either in principle or in
practice, nor were they in the Rwanda case. For instance, norms affirming the right of refugees to
return, and norms against using military force from outside the country to enforce that right
seemed to be at odds in practice. No similar incompatibility existed between the general strictures
against intervention, on the one hand, and the explicit legal right and associated moral obligation to
intervene to stop the mass slaughter of civilians according to the Genocide Convention.
If the first task of evaluation is to identify appropriate standards, a second is to assign causative
responsibility for decisions and their consequences. Different degrees of responsibility may be
assigned to different actors, and responsibility may vary over time. In the Rwandese case, that
meant, in the first instance, the Rwandese parties to the conflict. It also meant, however, that the
regional and European states most closely involved in the development of the conflict triggered by
the October 1990 invasion (especially Uganda and France as the principal associates of the major
Rwandese protagonists) bore significant responsibility. Third, regional actors (especially Tanzania
and the OAU) were moved by self-interest and formal obligations, and, in fact, played an active
and sustained mediating role in the pre-1994 period. Fourth, there were actors (other states, the
Catholic Church, NGOs) that chose to become involved for various reasons; they also had
commitments and responsibilities that varied with the extent of their involvement. Fifth, the UN
and its agencies became involved in a major way as the crisis in Rwanda threatened regional peace
and security. Finally, the media had a potentially important role, particularly in the development of
public awareness and public policy.
As for responsibility in the normative sense of obligation to act, the premise of this report is that
15
the principal though not exclusive responsibility for anticipating and managing conflict lies with
the parties themselves and their principal associates. In particular, insofar as the principals are
responsible for the development of the conflict, they have both a political/moral obligation and
some actual capacity for regulating it. Beyond that, international organizations have formal
obligations to assist in conflict management. In cases of genocide, those obligations are much
greater and apply to the members as well as the organizations.
It must not be forgotten that those most responsible were the perpetrators of the crimes. They have
been identified as extremists. In this context, extremists are presumed to have possessed two
essential characteristics. First, their actions were determined by an ideological view of the world –
in their case, the definition of Rwanda as a Hutu nation. Second, they regarded others who did not
fit into this definition and who threatened its realization as legitimate targets for murder.
Information gathering, analysis and communication, as well as the actions based on this, must not
be assessed only according to norms and attributed to agents based on their respective
responsibilities. These assessments must take into consideration the limitations at work. For
example, a system cannot be faulted for a failure at early warning if the information available was
small, inconsequential, of doubtful value, or ambiguous. Further, not all conditions within
Rwanda that led up to and accounted for the genocide were equally amenable to change. While
there are various theories about how and why the genocide occurred, most distinguish between
structural factors and situational variables. Structural factors, such as demographic density and the
shortage of land, can create a context conducive to social violence, but policy can only alter that
context in the long run and only with uncertain outcomes. Situational variables that had a
triggering or accelerating effect, such as arming militias or scaling down UNAMIR, were subject
to more immediate control through internal as well as external influences.
The same distinction between situational and structural variables is also applicable to those charged
with interpreting and responding to the crisis. A realistic assessment of the international response
to the Rwandese catastrophe must recognize the structural restraints of the existing international
system based largely on competitive nation-states. Current capacity at state, regional (OAU) and
international (UN) levels to identify and respond to conflicts of the kind that occurred in Rwanda
constitute structural constraints on action. The broader foreign policy interests of the major powers
provided the context within which specific policies towards Rwanda were formulated, both prior
to and during the 1994 violence, and represented structural constraints of a different kind.
Nevertheless, within these limitations, options were available; these constitute the focus of this
report. Structural problems (e.g. on the UN level) will be dealt with only insofar as they affected
the Rwanda case.6
The conclusions and associated recommendations are situation-specific. As such, they are not
necessarily relevant for warning or management of the conflict in Rwanda as it is currently
evolving, but rather to a set of circumstances and actions that have generic applicability.
Time frame for analysis
As recognized in the terms of reference, the period preceding the genocide is relevant for an
understanding of the signals emitted about the evolving conflict, and efforts to contain it through
diplomacy and military force. This report is organized around five phases of the conflict; these are
similar to those suggested in the terms of reference, but adjusted after closer examination of the
historical material.
Phase I covers the refugee problem in Uganda prior to October 1990. Chapter 1 examines how
the refugee issues became a source of renewed violence, given the failure of the Rwandese
government to permit return, the unwillingness of Uganda to accord full rights of settlement, and
whether there were early warnings of the invasion itself.
Phase II includes the period from the October 1990 invasion until the signing of the Arusha
16
Peace Agreement in August 1993. This period was characterized by an interlinked process of civil
war, civil violence and preventive diplomacy. The organized violence against unarmed civilians in
this phase constituted a kind of early warning to the effect that worse might be in store.
Simultaneously, the intense diplomatic activity resulted in a peace agreement between the
government of Rwanda (GoR) and the rebel Rwandese Patriotic Front (RPF). The political
democratization within Rwanda during this period was related to the dynamic of conflict and its
management.
Phase III focuses on the planning, authorization, and deployment of UNAMIR from mid-1993
until 6 April 1994, and its activities in early warning and conflict management against the backdrop
of the progressive unravelling of the Arusha peace accords.
Phase IV covers the two weeks from 6 April until 21 April 1994. The immediate aftermath of the
death of Rwanda’s President in a suspicious plane crash – when it soon became evident that largescale and targeted killings were under way – marks a distinct phase in terms of the international
response. The report focuses on the information available that genocide was immanent and had, in
fact, begun, and the corresponding responses, particularly as they relate to the question of what
role UNAMIR should play. The dynamic interaction of the management of the genocide and the
management of the conflict – or rather, the failure to distinguish between the two – is a central
concern. The analysis focuses on the decision-making process in the UN over the future of
UNAMIR, efforts by UNAMIR to carry out relevant parts of its mandate in the new situation, and
the eventual decision by the Security Council to scale down the force, leaving only a symbolic
military presence.
Phase V focuses on the subsequent decision to reverse withdrawal and re-engage the UN to
protect civilians. The study assesses the efficacy of this process and its general impact on both the
conflict on the ground and the violence against civilians. In this phase, another effort at retroactive
conflict management was undertaken, i.e. the UN-authorized French Opération Turquoise. This
section of the report relies on historical rather than sociological analysis, which would require
micro-level assessments in the field.
Phase VI examines the issue of the security of the internally displaced within Rwanda and the
establishment of militarized refugee communities in neighboring countries. This affected the
security of the refugees and led to a massacre of several thousand persons at the Kibeho camp
inside Rwanda. There were also wider implications in terms of conflict management. Social and
political elements of the defeated regime who were responsible for the genocide regrouped to
consolidate their power in exile. Thus, one element in a renewed cycle of conflict was formed.
17
Chapter 1
The Festering Refugee Problem
The invasion that triggered the disaster in Rwanda was itself the culmination of previous strife of a
type frequently seen in Africa: victims or losers in a conflict seek refuge in a neighboring country,
which then becomes a base for invading the homeland. In this case, Tutsi who were the principal
victims of political violence in Rwanda in the decolonization period (1959–1963) fled to the
surrounding countries, with large numbers settling in southern Uganda. The conflict that sparked
the original outflow, however, was not resolved but acquired an additional refugee dimension.
The norms for regulating problems of this kind were in place. With reference to African legal
instruments and other international law, the OAU and UNHCR had appealed to African states to
prevent refugee problems from generating renewed strife by following accepted principles of state
responsibility towards refugees: The country of origin must allow them full and free right of
return; failing this, receiving countries should facilitate settlement and integration. The OAU had
attempted to regulate problems associated with militant refugee communities in its 1969
Convention on Refugee Problems in Africa (1969), more generally in the African Charter on
Human and People’s Rights (1981) and also in the Charter of the OAU itself (1963). All
instruments unequivocally prohibit the use of the territory of one state for subversive activities
against another. The 1969 Convention specifically addresses this obligation to refugee
communities and to the receiving state by imposing restrictions on both.7
The specific refugee problems in the Great Lakes Region had long been recognized by UNHCR.
In the early 1960s, the agency extended its protection to Rwandese refugees in Uganda, including
some 80,000 who were receiving UNHCR assistance. UNHCR also facilitated joint talks between
Rwanda and Uganda to solve the crisis that arose when some 35,000 Rwandese refugees were
expelled from south-western Uganda in 1982–83.
The refugee problem seemed intractable. In essence, it stemmed from unwillingness of the
Ugandan government to accord the refugees full rights of settlement, while the Rwandese
government for its part denied them the right of return.8 Caught between the two denials, the
refugee community predictably generated a militant movement.
International instruments governing host country treatment of refugees lay down certain principles
designed to facilitate the integration of refugees in the host countries, but leave the states concerned
with considerable freedom of action.9 While a signatory to the relevant instruments, Uganda did
not have a clear policy of integration. Municipal law regulating refugees (the 1964 Control of
Aliens Act) was restrictive and subjective, giving the relevant ministers great discretionary power
over the status, movement and property of refugees (Khiddu-Makubuyu 1993, 1994). The
subjective nature of the law reflected the actual experience of the Rwandese refugee community in
Uganda. Conditions varied according to the political climate, and, as such, gave little security for
either the present or the future. Few availed themselves of the opportunity to acquire Ugandan
citizenship. Legal formalities were no guarantee of security, as illustrated in the 1982–83
“chasing” when Banyarwanda of both Rwandese and Ugandan citizenship were attacked. More
important, Ugandan citizenship would have terminated their right to return – and for that reason
was the optimal solution mainly for the Rwandese government. Many refugees nevertheless found
sufficient security, educational opportunities and mobility in Uganda to enable the second
generation to become a comparatively advanced community. That relative success made them a
18
target for local grievances on many issues, including land ownership and positions in the
government.
The intermittent invitations to return issued by the Rwandese government during the First Republic
(1962–73) had been dismissed by the refugees as propaganda or a trap. During the Second
Republic, the Habyarimana government changed course and, citing overpopulation, consistently
and unequivocally denied the refugees the right to return. This was graphically demonstrated in
1982– 83 when refugees chased out of Uganda were stopped at the border by Rwandese
government forces and became stranded under very difficult conditions. Not until 1986 did
President Habyarimana suggest the possibility of return, and then on an individual and conditional
basis.
In the refugee community, the militants of the early 1960s had by the 1980s been replaced by a
second generation. A new political movement was formed in 1979 as the Rwandese Alliance for
National Unity, and evolved into the Rwandese Patriotic Front, which crossed the border in
October 1990. Key RPF leaders had earlier been in the Ugandan National Resistance Army
(NRA), and were alongside Museveni when he ousted Uganda’s President Obote in 1986 and
seized power. They continued to hold senior positions in the Ugandan army.10 A series of events
within Uganda and Rwanda in 1989–90, however, apparently convinced the RPF leadership that
the time to move had come, and that the Habyarimana regime could be replaced by a political
system in which the RPF would have its proper place and enable the refugees to return (Otunnu
1995).
Habyarimana seized the initiative on the refugee issue, whether to score a propganda victory or
genuinely seeking to settle the issue (Watson 1991, Erny 1994, Guichaoua 1995). He reiterated a
limited offer of conditional return and, in 1989, established a Commission on Rwandese Refugees
in Uganda. A Rwandese-Ugandan interministerial committee on refugees was also revived and, in
cooperation with OAU and UNHCR, developed a plan for UNHCR to survey Rwandese refugee
settlements in Uganda in the fall of 1990, so as to assess the demand for repatriation and prepare
the modalities of return. Whether Habyarimana’s moves were sincere, or much too modest, or
simply a public relations exercise, it turned out to be too little too late. On October 1, the RPF
launched the invasion and thereby radically transformed the situation of the refugees.
It has later been suggested that a portion of the refugee population, including its more prosperous
and well-established members, might have been more interested in the formal right to return and
the right to visit freely, than in permanent repatriation.11 The views of the refugees themselves
were not put directly to the interministerial committee, which had no representatives from the
refugee communities.12 Nor were the preferences of the refugees themselves well identified in the
preliminary and poorly designed survey conducted by UNHCR in mid-1990.13 Hence, the
existence of possibly diverse views that would suggest a compromise solution was left unxplored.
The issue of return must be also be considered in a broader political context. The demand of the
RPF was not simply to return, but to change the political regime in the homeland. Indeed, the
refugee issue appeared as only point no.5 on their 8-point political program (Reyntjens 1994,
p.200). Nevertheless, had the Habyarimana regime formally recognized the right of return, this
might have given symbolic satisfaction to many, even in the absence of political reforms. Equally
important, it would have undermined the position of the militants in the RPF.
In retrospect, a possible window of opportunity to deal with the refugee question can be discerned
around 1989 and in early 1990. Habyarimana was making some conciliatory moves, and the
invasion that locked the conflict onto a new course was still in the future. There was no doubt
about the significance of the issue, both from a humanitarian perspective and for conflict
management. States and international organizations are fully aware that, if left unattended, refugee
problems of this kind spell futher conflict. Moreover, international norms are in place that provide
suitable guidance for action. That more pressure was not brought to bear on Rwanda and Uganda
in this case to implement their obligations reflected conditions that were to recur in later phases of
the conflict – limited regional capacity and limited international interest. The OAU had the right
19
norms in place, but had no mechanism to enforce them. That was also the case with UNHCR.
Western countries, which had significant potential leverage, were preoccupied with events in
Europe and the disintegrating Soviet Union, and seemed largely indifferent. The same applies to
Uganda’s involvement in permitting and, apparently, supporting the RPF invasion in various
ways.
Recognizing that RPF’s decision to return from Uganda by armed force constituted a violation of
international law, President Museveni immediately disavowed any Ugandan responsibility or
foreknowledge of the event.14 The official view in Kampala was that this was a purely internal
Rwandese affair, which consequently did not warrant international response. There is strong
evidence, however, that Museveni indeed knew about the RPF’s offensive and at least did nothing
to prevent it.15 Museveni clearly stood to gain from a successful return. He would be rid of a
troublesome refugee issue and, at best, have close friends and erstwhile clients in a position of
power in a neighboring state.
Quite possibly, Habyarimana anticipated the invasion on the basis of a series of signals16: the
visible movement towards the border of Tutsi soldiers and officers in the Ugandan army; an
abortive invasion that had been attempted in 1989; the sudden slaughter in refugee areas of 1,200
cattle to make provisions of smoked meat in July 1990; the fact that the training of Rwandese
military refugees on ranches in the Mbarrara border region was raised at the same time in the
Ugandan parliament; furtive fund-raising among Tutsi throughout the region; and endless reports
in Uganda about military mobilization by Tutsi in the NRA. This intelligence was available to
Habyarimana; it was certainly available to diplomats and other observers in the region.17
20
Chapter 2
Civil War, Civil Violence and the
International Response
(1 October 1990 – 4 August 1993)
The RPF invasion on the night of 1 October 1990 set off a series of responses, including regional
and international military intervention, major shifts in the internal politics of Rwanda, and highlevel diplomacy. During this phase, diplomacy vied with military assistance, human rights
concerns and development aid in a disconcerting tension that allowed opponents of the peace
process to develop strength even as the latter moved forward. The war resumed, and so did
reprisal killings of civilians in an early indication of what was to come. This “second track” of
violence – civil violence as distinct from civil war – ran parallel to the peace negotiations as the
extremists organized and consolidated.
2.1. Track I: civil war and the response
2.1.1. From invasion to negotiations
Both France and Zaire sent troops to aid the government of Rwanda at the time of the invasion.
Belgian units were also flown in, but only to protect their nationals, and they were soon
withdrawn. With the French backstopping the Rwandese army in Kigali and “advising” on the
handling of heavy artillery, and with Zairian paratroopers fighting alongside Rwandese troops, the
RPF invasion was soon halted. The RPF regrouped in preparation for a long struggle.
The invasion triggered an extraordinary diplomatic activity that demonstrated the vitality of
international mediatory structures (Jones 1995). Just two weeks after the invasion, Tanzania called
a regional meeting of Heads of State of Rwanda, Uganda and Tanzania to discuss the situation.
Realizing that a renewed conflict in Rwanda would at the very least cause a large inflow of
refugees, Tanzania remained actively involved and became host as well as “facilitator” for the
consequent peace talks. The OAU Secretariat was also active in recognition of the organization’s
principle that African states had a primary responsibility to solve regional conflicts. As a
Tanzanian, the Secretary-General, Salim A. Salim, coordinated closely with Tanzanian mediation
efforts. Soon other levels were informed or engaged – the (informal) Summit of Great Lakes
Region Heads of State, the European Union and more peripherally the UN. Also the governments
of Belgium, France and the United States at various times helped to move the process forward.
The Belgian government was visibly involved within days of the invasion, when a troika of top
ministers – the Prime Minister, the Foreign Minister and the Minister of Defense – toured the
region to promote reconciliation and presented the case to the European Union for support.
A positive communique was issued from the first summit in Tanzania, marking an exemplary
transition from war to diplomacy only two weeks after the invasion. Yet the initial success of
regional diplomacy was short-lived. A formal cease-fire signed at N’Sele on 29 March 1991 lasted
only to mid-April when fighting resumed. The first deployment of an OAU-sponsored Military
Observer Team (MOT), as had been agreed to in October-November 1990, was a failure. The only
states to contribute observers were Zaire, Burundi, and Uganda. Since none of them was
considered neutral by one or the other Rwandese party, the force became inoperative. The
Rwandese government insisted on prior clearance for every movement of the military observers, a
21
demand that greatly diminished their ability to monitor cease-fire violations. A second cease-fire
agreement (26 October 1991 at Gbadolite, Zaire) called for a revised monitoring force, to consist
of military observers from neutral countries. However, the African Neutral Military Observers
Group (NMOG) did not become operational until mid-1992.18
The limits of regional diplomacy were revealed when also the second cease-fire broke down in
early 1992. It required a push from major powers – The United States and France – to move the
conflicting parties to the negotiation table. Also the European Union, Canada, Switzerland, the
Catholic Church and others weighed in. Eventually, this resulted, in the summer of 1992, in the
launching of the Arusha talks towards a comprehensive peace settlement.
Both Uganda and later France pursued a dual policy of supporting their respective Rwandese
partners, yet encouraging negotiations. In keeping with international conventions, Uganda
formally denied it was supporting a rebel army – a claim maintained even after the RPF seized
power in 1994. In fact, Uganda served as a rear base for the invasion, enabling the Front to
regroup, recruit and mobilize among the refugee community, and evidently to funnel weapons and
supplies to the RPF-controlled area inside Rwanda. At the same time, once the invasion was a fact
but did not bring RPF a speedy victory, Uganda cooperated to get negotiations started. The fact
that Uganda held the chair of the OAU at the time also suggested cooperative diplomacy. President
Museveni agreed at the regional summit in mid-October 1990 to work with his Tanzanian
counterpart, Ali Hassan Mwinyi, to persuade the RPF to accept a cease-fire. He further promised
to explore the possibility for an African force to be deployed between the combatants once the
cease-fire took hold.
2.1.2. France’s dual policy
During the invasion and in its immediate aftermath, French policy was decided at the level of the
President. The first response was to send “a few boys to help old man Habyarimana”, as the head
of Africa Unit in the Presidential palace said at the time (Prunier 1995). France had previously
done the same for other allies in Francophone Africa and had paratroopers stationed in the region
for that purpose. In Rwanda’s case, the determination to assist was sharpened by the fact that the
rebels had come from and in some respects were part of Anglophone Africa. The social origins
and platform of the RPF also made the Front leaders anathema in some French conservative and
army circles where the pejorative term “Khmer Noir” was used, a label projecting the horrors of
Cambodia’s Khmer Rouge onto an African context.
As the military weakness of the government was revealed, however, the Africa Unit in the Elysée
Palace and the Military Assistance Office in the Ministry of Cooperation embarked on a policy to
rapidly build up the Forces Armées Rwandaise (FAR).19 With French technical and military
assistance, including credits for arms purchases, Rwanda’s tiny army tripled from 1990 to 1991
and by the time of the peace agreement (mid-1993) had at least 30,000 men. The French military
assistance unit in Kigali (DAMI) provided training and operational guidance under the 1975
military assistance agreement, as amended in August 1992. Numerically, the French were not a
major presence. Some 40 coopérants militaires were gradually increased to 100. The size of the
paratroop detachment, Force Noroît, first sent in Ocotober 1990, varied according to the rythm of
the war. After the first cease-fire in 1991, the force was reduced to about half (168 men), but
increased after the RPF offensives in 1992 and 1993 to reach a peak of some 700. According to
French authorities, Force Noroît had a strictly defensive mission, including defense of expatriates
and securing the airport, and was deployed in and around Kigali. French government spokesmen
have consistently and categorically denied reports to the effect that the paratroopers occasionally
participated in the war.20 The French Ministry of Foreign Affairs, moreover, has taken pains to
point out that the force also had humanitarian functions in addition to its military duties, including
vaccinating civilians and distributing food to displaced persons.21 Granted that the French military
presence was both small and had a restricted mission, the effects nevertheless had strategic and
political significance. By taking on a major role in training the rapidly growing Rwandese army,
and by sending in paratroopers every time the RPF launched an offensive, the French government
clearly signalled that it stood by the Habyarimana regime. This was the interpretation within
Rwanda as well, which contrasted with the hostile attitude shown by many regime supporters
22
towards Belgium, precisely because Brussels decided to abort its 1990 intervention rather than
turn it into a long-term military presence on the French pattern.
The role played by the Elysée Palace and the Ministry of Cooperation was increasingly questioned
by the Ministry of Foreign Affairs, which around mid-1991 recognized a compelling logic of
negotiation in a situation it characterized as follows: the RPF might win militarily but not
politically, while the government could not win militarily, though it might command the numbers
to win politically (Callamard 1995). According to this logic, France could best salvage its interests
in Rwanda by promoting a negotiated settlement on terms favorable to the Habyarimana regime. In
October 1991, the director of the Africa bureau in the Ministry of Foreign Affairs attempted to
bring representatives from the Rwandese government and the Front together in Paris.22 The
efforts did not bear fruit until late spring 1992, when the two parties agreed in Paris on 8 June to
open formal peace talks.
By giving military assistance to the Habyarimana regime in time of war, France followed
customary alliance politics, rather than the neutral international practice of not supplying arms to
belligerents during war, as observed in this case by Belgium. At the time there was no
international arms embargo on Rwanda. The three formal cease-fire agreements signed in 1991–92
prohibited the “infiltration of ... war material to the area occupied by each party”23, but all had
been violated by one or the other Rwandese party and hence lost much of their contractual force.
Until the signing of the Arusha agreement, therefore, France could argue – as indeed it did – that it
exercised the common right of sovereign states to give military aid to a friendly government facing
a rebel force. French authorities did not deny selling arms or giving credits to the Habyarimana
regime for purchasing weapons elsewhere, but refused to comment on reports of particular
transactions.24
From this point onwards, and especially as the situation deteriorated in 1994, French policy was
shaped by competitive interplay among domestic institutional actors, which had different interests
and perspectives. The result was a dual policy that supported negotiations but simultaneously built
up the Rwandese armed forces and embraced the regime politically. The policy helped to move
negotiations forward at the inter-state level, while at the same time provided political space and
resources for the regime to consolidate, and – indirectly – for the Hutu extremist fringe to develop.
2.1.3. The beginning of democratization
The RPF invasion had a transformative impact on Rwandese politics by giving significant
momentum to a process of democratization that had started previously, but so cautiously it was
hardly noticed. Responding to demands for democratization expressed at the Franco-African
summit in La Baule in 1990 and elsewhere, Habyarimana had started to liberalize the political
system in 1989–90. The RPF brought the process decisively forward by making radical demands
for power-sharing, backed by the force of an invading army. Foreign donors added to the
pressures by emphasizing that democratization with a multi-party system was necessary to end the
war, and also desirable in itself. Noticing the trend and sensing that the regime was yielding, the
domestic opposition became more insistent. Habyarimana responded first by forming a token
coalition government. As pressures visibly mounted throughout the spring of 1992, he agreed to a
second coalition that had significant representation from the newly formed opposition parties.
The formation of the second coalition government in April 1992 had significant and contradictory
consequences. The timing was in some respects not auspicious for democratization – the country
was at war, the economy was deteriorating, and ethnicity was re-emerging as a polarizing
dimension. In part, multipartisme became a cloak for particularist interests, encouraged ethnic
mobilization and fed political fiefdoms that usurped the civil administration, as a disillusioned
former minister wrote from exile (Gasana 1995:231-2). Ultimately more significant was the effect
on Hutu extremism. The combined pressures for peace negotiations and democratization provided
a double-bored attack on the existing power-holders. When political forces later polarized, the
extreme Hutu fringe translated this into an absolute struggle with only final solutions. In 1991 and
early 1992, however, the most visible political forces were: a “presidential tendency” that was
conservative, had a regional base in the north-west, was inclined to defend Hutu supremacy, and
23
relied on a strong party, the militia, the army, and Habyarimana to preserve its power; a second
“moderate tendency” composed of various newly formed political parties that sought to replace the
Second Republic by non-violent means and institute a vaguely formulated plural political system;
and a third, radical force represented by the RPF that aimed to establish a new order, ideologically
defined as political pluralism based on a social structure that denied the legitimacy of ethnicity in
political mobilization, but suspected by its enemies to entail “une inversion hégémonique en faveur
de l’ethnie minoritaire tutsi” (Gasana 1995:224).
The formation of a coalition government that included the “moderate tendency” had a salutary
effect on the peace negotiations. The first act of the Foreign Minister of the new coalition – a
member of the opposition party, Mouvement démocratique républicain – was to call for political
talks with the RPF. This call was picked up by inter-state diplomatic mechanisms and, after
intercession by the United States, France and the Vatican, led to the Arusha talks in the summer of
1992. Having included his domestic opponents in the government, Habyarimana permitted them to
lead the negotiations at Arusha.
2.1.4. The Arusha period
The Arusha process – as it came to be called after the resort area in Tanzania where the peace talks
were formally launched on 12 July 1992 – was pursued at a very high level in the region, drawing
in heads of state and foreign ministers. The core negotiations on a future peace agreement had
participant-observer delegations from five African states (Burundi, Zaire, Senegal, Uganda and
Tanzania with its “facilitation” team), four Western countries (France, Belgium, Germany and the
United States), and the OAU. The UN was brought in at the invitation of the OAU, and UNHCR
attended as observer. Britain, Canada, the Netherlands and the European Union – important
members of the Western donor community in the region – closely monitored the process from their
local embassies. Nigeria was represented at the Arusha-linked Joint Political Military Committee
(see below).
Formally acting as a “facilitator”, the Tanzanian government undertook a sustained and skilful
mediating effort that was critical in bringing about the final agreement.25 The OAU had earlier
organized a military observer group to monitor successive cease-fires, and maintained a high-level
presence at the negotiations, with either a senior political officer or Secretary-General Salim
himself representing the Organization. For Salim, the Arusha process was a challenge to
demonstrate that the OAU could make a critical contribution to conflict resolution in Africa. To this
end, Salim had just established a new OAU Conflict Resolution Mechanism that would deal with
internal as well as inter-state conflict. Formal approval was expected at the Cairo OAU Summit
meeting in 1993. Given the tradititional OAU sensitivities on internal affairs, it would be gratifying
to have a freshly mediated peace agreement to put before the Summit.
The Arusha process represented a multi-pronged strategy of conflict resolution. A preliminary
phase was designed to obtain a cease-fire. The earlier N’Sele and Gbadolite cease-fire agreements
were amended to form the basis of the Arusha cease-fire, announced on 12 July 1992. All ceasefires were violated, but the process permitted continuous communication and led to the
establishment of a monitoring machinery. In particular, the Arusha process implemented a
mechanism agreed to in earlier negotiations: a Joint Political Military Committee (JPMC) ran
parallel to the Arusha talks and provided an escape valve that allowed all parties to discuss
violations of the cease-fire without derailing the broader negotiations. The negotiations on
substantive issues of the conflict, which started in mid-1992, covered the rule of law, the
establishment of transitional institutions, the repatriation of refugees and resettlement of the
internally displaced, and the integration of the two armed forces into one integrated national army.
The central problem lay elsewhere. Given the starting point of an authoritarian regime, it was
evident that any progress in the talks would require significant concessions from the government.
The existing power-holders had already tasted loss when reluctantly agreeing to a coalition
government. This was nothing, however, in comparison with the radical changes spelled out in the
full Arusha accords. In effect, the Accords represented a veritable coup d’état for the RPF and the
internal opposition, as a close observer later noted (Reyntjens 1994).
24
The Accords moved Rwanda from a presidential to a parliamentary system of politics. Most of the
power of the Presidency would be transferred to a Council of Ministers. Habyarimana – who was
to occupy the Presidency during the transition period prior to elections – was left with a ceremonial
position. During the transition period, seats in the Council of Ministers of a Broad-based
Transitional Government (BBTG) would be divided between the government and the internal
opposition parties as well as the RPF. The ruling party (MRND) was left with only 5 out of 19
portfolios, and most were politically insignificant. The regime also would lose power in local
administrative structures and the judicial branch. The distribution of portfolios agreed to at Arusha
accentuated the dramatic nature of the change. Ideologically and politically, the BBTG represented
a frontal attack on the power base erected by the Habyarimana government during twenty years of
ruling the country _– a denial of authoritarian rule, of “Hutu power”, and especially northwestern-based Hutu power, which was the regional constituency and political backbone of the
regime. The terms of the Agreement signalled a pluralist state and a civil society, and the idea of a
people united by a common Rwandese nationality rather than divided along narrow lines into Hutu
versus Tutsi.26
Significantly, the hardliners who surrounded Habyarimana – senior ministers in the MRND party,
senior military figures in the armed forces and the Presidential Guard, and members of the newly
formed Coalition pour la défence de la république (CDR) – were denied power in the transitional
institutions. In spite of advice to the RPF from Western states that it would be a better tactic to
coopt the extremists than exclude them, the CDR was not included in the power-sharing formula
of the BBTG, nor allocated seats in the transitional Assembly. Habyarimana had demanded that
the CDR be represented; the RPF, correctly noting that the CDR was both extreme and a nonparty, was absolutely opposed and prevailed. The extremists responded by openly threatening to
bring about “an apocalypse”, as their spokesman, Col. Bagosora, declared during a session at
Arusha when the power-sharing protocol was negotiated.
The losses incurred by the existing power-holders in the political sphere were paralleled, and thus
sharpened, by equivalent losses in military matters. Hoping to secure at least continued control
over the armed forces, the government opened the bid in negotiations on integration of the armed
forces by offering the RPF a mere 20% share of the joint army. The Front rejected this outright,
and the government was eventually forced to accept a much modified scheme that split positions in
the joint command 50–50, not only at the high command level but all the way down to field
commanders, and narrowly favored the government in a 60–40 distribution of troops. Observing
the trend, a Western diplomat cabled home that the agreement had moved negotiations ahead of the
consensus and risked repudiation by forces in Kigali. He was to be proved correct. Related
military issues were also hard to accept for the army. Over 20,000 government troops would be
demobilized, more than twice the figure for the RPF.27 While the government soldiers had
demonstrated little enthusiasm for fighting, especially during the 1993-RPF offensive,
demobilization seemed even less attractive.
That the Arusha agreement produced such profound changes – which were registered as defeats by
the government and especially by hardliners in Kigali – can be attributed largely to the balance of
power between the Rwandese parties. The RPF negotiating team was extraordinarily strong, with
discipline, preparedness and commitment in stark contrast to the fractionalized, ill-disciplined, and
ineffective government team. Reflecting the divisions at home, the Kigali delegation was
composed of some Habyarimana allies, but led by opposition party members, and obstructed by
recognized members of extremist factions, which also were at Arusha. Equally, if not more
important, negotiations on the military issues were conducted at a time when the military position
of the government deteriorated sharply.
In February 1993, the RPF had launched a major offensive to break Habyarimana’s opposition to
the power-sharing formula of the BBTG protocol, and to protest the massacre of 300 Tutsi in
Bugogwe. The offensive shocked Kigali and threw the government forces into disarray. Rebel
troops fought to within 23 km of the capital, despite the fact that France sent a small contingent of
paratroopers to reinforce government troops. The Front’s victories clearly showed their military
25
superiority; anticipation of success on the battlefiled was possibly a factor in the RPF’s decision to
undertake the offensive on the eve of the scheduled talks on military matters at Arusha.
The Arusha process has been commonly referred to as preventive diplomacy and judged against
the catastrophic violence that followed. Held against this standard, it is clear that the process
failed. But the main objective at Arusha was not to prevent a clearly perceived future calamity.28
More conventionally, the aim was to end a war and construct a post-war peace agreement that
reflected a situation short of total victory or absolute defeat. This required a formula for immediate
power-sharing as well as agreement on a broad range of issues: the nature of the future political
process, rules for demobilization and integration of the respective armies, and a system to facilitte
the return of refugees and internally displaced persons. Similar agreements had recently been
negotiated to end wars in Central America (El Salvador and Nicaragua), in Southern Africa
(Namibia and Mozambique) and in South-east Asia (Cambodia). These, if any, were the models.
The closer the Arusha process came to end the war and to define the protocols that would structure
future politics, the closer – most participants believed – they had come to achieve peace. In this
calculus, the primary criterion for success was to obtain the signatures of both parties on a peace
agreement.
The final agreement dealt comprehensively with all the issues related to ending the war and
opening up the political system. As such, it was a “perfect” agreement, as some observers claimed
at the time. Yet the accords evidently did not rest on a new-found political consensus created by
defeat, exhaustion or the emergence of new political forces. At the time, some Arusha observers
suspected that the RPF was using the accords merely as a convenient way-station while preparing
for a coup. As for the government, Habyarimana had repeatedly announced his unwillingness to
sign, and disinclination to uphold, this “piece of paper”, as he disparagingly referred to the
accords in November 1992. Instead, the accords formalized what André Guichaoua has called une
paix militaire.29 The Accords were a victor’s agreement. The RPF had extracted major
concessions at the negotiation table by means of military strength. However, the Front had
demonstrated its potential military superiority rather than inflicting an actual defeat on the
government forces. The latter fell back but were not beaten. The coercive apparatus of the regime
not only remained intact, but – in the logic of an armed peace – was strengthened. The difficult
side-negotations on the modalities of providing security for the RPF members in the transitional
government further underlined the point.
In this situation, one aspect of the peace agreement appears as a critical weakness. The Arusha
process did not deal with the losers to the agreement. The extremists were not co-opted into the
political process, nor neutralized by other means. They remained in a position to wreck the entire
edifice.
The exclusion of the extremist CDR from the transitional government has become a major issue in
subsequent evaluations of the strength and weaknesses of the Arusha Accords. Clearly, the mantra
of the Namibia negotiators in 1988 – “in this negotiation, there will be no losers” – was not
invoked at Arusha. Whether the alternative South Africa-style solution would have worked in the
end remains conjectural, and opinions were divided at the time. Most observers at Arusha
concluded it was unwise to exclude a major force in Rwandese politics, as this faction would have
no stake in the successful implementation of the Accords. Other diplomats in Kigali shared the
RPF assessment that the extremists were not interested in sharing power but bent on destroying
the new political order that was drawn up. In this view, a speedy implementation of the agreement
seemed the best means to control its violent opponents.
The party with most immediate leverage on the RPF was Uganda’s President Museveni.
Apparently he tried but failed to sway his erstwhile allies on both the CDR issue and the military
protocol. During the course of the civil war, Museveni evidently had lost some of his influence
over the RPF. The Front’s repeated successes on the battlefield gave it a stronger basis for
autonomous action. A second factor was Museveni’s decision to distance himself formally from
any involvement in the civil war, and, in particular, deny that arms were crossing the border in
violation of international norms. During the Front’s February 1992 offensive, Museveni wrote to
26
the Security Council inviting the UN to deploy observers along the Uganda-Rwanda border to
verify that Uganda did not aid the offensive30. This request, which resulted in the establishment
of the UN Observer Mission Uganda-Rwanda (UNOMUR)(Res.846(1993)), was seen as a
betrayal by some factions in the RPF.
