Introduction: Researching Postcolonial African History ^{Footnote 1}

The historical trajectories and dynamics of postcolonial Africa have received increasing scholarly attention, and some Africanist historians have proposed frameworks for conducting historical inquiries in this area. ^{Footnote 2} Arguing that the past is not just “the embryo of the present,” the late Stephen Ellis suggested that we should investigate “unfulfilled ambitions and disappointed hope, ideas that once seemed important but that did not actually result in outcomes that are still with us today.” ^{Footnote 3} Frederick Cooper agreed that alternatives existed in the past and that the historian “has to move backwards before moving forwards” to put “process, choice, contingency, and explanation into the fore.” ^{Footnote 4} Others have highlighted the importance of examining the process and consequences of decolonization by paying attention to original aspirations and their shortcomings. ^{Footnote 5} A recent volume of History in Africa, which has six contributions that examine methodological issues in “writing the history of Africa after 1960,” ^{Footnote 6} is an indicator of this welcome trend, reflecting the recognition that contemporary Africa has been shaped by its history in various ways.

However, this research topic entails several methodological challenges, one of which is the unavailability of sufficient archival materials, or their dispersion across different sites. The colonial archives, which have been the primary documentary source for the study of Africa, provide minimal, fragmented, and sometimes even no material on the postcolonial period. ^{Footnote 7} Further, in general, archival records tend to be incomplete, because governments and various organizations often classify documents as being “sensitive,” and, therefore, essential for research, keeping them hidden from the public. ^{Footnote 8} Another research issue is that in many African countries, local archives are not always utilizable despite the fact that the condition of local archives and their relation to statecraft in postcolonial Africa are themselves important subjects of inquiry. ^{Footnote 9} Thus, the question of how to identify and utilize other collections of documents has become increasingly pertinent.

The Multi-Archival Route

One way to tackle this challenge is to conduct multi-archival, multi-continental, and/or multi-lingual research. Gabrielle Hecht used multiple archives maintained by governments, international organizations, companies, and Non-Governmental Organizations (NGOs) in Africa, America, and Europe to trace the history of “nuclearity” and Africa’s position in the post-war nuclear world. ^{Footnote 10} Erik Kennes and Miles Larmer used the national archives of Belgium, Portugal, Britain, and the United States, and those of the United Nations to examine the conflict in Congo. ^{Footnote 11} Mairi MacDonald used the American archives and those of international organizations, along with the French archives, for a study of postcolonial Guinea. ^{Footnote 12}

In particular, studies focusing on the decolonization period must adopt a historical approach that is multi-archival and international in scope because of the complicated and often messy process of decolonization involving multiple and dispersed actors. Matthew Connelly consulted archives located in Algeria, Egypt, France, Britain, Tunisia, and the United States to examine the decolonization process of Algeria. ^{Footnote 13} Ryo Ikeda consulted American, British, as well as French archives to examine the decolonization process in two other French colonies, Morocco and Tunisia. ^{Footnote 14} MacDonald also consulted Anglo-American documents for a study on Guinea’s decolonization process. ^{Footnote 15}

To better understand postcolonial African history in the aftermath of decolonization, we should follow this methodological route, and extensively consult archives, worldwide, particularly those that are more nonconventional. Based on the expedient categorization of existing archives into several groups, local archives would come first, followed by direct colonial archives (the former metropole) including commercial and personal archives. Examples include British archival material on Zambia, French archival material on Algeria, and Belgian archival material on the Congo. Other category is indirect colonial archives maintained by former colonial powers on the colonies of other colonial powers, such as British archival material covering Francophone Africa, French archival material covering Anglophone Africa, and British and French archival material covering the Belgian colonies. The archives of missionaries and international bodies have also been used as this kind of conventional archives. Relatively underused archives, however, are nonconventional ones such as those of the superpowers (the United States and the former Soviet Union), as well as those of other countries. The archives of the superpowers can be viewed as “quasi-colonial” archives, as the United States and the Soviet Union intervened in postcolonial African politics during the Cold War period. In addition, countries such as Cuba and China influenced postcolonial African politics. ^{Footnote 16}

Among these nonconventional archives, those of the National Archives and Research Administration (NARA) in the United States have proven useful for historians of diplomacy and international relations, and thus could also be utilized by scholars engaged in African studies. Even though the United States did not colonize Africa, after the Second World War, this country gathered intelligence from all over the world as part of its political strategy during the Cold War. ^{Footnote 17} Thus, the NARA archives can supplement conventional archives, enabling us to uncover further postcolonial historical material.

