23 Jun The Duclert Report and France’s Responsibility in the Genocide of Tutsis in Rwanda
27 years after the genocide that targeted the Tutsi population in Rwanda, and in the line of several others, two reports, ordered by France and Rwanda respectively, were published to decipher the role of France in the events.
During years preceding President Habyarimana’s death, on 6 April 1994, France had been securing an alliance with the Rwandan government against the Rwandan Patriotic Front (RPF). French President Mitterrand saw the former Belgian colony as key to a francophone Africa, that he perceived was threatened by British influence. However, the ally would turn to become the orchestrator of the genocide. The Duclert report, by accessing French national archives for the first time, sheds a new light on the facts and provides insights into French officials’ decisions at the time. While some archives remain unexplored, particularly military ones, it allows for a better understanding of France’s degree of knowledge and participation in the genocide of the Tutsi population. To conclude almost 1,000 pages of research, the report concludes:
Is France an accomplice to the genocide of the Tutsi? If by this we mean a willingness to join a genocidal operation, nothing in the archives that were examined demonstrates this. Nevertheless, for a long time, France was involved with a regime that encouraged racist massacres. It remained blind to the preparation of a genocide by the most radical elements of this regime.
How does this translate into legal terms? What would the International Court of Justice (ICJ) determine would it have to reflect on the role of France, a party to the Genocide Convention, during the genocide? This short essay aims at looking beyond the historian’s analysis to contemplate France’s responsibility in light of international law.
The main insight of the report perhaps lies in it showing how, from 1990 to 1994, French officials, including intelligence services, members of the embassy and armed forces, alerted Paris of violations targeting the Tutsi population specifically. Thus, the first instance of such killings is found in a note by intelligence services dated 26 October 1990 that affirms, although using the conditional tense, that several hundreds of Tutsis might have been killed, while President Habyarimana contended they were fighters disguised as civilians (Duclert report, p.71). Armed forces gave the same analysis in a note dated 9 November 1990, warning of the threat to Tutsis and moderate Hutus (Duclert report, p.73). Over the course of the four ensuing years, similar reports, documenting the increasing of racist discourse, arbitrary arrests and detentions, and killings, would be transmitted to Paris (Duclert report pp.92, 121, 127, 153, 222, 224). The first instance of the word genocide is, for its part, recorded to have been used as early as 15 October 1990, in a note authored by the French ambassador in Kigali, reporting that Tutsis feared a genocide might happen absent an intervention by European forces (Duclert report, p.93).
In reaction, France adopted an ambiguous position, that will remain as such until the unfolding of the genocide in 1994. It would oscillate between assistance to the Tutsi population and the continuing support to President Habyarimana for four years. Thus, the report reveals the organisation by some officials of small-scale missions aiming at rescuing relatives or acquaintances, as in the cases of the rescues of embassy staff, the partner of a French citizen, or a member of the political opposition (Duclert report, p.96). On a larger scale, it also recalls the organisation of the humanitarian mission Noroit in 1992 after massacres targeting the Tutsi population in an attempt to protect them (Duclert report, p.153). At the same time, however, the report shows how France failed to grasp an accurate understanding of the situation and kept analysing the violations as the result of a conflict in which it had an ally (Duclert report, p.98). From December 1990, when he decided to keep a presence in Rwanda at the request of Habyarimana (Duclert report, p.117), the French President would not cease its support to the Rwandan government, from meetings to trainings, materials and weapons (Duclert report, pp.119, 143, 169, 172, 174, 217, 218, 231). French support was also economic, as in the instance of the allocation, in March 1991, of a 70 million Francs grant to Rwanda (Duclert report, p.132). During this period, the main preoccupation of the French government would remain the supposedly critical threat posed by the RPF, supported by Uganda and Zaire, allied to the anglophone world (Duclert report, p.160, 177, 207, 219, 232).
The report therefore shows that violence targeting the Tutsi population started as early as 1990. The existence of genocide is not defined by the number of deaths, but by the targeting of the group and the perpetrator’s intention to destroy it, in whole or in part. The International Criminal Tribunal for Rwanda (ICTR)’s jurisdiction ratione temporis was limited by the United Nation’s Security Council’s resolution establishing it to prosecute violations committed “between 1 January 1994 and 31 December 1994”. Further, the Tribunal, in its findings, highlighted the 7 April 1994, the day after President Habyarimana’s killing, as the starting point for the genocide (see, among others, Nahimana et al., judgment, para. 121, and Akayesu, judgment, para. 112 and 126). However, a court ruling on France’s responsibility would not be bound by this finding and could place the beginning of the genocide as early as 1990, when massacres targeting Tutsis started being reported (Duclert report, p.45). The appreciation of France’s responsibility would be impacted by such a finding, given its longstanding relationship with President Habyarimana.
Given these elements, it appears the potential for a claim regarding France’s involvement in the genocide would arise from Article I, that enjoins states parties to prevent the occurrence of genocide, and Article III (e), that provides for the prohibition of complicity in genocide.
The International Court of Justice (ICJ), in its Bosnian Genocide case, gave clarification for what might constitute complicity according to article III (e), finding that two elements are required: support in perpetrating the genocide and full knowledge of the facts (para. 432).