2.1.5. The UN and the OAU: structural disjuncture
Since early 1993, OAU Secretary-General Salim A. Salim had pursued the idea of strenghtening
the small group of military observers fielded by African states in Rwanda through the OAU
(NMOG). The RPF offensive and resultant cease-fire in March 1993 had expanded the buffer zone
to be monitored. The OAU Secretariat felt that a rapid response was necessary to stabilize the
cease-fire and pave the way for a larger international presence as the prospects for a full peace
agreement drew near(Tekle 1995). The present NMOG of only 50 men could not effectively
monitor even the smaller buffer zone, but OAU was unable to raise funds for a larger force.31
Turning to the UN for assistance in March 1993, Salim twice pursued the matter in letters to the
UN Secretary-General, personally presented the case to Security Council members and presented a
detailed proposal to a special envoy of Boutros-Ghali who visited Addis Ababa. His goals were
modest: the maximum option was for an OAU force of 500 men at an estimated cost of US$2.5
million per month. After preliminary discussions with the Security Council, the response of
Boutros-Ghali was that the request for UN logistics and financing “could only be entertained if the
operations were under UN command and control.”32 This went against the OAU’s wishes to
develop a field presence so as to complement its mediatory efforts in Arusha, and demonstrate that
African states indeed took an active part in solving African problems.
Finances were not the only issue. The larger question concerned the role of the respective
organizations in an eventual peacee-keeping force in Rwanda. Since late 1992 the French
Ambassador to the UN had started to lobby the Security Council for a post-war UN military
presence in Rwanda. In early 1993 – and especially when the RPF offensive in Feburary pushed
Rwandese government forces back – French efforts had become so insistent that it was “a standing
joke in the Council”, according to one ambassador present. From a French perspective, such a
force would at best check the RPF’s advance and provide a breathing space for the government
both by positioning itself between the two belligerents and by monitoring the border between
Uganda and Rwanda. At the same time, a UN force would be an answer to increasingly insistent
demands from the RPF that French forces withdraw, as had been stipulated already in the 1991
N’Sele agreement. With a UN force in Rwanda and a French veto on UN peacekeeping operations
in the Security Council, Paris could still exercise some influence after its own troops were gone.
The government of Rwanda, which had been elected to the Security Council, was generally
ineffective, but on this issue vigorously supported France. The OAU was, with some reason,
regarded as partisan towards the RPF and at any rate unable effectively to monitor or maintain the
military status quo. By the same logic, and because France was promoting a UN force, the RPF
wanted an OAU command.
Travelling to Dar-es-Salaam for a summit meeting on 5–7 March 1993, Salim nevertheless
obtained approval from both the Rwandese parties to move ahead with an expanded OAU
monitoring force. Two days later, the initiative was effectively upstaged in the UN Security
Council. France introduced a resolution proposing a UN peacekeeping force for Rwanda,
operating “in conjuction with” the OAU. (S/25400(1993)). Non-aligned members of the Security
Council and some European states cautioned that French efforts might be viewed as an attempt by
Paris to salvage its influence in Rwanda and warned against sidelining the OAU. While the
languge of the resolution subsequently was softened, the signals given to OAU became clearer as
every effort by Salim to increase even the monitoring function of NMOG slightly, was
rebuffed.33 Boutros-Ghali sided with his own organization, despite formal deference to the
importance of regional initiatives in conflict resolution. Apart from insisting on UN command and
control, he added that any discussion of UN assistance would have to await the outcome of the
Arusha talks. In the meantime, plans for a UN peacekeeping force went forward.34
In face of these discouraging messages, OAU abandoned efforts to play a significant role in the
27
monitoring and implementation of the Arusha peace agreement. The development raises important
questions concerning the appropriate working relationship between the UN and the OAU and, in
this case, its impact on subsequent events in Rwanda. OAU admittedly had limited experience in
peacekeeping and had problems fielding even a small military observer group. It was wholly
dependent upon the UN to provide logistics and finance. Yet it was effectively discouraged by the
UN and key Security Council members from taking a larger role. This is particularly ironic given
the demonstrted ineffiency of the UN to deploy and develop the peacekeeping operation it had so
aggresively sought (see chapter 3).35 When the crisis later erupted, on 6 April 1994, it was the
European battalion of UNAMIR that withdrew and crippled the future options of the force; the
African battalion stayed. The political instinct among the veto-wielding members of the Security
Council was to abandon the mission. Prima facie, there is no reason to think that the OAU would
have performed worse.
2.2. Track II: civil violence and human rights conditionality
The extremists in Kigali had a different vision of the peace process. As the Arusha process laid the
groundwork for a virtual political revolution in Rwanda – the framework of a “new order”, as it
was said in Arusha – the Hutu extremists planned their own bloody, reactionary vision of change.
In retrospect, that much has become clear. But at the time, what did donor countries and the
international community more generally know about the extremists?
2.2.1. Early warnings
The political scene in Kigali in 1992–93 was admittedly complex, and a flurry of rumors and
propaganda further complicated the picture. Key questions remain unclear even today, including
the relationship between Habyarimana and the extremists. Was he a master or a victim of the
hardliners, and if he moved from one role to another, when did this happen? Outside analysts also
had to decipher the meaning of the new civil society that developed. The print media multiplied
across the political spectrum. So did NGOs and various political movements. The difficulty of
distinguishing rumours from facts and “hard” infomation from rumours was exacerbated by
diffusion of propaganda that all sides resorted to as a weapon of war and political competition in
this period.
Yet there was considerable and growing evidence of Hutu extremism. Human rights violations
were documented by human rights organizations and concerned activists. The French had a
military mission in Kigali, and the small foreign diplomatic corps observed the scene, reported
home, and, for the most part, shared information with each other. An informal grouping of
ambassadors from the United States, Belgium, Germany, France and the Vatican – also known as
The Five Musketeers – met frequently.
It was well known at the time that the combined peace and democratization process faced strong
opposition. Already in July 1991, a French intelligence assessment identified three critical circles
of power in relation to the peace process (not to be confused with the three political tendencies
discussed above), starting with the inner circle formed by the President’s in-laws and associates,
an outer circle of relatively more moderate members of the military and the cabinet, and, lastly, an
amorphous grouping of younger officers and intellectuals willing to entertain change. Since the
objective of the first circle was to retain power, any form of imposed negotiation would provoke
their resistance. The analysis found its way to other donor states and was soon amplified by
pointed statements and actions by the hardliners.
By late 1992 and especially in early 1993, a much more comprehensive body of evidence was
available and repeatedly discussed in the diplomatic community in Kigali. Killings of Tutsi were
more frequent, and there was increasing information about death squads as well as the existence of
a Network Zero around the President, which reportedly was plotting to exterminate regime
opponents and circulated death lists. The report of the International Commission of Inquiry on
Human Rights in early 1993 was widely circulated (FIDH 1993). The Belgian government, for
one, took it so seriously that it temporarily recalled its Ambassador from Kigali. The report
documented the involvement of the government in what was described as systematic killings
28
directed against the Tutsi, and estimated that about 2,000 persons had been murdered in the
1990–92 period. Briefing reporters on its finding, the Commission used the term “genocide” to
describe the killings. This claim was retracted in their official report published in March because of
fear that such a designation would be considered hysterical.36 It was, however, upheld in a report
by the UN Commission on Human Rights a few months later (see below).
The diffusion of hate propaganda in Rwanda was also noted as an indicator of increasing tension
and mounting “Hutu power”. However, interpretations of the message differed. As with the
notorious Radio et télévision libre milles collines (RTLM) – founded soon after the signing of the
peace agreement – the hate media in this period were so explicit and literal that some diplomats,
both Western and African, afterwards said they tended to dismiss it.37 Similar reactions were
entertained even by the RPF, as a leading Front member later recalled: “What they said was so
stupid; we did not take it seriously enough.”
There were other telling pieces of evidence. At least two revealing government documents were
leaked to the diplomatic community and circulated in Kigali. One was an internal report from a
senior military officers’ commission entitled “Definition and Identification of the Enemy” (21
September 1992). The report listed as enemies not only those Tutsi inside and outside Rwanda
who supported the RPF, but also members of mixed marriages and moderate Hutu who opposed
the hardliners within the government. The other was a letter from the (opposition member) Prime
Minister to the Defense Minister, dated 25 March 1993. The Prime Minister gave details of illegal
distribution of weapons to civilians and called for immediate action to stop it.
By this time, links between the civil violence and the peace process became more explicit. At every
important juncture and setback for the government, Tutsi were murdered. Clearly discernable in
retrospect (see Figure I: A Pattern of Violence), the coincidence between killings of Tutsi and the
rhythm of negotiations in Arusha was also recognized at the time by several Kigali-based
diplomats and variously interpreted. “We read the killings as a political negotiating tactic”, one
analyst later said. Others saw the violence more radically as an attempt to derail the entire peace
process.
Simultaneously, diplomats in Kigali noticed that the extremist opposition to the Arusha talks was
gaining strength. During the talks on power-sharing, Habyarimana was oscillating between
moderates in his own delegation and the hardliners at home. The talks were halted in September,
but when they resumed the Foreign Minister, Boniface Ngulinzira, was negotiating without the
authority of the President, as an observer delegation noted at the time. Habyarimana soon made the
point publicly by calling the accords merely “a piece of paper” in his famous 15 November speech.
At Arusha, as noted, the CDR representative predicted “an apocalypse” when the power-sharing
protocol was finalized. The message carried added significance since it was delivered by
Col.Bagosora, the person widely believed to be behind the work of the death squads, including
the Kibuye massacres in August 1992.
Signs clearly pointed to the consolidation of the power of the Hutu supremacists – through the
formation of what amounted to para-statal organizations. The CDR was formed in March 1992,
bringing together military and political extremists from circles inside and connected to the
Presidential Palace. The party militias (the interahamwe and impuzamugambi) were also formed at
this time with direct support from the CDR, the ruling government party, and the Presidential
Guard. Both operated openly, and diplomats in Kigali readily recognized both as instruments of
the ruling party. Yet their precise function was open to various interpretations. Some observers
accepted at face value that the militias were formed to defend the country against the invading RPF
troops, according to the doctrine of village self-defense. Subsequent intelligence information
suggested that up until possibly late 1993, this was indeed their main purpose. In a cable from
Kigali to New York dated 11 January 1994 – later to become known as The Black File – the UN
force commander in Rwanda relayed a statement from a high-level government informant as
saying that the “principal aim of interahamwe in the past was to protect Kigali from RPF”. With a
peace agreement in place and the arrival of a UN peacekeeping force to help implement it, the
militias started planning for the extermination of Tutsi (see chapter 3).
29
Figure I: A pattern of violence
Track I: Civil war and peace talks
Track II: Civil violence
Events in 1990
October 1:
RPF invasion
October 1–15:
6–7000 persons detained ca. half in Kigali
October 4–5:
An “attack” on Kigali
October:
Ca. 300 Tutsi killed in Kibilira
(in Gisenyi prefecture, NW)
October–November:
Numerous Tutsi–Hima disappear in the
Mutara range (N)
Events in 1991
January:
RPF raid on Ruhengeri prison
Events in 1992
March 14:
Habyarimana agrees to coalition govt.,
to negotiate peace with RPF and to setttle
refugee problem
Late January–early February:
“Hundreds” of Tutsi (Bagogwe clan) killed
in NW region
Mid–March:
Massacres in Bugesera (NW) 60–300
Tutsi (Bagogwe clan) killed, 10 000 displaced
April 16:
Coalition government formed
April 25–May 6:
Ca. 22 killed in assassinations and grenade attacks
and ca 100 injured in Kigali and 3 provincial towns
August 10–17:
First round at Arusha–Agreement on Protocol
on rule of law, political pluralism and respect
for human rights.
August 15–17:
Ca. 300 Tutsi killed in Kibuye (N)
September–October:
Negotiations on power sharing at Arusha.
GoR splits and talks halted
October 30:
CDR demonstrations in Kigali turn violent
November 15:
Habyarimana calls Arusha accords “a piece of paper”
November 15:
Opposition demonstrations in Kigali for Arusha
turn violent
December:
Arusha talks continue, GoR still split on power
sharing issue
December 25–31:
Bombs explode in Kigali and Butara; three Tutsi
killed in Kibilira, dozens of opposition members
threatened / injured / homes burnt
Events in 1993
January 10:
Arusha Protocol on power-sharing
February 8:
RPF launches offensive
January 20–26:
Ca. 300 Tutsi killed in Gisenyi
February:
25–50 Tutsi reported killed in north and central
regions
Source: FIDH 1993, Human Rights Watch (1993,1994), Reyntjens 1994.
There was a certain confusion about who constituted the principal target. Some diplomats,
including the Canadian Ambassador in 1994, read the violence primarily as politically directed
against regime opponents regardless of ethnicity, and were totally unprepared for the genocidal
30
dimension of the violence.38 Others saw Rwanda’s Tutsi minority as a hostage to the war between
the Habyarimana-regime and the RPF, and thus collectively being at great risk. This was the
pattern of the early 1960s, when attacks by a previous generation of Uganda-based Tutsi refugeesbecome-guerillas brought fierce retaliation against the Tutsi who remained in Rwanda.
2.2.2. The conditionality question
Acknowledging the widespread evidence of a deteriorating situation, donor countries made some
efforts to correct it. While the United States did not stress, and France did not include, human
rights issues in their policy towards Rwanda,39 other donors did. Belgium formally incorporated
human rights criteria in its aid policy after the Socialists came to power in 1992. Canada, another
major donor, had done so all along. Also the representatives of Switzerland and the Vatican spent
considerable time and efforts raising human rights issues, and made repeated representations to the
President and the Minister of Justice (when there was one). More generally, diplomats from states
concerned with democratization put their faith in political reforms: promoting multipartisme and
good governance, it was argued, would help address human rights abuses, including the critical
issue of legal impunity.
There was some progress towards democratization. The regime’s progressive move towards
multi-partisme was partly due to Western persuasion and pressure. High-profile pressure by
foreign diplomats and quiet work by ICRC helped to bring the release of nearly 10,000 persons
who were detained by the government when the RPF invaded in 1990. A few individual human
rights cases were addressed.
In principle, most Western donors made economic aid conditional upon observance of human
rights. In practice, however, most donors did not reduce aid with specific reference to human
rights violations, even though these were recognized as severe. For instance, the Belgian
government took steps considered extreme in the language of diplomacy when the international
human rights commission published its incriminating report in March 1993. The Belgian
Ambassador in Kigali was “recalled for consultations”, and the Rwandese Ambassador in
Brussels was told that Belgium would reconsider its economic and military aid unless steps were
taken to rectify the situation. However, Habyarimana made conciliatory statements and Belgian aid
continued. The Canadian government announced in Kigali that, largely for human rights reasons,
but also for budgetary reasons, Rwanda would be phased out as a major aid recipient from
1992–93 onwards. In Ottawa the decision was explained largely with reference to budget cuts and
changes in bilateral programming.40
Among the African countries, only the Ambassador of Tanzania took a strong, outspoken position
on human rights violations. As a result – and in consequence of Tanzania’s weak economic
leverage – Habyarimana asked in late 1993 that the Ambassador be replaced with another one.
Tanzania complied. The OAU’s Commission on Human and People’s Rights visited Kigali once
and generally “performed abysmally”, an OAU observer concluded.
31
Why was human rights conditionality – an arrow in the quiver available to the international
community – not pursued more decisively? Conditionality on economic policy was all along
precisely defined and consistently promoted. The principal demands for structural adjustment and
fiscal reform government to comply in the area of economic reform possibly reduced the incentives
for donors to insist on human rights issues. There were also other concerns. With a traditionally
efficient local administration and functioning infrastructure, Rwanda had a capacity for absorbing
aid that partly accounted for large transfers in previous years. In the early 1990s, it was still a
country where foreign aid bureaucracies could reasonably disburse money. Project feasibility
helps to explain the continued inflow from two traditional donors, France and Belgium.41
There was a more fundamental reason. International pressure to democratize and institute good
governance developed a special rationale in Rwanda: just as the Arusha process was a solution to
civil war, so democratization came to be seen as a solution to the growing problem of civil
violence. Support for democratization and the related peace process implied continuous economic
and diplomatic engagement in Rwanda. From this perspective, the threat of ultimately imposing
sanctions by withdrawing aid – as Western human rights organizations called for in 1992–93 –
was counter-productive.42 Donors thus became hostage to their own policies.
As the economic situation deteriorated in 1993 and renewed war plus drought increased the
number of internally displaced people to around one million, the international community pledged
large amounts of humanitarian aid. Simultaneously, some donors shifted development aid into
relief assistance. The issue of conditionality thereby became more remote. The call for pledges of
international humanitarian assistance went out in the UN system in March, just at the time when
the International NGO Commission published its report detailing the government’s involvement in
massive human rights abuses and calling for human rights conditionality in aid policy. The UN
Department of Humanitarian Affairs put the question on the table by asking in an internal
memorandum how the human rights issue should be handled at the forthcoming consolidated
appeal for humanitarian assistance. The answer suggested by the DHA was to distribute copies of
the report before the meeting.43
Given the problematic nature of sanctions, donors instead looked to “positive conditionality”, in
European Union terms, to promote democratization and human rights, and later to have the Arusha
Accords implemented. Aid was allocated to strengthen particular sectors. Thus Canada, the United
States and Belgium supported projects to improve administration in the Ministry of Justice with
advisors, seminars and training, and to strengthen civil society by supporting local human rights
organizations and encouraging a free press.
Despite these efforts, the quality of governance declined markedly. In the Ministry of Justice, for
instance, the assignment of foreign advisors could not make up for the fact that for most of 1993
there simply was no Minister of Justice. The one who was finally appointed towards the end of the
year had a shady reputation which was subsequently confirmed.44 Under the impact of
multipartisme and fiercely conflictual politics in late 1993 over the implementation of the Arusha
agreement, administrative agencies turned into political fiefdoms of a more extreme kind than
before. There was no accountability and project implementation had become impossible, USAID
concluded, suspending its program in 1994. Germany did the same, citing insecurity in the
countryside and the erosion of administrative efficiency. The European Union also put most of its
4-year aid allocation to Rwanda on hold.
By suspending aid with reference to bookkeeping and resource restraints rather than human rights
criteria, donors sent the message that human rights conditionality was preached but not practiced.
The same applied to the infamous Radio mille collines. Unable to agree among themselves on the
significance of the radio and how to silence it, the donors limited themselves to making yet another
representation to Habyarimana. The latter responded as usual by promising to look into it, and
there the matter rested.45 Half-hearted conditionality of this kind probably succeeded only in
eroding the credibility of diplomatic suasion.
By late 1993, the limits of diplomacy were becoming obvious, yet the prevailing sense among
32
concerned members of the diplomatic corps was that the only alternative was to soldier on.
Arguably, a window of opportunity for dealing with the extremists had existed around 1992,
before their strength developed in earnest and while the donors still had commitments to
development aid that could be adjusted as needed. Although information about the extremists at
that time was less alarming, human rights violations were significant indicators of trouble, as well
as important in and of themselves.
2.2.3. The UN human rights machinery
In this period, the UN was peripherally involved in the conflict management process in Rwanda,
while humanitarian agencies, as noted, provided significant aid. The UN also had a formal
mandate to monitor and pronounce upon human rights violations in member states. This it did in
Rwanda, but in a very limited fashion compared to the vigorous and thorough monitoring carried
out by NGOs.
The UN Commission on Human Rights discussed Rwanda in both 1992 and 1993 under its “1503
procedure”, the confidentiality clause used to initiate discussion on countries where serious
problems are noticed.46 Customarily, the 1503 procedure is a warning signal that alerts the
Commission to the fact that human rights conditions within a country merit particular attention.
As the discussion was protected by the confidentiality clause, little is known of what transpired
when the Commission’s Committee of Five examined Rwanda in 1992, and again in 1993. The
first year, only a pro-government spokesman appeared, a local judge known to Western diplomats
for his grossly inadequate handling of cases brought against government officials. By the time of
the second meeting, Rwanda had a coalition government, and a member of the opposition MDR
party was in the delegation, the then Minister of Primary and Secondary Education, Agathe
Uwilingiyamana. Mrs. Uwilingiyamana – who was the first high-level official to be murdered
after 6 April – admitted to human rights problems and said the government would try to rectify the
situation.47
Rwanda was also discussed in several thematic reports presented to the Commission as a whole in
1992 and 1993.48 In early 1993, human rights groups in Rwanda asked the Special Rapporteur
on extrajudicial executions, Mr. B.W. Ndiaye, to participate in an international commission of
inquiry into human rights violations in the country. While declining on procedural grounds,
Ndiaye was sufficiently concerned to visit Rwanda. His 10-day visit produced a report in August
1993 (E/CN.4.1994/7), which was presented to the Commission early in 1994, i.e. shortly before
the genocide took place. The report was given only routine consideration in the Commission. As a
political body composed of UN members, the Commission rarely discusses particular countries
except those singled out for attention by a Special Rapporteur. Not being singled out, Rwanda was
folded into a report the Commission discussed thematically rather than according to country cases.
Because the Ndiaye report raised the question of genocide, the failure of the Human Rights
Commission to give it more than routine recognition has later been criticized. One explanation lies
in the nature of the report. Ndiaye supported the substance of the allegations of the NGO
Commission (FIDH 1993), in particular concerning official involvement in the massacres of
civilians (para. 28). He further noted that the question of whether the massacres may be termed
genocide “has often been raised” (para.78). After a brief discussion – contained in 4 out of the
report’s 86 paragraphs – Ndiaye offered an “initial reply”: since Tutsi have been the victims in the
overwhelming majority of cases, and have been targeted solely qua Tutsi, the Genocide
Convention’s Art.II (a) and (b) would apply.
A Western diplomat in Kigali who read the report, and himself was nearly killed by the Hutu
extremists, said it hardly prepared him mentally for the massive slaughter that was to come.
Rather, the Ndiaye report was read as a technical, legal analysis, not an alert that raised the specter
of the killings of half a million or more people. The magnitude of deaths associated with genocide
in popular opinion was absent.49 Moreover, the report presented the unfolding violence in the
context of “business-as-usual”: “...the country has already experienced many massacres of an
ethnic character. Such acts of violence recur periodically, and the persons responsible, who in
33
most cases are known to everyone, go unpunished. The Special Rapporteur became aware ... that
this situation was part and parcel of everyday life for many Rwandese people, and that ethnic
violence had become a practice which, if not accepted, was at least firmly rooted in Rwandese folk
memory.” (para 46)
The failure by the international community to make effective demands on its relationship with a
regime whose involvement in organized killings of civilians was generally suspected and carefully
documented in two 1993 reports – one from a joint NGO mission and the other by a Special
Rapporteur of the UN Commission on Human Rights – was an act of omission that carries at least
some moral co-responsibility for subsequent events in Rwanda. Acts of commission carry heavier
responsibility.
French aid was justified with reference to the continuing war with the RPF. The nature of the
conflict meant, however, that France helped to strengthen the military apparatus of a regime that
was deeply implicated in serious human rights violations, and some of whose members
subsequently became deeply implicated in the genocide as well. French military aid also
strengthened the latter more directly in that some weapons provided to the FAR were diverted to
the militias. Milita members were believed to be among the army recruits French coopérants
militaires were training (Prunier 1995). Diplomats in Kigali observed French officers with
interahamwe units in a national park area.50 On the eve of the civil war, and even a couple of days
after the killings had started in April 1994, French planes reportedly flew in arms to the Rwanda
government (see 3.2.).
Information about the human rights situation was generally available; given its close relations with
Rwanda, France had better access than most to obtain intelligence on the extremists. Yet it is
striking that senior officials in the Ministry of Cooperation in 1993 regarded Rwanda as “the
Switzerland of Africa”, as one official later said. To the extent that human rights violations were
noted, they were considered as not particularly bad “by African standards”. Evidently, the Office
of Military Assistance (DAMI), which is located in the Ministry of Cooperation and reports to it,
either did not observe the mounting force of Hutu extremism – which is possible given its
preoccupation with monitoring the designated enemy, the RPF – or observed but did not report.
Either way, for the ministry most directly and broadly involved in Rwanda, it amounted to a
massive intelligence failure.
French policy in Rwanda was hardly the result of a grand design. More likely, it evolved as the
outcome of multifarious and partly uncoordinated initiatives, each of which was shaped by the
prevailing sense of France’s purpose and role in Africa. The result was to give encouragement to
forces that were essentially evil, yet, the process itself demonstrated in a sense “the banality of
evil” (Callamard, 1995). The absence of a grand design does not, however, obviate responsibility
for the consequences. The French have been their own most severe critics in articulating the
political and ethical implications of France’s Rwanda policy (e.g. Bayart 1994, Brauman 1994,
Verschave 1994). Even from the criteria of Realpolitik it seems questionable. As the noted scholar
Jean-François Bayart asked, if French policy aimed to stabilize the situation, how could that be
achieved by supporting a regime that permitted, or even organized, assassinations and mass
murder?
34
Chapter 3
Implementing and Subverting the Peace
Agreement
(4 August 1993 – 6 April 1994)
The transition from peace negotiations to implementation was accompanied by mounting tension.
One fundamental problem was related to the domestic dynamic: the closer the peace agreement
came to be realized, the more desperate grew the extremists. While the Accords provided for a new
government that many hoped would subdue the militant opponents, the installation of the BBTG
was repeatedly delayed by parties who manipulated the transition instruments for narrow gains. As
the implementation process floundered, unforeseen external events in neighboring Burundi
heightened ethnic tension and strengthened the appeal of the extremists, thereby making
implementation of a delicate political compromise even more difficult. The UN Assistance Mission
in Rwanda (UNAMIR) had been designed to oversee a smooth transition and was not equipped to
handle the situation. Nor was it strengthened despite increasing recognition that civil war might
resume and large-scale civil violence erupt.
3.1. Implementing the peace agreement
To assist implementation, the Arusha Accords provided for the establishment of a neutral
international peacekeeping force to be deployed in Rwanda within 37 days from the date of signing
(i.e. by 10 September). The UN promptly sent a technical mission to Rwanda for needs
assessment. Three issues arose out of that mission: the timeliness of deployment, the size of the
force, and the precise mandate as well as rules of engagement for the force. Their resolution
depended in part on the team’s assessment of the situation in Rwanda, in part on what the UN
members were willing to invest.
The Reconnaissance Mission, headed by the designated Force Commander, General Romeo
Dallaire, reported a mixed picture. It was a foregone conclusion that the operation would come
under Chapter VI peacekeeping, and the situational prerequisites were present: a peace agreement
was signed, and the local parties consented and cooperated. In the meantime, both sides were
observing the cease-fire agreement, which was monitored by both regional (OAU-NMOG) and
international (UNOMUR) observers. To those in the UN who looked for a success to counter the
troubled UN mission in Somalia, “Rwanda seemed like a winner”, as one participant noted. In an
unprecedented move that foreshadowed a an easy mission, a joint delegation from both sides of
the Rwandese conflict came to New York to lobby the UN to endorse the peacekeeping operation.
Yet Dallaire’s team was also warned of instability and dangers. The schedule for deploying
peacekeepers was unrealistically short, so time was of the essence. A transitional government
required the participation of RPF members as designated, but the RPF absolutely refused to enter
Kigali until the French forces had left, and the French would not leave until the UN arrived. Thus,
the UN force held a key to the entire transition.
Four standards are relevant to measure the rapidity of deployment. On the standard of need, the
troops were required as soon as the Arusha Agreement was signed. Closely following was the
expectation of the Agreement itself, i.e. 10 September, which was generally agreed to be an
unrealistic deadline. Third, there was the timeline the UN Secretariat set for itself as spelled out by
the Reconnaissance Mission. In some important respects it was not met.51 Fourth, however,
standard UN procedures meant that 3–6 months would be needed to assemble and deploy a
peacekeeping force, depending upon the size. According to this standard, UNAMIR was deployed
35
fairly rapidly. The UN Secretariat started assembling the Reconnaissance Mission a few weeks
before the Arusha Accords were signed on 4 August, and the first battalion reached Rwanda in late
November.
Delays in establishing a transitional government would increase the likelihood that the entire peace
agreement would be derailed, the Mission was told by several foreign diplomats and the Rwandese
parties to the Agreement.52 It was known that militant forces remained vehemently opposed to the
Accords, and Rwanda’s history of ethnic and political strife cast a long shadow. Recent violence
and the growing strength of paramilitary groups added to the formidable problems of demobilizing
some 35,000 soldiers in a society with a shattered economy, famine, and large numbers of
persons displaced by the war. In this situation, concluded the UN Secretary-General, it was
imperative that the international community demonstrate its commitment to implement the peace
agreement it had helped to mediate. Delays would “seriously jeopardize” the Accords
(S/24688/1993 para.65).
One military expert in the Secretariat estimated that ideally, a mission of this kind should have
8,000. Dallaire proposed 4,500 as his maximum option. In the end, the Secretariat recommended a
force only half that size, anticipating that this was the maximum that the Security Council would
approve. Having exercised an anticipatory veto, the Security Council on 5 October authorized a
force level of 2,548 military personnel without much discussion.
The decisive restraint on the overall size of the mission was financing. Being assessed 31% of the
costs of UN peacekeeping, the United States insisted on a minimal force. The Clinton
Administration had just started an executive review of its UN policy, and was sensitive to
Congressional concern over the mounting American share of peacekeeping costs, which had
increased a stunning 370% from 1992 to 1993.53 Emphasizing the bright aspects of the Rwandese
situation, the cost-conscious US delegation in New York suggested in September that a token
mission of some 500 men would suffice. The French mission in New York recommended a small
force of around 1,000 men, noting that the French contingent in Kigali was merely 6–700 men.
The end result of 2,548 was more than a token force, and at that time considered quite acceptable
by the Force Commander. UNAMIR was estimated to cost about US$10 million a month, a very
modest amount compared to other UN peacekeeping operations.54
The signals from the Rwandese parties concerning force size were conflicting. A skeptical RPF
reinforced calls in the Security Council for a minimalist force. Recalling earlier French lobbying at
the Security Council to have a UN rather than an OAU force in Rwanda, the RPF suspected that
France was using UN peacekeeping operations as a shield to protect its political allies in Kigali and
secure a foothold for the future. The RPF consequently claimed that a few hundred men would do.
By the same logic, the government of Rwanda wanted several battalions.
Despite a speedy start,55 the deployment of UNAMIR did not fully meet the schedule laid out by
the Reconnaissance Mission. The special representative of the Secretary-General did not arrive in
late November, or three and a half months after the Arusha Agreement was signed. The first
UNAMIR battalion entered Kigali shortly thereafter, enabling the French paratroopers to leave.
Putting together the rest of the mission also proved difficult. The force never received the
equipment authorized and required even at the reduced level (including an armored unit and
helicopters).56 UNAMIR’s budget was subject to the standard, lengthy UN decision-making
process and was not formally approved until 4 April, 1994.57 As a result, the mission was
constrained by numerous shortcomings in personnel, equipment and disbursable funds (including
petty cash), and even basics such as ammunition. Repeated field requests to New York to bring
the Mission up to authorized strength were to no avail.
The problem was partly structural, reflecting cumbersome UN procedures for budgeting and
deployment of peacekeeping operations generally, as well a the political system in the UN. For all
but one of the major powers on the Security Council, Rwanda was of peripheral interest. This
meant that UNAMIR lacked a powerful patron in the Council to help cut through the bureaucraticpolitical morass that governed normal procedures for UN peacekeeping. Without this political
36
pressure, UNAMIR moved to the bottom of the UN security agenda. Nuts-and-bolts issues as
well as more fundamental mandate questions were neglected. Only the French government had a
direct interest and considerable presence but, with some reason, claimed that such involvement
restricted France to a low profile in the Council deliberations on UNAMIR so as not to
compromise the neutrality required by a UN peacekeeping force.
The barebones UNAMIR had weak or no capacity in two areas that were significant for its
operation. With only a small civilian police unit and no human rights cell, the mission had very
limited ability to investigate violent incidents. The force also lacked an official intelligence unit, a
fact that was deplored by the Force Commander and led to some improvization on the ground (see
3.2.)
3.2. Civil violence and the subversion of the peace agreement
In early autumn 1993, the Rwanda situation did permit some optimism that the peace agreement
would be implemented in an orderly fashion. Yet underlying tensions soon surfaced to reinforce
earlier warnings that the process was being subverted. The splitting of political parties, which had
started already in late 1993, became more profound as factions competed for positions in the
transitional government, thereby undoing the formula for power-sharing so carefully negotiated at
Arusha and playing into the hands of those who opposed the Accords altogether. Repeated failures
to form a transitional government made the diplomatic community in Kigali and states associated
with the Arusha process renew their efforts to break the deadlock. The Tanzanian government took
the lead, but Belgium, Germany, the United States, Canada and the Vatican also weighed in with
strong language and sometimes high-level representations to repair the process. During February
and March, and even right up until the crisis erupted on 6 April, concerted efforts were made to
avoid a rupture.
The diplomats who labored to save the Arusha Accords saw implementation as a means to control
the extremists. With Hutu moderates and RPF in charge of key ministries in the planned
transitional government – including Interior and Justice – and a new national army to be formed,
many expected that the Hutu extremists could be subdued. Also leading members of the RPF
politburo appeared to take this view. However, the implementation process was fundamentally a
race against itself: the closer the Arusha Accords came to be realized, the more the extremists were
prepared to sink it in violence. The point was formulated unequivocally in a cable from the
UNAMIR force commander to DPKO/ New York on 11 January 1994: A “very important
government official” turned informer had told him that “hostilities may commence again if political
deadlock ends”.58 Previously, “hostilities” had often been interpreted to mean renewed civil war.
By early 1994, the term clearly conveyed the meaning of civil violence. It was quickly recognized
by the SRSG, Jacques-Roger Booh-Booh, who reported to the Secretary-General soon after his
arrival in Kigali on 23 October that the extremist adversary to the peace process was a powerful,
subversive Third Force – ruthless, well-organized, well-equipped and centrally directed.
Additional information provided in the UNAMIR cable of 11 January to the UN Secretariat in
New York was quite specific. The Rwandese government informant had revealed the extremists’
plan to assassinate politicians at the scheduled ceremony for swearing in of the transitional
government. In the process, they would provoke an encounter with Belgian UNAMIR soldiers,
expecting that by killing some, the entire UN contingent would leave Rwanda. As the RPF would
be instigated to resume war, 1,700 interahamwe who had been training in camps outside Kigali
were staged to sew insecurity throughout the city; teams of 40 each were organized within Kigali
to kill all the Tutsi who had been registered by the militias. The informer estimated his units could
kill 1,000 persons per 20 minutes.
Assassinations on this scale were no longer of the order of a political tactic but clearly genocidal.
The message dovetailed with a series of other developments, indicating that the situation was
rapidly deteriorating. Both at the time and looking back, foreign diplomats and UN officials
37
serving in Kigali clearly noted the worsening trend. The October events in Burundi represented a
watershed. Rwanda’s Hutu population projected developments in the neighbouring state onto their
own society: a newly-elected Hutu president was assassinated and tens of thousands of Hutu were
killed in the ensuing violence, many in situations where Burundi’s Tutsi-dominated army was
directly or indirectly responsible. With the RPF poised to share both governmental power and half
of the national army in Rwanda, the message of fear of the Tutsi being spread by the Hutu
extremists gained new credibility.
Additional information was picked up by the small, unofficial intelligence unit set up by the
Belgian 1st paratroop battalion serving with UNAMIR under Col. Luc Marchal. The unit was
financed directly by Belgium after the UN Secretariat declined to finance it, and reported to the
Belgian military headquarters in Evere outside Brussels as well as the UN. During early 1994, the
Belgian intelligence channels provided detailed secret information, including vouching for the bona
fides of a key informant, detailing secret arms caches, and spelling out plans by MRND leaders, in
cooperation with interahamwe units, to isolate, harass and undermine the work of the Belgian
UNAMIR contingent.59 Belgium also had another intelligence channel. The military intelligence
agency (SGR) of the Ministry of Defense ran an operation attached to, and under cover of, the
small military assistance unit that Belgium maintained in Kigali (mainly for medical training) until
April 1994 (De Morgen, 4 November 1995).
Hutu supremacist rhetoric became increasingly vocal and public. Radio mille collines (RTLM),
_broadcast hate messages that called on the people to rid the country of Tutsi. African diplomats
reported CDR members telling them that in the future there would be no Tutsi left in Rwanda.
Referring to the RPF as the “historical overlords”, the RTLM whipped up fear by claiming that
once the RPF army was in Kigali – as provided for in the Arusha Accords – Tutsi troops would
attack the Hutu population.
