Understanding Rwandan politics through the longue durée: from the precolonial to the post-genocide era

ABSTRACT

The transition from precolonial to colonial rule at the end of the nineteenth century, the 1959–61 revolution followed by independence in 1962, and the 1994 genocide followed by the RPF’s military victory are defining moments of modern Rwandan history. Each of these periods was a major break with the previous one. However, there are also striking continuities throughout the entire history spanning the precolonial to the post-genocide eras. Continuities include the concentration of power, intra-regime conflict, the salience of ethnicity, and the nature of the state. Discontinuities can be seen mainly in the role of the army as an institution and a source of values, and the role played by and the use made of ethnicity. A very distinctive feature is the re-emergence of militarisation in 1994 after a century-long break, thus reconnecting with the precolonial period. This longue durée view allows us to better understand the defining features of governance in present-day Rwanda.

KEYWORDS:

• Rwanda
• politics
• continuity
• longue durée
• authoritarianism
• ethnicity
• militarisation

At first sight, three fundamental ruptures occurred in modern Rwandan history: colonisation, starting at the end of the nineteenth century; the revolution of 1959–1961 followed by independence in 1962; and the 1994 genocide followed by the seizure of power by the Rwandan Patriotic Front (RPF). Of course, these are breaks with the past, and they have been presented as such both by the players themselves and by historians. However, this article argues that there are also striking continuities spanning the entire period, from the mid-nineteenth century to the 2010s. These include the concentration of power, intra-regime conflict, ethnicity, and the nature of the state. Another characteristic – the pervasiveness of the military institution and of military ethics – disappeared during colonial days and the first two republics, but resurfaced from 1994 onwards, thus resuming continuity after a century-long interval.

While Rwanda had been quite extensively studied in the past, scholarship exploded after, and because of, the genocide. Much of that research is of excellent quality, but most newcomer academics have shown limited interest in the history that preceded 1994. Yet, Newbury and Newbury have convincingly argued that “[i]nstead of seeing history exclusively through the genocide (…), one can only understand the genocide through an understanding of Rwanda’s history”.¹ This also holds true for understanding governance in post-genocide Rwanda. The contribution of this article is therefore fourfold. First, though some elements of continuity have been addressed in earlier literature, this is the first attempt to analyse continuities and ruptures across a large spectrum of defining features of Rwanda’s political evolution. Second, also for the first time, this article surveys the entire period spanning from the mid-nineteenth century to the 2010s. Of course, this broad scope prohibits delving in any great depth into the many areas studied in this article, but it allows us to go beyond contingent events and evolutions and trace broad defining lines. Third, this longue durée view is very illuminating, offering as it does a better understanding of crucial characteristics of governance in Rwanda today, both at home and in the region. Finally, although history construction for current political purposes can be found in all places and times, including under previous Rwandan regimes, this article shows how the RPF’s appeal to precolonial history serves to legitimise contemporary political goals and practices. The article thus shows the link between historical processes and their representation with governance practices today.² It first surveys the four periods: precolonial, colonial, post-revolution, and post-genocide. It then outlines continuities and changes between these periods, whereby continuities clearly outweigh ruptures.

The precolonial period: “Defeat is the only bad news”³

In most African countries, referring to precolonial history for understanding contemporary political practice would not be considered relevant. However, in the case of Rwanda it is, as the RPF routinely refers to it and both discourse and policy are claimed to find inspiration in ancient times.

It is necessary at the outset to acknowledge that Rwandan historiography has mainly focused on the centre, and the royal court in particular. However, Newbury notes that this is a skewed presentation of a more complex reality: “Central court allegiance and local cultural affinities were seldom aligned”.⁴ As information came predominantly from court sources, “[o]fficial accounts present Rwandan history as exclusively the history of kings”.⁵ Newbury therefore reconstructs precolonial Rwanda based also on regional history, ecology, ethnicity and the experiences of ordinary people.⁶ Informed in large part by the official history promoted by Alexis Kagame, who was embedded in the royal court, the RPF leadership is only aware of (and interested in) that centralised narrative. Despite Newbury’s appropriate warning that it is “imperative that we distinguish clearly between the history of the royal family and the lived experience of the people living under that monarchical power”⁷, the remainder of this presentation will therefore focus on the royal, “centralised” history.