Smaller countries in Africa, in contrast with larger and better-known countries like the Congo, were not considered “hot” topics during the Cold War, were not accorded much importance in international politics at the time and thus did not receive much scholarly attention. Research on these countries could benefit from the adoption of such a methodology, especially if the objective is to understand the complexity of decolonization processes. Taking Rwanda as an example, I first review the historiography of Rwanda and its methodology for understanding Rwandan ethnicity. ^{Footnote 18} Next, in relation to the political events that unfolded in Rwanda in January 1961, my paper reveals the limitations of conventional archives and the promising potential of using nonconventional archives for the historical studies of postcolonial countries.

Ethnicity and History in Rwanda

Here, I would like to provide a brief review of the historiography of Rwandan ethnicity and the associated methodology. The genocide that occurred in Rwanda in 1994 grabbed international attention. ^{Footnote 19} In stark contrast to the centuries-old “tribal” hatred blamed on media, Rwanda has a complicated and dynamic history of ethnicity. A more flexible and less socially significant ethnic category existed in precolonial Rwanda. However, as the kingdom of Nyiginya, which preceded the modern state of Rwanda, expanded with the strengthening of monarchic power, this ethnic category became more politicized. ^{Footnote 20} Its stratification and hierarchicalization under Belgian colonial rule subsequently fed into escalating violence during the period of revolution and decolonization in Rwanda from the late 1950s to the early 1960s. ^{Footnote 21}

In the postcolonial era, following the establishment of the First Rwandan Republic, the ruling Hutu regime introduced policies that discriminated against the Tutsi. Church authorities, who participated in colonial projects as well as in the revolution, retained a close relationship with the state in spite of its occasional and somewhat rhetorical criticism of ethnic violence, thus influencing ethnic relations even in postcolonial Rwanda. ^{Footnote 22} The school system and curriculum, classroom practices, as well as differentiated access to schools, all contributed to the continued salience of ethnic affiliations in Rwanda. ^{Footnote 23} Discriminatory policies were relaxed after the Second Republic was established under President Juvénal Habyarimana in the 1970s. However, ethnic tension intensified during the critical period of the late 1980s when the economy was in decline and international calls for democratization increased. The civil war that erupted in October 1990 further exacerbated ethnic tension. Land shortages, economic difficulties, and famines impoverished many ordinary Rwandans, most of whom were Hutu. Despite the mediating role of the international community that led to a ceasefire, the civil war resumed, culminating in the genocide that followed the crash of the president’s plane in April 1994. The extent of violence during this period showed regional variations across Rwanda, reflecting regionally defined historical, ethnic, political, and security contexts, as well as individual motivations. ^{Footnote 24}

These previous inquiries into Rwanda’s ethnic history have used both oral and written sources, and often a combination of both. Interviews as well as oral and life histories, often conducted during long-term and intensive fieldwork, have provided important and informative insights. Alexis Kagame, the best-known Rwandan historian, published court records in both Kinyarwanda and French. ^{Footnote 25} Jan Vansina’s study effectively demonstrated the use of oral history and data in African historical research, showing how diverse oral traditions were consolidated and presented as the standardized version of official Rwandan history. ^{Footnote 26} Other scholars, for example, Helen Codere, Alison des Forges, and Catharine Newbury, have followed this path to construct a dynamic history of ethnicity in Rwanda. ^{Footnote 27}

Of the written materials consulted by scholars researching Rwandan history, those in the Belgian colonial archives and the Catholic missionary archives have been the most utilized. In particular, the African Archives of Belgium, located in Brussels, are a rich repository of documents on Rwanda holding preserved telegrams, reports, letters, and photocopies of local newspapers. ^{Footnote 28} The Archives of the Royal Museum of Central Africa in Tervuren is another important source, holding the private papers of former colonial officers. ^{Footnote 29} A further source is the Catholic White Fathers’ Archives in Rome, which includes local newspapers as well as diaries and letters written by European missionaries stationed across Rwanda. These documents provide information on local politics and other relevant details. ^{Footnote 30} In addition to these archives, the biographies and autobiographies of the individuals who were in Rwanda enable us to retrace the thinking of various actors at the time. ^{Footnote 31} There are also several published collections of documents that offer insights into the perceptions and policies of both Rwandan and Belgian politicians during this period. ^{Footnote 32} However, at the local level, existing archives in Rwanda have not been consulted by many scholars; nor has their existence even been recognized to any significant extent. ^{Footnote 33}

There are some methodological limitations associated with historical research on Rwanda. First, in recent years, the current research environment has made the conduct of fieldwork and the use of oral history more difficult. While acknowledging the potential of oral history to democratize historical inquiry, Erin Jessee drew our attention to constraints associated with the use of oral history, particularly when conducting research in/on a highly politicized environment such as that of Rwanda. ^{Footnote 34} Second, the multi-archival trend, discussed above, and especially the use of nonconventional archives, has not yet emerged as a norm in Rwandan history.