In the streets, the violence was real. Assassinations of two high-level politicians in late February
made the CDR organize a four-day rampage in Kigali with interahamwe mobs (readily recognized
by the colors of their “uniforms”) roaming the streets and killing dozens of people. The mob
violence clearly targeted individuals, and even foreign diplomats considered “pro-Arusha” were
threatened. The Tanzanian Ambassador was almost killed on 5 January when a CDR crowd
attacked his car. The Papal Nuncio received grizzly death threats.
UNAMIR’s Force Commander responded to the deteriorating situation by repeatedly requesting
equipment to bring his force up to strength. In January, he sought guidance from the Department
of Peace-keeping Operations in the UN Secretariat (DPKO) on actions to be taken if the situation
worsened. In early February, he asked for clarification of the mandate to permit searches of arms
caches. In late February and again in mid-March, he requested reinforcements.60 Reports of arms
being distributed to para-military groups in the Kigali area were frequently discussed by UNAMIR
staff, reported by the Belgian military intelligence unit back to the Belgian government, and were
circulated in the diplomatic community. Some went public with their concern, including the
UNAMIR commander of the Kigali sector, Col. Luc Marchal, and the SRSG, Jacques-Roger
Booh-Booh. Noting that violence was about to overtake the peace process, Booh-Booh
complained to the press on 24 January that “weapons are distributed from arms caches around
Kigali and even inside town.”
Some arms shipments could readily be traced. On 22 January 1994, a planeload of arms from
France was confiscated by UNAMIR at the Kigali airport.61 The delivery was in violation of the
cease-fire clauses of the Arusha Accords, which prohibited introduction of arms into the area
during the transition period. Formally recognizing this point, the French government argued that
the delivery stemmed from an old contract and hence was technically speaking legal. More serious,
UNAMIR’s Kigali sector commander has reported that a planeload of arms arrived from France in
the early morning of 9 April, that is, shortly after the killings started. The French government has
categorially denied this shipment.62
Arms distribution affected the immediate security situation as well as the climate for implementing
38
the peace accords. It also went to the core of UNAMIR’s mandate to help make Kigali a “weapons
secure area” where weapons were strictly controlled. Failure to react to flagrant and illegal arms
distribution clearly would erode UNAMIR’s credibility. Immediately after sending the 11 January
cable detailing the interahamwe plans for killing Tutsi in Kigali, Dallaire requested permission to
seek out arms caches that the informant had offered to identify. In the UN Secretariat, however,
the DPKO denied permission. On three occasions in early February, Dallaire made similar requests
to carry out cordon-and-search operations to seize arms. Each time the DPKO turned down the
request. Similar calls from the Belgian government to permit UNAMIR to interpret its mandate
more proactively – issued after the Belgian Defence Minister Leo Delcroix had visited Rwanda in
mid-March – were likewise rejected.63
In retrospect, it is hard to fault the UN for failing to anticipate the impact of external events,
notably the October massacres in Burundi. The prescient and cumulative warning signals from
Rwanda itself were more compelling. They were taken seriously by many in the field: diplomats
redoubled their efforts while UNAMIR tried to respond. In New York, however, the Secretariat
was passive or cautious. The most critical warning from the field – the 11 January cable – was put
aside in the DPKO.64 Requests to bring the force up to strength, let alone provide reinforcements,
were rejected, although the deployment of the second battalion was speeded up. UNAMIR was
denied permission to interpret its mandate proactively and search for arms. When the SecretaryGeneral requested that the Security Council renew UNAMIR’s three-month authorization, he did
not mention the critical, new intelligence elements in the situation. The reasons why this
information apparently was not reported to the Council are unclear.
New York’s response reflected a cautious and politically sensitive interpretation of UNAMIR’s
mandate. The latter had been tailored to peacekeeping of a classic consent-and-cooperation kind.
The mission was to “contribute to the security” of the Kigali area, monitor the cease-fire, and
assist local authorities in demobilizing the two Rwandese armies and investigating violations of the
Accords (Res.872(1993)). It is important to recall that this mandate grew out of but differed from
what was envisaged in the Arusha Accords. In the central clauses defining UNAMIR’s role in
providing security, in protecting civilians, and in confiscating illegal arms, the Arusha Accords
were significantly broader than the terms of the final UN mandate.65 The disjuncture between the
mediation phase and implementation was again revealed: the signatories to the regionally-brokered
peace accords had evidently assigned an importance to the peacekeeping force that exceeded what
the UN was willing to provide.
In the fall of 1993, the UN Secretariat and the Security Council had considered Rwanda a low
priority, drafted a classic and minimalist peacekeeping operation, and hoped for the best. As the
situation in Rwanda progressively deteriorated, the Secretariat and the Security Council members –
in particular the United States – labored under the impact of the Somalia experience. The crisis in
Somalia, where US/UN troops became embroiled in a de facto war with a local faction and took
casualties, had been politically costly both for the US government and for the UN SecretaryGeneral. Not crossing “the Mogadishu-line” became a firm dictum in the Secretariat, as a highlevel official in the DPKO later put it. A proactive interpretation of UNAMIR’s mandate to permit
confiscation of illegal arms was much too close to the Somalia syndrome for comfort.66
Having ruled out a proactive and coercive response in Rwanda, the Secretariat and the Security
Council left themselves only one option in case of a crisis: withdrawal. The strategy was spelled
out by Boutros-Ghali in the form of a bargaining tactic: unless the peace agreement was
implemented, the UN would withdraw its peacekeeping force. The message of the SecretaryGeneral was reiterated by the Security Council when it renewed UNAMIR’s authorization in
January and again on 5 April.
As the crisis intensified, the range of options to save the peace agreement narrowed. Nevertheless,
in Kigali, UNDP proceeded with the planning of the demobilization program by bringing the RPF
and Rwandese army officers together to prepare the details, including the rate of demobilization
and measures to reintegrate the soldiers back into the civilian economy. The planning process –
which was sustained for a full five months – was in itself a confidence-building measure. In a
similar spirit, donors offered economic incentives by holding out the promise of large development
39
funds once the new transitional government were installed. Using a different tactic, the UN
Secretary-General and the Security Council threatened to withdraw UNAMIR, although this was
what the Hutu extremists wanted.
In retrospect, it is possible that by early 1994, Habyarimana himself was a prisoner of the
extremists. If so, putting pressure on the President would yield few results. The option of
strengthening UNAMIR and applying force against the extremists – which theoretically was
available to the UN _– might have been effective. This option was not considered; allowing
UNAMIR to show a bit of muscle was rejected. While this might not have resulted in the
installation of the BBTG, constraining the extremists could have modified the violent
consequences.
It is not a foregone conclusion that a stronger or more proactive UNAMIR would have altered the
course of events leading up to 6 April. The consequences of weakness, however, were reasonably
clear. Slowly and haltingly deployed, UNAMIR failed to project strength and decisiveness from
the beginning. As the struggle over implementation intensified, the force was constrained from
acting in matters pertaining to the security of Kigali and thus central to its mission. Its presence
gave a false sense of security to Tutsi as well as Hutu regime opponents and moderate coalition
members who, feeling themselves protected, spoke out against extremists and coup-plotters. All of
them, a close observer later wrote, “knew what the Hutu extremists were capable of, but many ...
were persuaded that the presence of UNAMIR would prevent things from going much further than
the repression in October 1990.”67 When the situation radically changed and widespread violence
erupted, UNAMIR lacked everything from sandbags to armoured personnel carriers to protect
either its own troops or civilians.
40
Chapter 4
Crisis and Withdrawal
(6 April 1994 – 21 April 1994)
4.1. Crisis
On 6 April at approximately 8:30 in the evening Kigali time, the plane carrying Habyarimana was
shot down as it was returning from Dar-es-Salaam, where the President had finally committed
himself to implement the BBTG. What followed in the next 24 hours was the almost simultaneous
occurrence of a military coup, renewed civil war, systematic political assassinations, and
commencement of genocide. At this stage, early warning with respect to forthcoming events was
irrelevant. What was required for rational decision-making was knowledge about the confusion in
the present. Who knew, what did they know, and to whom did they communicate what was
known?
By the evening of 7 April, the UNAMIR command had put together some major pieces of the
picture: the Prime Minister and other key moderate and opposition political leaders had been killed;
the well-known extremist, Col. Bagosora, appeared to be in charge of the coup; 10 Belgian
UNAMIR soldiers had been overpowered and subsequently murdered by Rwandese government
soldiers who accused the Belgians of having shot down the President’s plane; the RPF unit in
Kigali was breaking out of its quarters while the main RPF force in the north prepared to move
down through the DMZ; the militia were manning roadblocks and civilians were being killed.68
The Force Commander knew that a military coup and politicide (systematic political assasinations)
had taken place. He knew civilians were being killed, but did not recognize that genocide was
under way. He considered his main objective to be safe evacuation of the expatriates69 and
prevent the resumption of the civil war, while ensuring the safety of his own troops.
With the coup a fait accompli, the immediate options were limited. New York instructed Dallaire
and Booh-Booh to talk with the representatives of the “interim government” and the RPF in order
to obtain a cease-fire and restart the Arusha process. The UNAMIR Force Commander also had a
number of immediate tasks. With inadequate transport, he had to relocate his Ghanaian battalion to
Kigali since the RPF had ordered it out of the DMZ. Escort services were provided for evacuation
of expatriates, which began to be organized by the morning of 8 April. He had to deal with
French, Belgian, and Italian forces who had arrived, not to assist him, but to secure the airport
while evacuating their nationals. UNAMIR also provided protection to Rwandese nationals who
had fled to the Amahoro stadium and a nearby hospital. In all of this, the force had only one
working armoured personnel carrier, a demoralized Belgian battalion and an under-equipped,
below-strength unit from Bangladesh.
Between Friday and Monday, most expatriates had managed to leave. With the evacuation problem
less pressing, the UN started to focus on UNAMIR’s role in the new situation (Lægreid, 1995).
During the first week (April 7–13), much of the initiative lay with the Secretariat, which received
information from its newly-established Situation Center, was in constant telephone contact with
the Force Commander in the field, and was repeatedly asked by the Security Council to provide it
with information and options.70 Until 13 April, when Belgium formally told the Security Council
of its decision to withdraw from UNAMIR, the diplomatic situation was fluid, with a range of
actions available to the Secretary-General. There were daily, sometimes twice-daily informal
consultations between the Security Council and the Secretariat. The week, in other words, was a
window of opportunity for the Secretariat to define UN policy towards Rwanda.
41
The Secretariat proceeded with great caution. After discussing the situation in the SecretaryGeneral’s task force, DPKO officials suggested to the Security Council that the most likely
alternatives were withdrawal or a Chapter VII enforcement action. Under Secretary-General Kofi
Annan told the Council on 9 April and again on 11 April that if UNAMIR were to carry on,
additional resources and a new mandate were needed. On 13 April, Assistant Secretary-General
Iqbal Riza raised the question of humanitarian intervention to protect civilians, only to dismiss it.
Acknowledging that at the time UNAMIR was actually trying to protect civilians in Kigali, Riza
cautioned that this course of action would require more resources, and that the Council should
consider whether peacekeeping operations should involve such tasks.
The Security Council commented critically on the failure of the Secretariat to come up with options
outside the classic framework of Chapter VI versus Chapter VII, or a Somalia-type engagement
versus withdrawal, as the British Ambassador said on 13 April .71 One reason stemmed from the
nature of the conflict. In the dynamic interaction of the response to the genocide and the
management of the renewed civil war, there was a failure to distinguish between the two. Options
to reinforce UNAMIR were always put by the Secretariat in terms of an enforcement operation,
suggesting intervention between the two armies, rather than maintaining or increasing troop
strength to protect civilians. Even two weeks after the killings commenced, the Secretary-General
reported to the Security Council in terms that showed a preoccupation with the war between the
Rwandese army and the RPF (S/1994/470, 20 April 1994). An option to deal with the war on
civilians was never presented to the Security Council, and the latter did not strike new directions
on its own. The decision to withdraw the bulk of UNAMIR on 21 April was taken accordingly in
the context of a situation depicted by the Council as civil war with related “mindless violence”,
rather than organized genocide accompanied by a smaller civil war (Res.912(1994), 21 April
1994).72 This misinterpretation of the situation occurred despite the earlier warnings about plans
for organized, systematic killings received by the Secretariat in communications from UNAMIR
and elsewhere.
Despite incomplete information during the seemingly chaotic first days after 6 April, the UN was
receiving through its own channels information to the effect that the killings were both extensive
and organized. In a perceptive cable, Booh-Booh, the SRSG, reported from Kigali already on 8
April that “a very well planned, organized, deliberate and conducted campaign of terror” was
under way.73 Aggressive action had been taken against the opposition leaders, Tutsi were being
massacred, and UNAMIR and the RPF were attacked. There were indications of large-scale
killings. On April 11 the DPKO Situation Centre relayed reports of “many thousand deaths”. The
Nigerian draft resolution, which was circulated in the Security Council on 13 April (but never
tabled), expressed shock over the “death of thousands of innocent civilians”. In the following two
weeks, additional information about the scale and pattern of killings came in.
When the Secretary-General on 20 April formally presented the Security Council with options of
how to respond, he failed to recognize the organized and systematic nature of the violence
(S/1994/470). Instead, Boutros-Ghali saw anarchy and spontaneous slaughter. It was the same
distorted picture that much of the Western media had reported in April (see 4.3 below). Given the
significant role of the Secretariat at this juncture, the 20 April report merits closer attention.
Prefacing his options with a discussion of the situation in Rwanda, Boutros-Ghali claimed that
“reliable reports strongly indicate that the killings were started by unruly members of the
Presidential Guard” (not the Presidential Guard itself, and certainly not the extremists). Stressing
the initiative of the RPF in resuming the civil war, Boutros-Ghali reported that the RPF battalion
was breaking out of its compound in Kigali and “started to engage the government troops...despite
the best efforts of UNAMIR”, while other RPF units from the demilitarized zone were moving
towards Kigali. There was no mention of an organized conspiracy, and when the killing of
moderate cabinet ministers (starting at dawn the morning after the plane was shot down), of the
Prime Minister, and of the 10 Belgians were mentioned, those deaths were blamed on “unruly
soldiers”. There was no mention of the beginning of the organized murder of Tutsi. The
provisional government was depicted as disintegrating, as if it fell apart on its own. “Authority
42
collapsed”; (it did not, at least not the authority organizing the conspiracy). Instead of stressing the
need to stop the massacre of civilians, Boutros-Ghali declared that, “the most urgent task is the
securing of an agreement on a cease-fire in the hopes that this would lead to the resumption of the
peace process and reviving the Arusha Accords.” Further, UNAMIR could not be left at risk
indefinitely when there was no possibility of performing the task for which it was sent. The
implication was that its sole task was to observe the cease-fire. The statement made no mention of
UNAMIR’s efforts on the ground to protect civilians, a subject that had been discussed in both the
Secretariat and the Security Council.
A month later, on 25 May, Boutros Boutros-Ghali publicly confessed: “We are all responsible for
this disaster, not only the super-powers, but also the African countries, the non-government
organizations, the entire international community. There has been a genocide, and the world is
talking about what it should do. It is a scandal.”74 In light of the historical record, this statement
might appear as an effort to shift the responsibility around. Earlier misperceptions and inaction not
only reduced the impact of the statement, but possibly also reduced the subsequent and very
significant efforts by the Secretariat to reverse direction.
From 29 April and onwards, the Secretariat played a vigorous and innovative role vis-à-vis the
Security Council. Its inability to produce a more differentiated set of alternatives earlier, and the
near-paralysis that characterized the first week, is therefore puzzling. It is also important since
early action conceivably could have prevented the critical loss of power and credibility brought on
by withdrawal.
Evidently several factors were at work. There were institutional limitations. While expanding
rapidly between 1990 and 1993 to keep up with burgeoning mission activities, the Department of
Peace-keeping Operations in early 1994 had less than 200 officials to oversee 70,000 military and
civilians in the field.75 It had a one-person unit for policy analysis, another one-person unit for
evaluation. Moreover, nowhere in the Secretariat had information from the past been accumulated,
analyzed and structured to prepare policy options either for pre-emptive action or for crisis
management, so important given the general inability of bureaucracy to innovate during a crisis.
With limited previous experience in protecting civilians, and doctrinal poverty, the Secretariat’s
response was framed by the conventional dichotomy of Chapter VI versus Chapter VII action.
Those charged with leadership had a distorted view of events: both the Secretariat and the Security
Council drew a picture of Rwanda as a failed state in which rogue troops and spontaneous mobs
were killing Tutsi. The concept of failed state, of course, suggested the analogy with Somalia,
then uppermost in the consciousness of UN officials. Bureaucratic caution reinforced the
conclusion drawn from that experience: the UN could not afford another peacekeeping failure,
with failure defined as loss of UN peacekeepers in the field. Finally, the striking aspect of the first
week of crisis was the physical absence from New York of Boutros-Ghali, who was travelling at a
brisk pace in Europe and the former Soviet Union.76 During the fast-moving and critical first days
of crisis, the Secretary-General was unavailable to provide leadership for action.
4.2. Withdrawal
Until the middle of April, there were open divisions in the Security Council on the issue of
UNAMIR’s future. The non-aligned states, led by Nigeria, argued for strengthening UNAMIR,
and on 13 April circulated a draft proposal to that effect. By then, however, it was too late and not
backed by offers to contribute. The Belgian decision to withdraw its contingent, following the
murder of the 10 Belgian peacekeepers, was communicated to the Secretary-General on 12 April,
and to the Security Council the next day. UNAMIR was deprived of its strongest unit, and put the
rest of the force in a precarious position. The Secretary-General made the point bluntly in a letter to
Security-Council members on 13 April. Belgian withdrawal will make it “extremely difficult for
UNAMIR to carry out its task effectively...In these circumstances, I have asked my Special
Representative and the Force Commander to prepare plans for the withdrawal of UNAMIR.”77
Having decided to withdraw its own contingent, Brussels lobbied hard to persuade Council
43
members that conditions in Rwanda necessitated withdrawal of UNAMIR as a whole. The stance
was widely seen as an attempt to legitimize its own withdrawal, but the Belgians were pushing on
an open door.78 They were strongly backed by the Americans;79 the UK and France, though less
vocal, also favored withdrawal. No country came forward with troop contributions, and the
Secretariat claimed later that informal canvassing at the time had produced negative results.80 It
was obviously not a question of capacity; collectively or individually, several UN members had
the means to intervene decisively, as France and Belgium had shown by their efficient airborne
operations to evacuate expatriates.
The assessment from the field was that, even with the Belgian withdrawal, the rest of the mission
of around 1,200 should carry on as long as possible. “My decision to remain was a matter of
moral concern,” Dallaire later wrote.81 With the tide running the other way in New York,
however, alternate perspectives from the field were not forwarded to the Security Council in a
forceful manner. Instead, the options were formulated as full withdrawal versus leaving a token
force. In the “stampede to get out”, as one ambassador later noted, the Nigerian draft resolution
for strengthening UNAMIR was not even tabled.
While the Security Council debated options, UNAMIR’s position in the field deteriorated. The
Belgian contingent completed its withdrawal on 20 April. Dallaire redeployed the Ghanian
battalion from the DMZ to the airport, enabling the Canadian communications unit to operate.82
Otherwise, his force was not being re-supplied and had to work hard to secure even some
equipment from the departing Belgian contingent. UNAMIR was short of water and fuel, and was
taking indirect fire. Continuing negotiations with the RPF and FAR to respect a UN-supervised
neutrality of the airport did not succeed, rendering the airport and hence the exit route insecure. In
a self-fulfilling prophecy, UNAMIR had been weakened to the point where withdrawal
increasingly seemed the most rational option. The field mission still refused to concede the point.
On the eve of the Security Council’s final decision on 21 April to withdraw, Dallaire’s “cabinet” of
officers recommended as their first option that they stay on at present strength (i.e. without the
Belgian battalion), while listing scaling down and withdrawal as less desirable alternatives.
The formal decision to withdraw was taken by the Security Council in the context of a choice
between three options formulated by the Secretary-General. In reality, all but one option had
already been ruled out in the informal and officially unrecorded consultations in the Council and its
communications with the Secretariat, probably de facto almost a week earlier. As presented by
Boutros-Ghali on 20 April, the language clearly excluded the option to increase: the situation in
Rwanda could only be changed by “immediate and massive reinforcement of UNAMIR... which
would require several thousand additional troops and UNAMIR may have to be given enforcement
powers under Chapter VII” (S/1994/470. para.13). The alternative of complete withdrawal also
seemed difficult. Given the “dimensions of the violence and mass killings over the last two
weeks...the consequences of complete withdrawal, in terms of human lives lost, could be very
severe indeed” (para 19). That left the middle alternative of reducing UNAMIR to about 270
military personnel. These would take on a diplomatic role by promoting a cease-fire and, when
feasible, assist in the resumption of humanitarian relief. On 21 April the Security Council
approved this option over the recorded doubt, but with the votes, of its non-aligned members
(S/PV.3368, 1994).
The two weeks following the shooting down of the plane carrying President Habyarimana and
associates constituted a critical decision-making phase in the international response to the conflict.
If there were any chances of stopping the genocide before it fully unfolded, this was the time.83
Pulling out would mean abandoning the initiative to the forces on the ground. To re-enter would
require additional efforts to secure the airport. The implications were fully understood by the
UNAMIR officers who discussed options in the middle of the madness that was Kigali on 21
April. Reducing the force to 270 men, which the Security Council was about to decide, meant “we
can do little”, and “we will really not be able to come back”, the Deputy Force Commander
scribbled on a notepad.
The Security Council and the Secretariat saw withdrawal as a means to salvage a UN peacekeeping
44
operation that had been tailored to a situation that no longer existed. But by largely absenting itself
from the conflict, the UN simultaneously lost leverage to influence its future course – on the
ground and diplomatically. External conflict management essentially came to a halt. When the UN
subsequently reversed itself, re-entry proved slow, difficult and fundamentally too late.
The civilians, thousands of whom were being killed daily, were largely abandoned to their fate. A
symbolic presence at key points in the Kigali area enabled UNAMIR to provide protection for an
estimated 20,000 persons (at the Amahoro stadium, the Hotel Mille Collines, the Méridien Hotel
and the King Faysal Hospital). Initially, a Belgian platoon effectively protected several thousand
persons at the Ecole Technique Officielle; as soon as it was withdrawn, the killers closed in.
Protection of this kind relied on a combination of bluff, inertia or intersecting interests much more
than a projection of force. By their presence, ICRC and MSF personnel also provided limited
protective “space” in Kigali and elsewhere;84 they were later joined by UN agencies and NGOs
providing humanitarian assistance. Despite these efforts, the limited scope and vulnerability of
such protection meant that, for the vast majority of civilians, security as well as humanitarian
assistance came to depend primarily upon the advance of RPF troops. These came too late to save
many.
Given the importance of the Belgian decision to withdraw from Rwanda, several questions have
been raised in this connection. One set concerns the circumstances of the death of the 10 Belgian
soldiers and how the UNAMIR command handled the matter. This has been the subject of an
official Belgian investigation but remains a source of some controversy.85 Another question is
whether it was wise to have a Belgian unit in UNAMIR, given Belgium’s colonial past in Rwanda
and recent policies, which had caused hostility on both the government side (for failing to render
sustained support during the 1990 offensive) and that of the RPF (for having tried). The Belgians
clearly did not meet the formal UN requirement that peacekeepers should come from states that are
neutral relative to the country where they will serve. The wisdom of sending Belgians was for that
reason questioned at the time. However, no other country came forth to provide a fully equipped
unit, and quick deployment, it will be recalled, was considered important.
Others have argued that it was its European and not its Belgian identity that made the contingent
targeted and vulnerable to extremists. The latter clearly sought to replicate the American
withdrawal from Somalia by capitalizing on the low threshold in Western states for incurring
casualties in Africa. By contrast, Ghana did not propose to withdraw its battalion, nor did several
African and other states whose military observers were attached to UNAMIR when the old NMOG
was folded into the larger structure. For all of them, the decision to withdraw was made by the
Security Council.
The small UNAMIR force that remained (it turned out to be 540) had as one of its main tasks the
promotion of a cease-fire in the civil war, which had resumed alongside the genocide. For this
“obsession”, the UN has been severely criticized (African Rights 1995:1121). The argument is
that the UN focused on the least devastating conflict – the civil war – rather than the genocide,
and, by implication, sought to restrain the advance of the RPF which was the only force on the
scene able to stop the mass killing of civilians. The factual points are not here at issue. The
Security Council consistently demanded a cease-fire, and the UNAMIR mission in Rwanda
continuously sought to negotiate one. Motivations are more difficult to assess. The rationale for
demanding a cease-fire may have been politically motivated to halt the RPF, as the RPF claims.86
It was also a habitual UN response to wars of any kind. When the decision was made in May to
redeploy a larger UN force, a cease-fire was considered even more important.
How a cease-fire would have affected the level of killings is speculative. The extremists were
determined and well organized; rage and fear prevailed. This suggests that only a military defeat
would have put an end to the killings. On the other hand, there were divisions in the army and
moderate army factions might have restrained the killers if a cease fire had been obtained. The
Chief of Staff in the self-styled interim government, Gen. Bizimungu, told the UN Special
Rapporteur that the government would stop the killings only after the RPF agreed to a cease-fire,
45
implying that this was within its power. Also, a cease-fire might have reduced the fear and rage
among the Hutu that accompanied the advance of the RPF and possibly intensified the killings.87
As it was, the war and the killings continued until the middle of July, when the RPF proclaimed a
new government in Kigali. For almost three months after 6 April, RPF units were the only
significant force seeking to stop the killings as they advanced. The RPF consequently made a
cease-fire conditional upon an end to the killings.
OAU efforts in this period also focused on the need to obtain a cease-fire and restore the Arusha
process. There was no evident recognition of an organized genocide. An ambassadorial-level
meeting of the Central Organ on 14 April referred to “carnage and bloodletting”; Secretary-General
Salim’s letter to his UN counterpart on 21 April mentioned “massacres and wanton killings”. A
consistent theme of OAU communications to the UN, however, was that the UN ought not to
reduce the mission in Rwanda it so eagerly had sought the previous year.
4.3. The media
Had there been any forewarning in the media? Rwanda is a relatively small country on the
periphery of the industrialized world. As might be expected, during the first three months of 1994,
there was virtually no Western media coverage of events in Rwanda.88 In the early part of the
year, the main African story was the UN withdrawal from Somalia, which would prove prescient
in itself. Massive media attention then switched to the violent build-up to the South African
elections.
Surveys of the British, French and US media (Hilsum 1995, Johansson 1995, Livingston and
Eachus 1995, Verschave 1994), show that relatively little change occurred in the media coverage
after 6 April compared to the paucity before. There was a blip with the shooting down of the plane
and the reporting on the slaughters – generally portrayed as ancient tribal feuds – but with the
withdrawal of foreign personnel there was a precipitous drop in coverage. When the genocide was
accelerating, the Western press virtually ceased to report on Rwanda. The lack of coverage cannot
be blamed simply on the relative disinterest in Rwanda. The real danger, the genuine confusion on
the ground, the restricted mobility of the reporters, and the inability to fly out photos or videos
were major handicaps. In addition, American employers had ordered their reporters out for
reasons of safety, and possibly also because of costs.89 But some stayed and accurately reported
events, demonstrating all the more the failure of those who did neither.
There were, by chance, two reporters in Kigali at the time, Katrin van der Schoot, a freelance
Flemish reporter for Belgian radio, and Lindsey Hilsum, an experienced freelance journalist on
Africa who normally worked for the BBC, The Guardian, and The Observer, but was in Kigali on
_a special assignment for UNICEF. With the shooting down of the plane carrying Habyarimana,
the slaughter of the moderate Prime Minister and cabinet ministers, the massacre of the 10 Belgian
_peacekeepers, and the subsequent murder of Tutsi in very large numbers, the number of reporters
sometimes grew to a maximum of 15 (in contrast with 2,500 accredited to cover the elections in
South Africa at the time).
Mark Doyle of the BBC reported on the RPF advance and the scale of the civilian slaughter by
mobs, accurately depicting the murders as being instigated and led by the government and the
military. Jean Hélène, the Le Monde correspondent, provided detailed and accurate background
information to the Rwandese crisis. While reporting the mutual accusations levelled at each other
by the two sides, the implication seemed clear that it was Habyarimana’s enemies within the
government and armed forces who targeted the plane. On 12 April, the fear was expressed that the
Tutsi would all be systematically massacred before the RPF captured the capital.90 Catherine
Bond, The Times correspondent, reported on 12 April from Kigali that most of the killing was
probably not random. Michael Taylor, director of the British Christian Aid, wrote in a letter to The
Times that groups of armed youth loyal to different political parties were instigators of the
violence. If he had written that they were perpetrators of violence instigated by higher-ups, this
46
partial truth would have been more accurate. But it at least counteracted the impression of
spontaneous tribal violence that was the general theme of most stories.
Such accurate or partially accurate reportage was the exception, however. The initial reporting in
both The Times and New York Times had appallingly misleading reports: the downed plane was a
result of a Tutsi attempt to destroy the Hutu leadership in Rwanda and Burundi; “mobs” or a troop
rampage killed the Rwandese Premier and 10 Belgian soldiers; anarchy (not interahamwe with
roadblocks) reigned in the streets; “rival tribal factions waged vicious street battles”. On all the
critical points, these early reports were wrong.91 An interpretive piece in The New York Times on
9 April explained the events as a genocidal orgy (rather than a systematic organized genocide), a
continuation of a centuries-old feud. Adding to this fictional tale of Tutsi-Hutu orgiastic killing, the
paper on 12 April retribalized the murdered Prime Minister as a Tutsi.
The depth of Western stereotypical perceptions of Africa is indicated by the difficulties
encountered by informed NGOs to place op-ed pieces that had a different angle. The director of the
US Committee for Refugees, Roger Winter, who had worked with Rwandese refugees for years
and returned from Rwanda on 4 April 1994, wrote an article emhasizing the political and organized
nature of the violence. To present it as simply another case of “African tribal bloodletting”, was a
“fatalistically superifical interpretation”, he concluded. The piece was turned down by several
American papers, but eventually published in the Toronto Globe and Mail on 14 April.
US television coverage and the CNN erred on the side of vagueness, generally referring to
“unspeakable atrocities”, and “ethnic violence,” but picked up the theme of tribal or mutual ethnic
slaughter. It would not be until 7 May that ABC correspondent Ron Allen suggested that the
events were not a product of spontaneous tribal violence, but were a premeditated political act
intended as a final solution.
An exception to this kind of reporting was an op-ed piece by Alison des Forges of Human Rights
Watch providing an in-depth background to the Rwandese crisis and indicating Western complicity
and responsibility in the genocide. However, this was not the predominant Western version of
events, even if it was an authoritative one. The editorial line of The New York Times followed the
fictional interpretation of events and concluded that the cure for Rwanda’s “blood frenzy” was the
business of Rwandese.92 The Times did the same. Even if obtaining accurate information was
difficult – especially at the beginning of the crisis – more accurate analysis of the information that
gradually was coming in could easily have been made by fuller consultation with Rwanda experts
and human rights and refugee organizations. There was also more accurate reporting in some of
the media that could have been utilized by others less well informed.
In the French press, the genocide was by early May a given, and the emphasis began to be on
French responsibility for it.93 Print and television coverage of Rwanda began to increase also in
the English language press, partly because the election in South Africa had been held and Mandela
was sworn in on 10 May in a relatively peaceful way, and journalists were reassigned to Rwanda.
The Times had quoted Oxfam (29 April) that “the systematic killing of the Tutsi minority in
Rwanda amounted to a genocide,” and reported that the Kagera river filled with floating dead
bodies, but the focus by 2 May was still on the unprecedented 250,000 refugees who had flooded
into Tanzania in one 24-hour period.94 The refugees became the predominant story (See Fig II).
Only when reporters reached Lake Victoria and witnessed the logjam of dead bodies floating down
from the Kagera River did the Western media refocus on the massive killings. More accurate
reporting on the mechanism of the genocide began to appear in the American press (Time
Magazine, 18 May; Associated Press, 22 May), although Rwanda soon faded into the
background, crowded out by Haiti and the O.J. Simpson story at least until Goma (See Study III).
The misleading media coverage was echoed in the accounts of events in Rwanda given by both the
Security Council and the Secretary-General, explaining the withdrawal of UNAMIR on 21 April.
As discussed above, these were totally distorted. More generally, the Western media’s failure to
report adequately on the genocide in Rwanda possibly contributed to international indifference and
inaction, and hence the crime itself.
47
48
Chapter 5
Retroactive Conflict Management
(21 April 1994 – August 1994)
5.1. Changing the mission: UNAMIR II
Not until the end of April was there some acknowledgement from the United Nations that it ought
to respond more forcefully to the Rwanda conflict. The initiative did not come from the Security
Council, where some members had registered strong disapproval of the proposal to scale down
UNAMIR, but nevertheless had voted for it on 21 April. Rather, it was the Secretary-General who
turned around and already on 29 April called on the Security Council to take “forceful action” in
Rwanda. The move marked the beginning of a much more proactive role by the Secretariat
towards the crisis. Recovering from its near-paralysis in the immediate aftermath of 6 April, the
Secretariat assumed the role of a model executive branch: it asserted the moral and political need
for action, assessed the problem, formulated an innovative and bold response, and actively sought
to mobilize resources for its realization. This change in itself requires explanation.
As information about the situation in Rwanda and pressures on the UN to act were accumulating,
the Secretary-General took a high-profile role to re-engage the UN militarily. The media, as noted,
had scattered, inadequate coverage during most of April. Media pressure, therefore, was an
unlikely factor in the reversal. The pressure from the NGOs was mounting. The ICRC had
launched several appeals, although the principal humanitarian NGOs involved in Rwanda did not
identify the massacres as a genocide until early May.95 Probably more important, the SecretaryGeneral was under increasing pressure from African countries to demonstrate that the United
Nations took its African constituency seriously and did not give preference to European crises.
African members of the UN had called for “forceful action” during the discussion of UNAMIR’s
fate after 6 April, and maintained this position as the months went by and the death toll mounted.
Their demands were buttressed by growing information being fed to the Secretariat by UNAMIR
and DHA about the nature and magnitude of the massacres. The newly established Operations
Centre in DPKO/New York relayed graphic information about the slaughter. On 23 April, the
Under Secretary-General for Humanitarian Affairs, Peter Hansen, travelled to Kigali, becoming
the first New York-based high UN official to report back on the situation.
Elsewhere in the UN system, there was little impetus to act. The key human rights mechanisms
had not been visible in early April, nor did they figure in this reversal of policy. The new position
of High Commissioner for Human Rights was filled only one month before 6 April, and the new
Commissioner did not physically take office until the day before; it is understandable that he did
not immediately play an active role. A week after the Secretary-General had put Rwanda back on
the Security Council’s agenda, the High Commissioner, José Ayala Lasso, called on the UN
Commission on Human Rights to convene an emergency session to consider the killings.96 More
striking was the passivity of the Commission itself. Only on 25 May – that is, several weeks after
genocidal killings had commenced, one week after the Security Council had agreed to reverse
course and send in a military force to protect civilians, and after an urgent appeal by the new High
Commissioner for Human rights – did the Commission act. A Special Rapporteur was appointed
to report on the human rights situation in Rwanda, and on 28 June he confirmed in a detailed
assessment that genocide and other human rights violations had occurred (E/CN.4/1995/7). By
this time, the genocide had claimed most of its victims. The OAU’s African Commission on
Human and People’s Rights was completely inactive.
By taking a high-profile role in calling for a UN reversal on Rwanda, Boutros-Ghali put the onus
of rejection or compliance on the Security Council. The immediate response from the Security
49
Council on 30 April was cautiously supportive in principle, and even contained an indirect
reference to the possibility that genocide might be unfolding in Rwanda (S/PRST/1994/21). It was
evident, however, that the Permanent Five did not wish to engage directly. The Council passed the
initiative to the African states by asking the Secretary-General to consult with the OAU and report
back on further action. OAU’s Secretary-General responded positively, and on 10 May provided
firm offers of African troop contributions.97 In the meantime, the Secretariat was developing a
plan to expand UNAMIR and change its mandate to help protect civilians. The outlines of the
innovative plan was ready on 13 May and was authorized by the Security Council on 17 May
(Res.918 (1994)). In a surprise move just before the vote, however, the US delegation asked for
more detailed preparations from the Secretariat before agreeing to the force. The final approval for
establishing UNAMIR II came in the form of an implementing resolution on 8 June (Res.