The Nyiginya kingdom emerged in the seventeenth century with the reign of mwami (king) Ruganzu Ndori. This was the beginning of a centuries-long evolution, but reliable and detailed information on Rwanda’s history became available only from around 1780.⁸ The genealogy of kings proposed by Rwandan historians who trace the dynasty back to the eleventh century⁹ is “purely imaginary”¹⁰, a “fairy tale”.¹¹ However, enough is known about the previous period to allow Vansina to write that under mwami Ruganzu Ndori, the army – an innovation that he created –, along with the ubuhake clientship system, became the foundation of power in the kingdom. While the Nyiginya kingdom was but one of the many that emerged in the great lakes region during the seventeenth century, in the course of the eighteenth century it became very different from its neighbours when non-territorial, multiple and permanent armies were put in place.¹² The monarchy then took shape, linking military expansion with political centralisation. King Rujugira (reign ca. 1770-ca. 1786) structured the armies by installing them in permanent camps near the most threatened borders. Two-thirds of these armies were created between his reign and that of Rwabugiri, roughly between 1770 and 1895.¹³

The deepest effect of this new military organisation was “the institutionalisation of a glorification of militarism and martial violence that finally permeated the whole of Nyiginya culture as the armies became the foundation of the administrative structure of the realm. (…) [U]ltimately, all the inhabitants of the realm were incorporated in the military organisation”.¹⁴ The army constituted the administrative framework of the country, and the concentration of power in the hands of the military commanders was an essential step in the unification of the kingdom.¹⁵ The RPF military historian Rusagara noted that “it is the military that played the most central socio-political role in what became of Rwanda”.¹⁶

Vansina also finds that the recruitment and indoctrination of intore (chosen young men serving as soldiers) from about ten years of age “favoured [the] exaltation of violence, imposture, and the right of the strongest that became the universal theme of all literary and choreographic artistic forms”.¹⁷ Although his book is replete with the “fairy tales” denounced by Vansina, Sebasoni, an early RPF ideologue, states that itorero, where the intore were trained, was the “crucible of chiefs and warriors”, “a military school of sorts”.¹⁸ By the end of the eighteenth century, the part of Rwanda under the reach of central court power was characterised by “utter militarisation”: the military machine included some thirty armies with about 12,000 combatants.¹⁹

Under these cultural, logistical and institutional conditions it is not surprising that the history of the kingdom is coterminous with war and violence, at least for those areas targeted by the Nyiginya court or administered by its delegates. The current official historical narrative is based on the notion of continuous war and conquest, ku-aanda (“from which Rwanda derives its name”²⁰), literally “expansion or spreading out from the centre”: “the principle of ku-aanda, which involved annexation and subsequent integration of neighbouring territories, informed the continued expansion and growth of pre-colonial Rwanda”.²¹ All the kings mentioned by Rusagara are warrior kings, and the “Map of Ku-aanda” includes large parts of present-day Uganda and the Democratic Republic of Congo (DRC).²² Alexis Kagame quotes the saying “Rwanda attacks, it cannot be attacked” (Urwanda ruratera, ntiruterwa) attributed to king Rujugira (late eighteenth century).²³

Even the largely mythical narrative proposed by Alexis Kagame is a long litany of wars against neighbours, conquests, punitive expeditions against unruly regions, reprisal attacks, insurrections and their repression, and civil wars. Violence was not only directed towards external enemies and internal opposition; it was also a frequent occurrence within the court and among ruling circles. Kagame’s list of royal succession struggles, massacres of entire princely families and those of chiefs whose loyalty was in doubt, rumour mongering and revenge, poisoning and cruel torture, executions, score settling etc. is nearly endless.²⁴ Vansina notes that from the reign of Rujugira onward, “the country was almost continually in a state of war”.²⁵

Similarly, when addressing the more recent period, from the mid-nineteenth century onwards, which is known in quite some detail, all the events mentioned by Kagame are wars, massacres, intrigue and competition inside the royal court. Kagame called Rwabugiri’s reign (ca. 1860–1895) “bloody”: “It was impossible to find a family in the country of which he had not killed at least one member”.²⁶ The country was at war two years out of every three during his reign, and there were 13 military campaigns in less than 20 years.²⁷ In Vansina’s words, he was first and foremost a warrior, inclined to resolving all difficulties by applying brute force.²⁸ Newbury calls him a “quintessential military monarch”.²⁹ His reign was also important for the centralisation of royal power and the fragmentation of the aristocratic court.³⁰ Rwabugiri’s successor Mibambwe IV Rutarindwa, who became king in 1895 under contested circumstances³¹, was overthrown and killed in a coup d’état a year later. The legitimacy of his successor Yuhi V Musinga, enthroned in early 1897, was immediately challenged but he was saved by German protection. These struggles were by no means exceptional, as most successions in Rwanda’s history have been violent and sometimes led to outright civil war. An uncontested accession to the throne was such a rare event that when it occurred in 1786, with Ndabarasa succeeding his father Rujugira in an orderly fashion, the latter’s sons were called Abatangana, “those who agree with each other”.³² However, after Ndabarasa’s death in 1796, civil war again broke out when his sons violently clashed over the succession.³³

