Journal of Humanitarian Assistance
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 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 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 signals[16]: 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]