925(1994)) – i.e. almost two months after the massacres had started and over a month after the
Secretariat and the Security Council had started to reverse course.
The UN maintained a slow pace even though time was of the essence. The Security Council had
authoritative information to the effect that genocidal violence was unfolding at a rapid rate, with
possibly thousands being killed every day. Already on 30 April, the Council itself acknowledged
that “massacres and wanton killings...in a systematic manner” were occurring; these were further
linked to genocide although in carefully non-committal terms.98 The point was reiterated in midMay. The Secretary-General used progressively stronger language to describe the violence in his
communications to the Security Council, going from “massacres of innocent civilians ...on a
massive scale” (29 April) to unequivocally declaring “genocide” (May 31). Reporting on a special
mission he sent to assess the Rwandese situation during the second half of May, Boutros-Ghali
used the figure of 250–500,000 killed (S/1994/640, May 31). In Washington, US administration
officials had told the Congress in early May that an estimated 100–500,000 persons had been
killed, although they considered the latter figure to be on the high side. Despite this information
and the acknowledged severity of the situation, the Security Council itself did not push for action,
and the United States slowed down efforts by the Secretariat to move more quickly by insisting on
additional procedures before authorizing implementation.
Why was the UN so slow to act? Were the causes “systemic” characteristics of the UN, or the
result of particular actions (or inactions) by individual actors?
While slow relative to the needs on the ground, the process was not so tardy by standards of
multilateral diplomacy. The Secretariat had to identify troops, secure finances, and prepare a plan.
When objecting to immediate authoriztion on 17 May, the United States delegation claimed that
Boutros-Ghali’s plan was inadequately prepared and lacked field assessments99 ; some regarded it
as more of a public relations effort than a strategy for a serious mission. Yet neither the US nor
any other of the Permanent Five that had the capacity to energize the process did so, reflecting a
more fundamental, collective concern not to become involved in another UN operation in Rwanda.
The question of obligation to act was treated cautiously by Security Council members. With fastmoving violence on the ground, the UN’s earlier commitment to implement the Arusha Accords
had become irrelevant, at least in the short run. The genocidal violence against civilians did,
however, raise the question of an obligaton to act under the Genocide Covention. Key members of
the Security Council answered in the negative. The United States carefully sought to limit the
obligations by terminological adjustments.100 The British government “did not accept the term”
genocide either, and the British Foreign Office was inclined to see a discussion of whether or not
the massacres constituted a genocide as “ sterile”.101 The French Foreign Minister, Alain Juppé,
responded to the growing outrage in France over the killings by declaring it a genocide on 16 May,
making France the only one of the Permanent Five to do so.102 However, France did not translate
this into a policy of expediting the deployment of UNAMIR II. The subsequent French unilateral
intervention was launched as a humanitarian action that made made no reference to the Genocide
Convention (see 5.3 below).
By omission and commission, the role of the United States was critical. As the world’s last
remaining superpower, the United States had repeatedly demonstrated that it could move the
50
United Nations to take rapid and effective action. In the Gulf war (1990–91), the Administration
had done this both by committing a significant part of its own military force, and by mobilizing
other states through the US alliance structure and by exercising its formidable economic power.
By not utilizing its capacity to act, and its formal leadership role in the United Nations, the US
must take considerable responsibility for the overall failure of the UN to respond at this juncture.
The same applies, though to a lesser degree, to the other major powers on the Security Council.
France and Great Britain, in particular, had the capacity for rapid military action and considerable
political resources to energize the Security Council. Even so, Britain could only muster 50 fourton trucks at the height of the crisis. France demonstrated its capacity much later by undertaking a
unilateral “humanitarian intervention”, which the Security Council endorsed with utmost speed.
By not taking the initiative, the United States and others of the Permanent Five ensured that the
Security Council proceeded along its customary and laborious path of consensual decisionmaking. More specifically, the explicit unwillingness of the major Western powers to commit
troops to a revitalized UN force – which had been clear in the decision to scale down UNAMIR I
and was reiterated in early May when UNAMIR II was planned – meant that the UN force had to
be established on a mix-and-match basis. African countries would send troops, while
industrialized states were to provide equipment and finances. To constitute a force along these
lines was extremely time- consuming even under the best of circumstances.103
Problems of financing the force also caused delays and weakened the response. The role of the
United States in this respect was particularly important since the US was automatically assessed 31
% of the costs of all UN peacekeeping, thereby acquiring more than a formal veto in the matter.
US consent is essential to mount an operation whether or not American troops are involved. The
United States made it quite clear during May that on purely financial grounds it would not readily
endorse an expanded UNAMIR. US reluctance bogged down the UNAMIR discussion from the
start. At one point, US demands for careful planning, prior troop commitments, and security of
the mission – justified by the need to avoid wasting money on a failed mission – made the Security
Council delay final authorization of the force for an additional two weeks, from 17 May to 8 June.
While the structure of UN assessment gives the US an inordinate power to influence UN
peacekeeping, there are ways around this. In early May, the Secretary-General floated the idea of
establishing a voluntary fund to finance an expanded UNAMIR. Having obtained troop
commitments in principle from several African countries, Boutros-Ghali proposed an initial fund
of US$50–80 million to launch the force. None of the economically powerful UN members came
forward to subscribe, however, and the US received the idea coolly. By their inaction, other
members of the Security Council ensured that the regular assessment structure would apply,
thereby keeping the onus for action on the reluctant US.
US reluctance to support UNAMIR II reflected a progressively critical attitude of the Clinton
Administration towards UN peacekeeping operations. With great deference to a hostile Congress,
the Administration had just completed a review of UN peacekeeping operations that established
stringent criteria for US support (PDD 25). To support UN peacekeeping in Rwanda, the
Administration would have to determine that peace in this remote Central African country was
critical to US national interests. In conventional strategic terms, this would be difficult, and the
Administration did not try. Nor did it elaborate alternative criteria for national interest, e.g. to
reduce massive international refugee flows or the principled need to react to genocide.104
Congress had expressed concern over the escalating bill for UN peacekeeping when UNAMIR I
was planned in September 1993. Since then, the operation in Somalia – deemed a disaster from a
US perspective – had reinforced Congressional reluctance to finance UN peacekeeping anywhere,
with or without US troops.
In reality, the budget for the projected UNAMIR II was quite small when judged against the
“downstream costs” for relief and rehabilitation (estimated to US$1.4 billion in 1994), not to
mention the human toll. The estimated budget for UNAMIR II was US$115 million for 6 months
(or slightly under US$20 million a month).105 This was the same rate as UNAMIR I (about
US$10 million per month for a force of about half the size), and reflected attention to fiscal
51
stringency despite the changed situation. Both this budget and an advance voluntary fund of
US$50–80 million were well within the means of the industrialized states to finance.
The reluctance of the US and other members on the Security Council to embark on a military reengagement reflected additional concern about the risks and practicality of intervention. On the
ground, the Rwanda situation seemed confused and anarchic, raising the question of the mission’s
effectiveness, security of personnel, and fears that support operations might be needed for rescue
or reinforcement. Again, behind these concerns lurked the Somalia experience, although the two
situations in reality were quite different. There had been no systematic and massive killings of
civilians in Somalia. The proactive orientation of UNOSOM to disarm armed factions and to arrest
a prominent faction leader had been undertaken with a Chapter VII enforcement mandate. In
Rwanda, by contrast, the proactive function would be to protect civilians. Although force could be
used if necessary, a reduced-risk option was proposed whereby the units would move in from the
border and remain in the protected zones in outlying areas.106 On the other hand, there were
obvious elements of risk. Throughout May, when the Security Council deliberated on the nature
and mission of a military re-engagement in Rwanda, the members were fully aware that there was
no cease-fire in the country or in Kigali; the airport was not secure; the Ugandan army had stopped
all UNOMUR patrols on the Ugandan side of the border to keep arms flowing to the RPF; and it
was not clear that the parties to the conflict would welcome, let alone respect, a UN force as a
neutral, humanitarian entity.
The Rwandese parties were also responsible for the slow and hesitant UN deployment. During the
critical Security Council deliberations in early May, the RPF sent ambiguous messages. While in
principle accepting a humanitarian force, the Front feared that UNAMIR II at best would be
irrelevant, and at worst a hindrance to their advance or protection for the government forces.107
These concerns were consistent with their initial hostility to a UN force in early 1993 – which only
changed to a lukewarm attitude prior to 6 April 1994 – and further strengthened the forces of
caution in New York. On the government side, a UN force would prima facie be even less
welcome insofar as its mission would be to protect civilians, who were targeted principally behind
government lines.
5.2. Deploying UNAMIR II
UNAMIR was reauthorized at the level of slightly more than 5,000. The new mandate was to help
protect civilians by establishing humanitarian zones inside Rwanda, and by facilitating
humanitarian relief to displaced and other needy persons. While designed under a Chapter VI
mandate, the operation envisaged clearly went beyond conventional peacekeeping; for example,
the final resolution _authorized the force to “take action in self defence” against those who
“threaten protected sites and populations emphasis added” (Res.918/1994). In effect a “6.5
mandate”, the formulation permitted a flexible and firm response to protect civilians. It was a
significant formalization and a further development of the practice undertaken on the ground by the
rump UNAMIR contingent in Kigali, where UNAMIR soldiers were posted to sites that had
become sanctuaries for threatened civilians.
The mandate and design of UNAMIR II seemed appropriate to the situation. It is reasonable to
assume that UNAMIR II, if quickly and effectively deployed, could have had a significant impact
in terms of saving lives. Even the minimal efforts by the UN forces on the ground, it will be
recalled, had some protective effect. The most impressively cost-effective operation was at the
Amahoro Stadium in Kigali, where, after the Bangladeshis left, 12 UN “blue helmets” armed only
with hand weapons and barbed wire protected several thousand persons. Since the civilians mostly
were not threatened by organized, well-equipped army units (as in Bosnia), but mainly by paramilitary gangs and mobs, a relatively modest force could have a significant deterrent effect.
However, the intervention was delayed in two ways that rendered it irrelevant to the most critical
phases of the conflict. Over a month passed before the Security Council authorized the operation.
The actual deployment took another two months, with the result that the force arrived in Rwanda
52
after the civil war and the genocide were over, and after a unilateral French intervention with a
similar mandate had taken place.108
Despite the risks of the situation and the novelty of the mission, African countries responded
rapidly with troop commitments. OAU again emerged as a focal point for mobilizing support.109
By the first week of May, Nigeria and Ghana – both with large armies and peacekeeping
experience – had responded positively, and Tanzania, Senegal, Zimbabwe and Zambia had
indicated interest. In addition to regular financing, however, all needed various kinds of equipment
and logistical support; even Nigerian and Ghanian units lacked vehicles and heavy equipment.
As a result, the mix-and-match pattern of African troops and Western equipment developed. This
had been practiced in other UN peacekeeping operations but on a smaller scale and not for units
that formed the backbone of a force. Before being deployed, the troops committed to UNAMIR II
had to be trained and equipped within a multinational and often two-layered bureaucratic structure
(UN and bilateral). Donors of equipment grumbled that African countries used peacekeeping
operations to outfit their forces, and worried that their equipment would be misused. Combined
with “normal” bureaucratic procedures, this produced long delays.
The infamous APC-case is illustrative: The United States was to lease (with full renumeration from
the UN) 50 armored personnel carriers (APCs) to a second Ghanian battalion earmarked for
UNAMIR. The Pentagon received a formal UN request for these APCs on 19 May, which started
a lengthy process. Paperwork for the lease had to be completed before the vehicles could be
assigned. Several practical issues went back and forth regarding the type of vehicles (wheeled or
tracked?), repainting them (white), finding a training partner (the US requested Egypt but the UN
declined due to “extraordinary preconditions.”) It took a full week to transport the 50 vehicles
from a US base in Germany to Entebbe (Uganda), and when the last APCs were finally unloaded
at Entebbe (30 June), they remained on the tarmac under UN guards for a month before moving
on to Rwanda. Not until 30 July did the first APCs roll into Rwanda. By this time, the end of the
war and the installation of a new government in Kigali rendered their original mission of securing
the airport and protecting civilians quite irrelevant.
The “systemic” and individual factors that caused the crucial delays in establishing UNAMIR II
were the same that had produced a halting deployment of UNAMIR I. Requiring consensus from
key members of the Security Council and being dependent on national sources for men and
equipment, the UN needs 3–6 months to form a peacekeeping force of this size. A determined
political push by member states, however, can cut through the bureaucratic tangles to speed up the
decision-making process, or come up with missing force components from national sources. In
this case, strategic disinterest in Rwanda, combined with fear of a failed or costly mission, made
key members of the Security Council hold back. Since it is reasonable to assume that a more rapid
deployment of UNAMIR II could have protected a large number of civilians, the consequences of
delay are the equivalent loss of life, and, quite possibly, a larger outflow of refugees who
otherwise might have remained in protected areas, as happened in the French humanitarian zone
(see 5.2).
Critics have argued that the UN presence and activities in Rwanda throughout the crisis were
marked by great concern for neutrality (African Rights, 1995). Given that events in Rwanda
constituted genocide – as the Secretary-General called it on 31 May (S/1994/640), and as did the
Special Rapporteur of the UN Commission on Human Rights a month later (E/CN.4/1995) – then
neutrality is untenable on moral and legal grounds (Brauman, 1994). However, the more
fundamental issue was the lack of commitment: the main obstacle to an effective response in
Rwanda was not that the UN insisted on neutrality, but that it was not there.
53
5.3. Opération Turquoise
While the UN was struggling to mount troops and equipment for an expanded UN presence in
Rwanda, on 19 June France suddenly offered to send its own force, but under French command
and control. Officially designated as a humanitarian intervention to protect civilians and hinder a
mass outflow of refugees,110 Opération Turquoise was launched and executed in a manner
suggesting mixed motives. Partly for that reasons, the consequences were both humanitarian and
political in nature.
At the UN, the Secretary-General presented the French offer as a stop-gap measure until
UNAMIR II was fully deployed, then expected to take another three months.111 Critics have
argued that France could readily have speeded up this process by placing its Turquoise-designated
units at UNAMIR’s disposal. However, strong RPF opposition to accepting a UN force with a
French contingent made this difficult as long as UNAMIR operated under a Chapter VI mandate
requiring at least minimal consent from the local parties. When France decided to intervene
unilaterally, it was regardless of RPF permission and with a heavily armed force authorized to use
“all means necessary” under Chapter VII of the UN. The RPF chose not to challenge the
interventionist force.
By retaining national command, the French government was able to conduct and claim credit for
an operation that on purely military grounds was generally judged as well planned and superbly
executed under the command of General Lafourcade (Connaughton, 1994). The political credits
were also “national”; the ably run Opération Turquoise came to be juxtaposed with the uncertain
and fumbling UN peacekeeping operation. At the same time, the French had requested and quickly
obtained a Security Council endorsement of the intervention (Res. 929(1994), 22 June). The
operation consequently confirmed the growing precedent for unilateral interventions initiated by a
state and conducted under national command, but legitimized by the UN.112 The overall effect
was to further undermine the institution of collective international intervention as delegated to the
UN and envisaged in its Charter.
These wider consequences partly flowed from the judgement that the operation had been a success:
As planned, the units moved rapidly and mostly avoided armed confrontations with the RPF or
any of the other parties. They could claim with some reason that the intervention saved lives and
stabilized the situation in the south-western corner of Rwanda, thereby reducing the outflow of
refugees. The French presence facilitated relief work carried out by humanitarian organizations.
Moreover, the force accomplished all this without incurring any casualties and withdrew as
planned within the two-month time limit specified at the outset.
Each of these claims have some validity, although their significance must be judged in a wider
context. Opération Turquoise did save some lives if it is assumed that these otherwise would have
been killed before the RPF could reach the area. By 2 July, the French had established control
over the south-western corner of Rwanda, equivalent to one-fifth of the national territory, and
proclaimed it a “safe humanitarian zone”. The RPF claims that its forces could have “liberated” this
area within a short period of time, and that the Tutsi who survived for over two months to be
rescued by the French would equally have been there later if RPF units had arrived. As noted,
however, there was a dynamic at work that tended to speed up the killings of Tutsi behind
government lines as the RPF advanced. What would have happened in the south-west without the
French therefore remains unclear.
The numbers are also uncertain. French government claims that Opération Turquoise had saved
“tens of thousands” seem exaggerated. Probably the vast majority of the Tutsi population and
associated Hutu in the south-west had been killed by the time French troops entered.113 Some
13– 15,000 persons remained in camps and sites guarded by a hostile gendarmerie, including
11–12,000 in the Nyarushishi death camp in Cyangugu (Prunier 1995, ICRC, interviews 1995,
African Rights 1995:1147). These were freed by the French.
The issue of lives saved is central in evaluating the mission because it was launched and endorsed
54
by the United Nations as a “strictly humanitarian” operation (Res.929(1994). By that criterion, the
mission was not ineffective, but the timing made its potential underutilized. If a similar effort to
save lives had been undertaken in April rather than in late June, the number of lives saved
undoubtedly would have been much higher.
The late date of the intervention fuelled widespread suspicions that the main purpose of Opération
Turquoise was to save France’s erstwhile clients in Rwanda from total defeat. A careful analysis
of the planning of the operation suggests that this was not the principal political aim, at least
initially (Prunier 1995).114 Moreover, the intervention was first proposed by the Minister of
Foreign Affairs, Alain Juppé, evidently in response to mounting pressure from NGOs and the
media to stop the slaughter. Simultaneously, critics focused on France’s role in arming and
supporting the government on whose side the atrocities were being perpetrated. An intervention
designed to save lives and help bring relief supplies to displaced persons would be an atonement
of sorts. Media coverage was an important element in this respect, and official guidelines for the
operation directed the troops to “demonstrate great understanding towards journalists and
humanitarian organizations.”115
Commonly offered, the atonement explanation seems plausible. A (London) Times journalist
captured the sense when reporting on 1 July that “a week after starting the rescue mission, French
Special Forces finally discovered 250 Tutsi people yesterday.” It also helps to explain why the
French pulled out as promised after two months, leaving the RPF to take full control of the zone.
In this respect, French policy was consistent with the trend observable since 1990 of rendering
only half-hearted support to the regime: the Habyarimana government had received significant aid,
but not enough to hold back the RPF.
If Opération Turquoise was a humanitarian gesture for some, this did not exclude consequences of
another kind. As the mission was implemented, it became clear that it had the effect of giving some
protection to retreating government forces. By late June, government troops were falling back as
RPF units moved quickly into the south-east region, and there was a similar rout in the rest of the
country. Strong pro-FAR circles in the Ministry of Defense saw the French intervention as an
opportunity to help the Rwandese army, which for years they had been training and equipping
(Prunier 1995). Soon after a “safe humanitarian zone” had been declared on 4 July, French
commanders announced that RPF rebel forces would not be allowed to enter, and attempts to do
so would be met with force.116 To RPF Secretary-General Theogène Rudasingwa and more
neutral observers as well, this was clear evidence that the operation was exceeding its UN
mandate, which specifically prohibited an “interpositional force”. Force Commander General
Lafourcade sharpened the partisan edge of the intervention by declaring on 11 July that members
of the “interim government” – who were directly linked to the genocide – would be allowed to
seek asylum in the French zone. A contrary statement from the Ministry of Foreign Affairs
declared that they would be interned, but the military were on the ground and in fact made no
arrests. Instead, French forces were observed by journalists and foreign relief personnel escorting
members of the former regime out of the zone.
For a humanitarian intervention force of 3,000, the unit was heavily armed (including air support
from four Jaguar fighter-bombers and four Mirage ground attack planes based in Goma, Zaire).
This discouraged the RPF from challenging the intervention militarily, even though the RPF feared
that the French planned to divide the country and dig in. The RPF instead accelerated its advance
so as to secure Butare and Kigali (2–4 July) before the French could do so. After 4 July, a
dividing line between the two armies running just west of Butare was tacitly accepted. Apart from
two incidents, there were no military confrontations: the French ceased their advance, and the RPF
did not press forward into the zone. Effective and early communication between the two parties
was instrumental in avoiding an escalating conflict that neither wanted. Established in Paris just
prior to the operation and continued throughout, the communications structures permitted careful
management of a conflictual relationship.
Within the zone, the French operated by and large in a friendly environment; indeed, the embrace
of local Hutu leaders proved something of an embarrassment. This removed some of the security
55
concerns that had weighed on the planning of UNAMIR II. With a formal welcome from the Hutu
in the south-east and tacit agreement from the RPF, the intervention became a relatively low-risk
venture.117 The point is important since the efficient execution of Opération Turquoise is often
held up as a standard by which the more reticent UNAMIR II can be judged. In fact, the two
operated in different security environments, and as they also commanded very different firepower, they are not fully comparable.
Longer-term consequences of the intervention stem from the friendly relations between the French
and the people in their zone. French forces were instructed to disarm the militia.118 This they did
when encountering paramilitary groups, but apparently did not follow a consistent and proactive
disarmament strategy based on cordon-and-search operations (Prunier 1995). Importantly, there
were no official instructions to disarm government troops. As a result, the French zone facilitated
an orderly retreat of FAR units through the zone and into the South Kivu area of Zaire with their
weapons intact. Two major military camps – as distinct from designated refugee camps – were
formed near Bukavu, the main destination for refugees moving out of the south-west. A UNHCR
report estimated that the Bukavu camps held 10,000 soldiers, or about one-third of the Rwandese
army (UNHCR/ FRS/ A/04: para.5).119
The French did not “rescue” either the government forces or the militia in the sense claimed by
RPF and other critics. If the French zone had not been declared, both soldiers and militiamen
would probably have moved across the border sooner rather than later. A more fundamental
problem lay elsewhere. Neither the UN mandate for Opération Turquoise nor the objectives the
French government set for the mission included disarming soldiers. Yet the very concept of a safe
humanitarian zone implies a demilitarized site, and has developed as such in practice by
humanitarian agencies (ICRC, 1993). The failure of the French proactively to disarm government
troops within the zone must count as the cost of a significant opportunity lost, or rather,
deliberately bypassed. Even if partially successful, this would have helped the successor
government deal with the defeated enemy in an orderly manner and in accordance with legal
norms. Similarly, French policy in the zone was not to screen or arrest suspects in connection with
the genocide, nor was this included in the mandate requested from the Security Council. General
Lafourcade’s staff claimed to have compiled a list of suspects, but these were not turned over to
UNAMIR, which moved in when the French pulled out in late August (Human Rights Watch,
1995).
On the other hand, the French intervention clearly helped reduce the refugee outflow from the
south-west, thus accomplishing a major objective of the mission. This greatly facilitated relief
operations (see Study III and IV). Given the huge outflow of the time, this issue was so
significant that the success or failure of Opération Turquoise in large part came to be judged
against its capacity to stabilize the situation in the south-west and stem the outflow. The reverse
side of this dynamic was that militant communities were established inside the French-protected
zone instead. Opération Turquoise thus served to internalize the problem of “refugee-warrior”
communities. In some respects, these camps posed fewer problems for the new government than
the large concentrations in exile. The last and critical phase of dismantling them turned into a
disaster when RPF forces massacred thousands at the Kibeho camp (see chapter 6.3). The French
intervention made the Kibeho massacre possible, though certainly not inevitable.
Opération Turquoise also had more long-term and indirect consequences. In bilateral terms, the
intervention further worsened relations between France and the new government in Rwanda. The
uneasy mix between official humanitarian objectives and political motives at first worried
humanitarian organizations, which feared the political shadow would compromise their relief
operations. As the intervention turned out to facilitate their work, however, they came to applaud
it. Others concluded that it was a disingenuous coupling of national interests to a relief operation,
which served to discredit both the UN and the very concept of a humanitarian intervention. For
several African leaders, it was additional evidence that a major European power could manipulate
the UN and humanitarian operations to demonstrate its own power in the region. The five
developing countries on the Security Council marked their scepticism by pointedly abstaining on
the resolution authorizing the force. In this perspective, Opération Turquoise was yet another case
56
of powerful Western states using the UN to legitimize their interventions in the South.
The long-term costs of the mission are intangible. The demonstrative success of a national
operation further undermined the peacekeeping operations of the United Nations, possibly lending
support to potentially destabilizing unilateral interventions. Use of different mandates in the same
country added to the confusion and undermined a principled use of international law. To have free
hands, the French obtained from the Security Council a mandate that invoked a “threat to
international peace and security” under the UN Charter’s Chapter VII. However, the UN force
already in Rwanda, and the expanded UNAMIR being planned, had only a Chapter VI mandate,
the implication being that Rwanda was not for this purpose a threat to international peace and
security. The result may have strengthened the forces of cynicism and cast doubt on the very
concept of humanitarian intervention.
57
Chapter 6
Refugees and Internally Displaced
Persons:
The End and the Beginning of Conflict
(May 1994 – May 1995)
As RPF units rapidly advanced to seize power in mid-July, members of the old regime led a
massive flow of civilian refugees across the border to Tanzania and Zaire. Among them were some
30,000 government soldiers, militia members, local officials and former national leaders. In the
Bukavu area in South Zaire, government soldiers formed separate camps from the very beginning;
in the Goma area, the army, the militia and the civilian refugees were more thoroughly mixed. In
Tanzania, the military were not as visible among the refugees, but the militia and former officials
were.
These refugee settlements were problematic. There was little security in the camps. Food and other
relief supplies were diverted from the intended beneficiaries. The old leadership used its hold on
the camps to intimidate refugees and prevent repatriation. The areas became a sanctuary allowing
the defeated regime to regroup, rearm, and launch incursions across the border. With ex-FAR
officers openly proclaiming their intention to “kill all Tutsi who prevent us from returning”, the
camps became a key element in a new cycle of conflict.120
While “refugee-warrior communities” of this kind are characteristic of several exile movements,
the magnitude of the Rwandese refugee population, and the issues of guilt arising from the
apparent widespread participation in the genocide, made the situation particularly difficult with
respect to both relief and conflict management. The search for solutions was also precedent-setting
in many ways. For the first time, the UN Security Council discussed security in refugee camps as
a peacekeeping matter. While the Council did not come to a conclusion, the eventual solution
devised was innovative and represented a measure of success.
6.1. The refugee camps: humanitarian and security issues
It was soon painfully obvious that the perpetrators of human rights abuses and genocide were fed
and assisted in the camps. According to the Statutes of UNHCR and subsequent UN
authorizations, as well as the OAU Refugee Convention, all refugees from war are prima facie
entitled to protection, but the exclusion clause of the 1951 Refugee Convention applies if there is
reason to believe that a person has committed serious war crimes or crimes against humanity. The
legal norms, in a word, are unambiguous. To apply them in the Rwanda case was another matter.
Registration is necessary to determine eligibility, but the rapid influx and vast numbers involved
made it impossible to register the refugees upon arrival. Even when certain individuals were
known prima facie to be excludable, they could not easily be removed. International relief
organizations and UNHCR encountered the problem as soon as the first refugees arrived in the
Ngara region of Tanzania.
The rapid influx into Tanzania of some 250,0000 persons during a couple of days in April created
mass confusion. However, former provincial and commune officials soon reconstituted
themselves in the camps, thereby gaining considerable control over the food distribution as well as
the refugees in Ngara.121 Among them was a former bourgemestre and known killer. Attempts by
UNHCR to remove him from the camp resulted in a riot including several thousand people. The
58
incident _affected aid workers and refugees alike. Some NGOs, including MSF (Médecins sans
frontières) suggested that the only alternative was to leave the camps. This option was actually
carried out by MSF, which suspended most medical aid and some water services. UNHCR and
most NGOs decided to stay in recognition of the humanitarian imperative to protect and assist a
vast population of refugees, even if that meant assisting persons guilty of crimes against humanity.
The “Gatete incident” – named after the notorious bourgemestre – accentuated the immediate need
to police the camps. While policing would not address the problem of removing the excludables, it
would deprive them of some power and also improve security for the refugees. An agreement was
reached between UNHCR and the Tanzanian government that recognized the legal fact that the
formal authority and ultimate responsibility for policing refugee populations lies with the host
state. Tanzanian policemen, paid by the government but supported in other ways by UNHCR,
policed the camps and arrested sufficient numbers to establish the principle that violations could no
longer be carried out with impunity.
The problems in Goma were far more intractable. There, the refugee inflow was much larger – 1.2
million refugees fled into Zaire in a matter of days – and the soldiers and militia were more
numerous. The refugees established themselves in a corner of Zaire remote from central authority
and possibly any system of accountable governance. Soldiers and militia carried their weapons
with them; those who were disarmed by Zairian troops at the border were usually able to recover
or replace them with the assistance of other Zairian forces. Altogether, these factors led to an initial
condition of chaos and anarchic violence; later, it helped the former leaders to transform the
refugee community into a state-within-the-state.
Armed, organized groups of ex-FAR soldiers and militias were part of the sprawling refugee
settlements in Goma. In late August, after the devastating cholera epidemic had passed and the
camps were taking shape, a flare-up of violence occurred. The absence of any accountable
authority in the camps was one factor. The effort by the former leadership and armed elements to
assert their power over the refugees was another. The purpose was twofold: to prevent the
refugees from returning, legitimizing in that sense the new government in Kigali, and to secure a
popular base for the former regime in its continuing struggle for power with the RPF. The
refugees thus became hostage to the ambitions of the former leaders, as well as victims of ordinary
crime. Killings, threats, extortions, rape and other physical abuses were common in the Goma
camps throughout the autumn. Relief distribution was controlled by the military and the former
leaders who used the traditional Rwandese administrative community unit of cellule to divert food.
Refugees who tried to return home were intimidated or killed. While conditions within Rwanda
did not encourage return, violent harassment in the camps helped to bring spontaneous repatriation
to a virtual halt after the first wave of return ended in early August 1994. UNHCR attempts to start
registration in September–October had to be aborted when its personnel were threatened. Threats
to relief officials continued in the autumn, culminating in November when 15 international relief
organizations informed UNHCR that they would be forced to leave the camp if the security
situation did not improve.
Most of the retreating FAR units in the Goma area had established themselves with their families in
a separate camp (Mugunga). Later, other soldiers consolidated their presence in a second military
camp. This self-induced separation eased the humanitarian dilemmas for UNHCR and
organizations whose function was to assist only civilian refugees or military hors de combat. The
military could still move in and out of the refugee camps and assert their influence, but the more
immediate problem was the militia.They were present throughout the camps, believed to be behind
much of the violence, and – to the extent that they had played a major role in the genocide – were
not even entitled to protection and assistance as refugees. Local Zairian troops who were called in
to quell riots had proved to be part of the problem rather than the solution.
59
6.2. Searching for solutions
As the lead agency in refugee matters, UNHCR early on raised the issue of camp security with the
Zairian government. In September 1994, a joint mission from UNHCR and the government of
Zaire considered the feasibility of separating the militia from the rest of the population. The
mission estimated that around 100,000 persons, consisting of militia members and their families,
would have to be moved. The costs and problems of identifying, separating and relocating this
group – by force if necessary – were considered prohibitive. UNHCR fell back on a more modest
proposal to police the camp with a security contingent drawn from Zaire’s elite forces, backed by
international technical expertise.
The High Commissioner for Refugees, Sadako Ogata, obtained agreement from the relevant
Zairian authorities and forwarded the proposal to Boutros Boutros-Ghali at the end of September.
The proposal was almost identical in format to the later Zairian Camp Security Operation (ZCSO),
which finally was agreed upon. However, the agreement came only after the UN Secretariat and
the Security Council had debated and deliberated for over three months, rejecting the initial
proposal and examining counter proposals in the interval. In the end, UN/New York asked
UNHCR to unearth the original draft (S/1995/65). Meanwhile, security conditions in the Zairian
camps worsened – in one incident in Goma in late November, 21 persons were killed and 77
wounded; intimidation of refugees wanting to repatriate continued; and the former leadership
strengthened its hold.
The initial reaction of the Secretary-General to the UNHCR proposal was negative. In part,
Boutros-Ghali was disinclined to rely on Zairian forces due to the political sensitivities of working
with President Mobutu and, in effect, helping to finance his elite troops. More fundamentally, the
UN Secretariat viewed the question of security in camps as more than a protection issue and rather
a matter of peacekeeping since it involved separating civilians from armed groups and, in
principle, dealing with the latter as a threat to both the refugees and the security of Rwanda. So
defined, the matter clearly went beyond the mandate and competence of UNHCR and required the
involvement of the Security Council. The Secretary-General consequently took charge of the issue
and instructed DPKO to prepare options for a comprehensive solution.
While security problems in refugee camps were common and recurring issues, the Security
Council had not previously considered them to be a matter of peacekeeping. UNHCR officials
remained sceptical to this interpretation, citing tension between humanitarian concerns and military
operations. When General Dallaire floated a proposal to extend UNAMIR II’s operations to the
camps in Zaire, UNHCR opposed it on the grounds that UNAMIR had shown insufficient
understanding of the humanitarian aspects of the emergency and of voluntary repatriation. Some
countries that had contributed units to UNAMIR were also reluctant to extend the mandate to
operations across the border. As one official said, it seemed risky to “send our boys into the black
pits of the camps in Zaire.” A similar sentiment prevailed when the Security Council discussed the
options laid out in DPKO’s report of 18 November (S/1994/1308).
The Secretariat’s maximum option was to make a comprehensive assault on the problem by not
only separating soldiers and militiamen from civilians, but also disarming the first two (Jones
1995a). The operation was seen to require 7,000 well-armed troops, with 4,000 for the Goma
region and 3,000 for the Bukavu area, and a Chapter VII enforcement mandate. Having hesitated
to send fewer than that to Rwanda in UNAMIR I or II, the Security Council dismissed the option
as “fantasy”, as one participant later said. Option B was a slimmed-down version of A, involving
3– 5,000 troops and a Chapter VI mandate. It was received with more interest, but when members
were asked to contribute troops, only one out of 60 states approached responded in the
affirmative.
Simultaneous on-site investigation by a joint DPKO–UNHCR team concluded that the task, at any
rate, was more difficult than anticipated and would require more than 5,000 troops. That left the
third option of privatizing the security service. A British security company offered to train and
provide logistical support to Zairian troops. While attractive to some Security Council members,
60
who saw it as a way to depoliticize a sensitive issue, others argued that it amounted to shirking an
international public responsibility. UNHCR, which would have to raise the funds, judged it much
too expensive.
The Security Council failed to support any of the options tabled. In early January, Boutros-Ghali
acknowledged that his efforts to find a suitable plan for dealing with the situation had failed and
asked UNHCR to revive its original proposal of September 1994. This materialized in the form of
the Zairian Camp Security Operation (ZCSO).
6.2.2. The Zairian camp security operation
In a novel development, UNHCR and the government of Zaire signed a memorandum of
understanding on 27 January 1995, whereby the latter agreed to provide a contingent of elite
troops to police the refugee camps, working with an international civilian police unit. UNHCR
would pay their salary, food, health care and other incentives. The operation was carefully
designed to meet demands from the Zairian government that its troops would serve only under a
national command structure.122 The Zairian troops would be trained and liaise with an
international group of civilians with police and military backgrounds, but would remain under the
command and control of their own officers. The first group of 100 men from the prestigious
Presidential Guard arrived on 11 February; by the end of April the contingent was increased to a
total of 1,513.123
By mid-1995, the operation as narrowly defined had successfully established security in the
camps. The Zairians had been trained in refugee law and riot control, and worked closely with the
small international civilian police (CLSG). Their task was to improve law and order conditions in
the camps; prevent intimidation and violence against refugees wishing to repatriate; protect relief
personnel, infrastructure, equipment and supplies; and escort returnees to the border. Reports
from both UNHCR and other sources registered progress towards these ends. The numbers of
killings, theft, banditry, rapes, beatings, and other safety incidents had been greatly reduced, and
the associated tensions in the camps had diminished. Relief workers found it easier to perform
their duties. The critical impunity question had been addressed as people were arrested and
punished for criminal acts. Small arms were seized, although systematic searches were not
conducted or envisaged.124 _Ex-government officers who made political speeches in the refugee
camps were investigated and in some cases arrested (Halvorsen 1995). The effect on repatriation
was more uncertain. UNHCR figures for repatriation show a steady decline during the first three
months after the ZCSO was established, but, after a low point due to the Kibeho massacre in April
1994, a steady climb was recorded from May to July (See Study IV).
Given the importance of camp security issues in refugee situations worldwide, it is important to
consider the factors behind this relative success story. The most important of these appeared to be
the elite nature of the Zairian troops, their careful training and substantial remuneration, and their
close liaison with the international civilian contingent.