The colonial period: indirect rule and Pax Germanica/Belgica

German and, more so, Belgian colonial rule have built on and reinforced the existing political and administrative system, while at the same time weakening and eventually destroying its underlying ritual and moral legitimacy.

When Rwanda became a German protectorate in 1896, mwami Musinga was facing serious challenges, both within the court and the chiefly structure, where his legitimacy was challenged, and across the country more generally. The north in particular was very unruly, and Musinga’s authority there was only nominal.³⁴ It took the Germans until 1912 to quell several insurgencies, and they did so in a brutal fashion, slaughtering people, burning villages, and executing the ringleaders. So the extension of court power throughout present-day Rwanda was the result of colonial conquest rather than local political dynamics.³⁵ The result of the “punitive expeditions” was the strengthening of the authority of the mwami and the German administration.³⁶

For reasons also seen elsewhere³⁷, Germany introduced a system of indirect rule, meaning it ruled the country through the existing political and judicial system. The cornerstone of these institutions was the mwami himself, to whom the German resident guaranteed protection and who retained jurisdiction over his people. He was recognised as owner of all property: land, crops and animals.³⁸ All this may seem to denote respect for traditional structures and norms, but indirect rule was soon to become indirect in name only. Already in 1910 the Duke of Mecklenburg stated that “by degrees, and almost imperceptibly to the people and to the sultan (king) himself, he eventually would be nothing less than the executive instrument of the Resident”.³⁹ This evolution was to accelerate under Belgian rule.

In 1916, Belgian colonial troops conquered a sizeable part of German East Africa during a joint campaign with the British. After difficult negotiations with the UK, Belgium secured possession of Rwanda and Burundi. It was formally given a mandate over the territory of Ruanda-Urundi by the League of Nations in 1923, but had decided to practice a policy of indirect rule well before. An ordinance-law of 1917 provided that “the Sultan (king) exercises, under the supervision of the (Belgian) Resident, his political and judicial competences in the way set out by indigenous custom and by the instructions of the Royal Commissioner”. In 1920, the Belgian Minister of the Colonies decided that “in Rwanda (…), where there is a strongly established indigenous organisation with a powerful authority, the relations between the metropolis and (that country) will be based on indirect rule”. But Minister Franck also explained that, while the indigenous political institutions needed to be “respected”, at the same time they were to be adapted and used on the road to “civilisation”.

The “adaptations” were many and profound.⁴⁰ The monarchy was desacralised: the king lost his right to kill, appointments of chiefs and sub-chiefs and the attribution to favourites of landed domains needed the approval of the Belgian administration, freedom of faith and the separation between state and religion were imposed, and a number of ritual practices were forbidden. The source of authority gradually shifted from the court to the colonial administration, a shift that was in a sense made official when mwami Musinga was deposed and deported in 1931. His successor Mutara Rudahigwa was a Christian appointed by the Belgians and the Catholic Church without regard for the traditional rules of succession.

Major changes were also made at the local level. While situations had been very different across the country, with some regions having a great deal of autonomy from the central state, the Belgians set out to standardise the indigenous administration and to bring the entire country under the control of the royal court. Inspired by the “Hamitic Hypothesis”⁴¹, the Belgian administration –supported in this by the Catholic Church – increasingly reserved political and judicial functions for Tutsi. While in the past Hutu had occupied chiefly functions, by 1930 the tutsisation was complete. The chiefly function was increasingly bureaucratised, and a new generation of chiefs was trained in schools with a European curriculum. The abolition of the triple hierarchy of chiefs (one for the army, one for the cows, and one for the land) made the system more authoritarian. Ordinary people had to pay both traditional and colonial taxes, and the fiscal weight increased considerably. Other alterations affecting commoners included forced labour, compulsory crops, restrictions on movement, and a selective access to education and social institutions.