Of equal importance is the interpretation of historical sources, for which a careful reading of archival documents is indispensable. Previously, it has been argued that reports on colonial situations written for readers at home by Europeans, who viewed these situations through their own lens, entailed the possibility of “bias.” Therefore, even though these documents are useful sources for scholars researching African history, it is important to examine their contents critically to assess their “authenticity and credibility.” ^{Footnote 35} Recently, some scholars have suggested paying attention to the epistemological dimensions of colonial documents, notably uncertainty, doubt, misperception, absence, distortion, rumors, and dissatisfaction, thereby reading “along the archival grain” before reading “against it.” ^{Footnote 36} Gaurav Desai also pointed out that colonialism was rather messy, replete with contradictions and (mis)understandings, and productive of a complex series of interactions that occurred between the colonizers and the colonized. ^{Footnote 37}

Bearing these issues in mind, I now turn to a discussion of the Rwandan Revolution, highlighting how nonconventional archives could fill an evident gap in the conventional archives.

Revolution in Rwanda: What We Already Know

From the late 1950s to the early 1960s, Rwanda went through significant political changes that reached a climax in January 1961, when Hutu politicians carried out a coup. There are two main contested views on the causes of the political upheaval in Rwanda. On the one hand, some have claimed that the Revolution was caused and manipulated by the European colonial administration and the Catholic Church; this is the view of the current Rwandan government. ^{Footnote 38} On the other hand, contemporary European scholars have pointed to the Hutu leadership and their mass grievances against the Tutsi rulers. ^{Footnote 39} For these early academics, the conflict between the Tutsi and the Hutu was the inevitable consequence of Rwanda’s stratified society.

Later scholars have paid more attention to the reasons why the Hutu obtained power by focusing on the rise of Hutu elites, the reactions of the Tutsi leaders, Belgian policy shifts, the role of the Church, and the pressure exerted by the United Nations (UN). ^{Footnote 40} Beginning in the 1950s, the UN repeatedly demanded that Belgium take measures toward the “democratization” and decolonization of Rwanda. Cooperation between the Belgian colonial administration and Tutsi leaders began to deteriorate in the late 1950s when a new political movement was initiated in Rwanda. The Church, which had previously sided with the Tutsi nobles, began to sympathize more with the Hutu. As a result, many Hutu elites were educated at missionary schools and thus maintained a close relationship with the Church. After completing their education, many of these individuals sought administrative jobs but experienced discrimination. This situation fostered their discontent with the Tutsi system. However, the Tutsi leaders did not take the Hutus’ demands for social reforms seriously; they merely sought independence from Belgium and thus did not venture to change the political system.

The political situation consequently became more volatile, with violence erupting between the Tutsi and the Hutu in November 1959. The shift in Belgian support from the Tutsi to the Hutu at this stage aided the Hutu in seizing power. ^{Footnote 41} The Belgian government declared political reforms with the aim of conducting local and national elections and establishing a provisional government and an assembly as a step towards self-government. In line with this policy, communal elections were held in June and July of 1960. In October, a provisional government and assembly were instituted, and a national legislative election was scheduled for January 1961.

The road to self-government and independence was not, however, a smooth one. The Belgians were criticized and denounced for its colonialism by the Eastern Bloc as well as by African and Asian countries within the UN General Assembly (GA). On 20 December 1960, the UN GA Resolution 1579 was passed demanding the postponement of the election and the organization of a conference by the Belgians aimed at fostering “national harmony.” ^{Footnote 42} A conference was held in Ostend, Belgium, in early January of 1961 to discuss the timing or postponement of the national election, and a decision was reached at the conference to hold the election as scheduled. However, the Belgian government in Brussels announced the postponement of the election in accordance with the resolution.

Infuriated by this decision, Hutu politicians resorted to direct action. On the morning of 28 January 1961, about 3,000 burgomasters elected during the previous communal elections were summoned to Gitarama in central Rwanda. Residents in and around Gitarama were also assembled. During this meeting, the reign of the Tutsi king, Kigeri Ndahindurwa, was terminated, the birth of a republic was announced, a president was nominated, and the National Assembly was formed. The coup thus brought about a power shift from the Tutsi monarchy to the inception of a Hutu republic.