Yet the program did not deal with the security issues in a broader sense. The former government
soldiers were still in their separate camps; the militiamen and the military had easy access to the
regular refugee camps from which they recruited and trained new soldiers. Their equipment was
replenished with new arms supplies. Their ability to attack across the border and, in the longer
run, to mount a serious security threat to the border area and the new government of Rwanda was
unimpaired, indeed it strengthened over time.
The UNHCR solution, in other words, was based on a fragmentation of the problem rather than a
comprehensive assault as envisaged in the alternative peacekeeping approach. Separated out, the
military component of the security issue was left unattended. UNHCR had neither the mandate nor
the competence to take on a rag-tag army in exile. The Security Council had declined to act,
although the camps were clearly a matter of international peace and security. The OAU was silent,
even though the conflicts wrought by refugee-soldiers permeate the history of post-independence
Africa, including Rwanda’s. Zaire, itself in a condition considered precarious by many, made no
efforts to accept its minimal legal responsibility to prevent the exiled armed groups from making
61
incursions across the border.
While some of the troops retreating into North Kivu were disarmed by Zairian troops at the
border, many weapons were either stocked for later use or replaced by new ones. Virtually all
levels of Zairian authority were involved in channeling arms to the ousted Rwandese government
troops, including national and provincial authorities, the armed forces (FAZ) and semi-private
cargo companies.125 As documented by human rights organizations, supplies were delivered
across the border via Goma airport during the war (April-mid-July), and continued to be supplied
afterwards (late July and August 1994). Zairian authorities helped ex-FAR units and staff to
establish themselves in five sites along the border in the North and South Kivu areas. Thus, with
Zaire’s help, the defeated forces could regroup, retrain and rearm freely. The Zairian government
in this respect violated a number of international legal instruments – the UN arms embargo on
Rwanda ((S/RES/918(1994)), the 1969 OAU Convention that proscribes armed encampments of
refugee-soldiers close to the border, and international law prohibiting armed incursions across the
border – as well as the legal and moral principles to punish perpetrators of genocide.
Some NGOs and Western journalists reported that France violated the UN arms embargo by being
involved in arms deliveries to the retreating FAR units via Zaire in May–July 1994 (see Study 1).
Informed French military officials acknowledge that there undoubtedly was a clandestine arms
flow across the Zaire–Rwanda border in this period, but deny that France was involved. The
French Ministry of Foreign Affairs also has categorically denied all reports of French
involvement.126 None of the allegations are supported by documentary evidence – which at any
rate would be hard to obtain given the nature of the transaction – but rely on unnamed local
witnesses, or statements by named individuals that later were retracted. One crucial piece of
documentary evidence points in the opposite direction. About one month after the genocide
commenced, a delegation from the Rwanda interim government met with the head of the Military
Mission in the Ministry of Cooperation in Paris, General Huchon, and pleaded for political and
military support. According to the report the delegation sent back to Kigali, Huchon told them in
no uncertain terms that the French military was prevented by public opinion and the media from
giving any support; “their arms and legs were tied”.127 On the other hand, it is clear that the
French army and the military intelligence agency possessed the structural prerequisites for
providing clandestine aid, and that four years of close collaboration with the Rwandese army
would have given plausible motives for assisting an erstwhile client in distress.
The evidence implicating other countries in violation of the spirit, if not the letter, of the arms
embargo is less compelling, although actual or prospective supplies from South Africa and China
have been reported. Since the retreating government forces brought the reserves of the Central
Bank along with them, they had easy entry to a large and poorly controlled international market of
dealers in small arms. A UN expert commission was appointed by the Security Council in
September 1995 to examine these and other reports of violations of the arms embargo. Its first
report was expected in late January 1996.
In addition to a large external refugee population, war and genocide had created a large number of
internally displaced persons, especially in the French “safe zone” where an estimated 1.2–1.5
million people had fled towards the end of the war. For the new Rwandese government, the
concentrations of internally displaced represented an internalized version of the problems posed by
the camps across the border.
6.3. Internally displaced: Kibeho 128
In the aftermath of Opération Turquoise, up to 390,000129 internally displaced persons (IDPs)
were left in 33 camps in the south-west of Rwanda. In the autumn of 1994, the government
decided to close the camps by December 1994, using force if necessary.130 The NGOs, ICRC
and UN agencies131 insisted on voluntary return. In November, with the government, they
developed an Integrated Humanitarian Response. Opération Retour was launched in December of
62
1994. But it ground to a standstill in February 1995 because the IDPs were no longer willing to
return to their homes voluntarily because of the deteriorating situation in the home communes and
the intimidating actions of the hard core within the camps.132 An Integrated Operations Center
(IOC) working group was established by the government, UN agencies and NGOs on 6 February
1995, and an IOC was set up within the Ministry of Rehabilitation and Social Integration
(MINIREISO) in March of 1995.133 Increasing insurgency across the borders heightened the
security concerns of the government. RPA shootings, beatings, and harassment in camps had
increased in February.134 By the end of February, the IOC concluded that Kibeho camp, with
almost half of the remaining IDPs, was becoming a criminal sanctuary and possibly a center for
recruitment and training of insurgents.
The government once again determined to close the camps, using force if necessary. In early
March, the Integrated Task Force developed six options135 to reconcile the legitimate aspirations
of the Rwandese state, on the one hand, to establish stability on its territory and prevent genocidal
killers who sheltered in the camps from escaping justice, and, on the other hand, the concern of
the UN and the NGOs with protecting the rights of individuals and avoiding exposing innocents to
unnecessary risks. The first option focused on independent forceful action by the RPA; it predicted
that a mass chaotic dispersion would result, with large numbers heading across the border to
Burundi, and large numbers of deaths, particularly children. Such action would also violate the
agreement of cooperation between the international agencies and the government, instill distrust of
the government in the people, and create a rift between the international community and the
government. On the other hand, continuing with a plan of voluntary return would violate
government policy and was not likely to be effective, but was likely to consolidate extremist
control of the camps. The UN, NGOs and government agencies were faced with a real dilemma.
The first option was rejected. Four other options focused on various ways to expedite a voluntary
program, and they too were dismissed because, among many reasons, they did not satisfy
government policy and would lead to consolidation of extremist control over the camps. The other
option dealt with avoiding the use of force while also surrendering the principle of strict voluntary
return. A variation of this option was adopted.
The decision was made to close the camps in series by gradually reducing food supplies and
transferring the displaced to their home areas. Those who chose to remain in the camps after the
returnees left would be investigated as perpetrators of the genocide. The plan would be
accompanied by a program of information in both the camps and the home communes,
strengthened security and administration of justice in both places, and an escort service jointly
conducted by the RPA and UNAMIR to guarantee safe conduct, complemented by food stations
manned by NGOs along the way. The operation would begin about 6 April.
The commencement date was postponed until mid-April, then 18 April, and then there was an
_agreement to postpone again until the 24th, but there is confusion on whether the latter date was
communicated to the RPA.136 When the cooperative plan for vacating the camps seemed to break
down, there were late warnings that a number of casualties could be expected if the plan was
executed improperly.137 Given these late warnings, NGOs, though part of the IOC, subsequently
questioned why the UN had not sent in more troops. For some NGOs, the UN should have
known that massacres would occur given the large build-up of troops. However, as a UN official
pointed out, the fact of a troop build-up does not necessarily mean that the soldiers will be used in
precipitous action against the IDPs, and is not sufficient grounds for predicting a massacre.
On 18 April, the RPA surrounded Kibeho camp with two battalions and cut off the food supply.
The Minister said that the humanitarian community had agreed to that date.138 A population of
80,000 was squeezed by the cordon from five hills onto one hill.139 On 18 April, 8–11 children
were killed in a stampede. On 19–20 April, 13–22 IDPs were killed by the RPA opening fire when
IDPs apparently threw stones or started to snatch weapons. On 20 April in the late afternoon it
started to rain. A large group of IDPs, deprived of food, shelter and sanitation for three days, were
either stampeded or panicked and tried to break through the cordon. The RPA allegedly suffered
casualties from firing from IDPs.140 The RPA fired into the crowd; numerous IDP casualties
63
resulted. The cordon was restored. In late afternoon, when it was raining hard, the cordon was
breached again. There was more firing, more RPA casualties, more return fire by the RPA with
numerous IDP casualties. The RPA even used rocket-propelled grenades. On the night of 22
April, there was both sniper fire and machete attacks among IDPs.141 The evacuation of the camp
had deteriorated into a full-scale battle using innocent victims as expendable tools of war. On the
fifth day, the largest massacres occurred.
The estimated number of persons killed between 18 and 23 April ranged from a low of 300, given
by the Rwandese government, and a high of 8,000 according to Australian UNAMIR soldiers
who made a partial count. Official UN estimates were revised downward soon after 23 April to
4,000 and then 2,000. The Special Representative of the Secretary-General, Sharyar Khan, and
the UNAMIR Force Commander held that the figure was around 2,000. The deaths were caused
by gunfire, machetes and trampling; the International Commission of Inquiry found that many of
the dead had machete wounds. Since the RPA did not have machetes, this suggests that the hard
core elements in the camps were responsible. However, there was also evidence of summary
executions of IDPs by the RPA.
Though there was a serious problem in the implementation of the agreed plan,142 and specific
confusions over the starting date for the operation, there were deeper problems, specifically the
different priorities, perspectives and values of the various parties, and the impact of the genocide
on the perceptions the parties had of each other. The repeated shift in commencement dates
consolidated RPF suspicions about UN and NGO intentions. The UN had failed to comprehend
adequately the urgent security concerns of the government.
What or who was responsible for this appalling loss of life? There were many possible candidates.
The terrible circumstances played a key role: the internally displaced were kept in camps; Opération
Turquoise had failed to arrest the militants; the alleged perpetrators of the genocide continued to
use their own Hutu compatriots as shields and as potential cannon fodder to launch attacks on the
government and even used machetes to enforce obedience within the camp.
The underequipped and impoverished government, still threatened by the militant elements from
the defeated genocidal regime, could easily be faulted. There was a clear breakdown in
communications and authority among various governmental sectors.143 The International
Commission was clearly correct in faulting the RPA for its lack of communications, its
inexperience, its inappropriate equipment and training for what was essentially a police action. But
these were not the essential elements; none of them explain the RPA’s precipitous action,144 the
most immediate cause of the disaster.
Could the RPA have been stopped? UNAMIR II had a mandate to support and provide safe
conditions for displaced persons. The mandate included taking self-defense actions against those
who _threatened populations in protected sites and the distribution of humanitarian relief. But the
humanitarian agencies and UNAMIR II had agreed to the suspension of humanitarian relief, albeit
gradually rather than suddenly. Moreover, several UNAMIR units were at the time on joint patrols
to escort the bulk of the population to their home communes. Should the Force Commander have
abandoned the returnees, immediately consolidated his contingents from all over Rwanda, and
ordered the superior forces of the RPA to cease and desist? UNAMIR II was also tinged with the
critism from the RPF that the previous UN force had created only an illusion of security prior to 6
April, and had been withdrawn when the genocide commenced. If confronting the RPF at Kibeho,
the UN force would be subjected to the accusation of protecting the killers rather than trying to
arrest them. For these reasons, UNAMIR was impotent to do anything.
The NGOs understandably placed primary emphasis on the well-being of their humanitarian
charges, perhaps without giving adequate consideration to the predicament and need for both
security and justice, and, most importantly, the sense of urgency of the government. The
International Commission criticizes the NGOs for allegedly encouraging IDPs to remain in
Kibeho, but cites no evidence for this charge. Government sources close to the event, and whose
primary interests differed from those of the NGOs, found little to substantiate the Commission’s
64
conclusion.145 However, tension between the command and control structure of UNAMIR and
the anarchic system of NGO coordination did not help the efficacy of the return operation.
A more serious factor was the failure of the international community to invest adequate resources
in Opération Retour to initiate the information campaign in both the camps and the home areas as
well as rehabilitate the justice system and the home communes in preparation for the returnees,
particularly in the winter phase of the plan when the operation was still a voluntary one. There was
also a failure to act sooner and use the not-so-gentle persuasion of promising to withhold food
when the incentives to return home to take advantage of the January crop season were stronger,
before the extremists could organize the resistance, and when the IDPs were not subject to the
dismal feedback by returnees and new IDPs about the perilous conditions at home.
What was not at fault was the effort to achieve coordination and the willingness to devise a plan
that attempted to be both humane and realistic. The UN and the NGOs, however, were not realistic
enough in their lack of a sense of urgency and reluctance to employ not-so-gentle methods earlier,
while the RPA clearly lacked sufficient humanitarian sensibility for the well-being of the
innocents, particularly women and children. In the end, a clear division of responsibilities and
explicit lines of authority in the implementation of a coherently drawn-up camp evacuation might
have averted the tragedy.
65
Chapter 7
Conclusion
7.1. Early warning
The Puzzle
Did those charged with the responsibility for making decisions in the various agencies and states
that comprise the international community know that the assassination of Hutu political opponents
and genocide of the Tutsi in Rwanda would take place? No. Virtually no-one anticipated the
swiftness, scale, thoroughness and unique character of the genocide as it unfolded. In its horrific
enormity, it took almost all international observers by surprise.
If they did not draw such an extreme conclusion, did those with the capacity to prevent and
mitigate the genocide have the information upon which such a conclusion could be drawn? Yes.
Many knew that organized extremist forces existed; increasingly, they even gave public proof of
their existence by words and deeds. A pattern of violence was discernable, and the state apparatus
itself was clearly implicated in arms distributions to para-military groups and extremist propaganda
advocating the need to rid Rwanda of all Tutsi and their supporters. By early 1994, specific
information about plans and conspiracies towards this end was picked up by the UN system, most
significantly in the notorious Black File of January 1994.
Close observers commonly interpreted this information in the Rwandese context to mean that
large-scale ethnic violence against civilians was likely if the civil war was renewed, as seemed
likely in early 1994. However, in the higher echelons of state and international agencies, the early
warning signals were hardly heard at all, and news of the genocide was an even greater surprise.
But pieces of information were available to permit policy-makers on any level to draw the more
radical conclusion that both politicide and genocide might occur on a scale quite different from past
patterns of “just” hundreds or thousands of victims.
Should the alarming indications have resulted in more thorough assessments and preparations?
Absolutely, given the political commitment and actual involvement of the UN in Rwanda, the
large-scale violence expected by close observers, and the possibility of genocide. The legal right
and the normative obligation to act to prevent genocides according to the Geneva Convention, and
the enormous cost of a miscalculation, made contingency planning clearly imperative.
Then why did states, international organizations and other parties who had assumed some
responsibility for regulating the Rwandese conflict and had the capacity to act not draw the
appropriate conclusions? Since close observers did anticipate mass violence and advocated swift
interventionist action, analyzing the problems that afflicted the central decision-makers is critical to
understanding the failure to act.
The problems of receiving even clear and unequivocal signals are found in four areas:
contradictions in the international system; the UN structure; attitudes of senior officials towards
messengers and inadequacies in the messages sent; and interference.
66
7.1.1.Contradictions in the international system
Neutrality and Intelligence Gathering
Unlike nation-states, the United Nations does not collect and analyze information to protect itself
from ostensible enemies. The world organization has no clear-cut security agenda and is supposed
to be a neutral body. Yet in matters of threats to international peace and security, the neutrality
principle does not necessarily apply, as in Chapter VII enforcement actions. By the same logic, the
UN should not be neutral towards genocide, or towards parties threatening civilians whom the UN
has placed under its protection. Moreover, once the UN assumes responsibility for conflict
management, it needs a capacity for information collection and analysis dealing with military and
political issues of member states. Nevertheless, member states are reluctant for reasons of national
security to let the UN develop such an intelligence function.
Financial, Remote and Regional Interests
In the area of conflict management, the UN is particularly beholden to the United States, which
pays almost a third of the budget for peacekeeping. When regional states and organizations, which
are closer to the conflict and its consequences, and in this case were willing, indeed eager, to
involve themselves, they were not given the wherewithal to do so, including a structure to gather
and analyze information. The OAU, which was intimately involved in the diplomatic process to
_settle the conflict, only had a skeletal staff for these purposes. The lack of resources also
characterized Tanzania, in this case the most non-partisan regional state. The major motive of
Tanzania was to regulate the conflict, rather than further one party’s interests.
Zaire had access to the Rwandese regime but no interest in collecting, let alone sharing, any
information that could be used against the Rwandese government. Uganda had such an interest,
and through the RPF also had access to information, but its early warning capacity supported the
RPF. RPF early warning could be dismissed as propaganda. France had both interest in and a
significant capacity for intelligence collection on Rwanda, but its interests were clearly partisan and
led to defining the role of the RPF in adversarial terms. (This included even the French Ministry of
Foreign Affairs, which was committed to the Arusha process.)
The United States also had the capacity, but it was initially truly disinterested, in the sense of being
both objective and remote, and utilized its capacity only sporadically. The CIA undertook a
January 1994 desk-level analysis of the Rwanda situation as a worst-case projection of the course
of current events, which included scenarios of deaths in the order of half a million casualties. This
indicated that specialist analysts who focused on a problem area could use current information to
develop reasonably accurate scenarios. Covering a country of marginal concern to the US, the
report was not distributed widely, nor did it reach the higher decision-making echelons. As there
was no “smoking gun” or verification from personnel on the ground, it was not taken seriously.
Moreover, since Rwanda posed only a hypothetical problem rather than being an actual crisis,
there was little inclination by higher officials to pay attention.
For the UN, both the interest and capacity to cover a crisis in a peripheral state are critical. Caught
between the absolute disinterest of the major powers and the need to cater to dominant financial
interests, the UN has been unable to establish the appropriate balance between disinterestedness
and utilizing regional interests to advance the peace process.
7.1.2. Structure and culture
Though the UN was inhibited from systematically collecting and analyzing critical information and
communicating the analyses to those with the power to take action against the genocide, the
Secretary-General has had a mandate to engage in preventive diplomacy and peacekeeping even in
intra-state conflicts. He has also been given an explicit mandate to engage in early warning. That
mandate can be easily extended to cover the analysis required for preventive diplomacy and
effective peacekeeping. Why has this capacity not been appropriately developed?
The individual relevant units are thinly staffed, given the exponential increase in global
responsibilities assumed by the UN. For instance, only one person in DPKO consistently
monitored Rwanda, and that person carried a heavy weight of operational responsibilities. But
there are deeper reasons for the failure.
67
With a changed mandate since 1990, the Secretary-General reorganized the Secretariat, but in so
doing, sacrificed an independent information and strategic analytic arm separate from any
operational functions by dissolving the Office for Research and Collection of Information (ORCI),
because of its shortcomings. The Secretary-General distributed its responsibilities among the
Political, humanitarian, and Peace-keeping Affairs Departments, while the Office for Human
Rights monitors human rights abuses. There was no central unit in the Secretariat or elsewhere
charged with collecting even “soft” intelligence, including the information available in the vast
structure of UN agencies and organizations to translate assessments into policy options and
strategic planning. As the DHA task force evolved into the focal point for early warning in the UN
system for mitigating the consequences of conflicts, there was no parallel tool to generate early
warning signals required for contingency planning in preventive diplomacy and peacekeeping
operations. DPA, which was expected to undertake this task, was too understaffed and had no
systematic link to strategic planning.
Further, the cultural milieu of the Secretariat had reverted to being politicized in two senses. In the
Rwanda situation, the DPKO at a critical juncture seemed to be too subservient to the concerns of
the major powers. In this case, the DPKO proposed only “what the traffic would bear” in the
Security Council, as they put it, not what the situation on the ground demanded. Further, some
personnel suggested that others were carriers for dominant state players. Politicization versus
professionalism in a context of relative great power disinterest by one major power and a partisan
interest by another proved to be a devastating combination.
7.1.3. Messengers
In spite of these deficiencies resulting from the current mandate of the UN and its structural
problems, UN headquarters did receive crucial information that should have led it at least to
undertake some contingency planning. However, the UN Secretariat had a propensity to discount
the information and warnings received from within its own system while paying inordinate
attention to media analyses, if the Secretary-General’s 20 April report, which prepared the ground
for UNAMIR’s withdrawal, is any indication.
Using information the various humanitarian organizations obtained from their own field
operatives, an inter-agency task force in Geneva was assigned the task of sharing information
about impending human disasters. It had many problems. It was ad hoc. Each of the agencies
showed an understandable reluctance to collect and divulge sensitive political information that
might jeopardize its operations in a country. More importantly, the inter-agency task force was
oriented towards anticipating the humanitarian consequences of a crisis (e.g. population
movements, the requirements of relief operations, etc.) rather than development of the crisis itself.
The task force had no access to the sort of political-military information so critical in discerning
escalating conflict. Nor did it have adequate support from qualified analysts. As a result, while the
UN system has an early warning capacity, this capacity is inappropriate to questions of early
warning for conflict management purposes.
Even greater suspicions were cast on the human rights organizations, which discerned patterns of
violence they characterized as genocide. Part of the problem was the manner in which the
warnings were conveyed. These organizations did not put human rights abuses in a political
context necessary for understanding the nature and probable evolution of the conflict, including its
likely points of culmination, or translate them into strategic options. Further, the human rights
machinery of the UN itself was virtually irrelevant to the early warning process, not only lacking
the capacity to relate human rights violations to a dynamic analysis of the social forces that
produced them, but utilizing a monitoring process that was sporadic and provided no followthrough on reports.
Within the UN Secretariat, far too little attention was paid to early warning emanating from human
rights organizations. The Secretary-General’s report justifying withdrawal on 21 April reflected
the misrepresentations of the media rather than the analysis of informed or more professional
observers of Rwanda. When the Secretary-General by mid-May did perceive the situation
68
correctly, he lacked the widespread and sustained support for policy engagement that adequate
media coverage can generate. With a few notable exceptions, the media in general perpetuated the
misrepresentation of the slaughter in Rwanda as one of anarchic ethnic violence. It took almost a
month for most of the media to “get the story straight”.
The import of the messages from the Rwandese media was downplayed. Hutu extremist calls in
the hate media to rid the country of Tutsi were interpreted by some foreign observers as efforts to
put pressure on the RPF to make concessions at Arusha, or as exaggerated rhetoric. Even
members of the RPF politburo acknowledge in retrospect that they underestimated the significance
of the extremist radio broadcasts.
7.1.4. Interference factors
Functional contradictions and structural deficiencies compounded by discounting some
messengers and exaggerating the value of others are, however, only part of the explanation. The
Rwandese Patriotic Front was closest to the scene, had networks in Rwanda to provide
information, and was in close contact with human rights and refugee organizations as well as its
own diaspora around the world. If any party had a stake in anticipating mass murder and its
extent, and with developing contingency plans to prevent, or, at least, mitigate such a disaster, this
Tutsi-dominated rebel organization surely did.
True, the leadership issued public warnings accusing the government of severe human rights
violations and identifying those responsible for killings in 1992–93 as guilty of genocide. But
leading Front members acknowledged in retrospect that they did not anticipate the magnitude of the
genocide and the RPF did not develop contingency plans for such an eventuality. Why not?
Five factors are suggested that afflicted everyone involved in the Rwanda crisis to different
degrees: incredulity, mind-blindness, shadows, noise, and desensitization.
Incredulity
Genocide is rare. Its path of development was unfamiliar and difficult to discern. Despite the
precedent of Cambodia, many associated the concept of genocide with the highly mechanized Nazi
holocaust. The idea that hundreds of thousands could be slated for execution in a poor, agricultural
society, and that this could be carried out in a short time, seemed incredible.
Mind-blindness
RPF leaders admit that on the eve of their 1990 invasion they expected that thousands of Tutsi
might be killed in retaliation, as had happened in the early 1960s. But to consider genocide as both
a possibility and one they had very limited ability to mitigate, let alone stop, would have either
frozen them into impotence or forced them to accept that they had a measure of historical coresponsibility in the slaughter. Hence, a sort of mind-blindness developed, i.e. an inability to use
the information available to deduce the appropriate conclusions.
A similar blindness (but in the nature of “wishful thinking”) may help to explain the failure of
more remote actors to anticipate the genocide. For the French government to acknowledge that
genocide was in prospect would be to accept a far greater degree of responsibility for its own
involvement in backing the Habyarimana regime and its connected extremists. For the United
States, it meant having to confront the fact that it had a legal right and a moral responsibility to
intervene. For the UN Secretariat it would have meant preparing for a very different kind of
mission in Rwanda at a time when its first principle of action was “not to cross the Mogadishu
line”. The UN mind-blindness carried a double level of guilt and denial by instilling a false sense
of confidence among the Tutsi in Rwanda, who could not imagine that the world in general, and
the UN in particular, would flee the scene when a UN peacekeeping force was already in place.
Shadows
The tardy and inadequate response of the UN and its leading members was also influenced by
external blinkers, by previous events that haunt decision-makers, distort perceptions and constrain
their willingness to act. These are in the nature of shadows.
69
One such blinker was the shadow of hope. Arusha became an inertial force. The peace accords so
preoccupied almost all the players that they tended to shift into the background the growing
organization of extremists intent on undermining that peace. The pursuit of the agreement, the
hope invested in its effectiveness, and the eagerness to see it implemented dominated their attention
right up until 6 April.
The other blinker was the shadow of despair and the propensity to read the present through
traumas of the past. Somalia certainly directly colored the American propensity to view (and
deform) the Rwandese operation through the Somalia lens. The issuance of PDD 25, a new
directive in May of 1994 strictly limiting American involvement and funding of peacekeeping, was
a culmination of the shadow of Somalia. But Rwanda was not a failed state; on the contrary, it was
an example of a state that, having been taken over in an extremist coup d’état, was executing a
massive genocide.
The UN was also haunted by Somalia. Fear of taking causalities in a UN peacekeeping mission
made the Security Council and the Secretariat willing and ready to follow the Belgians out of
Rwanda once the hostilities broke.
In the shadow of Somalia, the American and UN observers misinterpreted events on the ground
and exaggerated difficulties into an impotence to respond effectively. Not only were the wrong
lessons learned from Somalia, the right ones were ignored, In fact, the lesson of Somalia should
have been that too few forces inserted tardily can undermine a whole operation.
Noise
This is a standard element often used to explain intelligence failures. At the same time as the
situation was particularly tense in Kigali in March of 1994, the UN was preoccupied with a long
list of crises or problem areas – in Russia, Angola, Burundi, El Salvador, Georgia, Haiti, Liberia,
Nagorny-Karabakh, Somalia, South Africa, Tajikistan, Eritrea-Sudan, Mozambique, Hebron,
North Korea, and especially the former Yugoslavia (including the security and safety of
UNPROFOR). Because of Rwanda’s peripheral relationship to the major issues and structures of
international relations, the signals from Kigali were weakly heard.
In a context crowded with actual crises, it was particularly difficult to get busy decision-makers to
shift their attention onto a potential crisis.
Confusion about the Message
Noise may overwhelm a message, but this propensity is exacerbated when the message itself is
confusing. At various points, different actors monitoring the conflict – human rights
organizations, the RPF, the Special Rapporteur of the Commission on Human Rights – used the
term “genocide”. But the term conveyed different meanings, which created confusion among the
listeners.
The problem with the term “genocide” as a signal comes in the different implications and
illustrations of the two uses of the term. The use of the legal definition of the term in an accusatory
sense arguably diminished the impact of the term in its function as a warning signal. If the killing
of 300 Tutsi constitutes genocide (in the legal sense), then warnings about potential genocide
signal the potential death of a few hundred more. The linking of the deaths of 300–1000 people to
the terms “apocalypse” and “genocide” diminished their impact as warnings. While significant in
and of itself, early warning about genocide defined in legal terms leads to very different thinking
about consequences and reaction than would a clear signal of an impending genocide as defined in
the popular sense.
In the popular consciousness, the concept of genocide is linked to the massive slaughter of
Armenians by Turkey at the turn of the century, the extermination of millions of Jews in Europe
by the Nazi regime, and the wiping out of an entire generation and class of people by the Khmer
Rouge in Cambodia. The numbers involved in each of these cases are measured in the hundreds of
70
thousand or millions. Unfortunately, events propelled Rwanda into the same tragic class.
Desensitization
Only six months before the genocide occurred, massacres in Burundi claimed the lives of an
estimated 50,000 to 100,000 persons. There was no worldwide outrage and little response. This
failure of the international system to be startled or to respond reflected a pre-existing propensity to
expect disasters out of Africa. The threshold for international response seemed to increase
accordingly: if 100,000 persons could be killed with impunity, the possibility of another massive
slaughter did not seem so extraordinary, nor did it require particular responses. Hence, there
developed a mental indifference to the possible consequences of the rising tensions in Rwanda.
This was reinforced by the tendency to view the struggles within Rwanda as the recurrence of
ancient conflicts between Hutu and Tutsi rather than as a product of a centrally led, political
murder machine. As one Belgian official later put it: “We’ve learned from our experience that
whenever there is trouble in Rwanda, the Hutu and the Tutsi kill each other.”
7.2. Conflict management
The international community might have responded better had the early warning systems generated
a clearer anticipation that genocide was on the horizon. Yet conflict management is a function of
capacity, interest and commitment as well as information. In the Rwandese case, the relevant
actors knew at a critical stage that the situation was unstable and dangerous. Unforeseen
detrimental events did indeed occur to give the situation a turn for the worse – in particular, the
October 1993 ethnic massacres in neighboring Burundi. But the major powers that controlled UN
peacekeeping operations paid only cursory attention to Rwanda and there were no contingency
plans except for withdrawal. The sustained and careful attention so necessary to successful conflict
management was lacking.
There was considerable international investment in preventive diplomacy in Rwanda, but when
this failed and genocide ensued, the international community effectively disengaged. Humanitarian
relief within Rwanda and to refugees across the border soon resumed, but these efforts were
designed to deal with the consequences of conflict rather than the violence itself. Re-interventions
to save civilians had very limited impact, as indeed might be expected from retroactive conflict
management.
The reasons for this fundamental failure to respond are found, ultimately, in a structural mismatch
between institutions and interests in the contemporary state system. Revitalized by the end of the
Cold War, the United Nations in the 1990s rapidly expanded its peacekeeping operations
throughout the world. Rwanda was added to the list in October 1993. The framework for
peacekeeping, however, was set by the distribution of power in the Security Council, which in
form still reflected the world as it was half a century ago. Except for France, the major powers on
the Council were basically uninterested in a small Central African country that was marginal to
their economic or political concerns, and peripheral to international strategic rivalries. Through
their power of veto and financial commitments, these states also controlled the peacekeeping or
enforcement operations of the United Nations. Preoccupied with crises elsewhere, especially in
Bosnia and Haiti, and haunted by the memories of Somalia, they decided not to engage in Rwanda
until it was too late. Other potential actors were either unwilling or, at that late stage, unable to
respond on their own. Hence, a principal lesson from the Rwandese conflict is that in a world of
multiple crises, even major disasters in a seemingly peripheral state fall victim to neglect.
Another lesson refers to the general principle that respect for international law and norms will tend
to diminish conflict, whereas violations will tend to stoke it. The behavior of state and presumptive
state actors in the Rwandese conflict was in this respect less than adequate, and mostly counter_productive. International law and associated principles designed to uphold international order
were repeatedly violated, including sanctity of national borders and arms embargoes. International
refugee law was not observed. The legal right and moral obligation to intervene to stop genocide
71
was acted upon. Human rights law was repeatedly and severely transgressed with impunity.
Donors continued to give economic aid, and, in one case, also substantial military assistance, to a
government linked to systematic violations of human rights. While this might be expected in a
world of competitive nation-states observing the rules of Realpolitik, it should also be recalled that
the consequences of lawlessness in this case were exceedingly costly – both for the Rwandese
parties and people, and for outside states concerned.
International efforts to manage the conflict – as distinct from addressing its humanitarian
consequences – were a failure when judged against international norms governing the prevention
or mitigation of genocidal violence, as well as other UN standards. By explicitly asserting the right
to intervene in a world otherwise governed by principles of national sovereignty, the Genocide
Convention implicitly affirms the associated moral obligation to act. The United Nations and its
members had already assumed some responsibility for conflict management in Rwanda by
establishing a peacekeeping operation to help implement the Arusha Accords. When the Accords
failed and the genocide commenced, the UN formally acknowledged an obligation to help protect
civilians. This recognition came belatedly in the form of a Security Council resolution on 17 May
1994 – i.e. six weeks after the genocide commenced. The tardiness in authorizing and deploying a
protective force weakened its operational effectiveness, but not the validity of the principle.
7.2.1. Critical actors at critical junctures
While policies were shaped by underlying structures of power and interest, the various actors
nevertheless operated within a framework of options that permitted a certain choice. Among these,
a few stand out as commendable efforts to reduce or mitigate the conflict; others are conspicuous
for failing at critical moments. Significant in this respect are:
° • The Tanzanian government and the OAU efficiently and patiently spearheaded a sustained
international effort to mediate the civil war. While the Arusha Accords in retrospect were not as
perfect as the sponsors claimed, the agreement did provide a reasonable basis for settling the civil
war. Partly undermined by unforeseen developments in neighboring Burundi, the peace agreement
also carried some seeds of its own destruction by failing to take care of the losers. Having been
excluded from the settlement and not dealt with otherwise, the Hutu extremists became an obstacle
to its implementation.
° • UNAMIR on the ground was an exercise in doing the most with as little as possible. The
Force Commander struggled in vain to bring the mission up to authorized strength and to alert an
unresponsive UN Secretariat (DPKO) to the fact that the situation in early 1994 was seriously
deteriorating. Once the civil war resumed and genocide commenced, UNAMIR was able to save
lives by protecting some 15–20,000 persons in the Kigali area, despite lack of supplies and heavy
equipment and a drastically scaled-down force. Its credibility and hence ability to modify the
conflict in other respects (e.g. by promoting a cease-fire) was critically undermined by the decision
to withdraw.
• Of the humanitarian agencies, ICRC in particular helped provide protection and assistance to
thousands of civilians.
• After a bungled start and initial paralysis by the DPKO, the UN Secretary-General recovered the
initiative by proposing that the UN re-engage itself to mitigate the conflict. The Under SecretaryGeneral of Humanitarian Affairs helped focus attention on the crisis during the early period and
was the first high-level representative from New York to visit Kigali after 6 April.
• While failing to anticipate that some 1.5 million persons would flee across the border (see Study
III), UNHCR soon recognized the need to deal with the problem of militarized refugee camps so
as to avoid the start of a new conflict cycle. The High Commissioner’s innovative proposal was
bogged down in discussions at UN/New York, where the Secretary-General sought to address the
problem comprehensively as a peacekeeping matter. This failed, however, and the result was a
critical delay of several months before UNHCR could move ahead to deal with the camps issue.
72
• Of the regional states, Uganda and Zaire at various times provided support that served to escalate
the conflict rather than reduce it, though Uganda also contributed to the mediation process.
• Through its military and economic assistance, as well as diplomatic support, France gave
significant and sustained aid to a regime that was linked to systematic human rights violations and
ultimately the genocide.
• The Belgian decision to withdraw its UN contingent at the time of crisis crippled UNAMIR and
drastically reduced its options for the future. Belgian lobbying for the Security Council to
withdraw the remaining force altogether helped shape the final decision to this effect.
• By acts of omission, the United States ensured that neither an effective national response nor a
collective UN effort to mitigate the genocide materialized. Citing financial restraints, the United
States wanted a bare-bones UNAMIR before 6 April, argued for withdrawal soon afterwards, and
delayed the authorization as well as deployment of an expanded UN force in May–June.
• The media by and large covered events in a fundamentally irresponsible manner. Notable
exceptions were reports from correspondents for the BBC, Le Monde, Libération and The Times
(London). The rest of the media reported in a highly selective and initially misleading way,
_although some later were quick to charge Western governments with complicity in the genocide.
Genocide and politicide were at first depicted as tribal-ethnic conflict with ancient and typically
African roots. Coverage did not become intense until the genocide was over, and huge refugee
flows streamed into Goma. The failure of the media to report accurately and adequately on a crime
against humanity significantly contributed to international disinterest in the genocide, and possibly
to the crime itself.
In a more general sense, the failure to respond was also linked to inadequate structures or
procedures in decision-making. Some problems arose from the relationship between the OAU and
the UN, and point to the importance of including – one way or another – the relevant regional
forces in the decision-making order to achieve a durable solution to a local conflict.