From the time the genocide is commonly understood to have started, in April 1994, it does not seem to stem from the report that France was still providing support to the Rwandan government, the perpetrator of the genocide. Support for Rwandan President Habyarimana seems to have ended with the agreement to supply him with weapons in February 1993 in order to curb the progression of the RPF. The contentious Operation Turquoise, that unfolded from June 1994 in order to halt the massacres, on the other hand, does not appear, from the Duclert report, to have provided support in the perpetrating of the genocide (Duclert report, pp.447-612). On the other hand, a court could place the start of the genocide as early as 1990. In that case, while the report reveals a co-occurrence between the awareness of massacres, that could potentially be defined as genocide as early as 1990, and political, military and financial support of France to the Rwandan government, it does not show an intention to dedicate the support provided to the perpetration of the genocide, nor an awareness that this would be the case. Therefore, it does not seem to reveal support in perpetrating the genocide as mandated by the ICJ to establish the complicity of genocide provided by Article III (e) of the Genocide Convention. However, France would face greater challenges showing it did not breach Article I, providing for the obligation for states parties to prevent genocide.
Breach of the Obligation to Prevent Genocide
From the Bosnian Genocide case, it appears two of the most significant elements that would be at play in determining whether France breached its obligation of preventing genocide would be on the one hand to find whether “the State was aware, or should normally have been aware, of the serious danger that acts of genocide would be committed” (Bosnian Genocide, para. 432), and, on the other hand, whether it took “all measures to prevent genocide which were within its power” (Bosnian Genocide, para. 430).
In Bosnian Genocide, the Court, in order to reflect on the complicity and the failure of the state to prevent genocide, relied on factual elements. While acknowledging that it had not been shown that the decision to commit genocide “was brought to the attention of the Belgrade authorities”, it took into account, in addition to international outrage, concerns of the European Union negotiator and his request to President Milošević to allow the UNHCR and the ICRC to access Srebrenica. It thus concluded that the authorities “could hardly have been unaware of the serious risk of it” (Bosnian Genocide para. 436-438).
As was shown above, French decisionmakers similarly had access to a wealth of information from their intelligence services, military, embassy and NGOs that, since 1990, had alerted Paris of the increasing targeting of members of the Tutsi population, to build a correct understanding of the ongoing events (Duclert report, p.45). Instead, it appears their judgment was blurred by a wrong appreciation of the circumstances, a disproportionate concern for the RPF, and a lack of clarity regarding the intents of its ally, the government of Rwanda. If demonstrated, France’s lack of awareness seems therefore to have stemmed from a refusal to see the Rwandan administration for what it was. Thus, as it has in the Bosnian Genocide case, the Court could conclude that aggregating the facts should have given French decision-makers the awareness a genocide was about to happen.
Determining whether interventions such as the 1992 humanitarian assistance to the Tutsi population (Duclert report, p.153), the – belated – suspension of the authorisation of exports of weapons to Rwanda in April 1994 (Duclert report, p.409), numerous meetings between French and Rwandan officials in Paris and Kigali (Duclert report, p.410), and, ultimately Operation Turquoise (Duclert report, p.447) were sufficient appears to be a task only a court, with the resources and access attached thereto, could perform.
Nonetheless, we identify two challenges France would have to navigate in preparation for its defence.
It first would need to interpret the influence criteria. In Bosnian Genocide, indeed, the Court balanced the duty of the state to have taken all measures to prevent genocide with the influence it had over the perpetrator. The Court also gauged this influence using an array of factors, including the “geographical distance of the State concerned from the scene of the events, and the strength of the political links, as well as links of all other kinds, between the authorities of that State and the main actors in the events” (Bosnian Genocide, para. 430). It thus appears that the greater the influence of a sate over the perpetrators of the genocide, the greater the measures to prevent genocide it must have taken. The relationship between the Serbian government and the perpetrators of the genocide in Srebenica is however much different from that of the French and the Rwandan governments at the time. France would therefore have to navigate new waters while preparing its defence in this regard. Additionally, it would have to account for the possibility of the Court ruling that the genocide actually started as early as 1990. Then, while its influence over the Rwandan government faded after the death of President Habyarimana, the four preceding years were marked by tighter links that could prove decisive in the Court’s assessment of the measures within France’s power to prevent the genocide.
Second, France would need to build a defence regarding its awareness of the genocide. Indeed, the only way for a state to explain its lack of taking all measures to prevent genocide is to demonstrate it was not and should not have been aware of the ongoing or looming genocide. Thus, a state preparing its defence would have to take the crucial decision to either demonstrate it not being aware of the genocide and trust it to be plausible enough to explain it did not take any measures, or, on the other hand, to admit to it being aware of the genocide, and demonstrate that it then took all measures within its power to prevent it. The report at hands nonetheless renders the task arduous for France. It portrays a government constantly in-between, at once being informed by its field officials of the responsibility of the Rwandan administration in the incitement of massacres against the Tutsi population as early as 1990 (Duclert report, p.71), yet not reacting clearly to these; organising, as early as 1992 a humanitarian mission in support of the Tutsi population (Duclert report, p.154), yet providing the Rwandan government with weapons as late as April 1994 (Duclert report, p. 806). This ambiguity would likely be the thorniest challenge of a defence, putting France in a situation where neither its lack of awareness nor its taking of all measures would seem easy to demonstrate to a court.
While the report does not provide elements revealing France’s complicity in the perpetration of the genocide, awareness – the actual one, and the one a state should normally have had – lies at the core of the obligation to prevent. By showing France did take some measures to protect the Tutsi population on the basis of information shared by its services, the Duclert report reveals the state had some awareness of the threat to the group. Nonetheless, it also casts doubt about whether the state took all measures within its power to prevent genocide. Instead, it appears to put France in a delicate position, where it seems to have done too much to deny having known, yet not enough to demonstrate it did all it could. It also highlights a limit of the convention, and how a state might rather decide to disengage entirely from a situation showing the first signs of genocide to come, than taking the risk of later seeing its responsibility incurred.