There are several important questions relating to the coup that occurred in January 1961. Were the Belgians aware in advance that the coup was being planned? Why and in what capacity did they support it? Moreover, why did the Belgian government postpone the legislative election in the first place, which appears to have triggered the coup?

Belgian complicity in the coup has always been assumed or suspected. For example, Alexis Kagame described the coup as “camouflage of the Trusteeship.” ^{Footnote 43} René Lemarchand alleged that the coup-planning developed during the Ostend conference, writing:

Moreover, after interviewing the Special Resident himself in 1980, Reyntjens wrote that Guy Logiest had given assurances to the Provisional Government that they might proclaim a republic, promising to arrange the appeal to Brussels and the UN. ^{Footnote 45} Thus, the consensus was that the Belgians, and especially the Trusteeship administration, were complicit.

Retrospectively, Logiest revealed in his memoir that Grégoire Kayibanda had visited him on 25 January 1961, a few days before the coup. He told Logiest, “The situation is really bad for us.” In his view, Hutu leaders could not trust the UN, because the resolution favored the Tutsi leaders. Consequently, they “want to organize a big coup which will convince Usumbura and Brussels that the games are done and for that we need your help.” ^{Footnote 46} Logiest replied that he could not support a coup, because this would be a hostile act against the UN, but he was sympathetic toward the “idea to liberate people.” Thus, he would feign ignorance and would offer passive support such as logistics, including transportation and stage venues. ^{Footnote 47}

These are the most explicit references proving Belgian involvement in the coup. However, it is understandable that no obvious documents have been declassified from the colonial archives, or at least I have been unable to find any so far that reveal such involvement. Moreover, to date there is no explanation as to why the Belgian government postponed the election in the first place, even though it had been agreed at the conference that it should be held as planned. The prevailing consensus is that Belgian government officials postponed the election while supporting the coup plan, as they thought they could deflect criticism from both the UN and the Belgians and Rwandans in Rwanda. However, the decisions of the colonial government to postpone the election and support the coup at this important juncture do not appear rational and have not been justified. At this juncture, nonconventional archives have a potentially significant role to play, revealing a more complex picture.

Revolution in Rwanda: What the American Archives Reveal

NARA in the United States contains several thousand documents on the Rwandan political situation during decolonization. Officers in the US embassy in Belgium, France, and the Belgian Congo and the consulate in Burundi sent back intelligence information, the minutes of meetings with Rwandans and Belgians, and photocopies of newspapers in Kinyarwanda with English translations to their government at home. This intelligence was gathered, because the Rwandan political situation was under scrutiny from the Cold War perspective.

A careful reading of the NARA documents reveals that it is not entirely correct to say that Belgium unanimously supported the coup, as some Belgian ministers were seemingly unaware of the possibility of a coup, or, at least, they had disagreements with the local administration. Moreover, the decision to postpone the election was not reached until the last minute; rather, the option of carrying out the election as scheduled received more support than that of postponement. The decision to postpone the election was finally taken in consideration of international politics.

The Belgian government did not want to postpone the election in spite of its acknowledgement of the risk of holding the election as scheduled. If the government had prioritized African issues and maintained the pace of decolonization by holding the election in January as scheduled, it would have been accused by UN members of not respecting the UN resolution. A telegram dated 17 January 1961, which sent from the American Embassy in Brussels to the Secretary of State in Washington, reported that the “principal officials concerned (including [those in the] Foreign Office and African Ministry)” in favor of immediate elections were “overriding” because the January 1961 election had been originally proposed by the UN, and then accepted by Belgians and welcomed by most Rwandan politicians. Accordingly, if Belgians postponed the election, that would be viewed as “treason” by most Rwandans. Indeed, Kayibanda reportedly remained in Brussels after the Ostend conference and voiced the threat of a “revolutionary situation” facing Belgium and the UN if Belgium postponed the election. Additionally, the Belgians anticipated more violence if the pre-electoral period was extended. Thus, Belgium was expected to announce on 18 or 19 January 1961 that the election would be held as scheduled. ^{Footnote 48}

However, Belgium was facing opposition from its allies. On the same day, the American Embassy in Paris sent a telegram to Washington. This telegram contained a report on a NATO meeting at which a Belgian delegate stated that the government “must decide soon (probably tomorrow) on [the] election date” and stressed the dangers of delaying the election. However, France, the United States, Britain, Netherlands, Italy, Denmark, and Greece all asked Belgium to consider accepting the UN resolution. ^{Footnote 49}