7.2.2. Decision-making structures
Regionalism and the Disjuncture between Mediation and Implementation
From the beginning (1990), regional states participated in diplomatic efforts to deal with the
conflict. While two of them were also involved in the war (Uganda and Zaire), that was
considered all the more reason to include them in the process, which was spearheaded by another
neighboring state (Tanzania) and the OAU. The latter’s interests were primarily in a higher rational
order that, among other things, would spare the region another large refugee problem. Over time,
all external parties to the conflict appeared to gain a stake in the success of the mediation effort,
and all signed the final Arusha document. There was, however, no continuity between the
mediation and implementation phase of the peace accords. The UN took over the peace operation,
essentially closing off the regional effort and cutting out the OAU. The consequent structural
disjuncture between mediation and implementation affected three critical aspects that made it
difficult to maintain the momentum of the Arusha peace process. These were:
• Mandate and expectations: The Arusha Accords presumed and specifically called for a
peacekeeping force with a mandate broader than the UN was willing to provide.
• Timing: The timetable of the Arusha process presumed a speedy international presence, yet this
did not sufficiently take into account, nor was it adequately coordinated with, the time-frame for
establishing UN peacekeeping operations.
• The extremists: The existence of an extremist group that was excluded from power in the peace
agreement, but which continued in positions of power in the interim, should have been dealt with
by other means. In shifting responsibility from one set of actors in the mediation phase, to other
actors in the implementation phase, the critical issue of extremists was neglected.
73
In important respects, Rwanda obtained a peacekeeping force rather different from the one the
negotiators of the agreement had anticipated and deemed necessary for the implementation of the
Accords. The disjuncture was partly caused by organizational competition between the OAU and
the UN. The OAU actively sought to obtain a leading role in the peacekeeping phase as well, but
the Security Council insisted that the UN would not pay unless it was given command-and-control
of the operation. Except for France, which lobbied hard to make it a UN force, and the Rwandese
government, which happened to be a member after January 1994, no other states in the Security
Council gave the issue much consideration. The UN Secretary-General weighed in on the side of
his own organization. Moreover, the dismal record of the OAU in managing its minor military
mission in Rwanda (NMOG I and II), gave ammunition to critics in the UN who maintained that
the OAU had neither capacity nor the required impartiality to play a major role in a peacekeeping
operation. As it turned out, however, UN Headquarters did not run an effective peacekeeping
operation in Rwanda either, and UN neutrality in the face of genocide became a matter of criticism
rather than approbation.
Issue of Stakeholders
While the United Nations had formal responsibility for helping to implement the Arusha Accords,
the main stakeholders in the peace agreement were in the region, not in the Security Council. The
latter consequently gave UNAMIR a narrow mandate, a limited budget, and scant attention. When
the Arusha Accords appeared to unravel, the Security Council threatened to withdraw rather than
strengthen UNAMIR. When the crisis erupted after 6 April, there was a “stampede to get out”, as
one member of the Security Council described the reaction of the chamber. At this critical juncture,
the department of the UN Secretariat most directly responsible (DPKO) showed neither initiative
nor an ability to rise above its assessments of what “the traffic would bear” in the Council.
African states showed more interest, partly reflecting the notion that African states had a special
responsibility for solving their own conflicts. States in the region had particularist interests as
likely receiving countries for massive refugee flows. Except for the military observers, the African
battalion in UNAMIR I was the only unit that stayed put during the crisis; Ghana decided to keep
its contingent while the Belgian and Bangladeshi governments withdrew theirs. During subsequent
UN efforts to manage the conflict retroactively by reintroducing a force, only African countries
offered to send troops. Operating under severe resource restraints, they required external financing
and some equipment, to which the major powers only reluctantly agreed.
In sum, the regional contribution to conflict management was undervalued and underutilized
despite recent and formal recognition by the UN leadership, including the Secretary-General, of
the need to involve regional organizations to promote international peace (cf. Agenda for Peace,
1992).
7.2.3. Decision-making process
The United Nations
UN procedures governing peacekeeping operations made for slow and incomplete deployment.
While this was not specific to Rwanda, the consequences were more acutely felt because the
country had no “patron” in the Security Council that could cut through the political and
bureaucratic morass. The Secretariat for the most part proved unwilling or unable to compensate.
Before the acute crisis erupted on 6 April, DPKO, which had responsibility for UNAMIR, made
no contingency plans or efforts to strengthen the mission’s preparedness for worse-case scenarios
despite clear evidence of mounting tension. The problem was partly due to limited institutional
capacity in the face of a rapid increase in peacekeeping operations worldwide. The restraints were
also more deep-seated, as indicated by the failure of DPKO to overcome a near-paralysis when the
crisis broke and Rwanda moved to the top of the agenda. DPKO’s communications to the Security
Council were tailored to expectations of what the Security Council would approve (thus giving the
Permanent Five anticipatory vetoes); options were formulated in terms of standard operating
procedures, rather than the unique needs of the situation; and instructions to the field were heavily
influenced by a concern to reduce risk so as to avoid “failures” – which were defined as the death
of UN peacekeepers. Only when the Secretary-General in late April decided to provide some
74
leadership did the Secretariat play an innovative and proactive role, based on more comprehensive
and independent assessments of the requirements of the Rwandese situation. At that point,
however, the limitations of retroactive conflict management, and inability of the UN to respond
with dispatch, rendered the operation largely irrelevant.
The critical importance in this case of leadership touches broader issues of accountability and
transparency in the continuing discussion of UN reforms. Given the large number of crises in the
world, the Secretary-General can not respond equally effectively to all. However, the criteria for
selection, and the process that leads the Secretariat to highlight some conflicts rather than others,
remain obscure to the public. Similarly, the fateful decision virtually to withdraw UNAMIR was
taken by the Security Council in informal consultations. In keeping with normal Security Council
procedures, only select formal statements made at the conclusion of the decision-making process
(21 April) were recorded in the proceedings and are thus in the public domain.
The effectiveness of UNAMIR also suffered from characteristic features of UN peacekeeping
operations:
• The field mission had very limited authority to make decisions; routine matters as well as issues
heavily dependent upon judgement of the situation in the field were micro-managed by New York.
• When deployed in late 1993, UNAMIR I had only a small investigative unit and no separate
human rights component designed to monitor and report on human rights violations. This limited
its ability to gauge a deteriorating situation. It also signalled to the Rwandese parties that the UN
placed low priority on human rights violations in implementation of the peace accords.
• The normal, slow process of assembling a peacekeeping force had monumental consequences.
Once the Security Council decided in May 1994 to re-engage the UN for the explicit purpose of
protecting civilians, a period of about three months elapsed before new UNAMIR units arrived in
Rwanda. In the intervening period, hundreds of thousands were killed. The deployment was
particularly slow because none of the industrialized states that could have provided the UN with
fully-equipped units at short notice were willing to do so. The solution was to match Western
equipment with African troops, a process that was time-consuming.
7.2.4. Policy issues
The international response to the Rwandese crisis also raised several policy issues of general
_concern:
National Peacekeeping Operations with a U N Umbrella. The French Opération
Turquoise was quickly deployed and efficiently run and some held it up as a model. It should be
noted, however, that the smooth operation was not only a function of its being national (as distinct
from cumbersome and multilateral). Effective communication with the RPF made it possible to
agree on the boundaries of the French “safe humanitarian zone” and, behind this line, French
troops were in generally friendly territory. The operation also raised some troublesome issues.
Mounted at the time the UN was struggling to obtain troop commitments for UNAMIR II, the
French initiative detracted from the collective effort and weakened the credibility of multilateral
interventions. The circumstances of the operation cast doubt on official claims that it was a purely
humanitarian intervention, thus degrading the latter concept.
Democratization. International insistence that Rwanda’s political system be democratized had
contradictory implications. The formal institutions of multipartisme provided a political framework
for accommodating the RPF and, in that respect, foreign donors promoted a formula that would
help terminate the civil war. On the other hand, democratization so defined did not address human
rights issues, and in some respects became a substitute for dealing with them. The results of
efforts to strengthen civil society were also ambiguous in that they provided space for both human
rights organizations and extremist groups. More generally, donor demands for democratization
added to the overall pressures experienced by a regime that was at war and faced national
economic collapse as well as mounting internal opposition. The combined effect was a sort of
75
system overload that enhanced the power and possibly the appeal of Hutu extremists. These
dysfunctional consequences were not sufficiently recognized at the time in the belief that the multiparty system entailed by democratization and endorsed by the Arusha Accords would take care of
the extremists in due course.
Preventive Diplomacy. In retrospect it can be clearly seen that the closer the parties at Arusha
came to end the civil war, the more Rwanda inched towards disaster. The inter-connectedness of
the two tracks – the civil war and the civil violence – was recognized at the time, but the Arusha
process was basically designed to settle a war, not to prevent a dimly perceived future catastrophe.
If there is a lesson here, it lies in the tested principle that a settlement to end one war can be the
beginning of a new one unless the peace agreement is also constructed as preventive diplomacy.
Refugees. Two principles of refugee policy are generally recognized in the international
community: refugees have the right to return, and those who cannot return should be given asylum
or resettlement elsewhere. Failure to observe such principles typically creates festering refugee
problems, and in many cases militant communities who seek to escape from their dilemma by
force. The phases of the Rwandese conflict considered here started and ended with festering
refugee problems. While cognizant of the problems preceding the 1990 invasion, UNHCR could
rely only on its good offices to promote a solution. Significantly, the agency lacked even the
capacity to assess adequately the magnitude of the problem, let alone other means to pursue what
in the early 1990s came to be known as a “comprehensive refugee policy”, that is, an approach
that addresses causes and solutions to refugee problems, not only immediate material and
protection needs (A/AC.96/799,1992).
The formation of new militant and militarized refugee communities in Zaire during the second half
of 1994 indicated the beginning of another conflict cycle. The response of the UN body
responsible for refugees, UNHCR, reflected increasing awareness of the problems posed by such
communities. _It was recognized that a solution, if any, requires disaggregating the problem. This
entails separating the refugees from the military and dealing with the former within a legal-political
framework of repatriation or resettlement. The latter constituted a military-security problem and
had to be dealt with as such. By early 1995, reasonably effective arrangements were made to
improve policing of the camps in both Tanzania and Zaire. Yet the issue of the armed militants and
their links with the refugees remained unresolved as the armed groups were free to rearm and
regroup near the border.
A solution to this problem required, in the first instance, greater cooperation from the authorities in
Zaire.
76
Chapter 8
Recommendations
A. General – humanitarian realism
The United Nations projects a vision of the unity of humans governed by norms that enhance
equality and justice, peace and the well-being of all. In the Rwanda case, these principles were
duly referred to and sometimes reflected in policy. Some states identified national security with
international security and greater world order, and placed peacekeeping units at the UN’s disposal
for this purpose. Other states invested in preventive diplomacy, recognizing that renewed conflict
would threaten regional peace and security and probably create large refugee flows. Some took
notice of persistent efforts by NGOs to flag human rights issues. On the other hand, the behavior
of states was also, and in some cases mostly, characterized by contests for power driven by
narrow concepts of self-interest. Policies of prevention and conflict management were subjected to
miserly criteria of risks and costs; only when the emphasis shifted to low-risk, high-visibility
humanitarian assistance did states generally show generosity. The UN as a collective actor was
unable to rise above its members in these respects, either in the Secretariat or the Security Council.
Despite the overall international failure to monitor and manage the Rwanda conflict, the efforts that
did take place to mitigate the crisis and its consequences suggest that states are not necessarily
bound to pursue a sterile Realpolitik. Sometimes, issues of self-interest coincide with higher
norms; at other times, it is necessary to examine whether self-interest requires a humanitarian
engagement. From the perspective of conflict management, this poses two critical challenges.
The issue is not choosing between vision and reality, between humanitarianism and realism.
Policies based on narrow self-interest are costly in terms of the conflicts that result and the
humanitarian expenditures that follow in spite of a commitment to hard-headed Realpolitik.
Further, a policy that may lead to shame is not realistic. On the other hand, an idealism that fails to
take account of the interests of states is a recipe for future failure. Policies must be based on
humanitarian realism. The recommendations offered below are based on the premise that it is a
contractual obligation as well as in the self-interest of states and international organizations to
apply the norms to which they formally subscribe.
The second task is to translate this most general philosophical approach into policy. This a a
challenge for all concerned with public policy. In part, this requires establishing structures that
include or otherwise take into accont the stakeholders in a given conflict. It also means developing
policy approaches and procedures which ensure that the stakes held are compatible with a rational
order of justice and peace. Some guidelines that embody the principles of humanitarian realism can
be derived from the Rwanda case.
The following recommendations are addressed to states as well as international organizations. In a
world of competitive nation-states, states hold the major keys to power, both in their capacity as
individual actors and as members of international organizations. By comparison, the latter are
weak in many respects. Indeed, if the present case study is any indication, international
organizations as presently constituted – both at the UN and at the regional level – seem unable to
deal effectively with violent political crises. While recognizing these structural restraints, it is also
important to identify desirable reforms.
77
B. Policy approaches to conflict management
Principled coherence
The Rwanda case demonstrated the need for greater policy coherence, but equally the need for
principled coherence. With respect to the latter, at least two major standards of international
conduct were violated at various times: the sanctity of borders and the rights of refugees.
1. a) Invasions across borders must be taken seriously by the international community,
initially at the sub-regional and regional level, to defuse and contain the resulting conflict.
b) Recognizing that festering refugee problems often produce new cycles of
conflict, the international community must support states most directly concerned to
ensure that refugees are not left in limbo, but within a reasonable time obtain secure
membership in a state. Given that solutions to refugee problems typically involve a choice
between inadequate alternatives, the least inadequate one in many situtions will entail
induced or “humane” repatriation.
There was a tendency to isolate human rights monitoring into the role of a moral watchdog rather
than to integrate it in anticipating and developing strategic responses to the crisis. At the same time,
organizations involved in economic development did not incorporate human rights as a specific
dimension of development. Peacekeeping operations did not acquire a human rights component
until after the crisis had erupted.
2. Human rights monitoring must be integrated with related economic development
planning, diplomacy and peacekeeping operations, as well as humanitarian polices to aid
refugees and displaced persons. A coherent approach must be developed in the work of
the United Nations, as well as bilateral donors, and exemplified in coherent plans of
action to avoid duplication.
Targeting
While all conflicts are important, in a world of limited resources and attention span, greater efforts
must be made to set priorities for purposes of early warning and conflict management. While
prioritizing among crises involves difficult trade-offs, similar rankings are made, for instance, by
legal scholars who seek to identify the most basic human rights in international law.
3. States that promote human rights in their foreign policies, the UN Commission on
Human Rights and the High Commissioner for Human Rights should develop a capacity
to relate human rights violations to a dynamic analysis of the social forces that produced
them, and strengthen procedures to flag crisis areas in terms of severity and urgency.
Procedures must be improved to investigate such situations and ensure that the results are
put before the relevant decision-makers.
4. Special efforts must be made to act on crimes against humanity. The international
community should remain on constant alert against perpetrators, with a need to determine
the crises that are both most urgent and amenable to corrective intervention.
Delegating
It will be recalled that the failure of the international community to mitigate the genocide in Rwanda
was neither a function of lack of early warning information, nor of malevolence, but essentially of
neglect. The case points to the inability of the UN, with its fragile structure and economic base, to
respond adequately in a world of multiple, simultaneous crises. Without a division of labor among
international organizations and state actors, the UN becomes easily overburdened and rhetorical
formalism develops.
In Rwanda, abstract principles were recorded in resolutions, but policy followed the interests – or
disinterest – of the major powers on the Security Council. By contrast, the regional organizations
and states, who were the most immediate stakeholders in both peace and war in the Great Lakes
78
region, were active during the mediation phase, but were not given the wherewithal to participate
effectively in the monitoring and peacekeeping phase.
5. In the developing relationship between the UN and regional bodies in matters of
preventive diplomacy and peacekeeping, regional bodies and parties most concerned
should be given a greater role and greater capacity to deal with local conflicts. Current
plans for strengthening OAU’s peacekeeping functions with the support of the UN should
be accelerated.
° A system of resource reallocation should take into consideration that most of the
world’s conflicts occur in regions where the neighboring states and regional organizations
have the fewest resources to deal with them.
° The UN should remain the body that sanctions action, sets parameters and monitors
implementation of forceful intervention and, where needed, helps finance and provision
such actions, but should be the body of last resort in the implementation of enforcement
action.
° Regional participation in preventive diplomacy should be carried over into
peacekeeping so as to ensure continuity between mediation and implementation.
Timing
The process of establishing and deploying UN forces in Rwanda was slow and inadequate relative
to the task at hand, though reasonably quick according to “normal” procedures for UN
peacekeeping. The consequences in terms of lives lost were significant, possibly enormous.
6. The international community must increase its capacity for rapid response, as
exemplified in several current proposals for a rapid deployment force. The UN should
undertake a comprehensive review of procedures to enhance rapid response, including
budget procedures. Existing plans to strengthen rapid response by African peacekeeping
forces should be carried forward.
Conditionality
Policy conditionality in Rwanda entailed several dilemmas. Attaching human rights conditions to
development aid required early action before conditions developed into a crisis where development
aid became irrelevant. Yet the nature of the conflict in its earlier stage was difficult to discern and
did not raise sufficient alarm. The relationship among different kinds of conditionality was another
problem. Conditionality on economic restructuring exacerbated social tensions and undermined
efforts to improve human rights through political conditionality.
7. A more general concept of “conflict conditionality” should be considered, where
different kinds of conditionality (economic, political, human rights) are treated flexibly in
terms of their likely contribution to, or impact on, violent social conflict. Policy should be
informed by systematic studies of past cases regarding the timing, nature and effect of
such conditionality.
Preparedness
An underlying philosophy and policy guidelines are mere abstractions unless the tools are in place
to respond to a crisis. Effective crisis management requires preparedness that permits a proactive
stance.
8. The UN and the states that assume multilateralism as an integral part of their foreign
policy must develop systems of planning and response to humanitarian crises that go well
beyond the ability to deploy troops with speed, but must include planning, strategies and
the preparation for deploying the wide array of personnel required for such emergencies.
79
C. Early warning
Early warning was less critical in the Rwanda crisis than the willingness and ability to respond.
Nevertheless, the failure to respond adequately was in part influenced by the failure to collect and
analyze the data that was available and to translate this information into strategic plans. Information
and analysis is critical, not only in assisting in anticipating a crisis, but in determining the
appropriate response in a particular situation.
As many UN resolutions have reiterated, there is a need to develop an early warning system within
the UN. A number of developments are under way. However, such an EW system must go
beyond simply developing a network or building on the new enchantment with data-based systems
and formal modelling for anticipating crises. Neither technique would have been adequate in the
Rwanda case. The UN and regional organizations must develop, both at headquarters and in the
field, a high-level capability for collecting and analyzing information in a particular crisis area and
translating this information into strategic alternatives. Further, the head of this early warning
analytic system must have direct and regular access to the chief executive.
The UN neither has the capacity nor the need to build such an early warning system on its own.
The organization needs to build on area expertise inside and outside the relevant regions – in the
Rwanda case, the Great Lakes Region of Africa – utilizing information officers to coordinate a
world network of specialists in various state agencies, academic institutions, rights monitoring
groups, etc.
The following characteristics of an ideal early warning system would have been extremely helpful
if even approximated in the Rwanda case: a) the presence of specialized units in the field, in
regional institutions (i.e. the OAU conflict resolution unit) and within the UN system, without any
operational responsibilities, but with responsibility for collecting and analyzing information and
translating it into strategic options; b) the backstopping of that unit with a network of experts,
many of whom are in academia, electronically linked to participate in the provision of information
and analyses; c) the institutional assignment of lead responsibility for regions to ensure that even
states considered peripheral in the international system are not ignored, lest other Rwandas fall into
the cracks of an intelligence analysis system; d) fitting human rights monitoring within a larger
information and analytic structure that can process the information in terms of complex social
conflict (not currently undertaken by human rights organizations) and communicate the analysis to
the policy planning levels; e) the development within the human rights monitors and any early
warning system of a method of prioritizing and flagging in order to bring significant attention and
resources to important cases; f) linking early warning with contingency preparedness, since even
under the best circumstances it is impossible to pinpoint specific future outcomes of a complex
social conflict, and since the absence of any contingency planning will limit what will be heard.
9. The UN should be at the center of the development of a network of experts focused on
different geographical areas for collecting and analyzing information and developing
strategic alternatives to guide preventive diplomacy and peacekeeping operations. A
specialized unit without any operational responsibilities, preferably building on an
existing UN structure, should be designated to collect and analyze information available
in the international system and translate that into strategic options. The head of such a unit
should have access to the highest levels of _decision-making within the UN and in the
various parts of the network. As participants in this system, regional organizations should
develop specialized area early warning systems.
These early warning networks would help relieve what seems to be an over-reliance on media
sources of information. The inadequate media coverage of the genocide in Rwanda is a separate
and serious matter of concern.
10. The media should retrospectively examine their coverage of Rwanda to explain and
80
draw _lessons from both the overall failure to report early and accurately on the unfolding
events, and on the excellent exceptions.
D. The Parties
In addition to the above general recommendations, more specific recommendations emerging from
this study are addressed to various actors concerned with complex humanitarian emergencies.
The refugee issue was central to the Rwanda crisis and remained a critical factor to later
developments. To deal effectively with the refugee issue before the 1990 invasion, UNHCR
would have needed: i) a plan of action to settle the problems within a reasonable time-frame, and
ii) support from states and financial institutions with leverage that could be brought to bear on the
situation.
11. a) In areas of unresolved refugee issues, the underlying political dimensions must be
recognized and dealt with through strategic plans of action. UNHCR should take the lead
in drawing up action plans with specific time lines for implementation, recognizing that in
every case optimal justice is a chimera, and choices will have to be made between lesser
evils.
b) Armed refugees must be separated from civilian refugees and dealt with as a political
issue in accordance with international law. Host states have an obligation to cooperate
with the UN agencies concerned to provide security for refugees and must be held
accountable. Given the relative success of the UNHCR-initiated Zairian Camp Security
Operation (ZCSO) in 1995, the arrangement may serve as a model in similar situations
elsewhere.
A central problem that was evident in Rwanda concerns the fundamental issue of the effect of a
system of condemnation without any penalties. This reinforces a sense of impunity on those who
violate the human rights of others.
12. The present nascent legal system for investigating and prosecuting crimes against
humanity must be developed from one rooted in specific crises. Support should be given
to current efforts in the UN system to develop a standing system of investigation,
prosecution and punishment for those engaged in such crimes.
Outside arms suppliers contributed to and exacerbated the conflict in Rwanda in violation of the
Arusha Accords and the spirit, if not the text, of the UN arms embargo.
13. Breaches of arms embargoes established by treaties or instituted by the UN should be
investigated, and member nations breaking such agreements should be penalized.
The Rwanda case demonstrated the need for authoritative determination of genocide while it is in
progress. A non-state body should hold hearings to determine whether violations of human rights
and humanitarian law constitute genocide and permit corrective actions according to the 1948
Convention. Since time is typically of the essence in such cases, a system of rapid determination
as well as rapid police action is needed.
14. The World Court or another independent international tribunal should be assigned the
task of holding emergency hearings in order to determine whether a prima facie case of
genocide is under way, thus justifying international intervention.
Questions were raised about the absence of a professional ethos in the UN Secretariat, particularly
in the DPKO, where analysis and options in a critical phase were based less on the dictates of the
information and an analysis of options for action than on the willingness or, more correctly,
81
unwillingness of Great Powers to endorse such actions. The weakness of the international civil
service in this respect is related not only to the recruiting and promotion system, but also to the
insecure financial base of UN peacekeeping. Peacekeeping has long been a deficit operation and
subject to prolonged struggles for states to pay up. By being assessed a stiff 31% of all
operations, the United States has a disproportionate influence in peacekeeping matters.1
15. Attention should be given to promote conditions conducive to greater professionalism
in the UN Secretariat. Officials should be trained as international civil servants and
assessed in terms of such professionalism.
The issue of UNAMIR’s mandate prior to April has been exaggerated. Far more important were
the rules of engagement and the narrow strictures within which the force was ordered to operate.
Nevertheless, the dichotomy between Chapter VI and VII operations sows confusion because it
conceals a range of actions that are neither pure peacekeeping (VI) nor classical enforcement
_against an aggressor (VII). Colloquially identified as Chapter six-and-a half actions, these include
a variety of circumstances and responses that need to be specified separately. Doctrine needs to be
developed for their utilization instead of relying on ad hoc innovation.
16. The UN, particularly DPKO, should encourage the development of doctrinal clarity
by elaborating standard operating procedures for peacekeeping actions that are neither
classical Chapter VI nor Chapter VII.
Sound principles warn against micro-managing a military or political operation in the field. In the
Rwanda case, the peacekeeping mission operated within much too narrow limits of discretionary
power. The Force Commander and the SRSG required authorization from Headquarters for even
petty financial matters, let alone more substantial issues. The strictures against micro-management
are even more pertinent when the chief executive carries ultimate responsibility for responding to a
number of crises at the same time.
17. Field officers, both political and military, must be given terms of engagement
sufficiently broad to permit them to respond to changing circumstances with innovation
and dispatch. Greater budgetary autonomy and sufficient field staff are also necessary to
strengthen an effective field presence.
The UN failed to build on the lessons of the old world disorder and the need to identify the
perpetrators of violence in order to confront them with diplomatic isolation and sufficient military
force to act as a deterrent.
18. The UN should review the doctrine of neutrality to clarify its meaning and the range
of situations where it is appropriate and where its invocation is inappropriate.
1 The US Congress passed legislation in 1993 to reduce the US share of the peacekeeping budget to 25%, to be effective as
of 1 October 1995.
82
Endnotes
1. The terms of reference asked us to examine issues of early warning and conflict management.
Events that took place in Rwanda in 1994 are referred to as an “emergency”, as “the Rwanda
experience”, and as a “conflict”. Not once is the term “genocide” used. (Emergency-Prevention
and Conflict-Management Efforts, Terms of Reference for Study II, January 1994.) “Emergency”
and “experience” are euphemisms to refer to what happened. “Conflict” is partly misleading
because it implies that two sides are fighting one another, which covers the civil war in Rwanda
but not the genocide.
2. An alternative view of early warning is rooted in a _single standard of morality that requires a
decisive response when a certain threshold is crossed. The purpose of early warning is simply to
identify the crossing of that threshold.
3. The relevant clauses of the Genocide Convention are Art. II, (a) and (b): “...genocide means
any of the following acts committed with the intent to destroy, in whole or in part, a national,
ethnical, racial or religious group, as such: (a): Killing members of the group or (b) Causing
serious bodily or mental harm to members of the group.”
4. The African Charter on Human and People’s Rights (1986), Art. 12(2) states that, “Every
individual shall have the right to... return to his country. This right may only be subject to
restrictions, provided for by law for the protection of national security, law and order, public
health and morality”. The same right is affirmed in Art. 13 of the Universal Declaration of Human
Rights (1948), though not qualified as in the African Charter. (“Everyone has the right to leave
any country, including his own, and to return to his country.”). The relevant clause in the
International Covenant on Civil and Political Rights (1966) is similar: “No one shall be arbitrarily
deprived of the right to enter his own country”. (Art. 12(4))
5. The preamble to the 1969 Convention on the Refugee Problems in Africa affirms that the
signatories are “determined to discourage” refugees from using their status for subversive activities
(paras. 4 & 5). Art.III deals in its entirety with “Prohibition of Subversive Activities”, prohibiting
refugees from engaging in subversive activities against any member state of OAU (III.1), and
requires that the host states undertake to “prohibit refugees residing in their respective territories
from attacking any State Member... by use of arms, through the press, or by radio”. (III.2) To
further ensure that these conditions are met, Art. II (6) advises that, “for reasons of security”,
refugees shall settle “at a reasonable distance from the frontier of their country of origin”. These
provisions are unique to African regional instruments of international refugee law. More generally,
the Charter of the Organization of African Unity expresses “unreserved condemnation” for
subversive activities on the part of neighboring states or any other state. (Art.III(5) The African
Charter on Human and People’s Rights states unambiguously that, “territories of signatory states
shall not be used as bases for subversive or terrorist activities” against another party. (Art.23(2)b)
6. Recent studies discussing reform of the UN relating to peacekeeping and preventive diplomacy
include, The Canadian Committee for the Fiftieth Anniversary of the United Nations, Canadian
Priorities for United Nations Reform: Proposals for Policy Changes by the United Nations and the
Government of Canada, Ottawa: United Nations in Canada and the Canada Communications
Group, 1994; Gareth Williams, (the Foreign Minister of Australia in 1993), Cooperating for
Peace: The Global Agenda for the 1990s and Beyond, NSW, Australia: Allen and Unwin; Boutros
Boutros-Ghali, An Agenda for Peace, UN 1992 ; cf. also Foreign Affairs, September/October
1995.
7. In recognition of the importance of these problems in Africa, the African refugee convention –
unlike other instruments of international refugee law – deals at some length with the problem of
refugees exercising their “right of return” by means of armed force. (See note 5).
83
8. The Ugandan government’s position was that Rwanda first must acknowledge its fundamental
responsibility for its nationals by permitting return; Uganda would then grant full settlement rights
to those who nevertheless wanted to stay. This view was reiterated during discussions on the
refugee issue in 1991 and during the Arusha negotiations.
9. The general principle governing host state treatment of refugees under the 1951 Convention
(chapters II–IV) is to provide treatment to refugees as least as favorable as that accorded other
aliens, i.e. non-discrimination, and in some areas (e.g. primary education, labor conditions) that
of nationals. In matters of citizenship, the Convention encourages, but does not oblige, the host
state to “make every effort to expedite naturalization proceedings”(Art.34). The OAU Convention
incorporates these principles insofar as it is “the effective regional complement” to the 1951
Convention” (Art.VIII(2)).
10. The link between high-level positions in the Ugandan army (NRA) and senior positions in the
RPF reflected the important role the Rwandese refugees had played in Museveni’s struggle for
power. RPF Commander Fred Rwigyema had been the Chief of Staff of the NRA and briefly also
Deputy Minister of Defense in the National Revolutionary Council. Paul Kagame, who replaced
him, rose to the ranks of Deputy Chief of Military Intelligence in the NRA. Several other senior
RPF officers had served under Museveni.
11. While there are no systematic studies of the subject, observations to this effect were made by
social scientists in Kampala who were close to the refugee community (August 1995
communication). The point is also made by Watson (1991) and Braeckman (1994).
12. The mistake of excluding representatives from the refugee communities was later rectified. At
the Arusha talks, refugees from Burundi, Uganda, and Zaire (mainly “fifty-niners”) were present.
The refugee participants were close to the RPF, which helped to select them. UNHCR covered the
costs. In 1995 it proved more difficult to select refugee representatives to the inter-ministerial
committee on refugees, composed of the countries concerned, UNHCR and OAU.
13. According to a Kampala-based social scientist close to the exercise, the lead question in the
preliminary survey was: “Do you wish to return to Rwanda?” The typical answer was: “Who
would not want to return to Rwanda?” The survey was conducted in the refugee settlements by
two local employees of UNHCR. (Interview, Kampala, August 1995).
14. Museveni denied giving either military, logistic or financial support to the RPF, and charged
the RPA with stealing Ugandan army equipment. He also assured the Rwandese government that
the borders had been sealed, that Uganda did not provide RPF with weapons, nor would it allow
them to retreat back into Uganda, and would arrest them if they returned. Cf. Briefing given by the
Minister of State for Foreign and Regional Affairs, Hon. David Omara-Atubo, to Diplomatic
Representatives Resident in Kampala on 4 October 1990. See also Ambassador Katena-Apuli’s
letter to Stephen Goose of 26 August, 1993, in which he asserts that, “On the day of the invasion,
October 1990, the Uganda Government declared all Rwandese who had left NRA to attack
Rwanda as Deserters under the Operational Code of Conduct. That means on conviction by a
Court Martial, they would be punishable by death”. There is no evidence that any were arrested.
On the contrary, RPF Commander Paul Kagame travelled often and openly to Kampala where he
met with journalists, foreign supporters and diplomats throughout the war, but was never arrested
(Human Rights Watch Arms Project, Arming Rwanda: The Arms Trade and Human Rights
Abuses in the Rwandan War, Washington, January 1994, 20).
15. “The charitable view”, wrote the well-informed Africa Confidential, was that “Museveni has
been unable to control his army. The cynical view is that he has been playing a double game,
allowing... RPF-commander... Rwigyema to build up his expeditionary army while professing
friendship with his neighbors.” (Africa Confidential, vol.31, no.20, p.2.) Close observers
conclude there was significant assistance, pointing to the RPF’s use of Ugandan territory and
shifts of equipment from the Ugandan Army to the RPF at the time as well as later (Human Rights
84
Watch 1994, Prunier 1995). The Human Rights Watch report Arming Rwanda (1994) concluded
there was “institutional complicity” based on findings that “...Uganda provided weapons,
munitions and other military supplies to the RPF. These included munitions, automatic rifles,
mortars, artillery and Soviet-designed Katyusha multiple rocket systems... and that Uganda
allowed the rebel movement to use its territory as a sanctuary for the planning of attacks,
stockpiling of weapons, raising of funds and movement of troops” (p.10).
16. See Otunnu (1995) for a long list of telling signals, also Prunier (1995). Foreign diplomats
and journalists commented on the unusual movement of troops towards the Rwanda border;
departing NRA members took farewell with relatives and friends; and a few days before the
invasion, Maj. Gen. Rwigyema told inquisitive onlookers in Mbarara that he was taking troops to
prepare for celebrations on October 9, Uganda’s Independence Day.
17. There are also reports that the invasion triggered a purge of officers in the Rwandese armed
forces on 2–3 October. The officers appeared to have planned to carry out a coup in conjunction
with the RPF offensive, but were caught by Habyarimana’s intelligence network. If so,
Habyarimana probably had information about the planned invasion as well.
18. NMOG was headed by a Nigerian officer. The amended N’Sele Ceasefire Agreement of 12
July 1992 specified that military observers would come from Nigeria, Zimbabwe, Mali and
Senegal.
19. While not a major state in Francophone Africa, Rwanda had entered that inner circle called
pays du champ – states with which France had a special relationship. The ties had developed
steadily since President Habyarimana was received at the Elysée Palace in 1975. As with other
pays du champ, relations were managed not only or primarily through the Ministry of Foreign
Affairs, but also through the Presidential Palace (the Africa unit), and aid (including military
assistance) was channelled through the Ministry of Cooperation. During periods when the
relationship was non-controversial and involved matters of routine, Rwandese affairs were
handled largely through the Presidential palace in cooperation with the Military Mission Office in
the Ministry of Cooperation (Prunier 1995).
20. Reyntjens cites AFP and Reuter dispatches in February 1993 saying a spokesman for the
neutral military observers group (NMOG) accused French troops of bombarding RPF positions
(Reyntjens 1994, p. 176). American aid personnel have reported that French officers were
manning artillery positions (Interviews, Washington, May 1995). In a report in Le Monde
(“L’armée francaise dans le piège rwandais”, 22 September 1994), Herve Gattegno cites various
evidence of direct involvement, including a memorandum from the head of the military mission in
Kigali recommending 21 members of the parachutist regiments for medals for bravery during the
fighting in January 1991 (Reprinted in Guichaoua 1995:720-721). According to French
authorities, however, the mission of the paratroopers was to evacuate French and other Europeans
from Ruhengeri during the offensive, and thus within their strictly defensive role. The medals
were given in appreciation of their rescue mission. As for other reports of French soldiers
manning artillery positions etc., this could have referred to hands-on training sessions by DAMI
personnel, some of whom were in French uniforms. Interviews, Ministry of Cooperation, Office
of Military Cooperation, Paris, January 19, 1996.
21. Correspondence to the authors from the Ministry of Foreign Affairs, 14 November 1995.
22. The director at the time was M. Dijoud. Possibly the French were piqued that a high-ranking
American official, Deputy Assistant Secretary Hicks, had organized a meeting between the RPF
and the government of Rwanda in Harare in July. A subtle competition between Washington and
Paris developed, driven on the French side by a sense that diplomatic initiatives in Central Africa
properly belonged to a French sphere. In Washington, there was generally little interest in
Rwanda, which gave lower-level officials greater leeway for initiatives. The only exception was
the personal interest in promoting African conflict resolution and activist stance of Herman Cohen,
the Assistant Secretary of State for Africa.
85
23. The N’Sele ceasefire agreement of 29 March 1991, Art.II(7), as amended at Gbadolite, 16
September 1991 and at Arusha, 12 July 1992.
24. A widely publicized case implicated the French government, via the bank Crédit Lyonnais, in
helping the Rwandese government secure finances for a US$6 million weapons purchase from
Egypt. Documents in the case were originally obtained by Human Rights Watch (1994). The
shipment itself was no secret. American diplomats in Kigali, for instance, were appraised of the
shipment in advance via American embassy staff in Cairo. (Interviews, Washington March 1995).
25. Tanzania’s brokerage became a significant budget item for the Ministry of Foreign Affairs.
Tanzania’s delegation, headed by an ambassador, worked practically full time on the conflict for
10 months and the hotel bill for this period in Arusha was payable in hard currency.
26. The power-sharing formula in the Arusha Accords distributed the portfolios in the transitional
Council of Ministers as follows:
° The ruling MRND would retain Defense and Planning but otherwise received marginal
ministries, including those for family planning and promotion of women, and enseignément
supérieur et culture.
° RPF got the Ministry of Interior, which gave it control over local administration, the Ministry of
Youth, which meant power to organize and form the next generation (and thus possibilities for
establishing structures competing with MRND’s educational portfolio), and the Ministry of
Rehabilitation and Social Integration, which entailed control over significant resources.
° The principal opposition party, Mouvement démocratique républicain (MDR), got three choice
portfolios in its batch of four: Prime Minister, Minister of Foreign Affairs, and Minister of
Primary and Secondary Education
° The Liberal Party (PL), also in opposition, got the Ministries of Justice, Commerce and Labor –
all three “heavy” portfolios;
° The opposition Social Democrats (PSD): Finance, Public Works and Agriculture.
27. This is based on the figures in the report of the UN Reconnaissance Mission to Rwanda in
August 1993. If the higher figures sometimes cited of FAR having 30, 000 men and RPF 15, 000
are more accurate (see Study I), the difference in demobilization requirements becomes even
greater, i.e. about 24, 000 on the government side and 9, 000 for the RPF.
28. Current definitions of preventive diplomacy have been formulated by UN Secretary-General
Boutros Boutros-Ghali in his 1992 Agenda for Peace as efforts “to identify at the earliest possible
stage situations that could produce conflict, and to try through diplomacy to remove the sources of
danger before violence results” (para.15). During the Cold War, “preventive diplomacy” was
customarily used in the sense pioneered by former UN Secretary-General Dag Hammarskjöld to
mean the prevention of local conflict from developing into confrontations between superpowers.
29. Communication with the authors, 27 November1995.
30. The letter of 22 February 1993 suggested 20–30 observers, surely a symbolic force for a 150
km long border.
31. By its own account, NMOG was much too small to carry out its mission. The head of field
operations, the Nigerian Col. Ihekiré, told a UN good-will mission sent to Rwanda by the
Secretary-General in early March 1993 that he needed at least 400 men to do the work effectively.
32. Report of Meeting between H.E. Dr. Salim Ahmed Salim, Secretary-General and Mr. James
Jonah, UN Under Secretary-General for Political Affairs. OAU Headquarters, Addis Ababa, 25
May 1993, p.2.
33. The US, Belgium and the UK suggested to the UN Secretariat they were ready to support the
86
OAU on a bilateral basis. France was said not to be enthusiastic, making reference to the UN role
in Rwanda. The OAU Secretariat, for its part, was sceptical. Earlier promises of bilateral
assistance to NMOG and its predecessor MOT had either not been realized, or funds were tied to
conditionality that prevented their disbursal. Report of Meeting, Addis Ababa 25 May 1993,
op.cit.
34. Letter to Salim A. Salim dated 1 April 1994. Boutros-Ghali did not mention that, on the very
same day, he had effectively upstaged his OAU counterpart by ordering his top military advisor,
General Maurice Baril, to travel to Rwanda and Uganda to prepare for a UN monitoring force in
the border area. The Baril mission resulted in the establishment of the United Nations Observer
Mission Uganda-Rwanda (UNOMUR), as approved by the Security Council on 22 June (Res.
846(1993)). NMOG for a while limped along and monitored the cease-fire within Rwanda, and –
thanks to Salim’s continuous efforts – increased its capacity slightly. It was later folded into
UNAMIR, as was UNOMUR.
35. The other Rwandese operation mounted by the UN was UNOMUR, designed to monitor the
supplies going to the RPF across the Rwanda-Uganda border. Also _UNOMUR had a checkered
efficiency record. Established by a UN resolution in June 1993, UNOMUR had only 81 military
observers and lacked equipment for aerial surveillance of the 150 km long border between Rwanda
and Uganda. The Security Council evidently made cost a primary consideration when deciding that
UNOMUR would focus its control and verification activities on main roads and tracks (Res.
846(1993)). American military experts regarded the force as largely symbolic.
36. “The majority of the victims have been members of the minority group, the Tutsi, and they
have been killed and otherwise abused for the sole reason they are Tutsi.” But the report
continued, “While the casualty figures established by the Commission are significant, they may be
below the threshold required to establish genocide... These technical matters aside, the tragic
reality is that for the sole reason of belonging to the Tutsi group, many Rwandans are dead, have
disappeared, have been seriously injured or mutilated, have been deprived of their property, or
have had to flee their homes and been forced to hide or live in terror” (FIDH 1993:29).
37. “The question of Radio Mille Collines propaganda is a difficult one. There were so many
genuinely silly things being said on the station, so many obvious lies, that it was hard to take it
seriously. It was like relying on the National Enquirer a supermarket tabloid to determine your
policy in outer space. Nevertheless, everyone listened to it, I was told by Tutsis sic, in a spirit of
morbid fascination and because it had the best music selection.” (Canadian Ambassador Lucie
Edwards, communication to the authors, 27 October 1995).
38. The Canadian diplomat Lucie Edwards, who was accredited as Ambassador to Rwanda in
January 1994, recalls a meeting with human rights organizations exactly one year before, in which
the growing human rights violations in Kigali were discussed. “They agreed it was El Salvador
style death squads, supported by the youth wings of the MRND and CDR, tasked with killing the
regime’s political opponents and disrupting opposition meetings. They were unanimous that the
primary target was Hutu opposition figures, including human rights activists and journalists. No
one mentioned systematic attacks on Tutsi. No one predicted genocide.” (Communication to the
authors, 27 October 1995).
39. An official who served in the Africa bureau of the French Ministry of Foreign Affairs during
the tenure of Ambassador George Martre could not recall any mention of human rights issues in
the communications between Paris and Kigali. Considered a strong supporter of “the Presidential
tendency”, Martre served from September 1989 until March 1993. He was replaced by JeanMichel Marlaud, who was more attuned to multipartisme, but did not raise human rights issues
because he considered the evidence too flimsy or feared it would play into the hands of the RPF.
Reflecting tacit principles of division of labor during the Cold War, the United States had for some
time considered Rwanda as part of a French sphere of influence in Africa. To the extent that
Rwanda figured at all, US policy was to emphasize democratization and good governance as a
means to deal with human rights. Both Ambassador Robert Flaten and his successor, David
87
Rawson, observed this priority. (Interviews Kigali and Paris, August 1995; Washington, March
1995).
40. Canadian aid to Rwanda had reached a peak of C$22 million in 1990–91, declined to 18
million in 1991–92, and was cut by one-third to just over 11 million for 1992–93 and 1993– 94.
(CIDA, ODA, Country-to-Country Report, 1994, p.3). Further cuts did not occur because there
was a significant shift from development to humanitarian aid.
41. Under the guidance of two “old Rwanda hands”, in the Ministry of Cooperation and its
implementing agency (ADCD), Belgium had chosen Rwanda for a pilot project in Belgian foreign
aid based on new guidelines. Bureaucratic incentives to maintain the aid help to explain why
Minister of Cooperation André Geens strongly advocated not only renewing but actually
increasing aid in late 1991, despite his admission that the government had not undertaken the
expected economic and political reforms (William 1995, pp. 436-439). Another Belgian official
who accompanied Geens on his tour of Rwanda in August 1991 noted that he was “very keen on
aid”. (Interview, Brussels, July 1995). The French Ambassador in Kigali, George Martre, was
known as an ancien du Rwanda. Martre was not a diplomat but came from the Ministry of
Cooperation and was described by a close observer as a “projects man”. (Interview, Washington,
March 1995). Martre is considered to have wielded considerable if indirect
influence on French policy towards Rwanda during his tenure as ambassador (September
1989-March 1993).
42. In Canada the issue came to the forefront in early 1992. Attempts by human rights activists to
portray a cut-off in Canadian aid to Rwanda as a reaction to human rights violations were in fact
resented by Ottawa officials who feared this would complicate Canadian efforts to promote peace
and democratization.
43. Memorandum from DHA/Geneva to DHA/New York, 19 March 1993.
44. Agnes Ntamabyaliro. Considered both corrupt and of “Hutu power” persuasion, she appeared
in the self-appointed “interim government” after 6 April.
45. Even such a concrete and seemingly clear-cut issue as hate radio was too problematic for the
international community to deal with. The Rwandese government had formally committed itself to
abstain from “propagande radio nuisible”, as the confidential annex to the Dar-es-Salaam joint
communique put it (7 March 1993). Although the vitriolic Radio Mille Collines (RTLM) clearly
qualified as nuisible and attacked even members of the diplomatic corps in Kigali, there was no
decision to take forceful measures to silence it. Both the French and the American Ambassadors
opposed such action. Ambassador Rawson claimed that Radio Mille Collines was the best radio
for information, and its euphemisms were subject to many interpretations. The idea of jamming the
radio was discussed in a preliminary fashion in the US Department of Defense, but never passed
“first base”, as one participant later noted. The reasons cited were many: It was illegal to interfere
in the internal affairs of another state, Rwanda was not considered an enemy or even an adversary
state, and jamming posed practical problems by requiring the application of sophisticated
equipment from the air or a nearby base. The United States was in principle wedded to a broad
view of freedom of speech. (Interviews Washington, March 1995; Kigali, August 1995).
46. Under the 1503 procedure, discussion is conducted by a small Committee of Five, and is a
means to approach the government in question to deal with the situation before it is brought onto a
public agenda. The confidential discussions are seen as one step in a process that – in the most
severe cases – might lead to the establishment of a Special Rapporteur with a country mandate.
Only on 24–25 May 1994 did the Commission call a special session and appoint a Rapporteur to
investigate the genocide in Rwanda.
47. Mrs. Uwilingiyamana appeared at the 8 March meeting at the specific instructions of the
Foreign Minister, Boniface Ngulinzira – who also was murdered after 6 April. Her presence
before the Commission visibly angered the MRND member who remained in the delegation.
88
48. In 1992, the Special Rapporteur on “extrajudicial, summary or arbitrary executions” devoted 6
paragraphs to Rwanda in his report to the Commission (E/CN.4/30: 461-467). The thematic
rapporteurs on torture and the Working Group on Enforced and Involuntary Disappearances
included Rwanda in their general reports to the Commission in 1993.
49. The report estimated that at least 2, 000 civilians have been “victims of extrajudicial, summary
or arbitrary executions” in the period October 1990 to January 1993. Nor, for those who knew
Rwandese history, would these figures suggest a turn for the worse: several thousand persons
were victims of the 1959 “social revolution”, and thousands were killed in retaliation for the
attacks in December 1963 by Tutsi guerrillas. At that time, as in 1992–93, the killings took place
in the context of a war, and the victims were primarily local Tutsi who were identified as the
internal enemy associated with the externally-based Tutsi rebels.
50. Diplomats from the region report having seen French officers in the Gabiro national park with
interahamwe units. The latter were easily recognizable from their uniform (kanga). It was one of
the “open secrets” of Kigali that the militias were training in the national parks (Interviews,
Geneva, March 1995; Kigali and Dar-es-Salaam, August 1995).
51. On 10 September – the acknowledged unrealistic date for deployment stipulated in the Arusha
Accords – the Security Council had not even approved the force. According to the recommended
timeline of the UN Reconnaissance Mission, the military and political leaders of the mission were
to be in place one week after the Security Council had given its approval (M+7). In fact, the Force
Commander, General Dallaire, arrived on 22 October (M+17), while the Special Representative of
the Secretary-General and head of the entire mission, J.-R. Booh-Booh, arrived on 23 November
(M+38). Half of the mission staff were supposed to be in place in Kigali on M+30; in fact, it took
over two months to assemble 40% of the staff. Report of the UN Reconnaissance Mission to
Rwanda, New York, September 1993.
52. The French Ambassador Marlaud warned most strongly of all that delays would give the
opponents of the Accords opportunities to derail the peace process.
53. The review had started out in mid-1993 on the premise that UN peacekeeping was a valuable
instrument of US foreign policy, but soon took a critical course. Announced in May 1994, the
Presidential Decision Directive 25 (PDD 25) held that the United States needed to apply stringent
criteria of national interest before supporting UN peacekeeping operations, whether or not this
involved US troops (National Security Council, The Clinton Administration’s Policy on
Reforming Multilateral Peace Operations, Washington: May 1994). The increased activities of UN
peacekeeping after the end of the Cold War had dramatically increased the assessed US
contributions, from a range of US$29–47 million in 1985–89, to a sizable US$460.4 million in
1993, the latter figure representing an increase of 370% from the previous year. Under a new
procedure laid down by PDD 25, the costs of all Chapter VII operations plus those in which US
troops were involved would be charged against the Pentagon’s budget, which inclined the latter to
oppose such operations (Rosner 1995:65-91).
54. At the very high end was the UNTAC operation in Cambodia, which had an international staff
of 22,000, including 16,000 blue helmets, whose cost was estimated to US$60–70 million per
month (Schear 1995, UN/DPI 1994).
55. The Arusha Accords were signed on 4 August. The UN Reconnaissance Mission
(DPKO/DPA/DHA) travelled in Rwanda during the last two weeks of August. Upon its return, it
prepared its report, had it translated, and submitted it to the Security Council on 24 September.
Authorization for deployment was received on 5 October (S/Res.872). General Dallaire arrived in
Kigali with an advance party on 22 October.
56. No country came up with a unit of Armored Personnel Carriers (APCs), as UNAMIR plans
approved by the Security Council called for. As a result, the DPKO had to search for left-overs
89
from other UN operations (and did scrounge eight APCs from Mozambique); the rest had to be
obtained through civilian contracts. Since the commercial market for APCs was limited, and
UNAMIR’s budget at any rate was not fully approved until April 4, 1994, the force never got its
armored unit. The APCs assembled from the UN operation in Mozambique rapidly broke down.
As UNAMIR had no repair facilities, only one was operational by the time of the April crisis. A
similar fate befell the small helicopter unit authorized by the Security Council. No country offered
even half a helicopter squadron; a commercial contractor who eventually was found withdrew the
machines in April when violence broke out. (Interviews, UN/DPKO, UN Permanent Missions,
April, June 1995).
57. The mission had a temporary budget until the end of 1993. The first formal budget proposal
was released by the Secretariat (DAM) on 3 January, 1994 (A/48/837), translated into 16
languages, passed to the Fifth Committee sub-committee dealing with budgets (Advisory
Committee on Administrative and Budgetary Questions, ACABQ), which began consideration of
the budget and released it on 17 March (A/48/908). On 2 April, it was formally approved by the
General Assembly (A/48/828 Add.1) Member states received requests for contributions on 18
April 1994. The process has since been speeded up. Previously the Secretary-General could spend
no more than US$10 million per year without an approved budget from the General Assembly,
and then only with approval of the ACABQ. Now, the Secretariat can go straight to the ACABQ
for a temporary budget of US$50 million once an operation is set up. Formal budgets are prepared
for fiscal years rather than mandated periods (annual budgets for stable periods and 6-months
budgets for less stable periods), which harmonize better with regular national and UN budget
processes. (See A/49/375 and A/48/945, also Durch 1993, 1995).
58. Cable from Dallaire, UNAMIR\Kigali to Baril\DPKO\Nations, New York, regarding "Request
for Protection for Informant", 11 January 1994. To draw attention to the cable a DPKO official put
it in a file with black colour before circulating it.
59. Intelligence cables, UNAMIR to Brussels, doc. no.6-16 (1994), Alexandre Goffin, Rwanda.
7 avril: dix commandos vont mourir. Brussels, ASBL, 1995, De Morgen (Brussels), 6 November
1995.
60. In addition to continuous requests for equipment to bring the force up to authorized strength,
General Dallaire twice requested reinforcements of a company of 150 men and 5 Canadian
bilingual officers, first in late February after the civil violence in Kigali, and again on 13 March.
61. A DPKO official in New York noted on 22 January 1994 the arrival of a French DC-8 with 90
boxes of 60 mm mortar.
62. The Belgian Col. Luc Marchal told the BBC, and later confirmed to Le Monde (23 August
1995), that one of the French planes supposedly participating in the evacuation operation arrived at
0345 hours on 9 April with several boxes of ammunition. The boxes, about 5 tons, were unloaded
and transported by FAR vehicles to the Kanombe camp where the Rwandese Presidential Guard
was quartered. The French government denied this, saying that the planes carried only French
military personnel and material for the evacuation (20 August 1995, reported in Le Monde, 22
August 1995).
63. Delcroix apparently was alarmed when finding that Rwanda was "awash with weapons" while
UNAMIR stood powerless by. The Belgian UN commander of the Kigali sector, Col. Luc
Marchal, told Reuters news agency that since the Belgian battalion's arrival three months earlier, it
had confiscated only 16 weapons and just over 100 hand grenades. Marchal was unequivocal: "It
is a problem with the mandate. Stopping and searching people for weapons is forbidden by New
York," he said. (Reuters dispatch from Kigali, 13 March 1994).
64. The file was seen in DPA, DHA and DPKO. Senior DPKO officials discussed the cable
several times, but questioned its significance and the credibility of the informant. DPKO sources
later claimed that information about the January cable was shared with key members of the
90
Security Council. The main response of the Secretariat, however, was to rely on the diplomatic
suasion of Western embassies in Kigali. Dallaire was instructed to share the information with the
French, Belgian and US embassies in Kigali and request that they take it up with Habyarimana in
yet another diplomatic representation. Le VIF/L'Express (Brussels), 11 January 1996. There the
matter rested. Although the significance of the cable became clear in retrospect, senior DPKO
officials later denied they had received any information suggesting a genocide was being planned.
A special search of DPKO archives was undertaken but failed to unearth any evidence, officials
said. As for Dallaire's request for reinforcements and a proactive mandate interpretation, the
DPKO officials said such requests were not forwarded to the Security Council because the Council
was expected to turn them down anyway . (Interviews, New York, April and June 1995; Kigali,
August 1995).
65. Three documents deal with the mandate for UNAMIR (which in documents preceding the
formal proposal to the Security Council is called a Neutral International Force (NIF)). These are
the Arusha Accords (hereafter AA), the report of the UN Reconnaissance Mission, headed by the
future Force Commander, and which visited the region from 19 August until 3 September 1993
(hereafter "Report"), and the Security Council resolution 846 of 5 October, 1993 (hereafter
UNSC). A comparison of the three shows significant differences on key issues:
AA: "guarantee overall security of the country"((B)1), in the Report becomes "establish
security zone in and around the capital city area of Kigali", and in UNSC: "contribute to the
security of the city of Kigali inter alia within a weapons-secure area established by the parties in
and around the city" ((3)a), i.e. a progressively weaker mandate.
AA: "assist in catering for the security of civilians" ((B)3); in the Report becomes "To
monitor the civilian security situation through the verification and control of the Gendarmerie and
the Communal Police". This is subsequently specified as monitoring with unarmed UN Police
Observers; and in UNSC "to investigate and report on incidents regarding the activities of the
gendarmerie and the police" ((3)h). Here the mandate becomes more delimited and specific, and to
that extent weaker.
The Accords have two strong provisions for confiscating illegal arms: (B)4 - "Assist in the
tracking of arms caches and neutralization of armed gangs throughout the country"; and (B)6 "Assist in the recovery of all weapons distributed to, or illegally acquired by the civilians". The
Report lists such activities as means of achieving the principal goals identified for NIF,
notably:"Assist in tracking arms and neutralizing armed groups (with armed UN Military Forces)"
and "Assist in recovering arms in the hands of civilians" with armed UN Military Forces and
unarmed UN Police Observers. Significantly, the Security Council resolution has no provisions at
all for confiscating illegal arms.
All three sources are similar for the other main provisions: monitor cease-fire, mine clearance,
monitor demobilization (cantonment points), integration, and protection for security for
humanitarian relief.
66. In Somalia, Boutros-Ghali had taken the lead in proposing a proactive and eventually Chapter
VII mandate which gave the UN broad powers to confiscate weapons and enforce peace. The first
debacle in Somalia occurred in early June 1993, when a UN Pakistani contingent of 23 men was
lost, the death of 17 US soldiers in a related mission two weeks later jeopardized the entire
operation and was widely considered as a major defeat for UN peacekeeping. In his
recommendation to the Security Council on 3 March for an expanded mandate for UNOSOM,
Boutros Boutros-Ghali had argued that the disarming of local factions had to be enforceable to be
effective. He was backed by Res. 814, which stipulated that factions or personnel who failed to
comply with the disarmament process would have their weapons and equipment confiscated. The
Council also supported him on this point after the Pakistani contingent was ambushed (Res. 837 of
6 June 1993). The main difference between Res. 814 authorizing UNOSOM II and the mandate of
UNAMIR was a Chapter VII provision to cover the "responsibility" of the UN force:"the
consolidation, expansion, and maintenance of a secure environment throughout Somalia on an
expedited basis". By contrast, UNAMIR was to "assist" local authorities, and only with respect to
the security of the capital by establishing a weapons-secure area in Kigali (Makinda 19954:76-81).
91
67. This is the assessment of the Rwanda expert, Andr‚ Guichaoua, who met with a number of
opposition figures on 6 April. The presence of "white" soldiers - professional, well-equipped, and
with a UN badge - created a sense of security. The Prime Minister, who was assigned UN guards
on the night of 6 April, had dared to be particularly outspoken. (Guichaoua, communication to the
authors, 22 November 1995).
68. Trying to find out what was happening, Dallaire rushed into a meeting of the top officers on
the evening of 6 April. Col. Bagosora was in charge. When Dallaire asked him to help maintain
stability by recognizing the Prime Minister as the acting head of state, Bagosora put down the idea,
contending she was inept and untrustworthy. By the next morning she had been killed. Dallaire
also learned of the murders of 6 other moderate cabinet ministers. On the evening of 7 April, he
went to the hospital to identify the bodies of the 10 Belgian soldiers.
69. Initially, the focus was on the safety of UN personnel. In a letter to the Security Council dated
8 April, the Secretary-General raised the possibility that UNAMIR might have to take main
responsibility for evacuating UN civilian personnel, in which case an expanded mandate and
strength would be required. A further deterioration of the situation could justify an evacuation of
UNAMIR itself, and if so he was ready to make that decision. While the Secretariat and the
Council were discussing these options, France, Belgium and Italy launched evacuation operations
of their own, the French giving New York and the UNAMIR Force Commander 45 minutes notice
before landing at Kigali airport on 9 April.
70. Smaller states on the Council complained in retrospect that the Secretariat provided insufficient
information. (Interview, New York, March 1995). Also, other members complained that the
Secretariat was inactive. The delegation of Ghana, which had contributed one of UNAMIR's two
battalions, found that its calls to DPKO were not even returned. The informal consultations in the
Council reinforce the impression of a Council ready to give the Secretariat room for initiatives, but
the latter did not respond. For instance, on 12 April - or five days into the crisis - several members
of the Security Council explicitly asked for the views of Force Commander and the SecretaryGeneral by the next morning; they also sought to hear the views of major troop contributors to
UNAMIR. DPKO officials briefed the Council next day, but members were still impatient and
asked for specific options. When the Secretariat came back on 14 April with options, the UK and
others complained about the lack of information in order to assess options. Argentina reiterated
earlier demands that troop-contributing countries be consulted.
71. The Secretariat came back on 14 April with two options. Both were premised on a cease-fire,
an assumption that the Security Council found unrealistic. The logic apparently was to threaten
withdrawal of the entire force unless the RPF and FAR agreed to a cease-fire. That threat would
hardly persuade any of the parties to stop either the war or the killings.
72. The Secretary-General's report of 20 April uses the term "widespread violence", not
"genocide", but admits that "possibly...tens of thousands" had been killed (S/1994/470). The
Security Council resolution (Res. 912(1994)) is mainly oriented towards the civil war, which is
addressed in conventional inter-state terms. For instance, Clause 4 claims: "Obviously, a cease-fire
agreement is the first step in establishing a stable and secure environment in the country, thus
allowing the organized, coordinated and secure delivery of humanitarian assistance and the
reactivation of the Arusha peace process." Item 6 "demands an immediate cessation of hostilities
between the forces of the Government of Rwanda and the Rwandese Patriotic Front and for an end
to the mindless violence and carnage which are engulfing Rwanda; Item 7 "commands the active
role of the Special Representative of the Secretary-Genetal and of the Force Commander to bring
about a cease-fire and to mediate between the parties in order to bring about the earliest resolution
of the Rwandese crisis." Item 8 adjusts the mandate of UNAMIR to serve as an intermediary to
bring about a cease-fire.
73. Cable from Booh-Booh, UNAMIR, to Annan/Goulding, UNATIONS, New York, 8 April
1994.
92
74. Agence France Presse. "Un scandal dont tout le monde est responsable," Le Monde, 27 May
1994.
75. Commenting on the work-pace when the crisis broke, one official in the Secretariat wrote:
"After April 6, UNAMIR and the DPKO desk officer were flooded with inquiries from
governments, NGO's, etc. for information about their personnel in Rwanda and for possible
arrangements of search-and-rescue missions which, though... important, reduced their
effectiveness as the inquiries were time-consuming. The botton line here is that there was no
organizational rearrangement in the Secretariat to deal with the crisis. Those involved just worked
harder and longer hours." Communication to the authors, 30 October 1995.
76. On 7 April, when the crisis broke, Boutros Boutros-Ghali was in Minsk. The next day he flew
to Geneva; when the Belgians decided to pull out (12 April) he was in Bonn; when the Security
Council debated what that meant for UNAMIR, he was in Madrid. With this travel schedule, it
was impossible to stay on top of developments in a week of complex and fast-moving events. The
essential decision-making during the first critical week fell to two high-level officials in the
Secretariat, Kofi Annan and Iqbal Riza. Lacking decisive leadership from the Secretary-General,
the DPKO seemed caught in a political culture that made it inappropriate to confront the Security
Council and futile to present proposals it would not readily approve.
77. Letter from the Secretary-General to the President of the Security Council, 13 April 1994. The
letter was not well received by the Council, where some members saw it as an attempt by BoutrosGhali to place the responsibility for UNAMIR withdrawal squarely on the shoulders of Belgium.
In a letter to the Security Council the same day, the Belgians had publicly pointed to "the chaos in
Rwanda." The Belgian government concluded that it was "obvious that under these conditions the
continuation of the UNAMIR operation has become pointless...and it is imperative to suspend the
activities of UNAMIR forces without delay" (S/1994/430). DPKO reinforced this interpretation by
telling the Security Council (13 April) that with the Belgians about to leave, the Force Commander
could no longer guarantee the security of the airport or that of even his own men. DPKO did not
relay to the Council the more differentiated appreciation of the situation made by the Force
Commander in Kigali.
78. Invoking NATO solidarity as well, the Belgian Foreign Minister phoned his counterparts in
Washington, London and Paris to plead for withdrawal. "Willy Claes panicked a bit", a Belgian
official later commented. Outcries from the public and the press over the brutal murders of the
Belgian soldiers had placed enormous pressure on Belgian political leaders to withdraw, similar to
that faced by the Americans in Somalia half a year earlier.
79. The United States had been traumatized by the Somalia crisis, which had been imprinted on the
minds of American TV-viewers by the picture of a mob dragging the body of an American soldier
through the streets of Mogadishu. Although no American soldiers were in UNAMIR, the
experience had given peacekeeping a bad name in the United States. Washington apparently
considered total withdrawal as soon as the Belgian peacekeepers were killed. When the Belgian
decision was known, the US position in the Security Council progressively favored withdrawal.
In the informal consultations on 12 April, the American Ambassador expressed serious doubts
about the viability of UNAMIR in the circumstances, noting that it could not carry out its original
mandate and might even be a de-stabilizing factor. The next day, he suggested withdrawing,
leaving only a skeletal force. On the 14th, the US claimed the Security Council needed a resolution
for orderly evacuation of the force; on the 15th there was no doubt: the US delegation expressed
firm opposition to keeping UNAMIR in place.
80. Reports from the informal consultations in the Security Council make no mention of such
canvassing. It should be noted, however, that even Nigeria, which circulated a draft resolution to
strengthen UNAMIR, did not itself volunteer troops. Although a major military power in Africa,
and a contributor to other peacekeeping operations, Nigeria claimed it lacked logistics to send
reinforcements. However, Nigeria had military observers in UNAMIR and made no move to
withdraw them.
93
81. While recognizing that the Belgian departure was leaving the force quite vulnerable, General
Dallaire and his deputy, the Ghanaian Brigadier-General Henry Kwami Anyidoho, were making
dispositions to dig in, did not recommend withdrawal, and asked for emergency supplies to carry
on. The Ghanaian battalion, considered the most professional and second in strength to the Belgian
unit, was redeployed from the DMZ to Kigali to protect the airport. On 13 April, one day after
Dallaire was informed by the Secretary-General's office that the Belgians were pulling out and that
the Secretary-General was inclined to withdraw the entire force, UNAMIR sent DPKO a long list
of requests for emergency supplies, including medicine, water, fuel and heavy machinery, plus
cargo trucks and 1, 900 flak jackets. On 14 April, Dallaire noted in his log the ironic differences in
perception: the Belgians were telling New York that the UNAMIR staff in the field did not realize
how the situation was deteriorating by the hour; "they say we are too optimistic here."
82. Setting up an airlift capacity in Kigali, Canada was the only country to send armed forces
personnel to help UNAMIR in this period.
83. Whether a larger - or only a much larger - force would have made a difference is a matter of
considerable controversy and remains speculative. If the symbolic force of 540 that stayed on
helped to protect thousands of civilians, the force of almost 2,000 which pulled out after the
Belgians withdrew would arguably have had a somewhat greater impact. The force, however, was
targeted intermittently by both side (including the militia) as well as taking indirect fire.
84. For example, outside Kigali the ICRC provided protection at the Sainte Famille and Saint
Michel church precincts, the church complex Kabgayi, the stadium at Cyangugu, and the camp at
Nyarushishi.
85. See Goffin, 1995, op.cit., and a series of articles in De Morgen (Brussels), November 1995.
86. The RPF maintains that the Security Council was following French interests by seeking to
stop RPF advances and save the remnants of FAR behind a cease-fire line. (See RPF
communiqu‚s 10-16 April, 1994).
87. Both ICRC delegates and UNAMIR staff have commented on the apparent dynamic of
killings, whereby retreating government soldiers and Hutu civilians who were fleeing the
advancing RPF armies pounced on Tutsi behind the lines with extra vengeance. Close UN military
observers claim that a cease-fire would have freed up government troops to rein in the militia. This
line of reasoning is strongly refuted by the RPF and sympathetic NGOs (African Rights, 1995:
1068).
88. There were, of course, articles in the specialist periodicals of human rights organizations and
those specializing on Africa. The only US television story on Rwanda during the period was on
the mountain gorillas. The New York Times did cover the UN Security Council decision to renew
UNAMIR and bring it up to full strength. Two and a half weeks later (26 January), in the same
paper, a very short story appeared on the Human Rights Watch Arms Report on the French role in
arming and training the Habyarimana forces. It was a month later (23 February) when Reuters
reported on the assassination of Public Works Minister F‚licien Gatabazi and the subsequent
murder of the head of the hardline CDR Hutu party. This report implied that Gatabazi was part of
the Hutu-Tutsi "tribal" killings, ignoring the fact that Gatabazi was Hutu. Le Monde did carry the
Agence France Presse report on the delay in setting up the BBTG on 27 March and the potential
this had for renewal of the civil war. For the reader of The New York Times, only a brief report on
13 March that the US Department of State had issued a travel alert for Rwanda provided a clue to
the looming crisis.
89. Livingston and Eachus (1995).
90. Le Monde, 12 April 1994, "Le Rwanda … feu et … sang."
94
91. The Times, 8 April and The New York Times, 9 April. Four days later, however, The New
York Times had a dispatch by William Schmidt quoting Colonel Alexis Kanyarengwe, the Hutu
chairman of the RPF: "This is not an ethnic war. It is a war against a dictatorship." The consensus
by scholars is now that the evidence points to the Presidential Guard itself as the source of the
missile that shot the plane down. The Presidential Guard also killed the Premier. The Belgian UN
soldiers were killed by the members of the FAR reconnaissance battalion after being told that the
Belgians had shot down the President's plane.
92. The New York Times, 10 April 1994 and The Times, 11 April 1994.
93. Lib‚ration 18 May, "Rwanda: les amiti‚s coupables de la France"; L'Humanit‚, 20 May, "Les
responsabilit‚s fran‡aises dans le drame rwandais". See also Prunier, Report prepared for Study II
(1995) and Verschave (1994).
94. Associated Press, 20 April 1994; Michael Binyon in
The Times, 2 May.
95. The U.S. Committee for Refugees used the term "genocide" in a public appeal on 3 May
1994, the (French) M‚decins sans fronti…res did so on 15 May.
96. After travelling to Kigali in mid-May, Ayala recommended that the Commission appoint a
Special Rapporteur for Rwanda, to be fully supported by a field operation (E/CN.4/S-3/3, 19 May
1994).
97. In Johannesburg for Nelson Mandela's inauguration, Salim A. Salim met on 9 May with
several African heads of state to discuss troop and other contributions to Rwanda. All (Mugabe,
Rawlings, Abacha, Zenawi, Chiluba, Mandela, Nujoma) expressed their support. The next day
Salim met with Boutros-Ghali and told him he had firm offers: "The issue of troops is settled."
(Interview, Copenhagen, 15 November 1995).
98. In a letter of 30 April (S/PRST/1994/21) and a resolution of 17 May (918(1994)), the Security
Council "recalls that the killing of members of an ethnic group with the intention of destroying
such a group in whole or in part constitutes a crime punishable under international law."
99. During the second half of May, the Secretary-General sent a high-level mission led by his chief
military advisor, General Maurice Baril, to Rwanda to prepare an assessment
100. The US delegation to the UN was instructed to use the term "acts of genocide" so as to
downplay the relevance of the 1948 Genocide Convention and associated obligations. The term
was adopted by the Security Council in its implementing resolution for UNAMIR ("acts of
genocide have occurred", Res.925/1994, 8 June 1994). Initially, Council resolutions and
statements avoided the question of whether or not genocide had occurred in Rwanda (see e.g.
S/PRST/1994/21, 30 April and Res. 918/1994, 17 May) by using the conditional form. The
Secretary-General declared on 31 May that there had been genocide in Rwanda (S/1994/640).The
Special Rapporteur for Rwanda appointed by the UN Commission on Human Rights likewise did
so in his report of 28 June (E/CN.4/1995).
101. Communication, British Foreign Office, United Nations Department, 7 July 1995.
102. M‚decins sans fronti…res claims Jupp‚ made his statement because MSF the previous day
had appeared on French television, calling the killings genocide and critizing the French
government. (Interview, Washington, March 1995).
103. For UNAMIR II, the Belgian government outfitted one Malawi company consisting of 135
men. With training, paperwork and transfer of equipment, the process took one month, which the
Belgians prided themselves on as beeing efficient.
104. Only one Congressman was recorded in May as arguing that it was in the enlightened self-
95
interest of the US to stop genocide wherever it occurred (Hearings, 5 May). Even a significant
refugee outflow (200,000) in late May did not lead to the conclusion that by reducing violence
within Rwanda, refugee outflows would probably also be reduced, hence touching more directly
on US interests. The point was demonstrated only after Goma (mid-July), when the United States
responded with a major and costly humanitarian relief operation.
105. United Nations Peace-keeping, UN/DPI, December 1994.
106. This was the proposed American strategy of "outside-in", designed to reduce the chance of
military encounter with either of the combatants. The original proposal of the DPKO, by contrast,
was an "inside-out" move whereby UNAMIR II units would fly in to Kigali and fan out from the
capital. In the end, a combination of the two strategies were adopted. S/1994/565, 13 May, 1994
and The New York Times, 14 May 1994.