The following day, on 18 January, the Belgians had still not reached a decision. However, they did reject two “extreme solutions” that were assumed: one entailed “fly[ing] directly in face of UN resolution” by holding the legislative election immediately and the other was to “yield completely to [the] UN” and postpone the election until GA decided it should be held. However, they wavered between two more middle-ground options. The first option, favored by the African Minister, Harold d’Aspremont-Lynden, and some Foreign Office staff, was to hold an election “of more limited scope with [the] promise [of] more definitive elections later at [a] time set by [the] UN.” The second option, preferred by the Foreign Minister, Pierre Wigny, was to postpone the election “temporarily” in order to appease the UN and hold it by early April after the GA had resumed in late March. However, the government sought the opinion of the Americans on this. ^{Footnote 50} The Department of State instructed the American Embassy in Brussels to tell Belgians that the latter option “appears [to be] far [more] preferable [than the former option] according [to] our assessment of [the] situation and apparently according [to] that of other allies as expressed at [the] NATO meeting.” ^{Footnote 51} These accounts suggest that the Belgian government was indecisive and was not confident about the outcome the postponement would produce.

Ultimately, the Belgian government decided to postpone the election because maintaining its relationship with its allies, as well as its international reputation, was perceived as crucial. ^{Footnote 52} Indeed, it seems that Wigny, the Minister of Foreign Affairs, pushed for the decision to postpone the election against the wishes of the Ministry of African Affairs and others in the Cabinet who favored an immediate election, because he wanted to placate NATO allies. The Belgians anticipated that the UN would agree to the holding of the election in late March. They simply intended to notify Rwandan leaders that they planned to hold the election at the end of March and managed to “obtain UN approval to this end,” which suggests that they wanted to reassure Rwandan politicians of political progress in Rwanda. ^{Footnote 53}

As expected, Hutu politicians and the Belgian trusteeship administration were infuriated by the Belgian government’s decision. On 25 January 1961, Jean-Paul Harroy announced his decision to grant internal autonomy to Rwanda. ^{Footnote 54} Moreover, he criticized the government during an interview, the transcript of which was published on 27 January 1961, in the Belgian magazine, Pourquoi Pas?:

On the 28 January 1961, which was the day after the interview transcript was published, Hutu leaders staged a coup d’état in Gitarama.

However, some, if not all, of the ministers in Brussels were not informed about the planned coup in advance. A telegram sent from Brussels to Washington stated that “it is apparent [that the] Belgian administration [in] Ruanda (specifically Harroy) undoubtedly connived at January 28 ‘coup d’état’ without consulting Brussels.” ^{Footnote 56} The Belgian government wanted to recall Harroy, as he seemed to be responsible for the coup, “of which Foreign Office officials maintain (probably with truth) they had no previous knowledge and on which Harroy reported almost nothing.” Furthermore, the abovementioned interview was not “cleared or even known about [in] Brussels prior [to] publication.” ^{Footnote 57}

Ultimately, the Belgian government’s plan to recall Harroy did not materialize, the UN’s attitude toward this government hardened, and the re-scheduled legislative election was not held in March. A new Belgian government was formed in April 1961, with Paul-Henri Spaak, former NATO Secretary-General, as the Minister of Foreign Affairs and African Affairs. The Belgian government attempted to regain its international reputation by cooperating with the UN while trying to control the local administration. The election was finally held in September 1961 in the midst of continuous ethnic violence that produced ever-increasing numbers of refugees fleeing to neighboring countries. Though Rwanda gained independence in July 1962, it did so without achieving “national harmony.”

Thus, a reading of American archival documents “along the grain” reveals evident frictions between Usumbura and the government in Brussels, as well as within the Belgian government, and the government’s inability to exert sufficient control over the local administration. Thus, their policies sometimes appeared contradictory, particularly regarding the timing of election. I have no intention of downplaying the responsibility of the Belgian government, but it does seem to be the case that not all Belgian government officers had knowledge of the coup in advance and, further, not all of them supported it. Another conclusion is that the Rwandans had a greater influence on policymaking in Belgium than was previously thought. Additionally, Rwandan politics were much more closely intertwined with Cold War international politics of the time. These findings, which reflect a shift from a “black-or-white” interpretation to a more comprehensive and dynamic one, modify our understanding of the coup in Gitarama.