107. According to the US Ambassador to the UN, Madeleine Albright, the RPF called her on 4
May and said they did not want UN forces in Rwanda (Statement to the Congress, House
Appropriations Committee, 5 May 1994). On the eve of the 17 May resolution authorizing
UNAMIR II, both Rwandese parties said they would accept a new UN force if it was only for
humanitarian purposes, but RPF warned that it would attack UN forces if they came between its
own units and the government troops (NET 18 May 1994). Towards the end of May, when the
Security Council went an extra round to give a second authorization to UNAMIR II, the RPF
reiterated that it would cooperate with UNAMIR II if it had exclusively humanitarian functions, but
was adamant they would have nothing to do with the UN mission if the latter talked to the "interim
government" (NET 25 May, 1994.)
108. On 25 July UNAMIR's troop strength was still around 500 - i.e. unchanged since the
scaling down decided on 21 April.
109. On 30 April, the Security Council requested that Boutros-Ghali consult with OAU for troop
contributions to UNAMIR II. The same day, Boutros-Ghali discussed the matter with the OAU
Chairman, President Mubarak, and approached his OAU counterpart, Salim A. Salim, as well as
leaders of African states who previously had contributed troops to UN peacekeeping operations.
(S/1994/530, 3 May 1994).
110. In the words of General Lafourcade, the Force Commander, the operation "‚tait une
intervention militare … but humanitaire et non une simple mission d'aide humanitaire. La mise en
ouevre de moyens militares avait pour but d'arrˆter les massacres et de cr‚er les conditions de
s‚curit‚ permettant l'exercise de l'action humanitaire internationale." "L'Op‚ration Turquoise au
Rwanda", speech, Geneva, 29 February 1995 (Document/DAM, Paris).
111. By this time, the Secretary-General had firm troop offers from 8 African countries, totalling 5
battalions and 4 companies, and a modest equipment list from 4 Western states, totalling 2 aircraft,
50 trucks and 50 APCs (S/1994/728, 20 June 1994).
112. A small contingent of 500 African troops, mainly from Senegal, was added to the French
core unit (2,500 men) largely to ward off criticism that this was a French rather than a
multinational intervention.
113. The human rights organization, African Rights, estimated that 85-90% of the Tutsi in the
pr‚fecture of Cyangugu had been murdered by the time the French arrived (African Rights, 1995:
455).
114. According to Prunier, who was a participant-observer in the process, confusion rather than a
clear strategy marked the preparation of the campaign. The Ministry of Defense at first considered
moving into the north-west, Ruhengeri and Gisenyi, but was informed that this would appear as
collusion with the government forces since the area was the stronghold of the old regime. As a
96
result, the planning shifted to focus on the south-west, also because it was believed that surviving
Tutsi might be trapped in this area and thus enable Op‚ration Turquoise to fulfill its humanitarian
objectives (Prunier 1995).
115. "Op‚ration Turquoise. Rwanda. June 22-August 11, 1994". Official briefing note. Mimeo.
116. Col. Didier Thibaut told Reuters on 5 July that the head of the southern command, Col.
Jacques Rosier, had given him orders to stop the rebels if they tried capturing Gikongoro or going
beyond. A statement from the French Ministry of Foreign Affairs on 5 July affirmed that French
units would stop RPF rebels from entering the zone in order to protect the refugees (Financial
Times, 6 July, 1994).
117. Force Commander General Lafourcade emphasized that his men were not wearing helmets or
flak jackets - in contrast to UNAMIR (Connaughton 1995). In part, this reflected a relatively more
friendly environment.
118. "Op‚ration Turquoise. Rwanda. June 22-August 11, 1994". Official briefing note. Mimeo
119. UNHCR SitReps for August 1994. Some of these soldiers could have retreated before the
French entered on 23 June, but this is unlikely since the war was then being fought in the east
(causing an exodus into Tanzania by late May). The next major refugee flows occurred in midJuly, both in the north into Goma and the south-west into Bukavu. On 19 July, UNHCR reported
that about 100, 000 had arrived (UNHCR SitRep, 20 July 1994). By that time, the French zone
had been in existence for over two weeks.
120. Col. Musonera, Bukavu, in interview with human rights investigators. Human Rights Watch
Arms Project, Rwanda/Zaire. Rearming with Impunity. May 1995, vol.7. no.4. p.3.
121. A UN official who coordinated relief in Ngara described a typical tactic as follows: commune
- level officials would provoke unrest or riots in the camps; UN and NGO personnel would be
faced with an unmanageable situation; the same persons would approach them and offer to help
control the population. Having no ready alternative, the agencies would agree and hence treat them
as camp leaders for purposes of distribution and maintaining order.
122. In November 1994, UN/DPKO had proposed a variation on the ZCSC in the form of an
International Police/Military Observer Group, where Zairian troops would serve under an
international military/police supervisory unit. The Zairian government, however, refused to place
its troops under the authority of officers from another state. Zaire did not object to the DPKO
proposals to send a peacekeeping force to the camps on its territory.
123. Simultaneously, UNHCR launched an appeal to donors asking for pledges to finance the
operation, and for personnel to form the international civilian contingent, the Camps Security
Liaison Group (CLSG). The Dutch government responded quickly with an offer of 16 policemen;
other European and four West African countries followed suit to provide a total of 45. UNHCR
assigned a senior official and five support staff to the task, among other things, of training the new
camp police in refugee law and doctrine. The operation was first planned for five months, but was
later extended to the end of 1995.
124. By the first week of June, small arms had been seized, including 23 rifles or pistols, 30
grenades, four mines and a variety of ammunition.
125. Detailed evidence is provided by Human Rights Watch Arms Project, Rwanda/Zaire,
vol.7.no.2, New York: May 1995, and Amnesty International, Rwanda. Arming the perpetrators
of the genocide. London: 13 June 1995.
126. Interviews, Paris, January 1996. In a brief statement commenting on the allegations of
French violations of the UN arms embargo made by the Human Rights Watch Arms Reports of
97
May 1995, a spokesman of the Ministry of Foreign Affairs said: "We categorically deny the
allegations on this subject contained in the report." Porte-parole adjoint du Minist…re des Affaires
Etrang…res, 30 may 1995.
127. Report on visit to the military cooperation mission in Paris, 9-13 May 1994 from Ll. Col.
Ephrem Rwabalinda, to the minister of Defense. Gitarama, 16 May 1994. The document was
obtained by the Belgian journalist Colette Braeckman and is reprinted in Dossier noirs de la
politique africaine de la France, no. 1.5, Paris: Survie-L'Harmattan, 1996, pp. 23-26.
128. This interpretation of what occurred at Kibeho is derived from interviews with government
and UN officials, and documents prepared prior to the incident, such as: the "Integrated Task
Force, Briefing Note, 6 March 1995," the "Strategy for the Southern IDP Camps, 5 April 1995,"
"The Integrated Humanitarian Response, Concept Paper, DHA," "Appeal for Rwanda 1994
Review," Rwanda Humanitarian Situation Reports and assorted letters and cables. Further, a
number of retrospective studies on Kibeho were extremely helpful including: the "Preliminary
Report on the April 1995 Closure of IDP Camps and the Integrated Humanitarian Response" by
UNREO, the "Report of the Independent International Commission of Inquiry on the Events in
Kibeho, April 1995," and Randolph Kent's unpublished manuscript, "The Integrated Operations
Centre in Rwanda: Coping with Complexity," 1995.
129. On the basis of food distribution figures, the estimates in November yielded a figure of
350,000. Using health and medical figures, the estimate was 280,000. Given the daily flow in and
out of the camps, and the estimated 80, 000 who had returned to their home communes, in midApril, the camp population of the remaining 9 of the 33 original camps was estimated to range
between 180,000 and 250,000, with greater verity attached to the lower figure. Almost half, up to
120,000, were in Kibeho.
130. Paul Kagame, Vice-President and Minister of Defense, who was viewed as the hero
responsible for the defeat of the former government and as the real power in the existing
government, initially set 30 November as the deadline for camp closure. He was also the one who
took responsibility for the use of force on 18-22 April 1995.
131. The UN agencies operating in Rwanda included FAO, UNDP, UNHCR, UNICEF,
UNIFEM, UNESCO, UNV, WFP, WHO and DHA's field team for coordination, the United
Nations Emergency Relief Operation (UNREO). As of November 1994, 114 international and 20
Rwandese NGOs were operating in the country. "From the outset of the April-July events, the
international community has been particularly preoccupied by the fate of the displaced populations
in Rwanda... During the latter part of 1994, the displaced persons throughout Rwanda have been
among the main beneficiaries of programmes carried out under the July-December Appeal. Parallel
to the provision of relief assistance, considerable energy has been expended by the international
community as well as the Government, to encourage and organise the return of the displaced to
their home areas." Appeal for Rwanda, 1994 Review, 5.
132. The voluntary return had ground to a halt because of a combination of: the deteriorating
security situation in home areas; the high numbers of arrests of returnees, including many who
were widely believed to be innocent even by government officials; the consequent fears for safety
if the IDPs return home; and the illegal occupation of homes in communes. The return of some
people from home to camps reinforced this perception. But these factors were greatly exacerbated
because of the misinformation, intimidation within camps, and killings by the agents of the former
government, and the strengthened grip of hard core elements on the camp population. There was
enormous and understandable concern by the government over security since the camps were
believed not only to contain genocidal killers but members of the interahamwe who used the camps
for recruitment and training. The government saw the camps as a potential "spearhead" for the
fallen government in Zaire. Cf. Integrated Task Force Briefing Note, 6 March 1995.
133. There were 59 employees in the IOC, 21 full-time and 38 part-time. The task force
governing the IOC consisted of the Director General of the Ministry of Rehabilitation and Social
98
Integration; representatives of the Ministries of Interior, Defense and Justice and from the UN
Operational Agencies (UNICEF, WFP, UNDP, etc.), UN Human Rights, IOM, and senior
officials of IOC. Cf. Kent, 1995.
134. The reports came from medical NGOs and MILOBs. Cf. Integrated Task Force Briefing
Note, 6 March 1995.
135. The six options included three main ones: 1 - The RPA goes in alone to close the camps by
force; 2 - Close camps in series beginning with Kibeho and Ngado (the largest camps) by ceasing
food distribution; 3 - Camps are consolidated and voluntary return continues with the remaining
population receiving basic food until a judicial system is in place. There were three others based on
reorganizing the camps into more controllable units as variations of the third option: 4 - Population
would be transferred to small camps in home communes (a proposal with increased logistical
problems and one that did not fit government policy); 5 - Reorganize the camps by communes after
closing smaller camps (chaotic and also does not meet government policy, with the additional
prospect of new intimidation); 6 - Population regrouped by communes in 4 camps (same problems
as above).
136. The source of the initiative and confusion over starting dates in the micro-management of the
operation was an immediate precipitating cause of the disaster. A 10 April agreement between
Randolph Kent, head of UNREO, and the Minister set a starting date of 18 April. The prefect of
Gikongoro and the local RPA agreed that camp closures would begin 16 April without the
elements agreed to between Kent and the Minister on 10 April meeting, that is, the coordination of
the information plan in the communes and the camps, increased security in both places, and the
presence of an enhanced justice process. On 13 April, the RPA drove people by force out of a
small satellite camp, Kivu Gizza. That is when Kent sent his warning to UN headquarters, but he
still believed that the situation could be retrieved and the populations returned in a peaceful and
humane way. On 14 April, a UN strategy position paper was adopted, "Resettlement of Internally
Displaced Persons in Rwanda," spelling out a plan insisting on no rapid closure of camps, but an
intensified information campaign plus quick action to deal with security in home communes. This
was read as simply another delay and unwillingness of the UN and the NGOs to go along with the
government policy or the previous agreements to expedite rapid closing of the camps. Between 1516 April, an IOC team visited Gikongoro to outline a plan of action with the prefect and army
commander. On 16 April, the pr‚fet and the IOC agreed on a new start date of 24 April, with an
end date set of 13 May. The pr‚fet agreed to telephone the RPA commander on the new agreed
date. Whether the pr‚fet failed to communicate the message, or whether the RPA just saw this as a
new stall and breach of previous agreements by the UN, or whether the RPA determined to take
the initiative on its own, we do not know.
137. On 14 April, Oxfam representatives reported a large influx of heavily armed RPA soldiers
into the region, and predicted the massive use of force. Randolph Kent, who headed UNREO, the
coordinating body of DHA in Rwanda, in a mid-April cable to headquarters warned that, "There
may well be at the same time serious pressures for public condemnations of any acts that are seen
as overly brutal," because "the IDP operation could put the UN in an embarrassing position, one
which will necessitate the distancing of UN agencies from possibly unacceptable initiatives by the
Government." The warning was given, not in order to allow the UN to take steps to prevent
anything happening, but so that the UN could be prepared to explain what happened and to put it
in a context for outsiders. "I hope any statement will emphasize the complexity of the situation,
including the security threat posed by the present IDP camps."
138. 18 April had been set as the date for closing the camps. Agreement had been reached on most
points, but a meeting had been scheduled for 10.am. on 18 April to settle the final points and
commence the closing. However, the camp had been surrounded three hours earlier. Though the
commencement of the RPA operation took the NGO and UN community by surprise, it was
evidently no surprise to the extremists, who began leaving the camp before the cordon could be
drawn around it. It is estimated that up to 20,000 of the hard-core elements had fled the camp by
16 April.
99
139. There have been many suggestions in the aftermath questioning the true intentions of the
RPA and second guessing the plan that was developed. Alison des Forges, one of the most
dedicated, knowledgeable and insightful Rwanda watchers, was in Kibeho and the three other
camps from 18 to 21 April, and on 23 April. She writes: "Why did I witness (and hear similar
accounts from others) of RPA soldiers driving people into the camps on the morning of 19 April people who came from hills adjacent to the camp where they were permanent residents? Why did I
witness soldiers refusing to let people leave the camps when all they wanted was to be allowed to
walk home?" (Communication to the authors, 20 November 1995). Robbery (based on perceived
incidents) and selective elimination of the strong, have been suggested as motives. As for the
actual plan, it has been suggested that as the 24 camps were closed earlier, Kibeho should have
been surrounded to prevent the extremists from consolidating in that camp. But surrounding a
camp spread out over six hills would have been extremely difficult if not impossible logistically. It
would be akin to setting up a closed military line around a city of 100-120,000 people.
Alternatively, consolidating the camp in a closer perimeter would have produced precisely the same
conditions as occurred on 18 April 1995. In fact, it is estimated that up to 20,000 extremists were
able to escape from the camp before the cordon around it was tightened. Similarly, options
suggesting that the police (Rwanda already had an acute shortage of police) should have been used
instead of the RPA, or that UNAMIR II (it already had to cover the whole country and the border
areas) should have been the military body to implement the plan fail to recognize the severity of the
perceived security threat that Kibeho posed. Arguably, the geographical and human conditions
were such that the options developed by the IOC were the only realistic ones. Further, if the
motives of the RPA included, in fact, robbery and the murder of the strong, then organizing any
co-operative plan with the RPA was a mistake in the first place.
140. "I interrogated the officer who was second in command at Kibeho and he couldn't give me
the names or companies of the soldiers who had supposedly been shot." Alison des Forges,
communication, 20 November 1995. The ICRC has also raised doubts about this claim.
141. "During the night of April 20, several people attacked the displaced with machetes, killing at
least two and wounding another 30." New York: Human Rights Watch, Press Release, 24 April
1995.
142. The International Commission of Inquiry found that there were RPA deficiencies in
communications, equipment, training, experience, and an intelligence failure in anticipating the
reaction of the hard-core, but they did not add the fact that the RPA unilaterally precipitated the
closure without the gradualism provided for in the original plan. There was little evidence that the
NGOs overtly tried to subvert the evacuation, as the Inter national Commission indicated. A
representative of the government indicated that one person from an NGO purportedly urged the
camp population not to comply with the government orders. Alison des Forges of Human Rights
has questioned the claim of a lack of communication. While at Kibeho during the operation, she
saw many officers chatting away on their hand-held radios.
143. The government was ridden with rivalry among the Ministry of Rehabilitation and Social
Integration, the Defense Ministry and the Foreign Ministry, as well as between Kigali and
Gikongoro.
144. For as long as Kibeho was perceived to be a large security threat, no one else but the RPA
could have tackled it. If anyone else had, perhaps the up to 20,000 extremists who managed to
escape the camp before the cordon was completed would have stayed, and there would have been a
two-sided war.
145. A highly-placed MINIREISO official said that only one NGO purportedly urged the camp
population not to comply with government orders. (Interview, Kigali August 1995)
100
Appendix
References
1. Interviews
The team leaders (Adelman and Suhrke) interviewed officials of the following
organizations/agencies:
United Nations/New York
• The Secretariat, Department of Humanitarian Affairs, Department of Peace-keeping
Operations, Department of Political Affairs.
• The Security Council, national delegations or observers from: the United States, France,
Nigeria, New Zealand, Canada, Sweden, Norway
• UNDP, UNICEF, UNHCR
United Nations/Geneva
High Commissioner for Human Rights
Center for Human Rights
National delegations to the Commission on Human Rights from Finland and Norway
UNHCR
DHA
United Nations/Kigali
UNREO
United Nations /Nairobi
UNHCR
OAU
The Secretariat (Addis Ababa), ex-Military Observer Team in Rwanda, OAU/Kigali
UNAMIR
Former Force Commander and Deputy Force Commander
Members of the diplomatic corps, Kigali 1990-94; participants/observers at
Arusha from:
Belgium, Canada, France, the United States, Tanzania, the Vatican, and the OAU
EU/Commission, DG-VIII
NGOs
M‚decins sans fronti…res
Africa Watch
US Committee for Refugees
ICRC Geneva
101
National actors
The United States:
U.S. Congress, Senate Foreign Relations Committee, National Security Council, CIA,
Department of State, Department of Defense
Canada
Department of External Affairs, CIDA
France
Ministry of Foreign Affairs
Ministry of Cooperation
Ministry of Finance
Tanzania
Ministry of Foreign Affairs
Uganda
Ministry of Foreign Affairs
Rwanda
Office of the President, Office of the Vice-President, the Ministry of Rehabilitation, the Secretariat
of the RPF.
2. Bibliography
A: Books, articles and unpublished papers
Adam, Absalom M. (1994)
"Early Warning and Conflict Resolution" (Review Article) in Refuge. Vol. 14, No. 4
(August-September): 15-17.
Adekanye, J. Bayo (1995)
"Rwanda/Burundi: 'Uni-Ethnic' Dominance and the Cycle of Armed Ethnic Formations".
Oslo, Norway: International Peace Research Institute.
African Rights (1994)
"Rwanda: Who is killing, who is dying, what is to be done". A discussion paper. (May).
Rwanda. Death, Despair and Defiance. London: 1994 and rev. 1995 ed.
"Humanitarianism unbound? Current Dilemmas facing Multi-mandate Relief Operations in
Political Emergencies". Discussion paper no. 5. (November)
Alker, Hayward R., Ted R. Gurr and Kumar Rupesinghe. (1995)
Conflict Early Warning Systems: An Initial Research Program. Paper, International Studies
Association, Chicago: February 21-24.
Amnesty International (1992)
World Conference on Human Rights: "Facing Up to the Failures: Proposals for
Improving the Protection of Human Rights by the United Nations". December.
Bayart, Jean-François (1994)
"Meurtres sous pavillon français" in La Croix (May 23).
----(1994).
102
La politique africaine de François Mitterand. Paris: ed. Karthala.
Braeckman, Colette (1994)
Rwanda. Histoire d'un génocide. Fayard.
Brauman, Rony (1995)
Rwanda. Devant le mal. Paris.
Bryden, Michael (1995)
"Somalia: The Wages of Failure", Current History. Vol. 94, No. 591: 145-151.
Burkhalter, Holly J. (1994/5)
"The Question of Genocide: The Clinton Administration and Rwanda," World Policy Journal, Vol.
II (Winter): 44-54.
"Burundi and Rwanda Mission: Finding/Recommendations," Refugee International.
Washington, D.C. (November 28 1994)
Callamard, Agnes (1995)
French Policy towards Rwanda. Toronto: Report prepared for Study II.
The Carter Center of Emory University (1992)
State of the World Conflict Report. Georgia: International Negotiation Network (1991-1992).
Childers, Erskine and Brian Urquhart (1992)
"Towards a more effective United Nations," Development Dialogue. 1-2. Uppsala.
Chrétien, Jean-Pierre (1993)
"Rwanda: 'Hutus modérés' ou Hutu power? Note sur l'enjeu politique au Rwanda, en date du 8
Mars 1993 … l'occasion de la présentation du rapport de la mission de la Fédération Internationale
des Droits de l'Homme au Rwanda.
Cohen, Herman J. (1993)
Africa and the United Nations "Agenda for Peace", unpublished paper. Washington, D.C.
Connaughton, R.M. (1995)
Military Support and Protection for Humanitarian Assistance. Rwanda, April - December 1994.
London: Report prepared for Study III.
de Moal, Yvon (1994)
Promotion of Human Security in Rwanda. Unpublished paper, Kigali.
de Waal, Alex and Rakiya Omaar (1995)
"The Genocide in Rwanda and the International Response," Current History. Vol. 94, No. 591:
156-161.
Destexhe, Alain (1994)
Rwanda: essai sur le g‚nocide. Paris: Ed. Complexe.
Durch, William J. (1995)
"Structural Issues and the Future of UN Peace Operations," in D.C.F. Daniel and B.C. Hayes
(eds.) Beyond Traditional Peacekeeping. New York: St. Martin's Press.
----(1993)
The Evolution of Peacekeeping. New York, St. Martin's Press.
Erny, Pierre (1994)
103
Rwanda 1994. Paris: L'Harmattan.
Gachuruzi, Shally B. (1995)
The Role of Zaire in the Rwandan Conflict. Toronto: Report prepared for Study II.
Gasana, James K. (1995)
"La guerre, la paix et la d‚mocatie au Rwanda," in André Guichaoua, (ed.), Les crises politiques
au Burundi et au Rwanda (1993-94). Paris: Karthala/Université de Lille: 435-447.
Gattegno, Herv‚ (1990)
"L'armée française dans le piège rwandais," Le Monde, 22 September 1990, in Guichaoua (1995).
Ghali, Boutros Boutros (1992)
An Agenda for Peace. New York: The United Nations.
Guichaoua, Andr‚ (ed.) (1995)
Les Crises Politiques au Burundi et au Rwanda, 1993-94. Université des Sciences et Technologies
de Lille.
Guilmette, Jean H. (1995)
Beyond emergency assistance. Early Warning, conflict prevention and decision-making.
Ottawa: CIDA. Unpublished paper.
Gurr, Ted Robert and Barbara Harff (eds.) (1994)
"Early Warning of Communal Conflicts and Humanitarian Crises" in The Journal of EthnoDevelopment (special ed). Vol. 4, No. 1.
Halvorsen, Kate (1995)
Refuge camps in Zaire: Security and Humanitarian Issues. Bergen: Report prepared for Study II.
Hanrahan, M.E. (1994)
Rwanda: An Introduction. Prepared for Canadian Peacekeeping Mission to Rwanda.
Human Rights Watch, Africa (1994)
"Rwanda, A New Catastrophe?" New York.
----(1993)
"Beyond the Rhetoric. Continuing Human Rights Abuse in Rwanda," Human Rights Watch,
Vol. 5, No. 7.
"Representatives of Rwanda's Genocidal Government Expected in New York: Militia Groups
Ordered to Halt the Slaughter During Kigali Visit by UN High Commissioner for Human Rights,"
Human Rights Watch, May 11.
Human Rights Watch Arms Project (1994)
"Arming Rwanda: The Arms Trade and Human Rights Abuses in the Rwandan War," in
Human Rights Watch, Vol. 6, No. 1.
----(1995)
"Rwanda/Zaire: Rearming with Impunity. International Support for the Perpetrators of
the Genocide." Human Rights Watch, Vol. 7, No. 4.
Institute for Multi-Track Diplomacy (1993-1994)
Peace Builder. Vol. 1, No. 2-3.
International Alert (1995)
104
"Burundi: Towards A Common Action Plan: Report of the Colloquium held in London, United
Kingdom" in International Alert. February 3.
International Federation of Human Rights(FIDH), Africa Watch, Interafrican Union of Human
Rights, International Center of Rights of the Person and of Democratic Development (1993)
Report of the International Commission of Investigation of Human Rights Violations in Rwanda
Since October 1, 1990 (January 7-21, 1993). Paris/Washington, March.
International Federation of Red Cross and Red Crescent Societies (1994)
"Under the Volcanoes: Special Focus on the Rwandan Refugee Crisis," in
World Disasters Report. Geneva.
International Organization for Migration (1995)
Executive Summary of IOM's Programming in Rwanda (January).
ICRC. International Committee of the Red Cross (1994)
Update No. 1 on ICRC Activities in Rwanda. (February 18)
Update No. 2 on ICRC Activities in Rwanda. (March 25)
----(1994)
L'‚tablissement de zones de s‚curit‚ pour les personnes d‚plac‚es dans leur pays d'origine. Paper
prepared by Yves Sandoz.
Issue. 1995.
Special issue on Rwanda. ed. David Newbury.
The African Studies Association, Vol. XXIII/2.
Johansson, Peik (1995)
International Press Coverage of the Rwanda Conflict. Helsinki: Report prepared for Study II.
Jones, Bruce (1995)
The Arusha Process. London: Report prepared for Study II.
----(1995a)
Refugee camps in Zaire: Security issues. London: Report prepared for Study II.
Kent, Randolph (1995)
The Integrated Operations Centre in Rwanda: Coping with Complexity.
Unpublished paper. Kigali: UNREO.
The Aid Impasse: Lessons to be Learned from Rwanda. DHA/UNREO
Khiddu-Makubuya, E. (1993)
The Legal Condition of the Refugee: National Report for Uganda. Unpublished paper, Kampala.
----(1994)
"Voluntary Repatriation by Force. The case of Rwandan Refugees in Uganda"
in H. Adelman and Sorensen (eds.) African Refugees. Boulder: Westview: 143-158.
Knight, Andy W. (1995)
Recommendations of the Core Working Group for the Canadian Government on
Developing a Rapid Reaction Capability for the UN system.. 20p.
Krop, Pascal(1994)
Le génocide franco-american: faut-il juger les Mitterand? Paris: Jean-Claude Lattes.
105
LaRose-Edwards, Paul (1994)
The Rwandan Crisis of April 1994. The lessons learned. Ottawa. Department of Foreign Affairs
and International Trade.
Legum, Colin (1994)
"A second, possibly greater tragedy waiting to happen, three months between now and renewed
genocide" in Third World Reports. London: December 17.
Lemarchand, Ren‚ (1995)
"Rwanda: The Rationality of Genocide." Unpublished paper.
Linden, Ian (1994)
The Genocide in Rwanda. Upublished paper. London: Catholic Institute of
International Relations.
Livingston, Steven and Todd Eachus (1995)
U.S. Television Coverage of Rwanda. Washington: Report prepared for Study II.
Longman, Timothy Paul (1993)
Christianity and Democratization in Rwanda: Assessing Church Responses to Political Crisis in the
1990s. Paper presented at conference on Christianity and Democratization in Africa, University of
Leeds, September.
Lucas, C. Payne and Donald M. Payne (1994)
Report to the President of the United States: U.S. Delegation to Rwanda, Burundi and Zaire.
August 26-31.
L‘greid, Turid (1995)
Mismanaging the Rwanda Tragedy. Oslo: Report prepared for Study II.
Makinda, Samuel M. (1993)
Seeking Peace from Chaos: Humanitarian Intervention in Somalia. London: Lynne Rienner.
Marks, Edward and William Lewis (1994)
Triage for Failing States. National Defense University: Institute for National Strategic Studies.
Washington, D.C.
McCormick, Shawn H. (1995)
"The Lessons of Intervention in Africa," Current History. Vol. 94, No. 591: 162-6.
Minear, Larry, Thomas G. Weiss, and Kurt M. Campbell (1991)
Humanitarianism and War: Learning the Lessons from Recent Armed Conflicts.
Providence, R. I: Brown University.
Neack, Laura (1995)
"UN Peace-keeping: In the Interest of Community or Self?" Journal of Peace Research,
Vol. 32, No. 2: 181-196.
Pottier, Johan (1995)
Rwandan Refugee Camps in Zaire. London. Report prepared for Study III.
Prunier, G‚rard (1995)
The Rwandese Crisis (1959-1994). London: C. Hurst & Co.
----(1995a)
Opération Turquoise. Paris: Report prepared for Study II.
----(1994)
106
"La crise Rwandaise: Structures et d‚roulement," in Refugee Survey Quarterly,
Vol. 13, No. 2-3.
Renard, Robrecht and Filip Reyntjens (1995)
"Aid and Conditionality: The Case of Belgium, with Particular Reference to Burundi, Rwanda and
Zaire," in Olav Stokke (ed.). Aid and Political Conditionality. London: Frank Cass.
Reyntjens, Filip (1994)
L'Afrique des Grands Lacs en crise. Rwanda, Burundi: 1988-1994. Paris: Ed. Karthala.
Roberts, Adam (1993)
"Humanitarian War: Military Intervention and Human Rights," International Affairs. Vol. 69, No.
3: 429-449.
Rosner, Jeremy (1995)
The New Tug-of-War. Congress, the Executive Branch and National Security.
Washington, D.C.: Carnegie Endowment.
Ruzindana, Augustine (1994)
"Banyarwanda must be protected," The New Vision (August 17): 13.
Rwandan Churches (1993)
"Réunion sur la r‚cherche d'une paix durable au Rwanda." Report.
Mombasa 27-30 November.
Sesay, Max Ahmadu (1995)
"Collective Security or Disaster? Regional Peace-Keeping in West Africa," in
Security Dialogue, Vol. 26, No 2.
Smouts, Marie-Claude (1995)
"The Meanings of Violence and Its Role in Legitimation," in Mershon International Studies
Review, Vol. 39, Suppl. 1: 111-115.
Speth, James Gustave (1994)
Africa: Conflict Prevention and New Development Initiatives. New York: The African-American
Institute, May 24. 9p.
Tekle, Amare (1995)
The OAU and Conflict Prevention, Management and Resolution. Asmara: Report prepared for
Study II.
U.S. Committee for Refugees (1994)
A Selected Chronology of the Rwanda Crisis, April 5 to September 30, 1994.
----(1994)
Genocide in Rwanda. Documentation of Two Massacres during April 1994.
United States Institute of Peace (1994)
Rwanda: Accountability for War Crimes and Genocide: A Report on a United States Institute of
Peace Conference. Washington, D.C.
----(1995)
The U.S. Contribution to Conflict Prevention, Management, and Resolution in Africa.
Washington, D.C.
Verschave, Fran‡ois-Xavier (1994)
Complicit‚ de g‚nocide? La politique de la France au Rwanda.Paris: La D‚couverte.
107
Watson, Catherine (1991)
Exile from Rwanda. Background to an Invasion. Washington D.C.: U.S. Committee for
Refugees.
William, Jean-Claude (1995)
"La Belgique et le 'Muyaga' des Grand Lacs," in André Guichaoua, (ed.), Les crises politiques au
Burundi et au Rwanda (1993-94). Paris: Karthala/Universit‚ de Lille: 435-447.
World Bank (1995)
Rwanda.
3. UN documents
DHA
Consolidated Inter-Agency Appeal for Persons Affected by the Crisis in Rwanda. Volume 1: The
Rwanda Perspective. January-December 1995.
Report on the Coordination of Humanitarian Activities in Rwanda. New York: November 1994.
UNREO, Kigali: Updates: Kigali, April, 1994.
DPKO
Improving Preparedness for Peace-keeping in Africa: An Informal Note. March 1995.
Peacekeeping. 1995.
UN Commission on Human Rights
Report on the situation of human rights in Rwanda submitted by Mr. R. Degni-S‚gui, Special
Rapporteur of the Commission on Human Rights, 25 May 199,. with add. E/CN. 4/1994/7/Add.
1, 11 August 1993; E/CN. 4/199570, 11 November 1994; E/CN. 4/1995/71, 17 January 1995.
The General Assembly
Administrative and Budgetary Aspects of the Financing of the United Nations Peace-Keeping
Operation. Report of the Secretary-General. May 1994.
Financing of the United Nations Observer Mission Uganda-Rwanda. Financing of the United
Nations Assistance Mission or Rwanda. March 6 1995.
UNHCR:
95/FRS/A/04 (9 February 1995)
Situation Reports, Zaire, June-August 1994.
The Security Council
Report of the Secretary-General on the Establishment of the Commission of Experts, 26 July
1994.
Second Report of the Secretary-General on Security in the Rwandese Refugee Camps, 25 January
1995.
The Security Council, Provisional Verbatim Records of Meetings
S/PV. 3377(1994), 3368(1994)
Reports of the Secretary-General
S/1993/25810, S/1993/26488, S/1993/26927, S/1994/360, S/1994/565, S/1994/640, S/1994/470
Letters from the Secretary-General to the Security Council
April 13,1994; S/1994/518; S/1994/518
108
Letters from member states to the Security Council
4 March 1993 (from Rwanda), 13 April 1994 (from Belgium), S/1994/531, S/1994/552,
S/1994/553, S/1994/586, S/1994/648
Statements by the President of the Security Council
S/PRST/1993/264, S/PRST/1994/21
Resolutions of the Security Council
812 (1993), 846 (1993), 872 (1993), 891 (1993), draft res. S/25981 (1993), 912 (1994), 918
(1994), 925 (1994)
Statements by the Secretary-General
SG/SM/5280 (1994), SG/SM/5289 (1994), SG/SM/5292
Issued as UN Documents
The Arusha Accords, 1993.
Status of Forces Agreements between the United Nations, Rwanda and (for UNOMUR) Uganda,
1993.
4. Other official documents
France
Opération Turquoise, Rwanda, June 22 - August 22. Briefing paper.
Opération Turquoise au Rwanda. Statement by General Lafourcade. Geneva 28 February 1995.
The Netherlands
Humanitarian Aid between Conflict and Development. Ministry of Foreign Affairs, The Hague,
1993.
Rwanda
Letter from the President to M. Boniface Ngulinzira, Minister of Foreign Affairs and Cooperation,
Kigali (26 December 1992).
Letter from the Prime Minister to the Minister of Defense. Distribution ill‚gale des armes … la
population civile. Kigali (25 Mars 1993).
Mission Permanente de la R‚publique Rwandaise Aupres des Nations Unies.
Allocation prononcée par son excellence le General-Major Habyarimana Juvenal, President
de la République Rwandaise devant la 48ème session de l'Assemblée Général des Nations Unies.
New York 6 October 1993.
Discours prononcé par le President de la République … l'occasion de la présentation du
programme gouvernemental pour le mandat 1989-1994. Kigali, 15 January, 1989.
Rwandese Patriotic Front. Press Releases: November 1991; March, August, October, November
1992; March 1993; April 1994.
République Rwandaise Services du Premier Ministre, Comité de Crise pour les Deplacés de
Guerre. Evaluation des besoins des déplacés de guerre du Rwanda. Dossiers Preparatoires.
Kigali, April 1993.
Rwandese Patriotic Front. Repatriation of Rwandese Refugees and Assistance to War Displaced
Persons in Rwanda. Rwandese Patriotic Front statement to the 44th session of the Executive
Committee of UNHCR, October 4-8, 1993.
109
Service National de Recensement, République Rwandaise. "Recensement Général de la population
et de l'habitat au 15 Août 1991." July 1993.
The United States
U.S. Congress (1995). Hearings on the Crises in Rwanda and Burundi. Subcommittee on Africa.
Committee on International Relations, U.S. House of Representatives. April.
----(1992). Somalia and the Question of U.S. Intervention. A Staff Report to the Committee on
Foreign Relations, U.S. Senate.
U.S. National Security Council, (1994). The Clinton Administration's Policy on Reforming
Multilateral Peace Operations, Washington, D.C.: May.
5. Newspapers and periodicals
Africa Confidential (1994)
Africa International (1994)
CBC Radio Reports (1994)
Jeune Afrique (1994)
Lib‚ration (1994)
International Herald Tribune (1994)
Le Monde (1994)
The Times (London) (1994)
New York Times (1994)
Reuters (1994)
The Economist (1994)
Toronto Star (1994, 1995)
Washington Post (1994)