The dual administration – indigenous and European – exhibited similar characteristics. Both were authoritarian, there was no separation of powers, and a recourse for citizens whose rights were violated was non-existent. Until the very last years of colonial rule, electoral legitimacy was absent. While indigenous absolutism was checked by the Belgian administration, the latter was an absolute ruler, barely controlled by the far away Belgian government and parliament. The indigenous political system experienced considerable strain from the mid-1950s onwards, and the Belgians switched their support from Tutsi elites to Hutu counter-elites. The Tutsi-dominated monarchy and the chiefly structure disappeared during the 1959–1961 revolution, and Rwanda became independent in 1962 as a Hutu-dominated republic.

The Hutu republics: the Thermidorian syndrome

Although the revolution was of course a major upheaval that replaced one elite group by another, political practice remained largely the same. Both authoritarianism and the distance between the core power holders and the wider population continued to be defining features.

When explaining the conflicts that shook the first republic (1962–1973), Lemarchand refers to the “Thermidorian syndrome”, involving “a partial restoration of the very order of things which the revolution at first sought to destroy”.⁴² Parliamentary reports published in 1964 and 1968 offer many examples of mayors (bourgmestres), governors (préfets) and local and regional leaders of the MDR-Parmehutu ruling party assuming many attributes, attitudes and values associated with chiefly function under the old regime.⁴³ At the central level, President Kayibanda became a “republican mwami” who relied on the old royal ways to relate to the popular masses. “Father of democracy”, “Father of the Nation”, “Beloved Chief”, and “Enlightened Guide of the Nation”⁴⁴, he gradually built a court in which the intrigues were reminiscent of the royal court of yesteryear. Without fully achieving this, Kayibanda aimed to establish what Apter has called a “presidential monarchy”, “with the presidential monarch embodying both non-dynastic aspects of the role and dynastic aspects associated with the ceremonial and ritual functions of kingships”.⁴⁵ Clearly the ideational impact of the revolution on Rwandan society was limited. There was a reversal of roles, but not of the structuration of those roles⁴⁶: a small group of Tutsi rulers was replaced by a small group of Hutu rulers.

By 1965, when the MDR-Parmehutu was the only party to field candidates for the parliamentary elections, Rwanda had become a de facto one-party state. Separation of powers was only nominal, and the real decision making took place increasingly within the ruling party. After the elimination of the Tutsi as a political force, the further narrowing of the political base was an intra-Hutu affair. Although they had played an important role in the establishment of the party, the regions of Butare, and later of Gisenyi and Ruhengeri were side-lined. By the beginning of the 1970s, the concentration of power in the hands of a small group of politicians from Gitarama (centre of the country) was almost complete. At the same time, the regime faced increasing internal instability linked to regional and ethnic tensions, and to the very functioning of the state. Isolated and unable to manage its own contradictions, the regime attempted at the beginning of 1973 to resort to ethnic manipulation, which then spun out of control. In July that year, the army staged a non-violent coup. The chief of staff and Defence Minister, General Juvénal Habyarimana (Gisenyi), became the new president, thus shifting the geographical base of power from the centre to the north.

While the new regime achieved a degree of ethnic pacification, in most other fields it continued old practices. It went on claiming the legitimacy of the 1959 “social revolution”, but eliminated its leaders, some politically, others physically.⁴⁷ Just two years after the coup, a single party was created, the Mouvement révolutionnaire national pour le développement (MRND), whose political monopoly was anchored in the 1978 constitution. The concentration of access to power, material resources, jobs and scholarships now benefited the regions of Gisenyi and Ruhengeri, which also secured almost all command functions in the army and the intelligence services. Through the MRND, the president controlled the three branches of government; separation of powers was practically non-existent. At the same time, the “egalitarian republic” increasingly became a myth, as a fourfold (military, political, merchant and technocratic) bourgeoisie emerged, and the gap between the urban and rural worlds widened.

Towards the end of the 1980s, the regime showed signs of exhaustion. The fall in coffee prices caused a fiscal problem, (sub-)regional conflicts came out into the open, civil society and the press became bolder and started to criticise certain aspects of regime behaviour openly.⁴⁸ In line with other authoritarian regimes in Africa, Habyarimana faced the “winds of change” after the fall of the Berlin Wall and reluctantly embarked on political reform. Democratisation coincided with the beginning of the civil war, when the RPF invaded the country in October 1990. After a failed peace process and the genocide against the Tutsi in 1994, the RPF took power in July 1994.

RPF rule: back to the Nyiginya kingdom?

The RPF was born in the Tutsi diaspora and viewed the precolonial period with nostalgia. This view was idealised and did not correspond to historical reality, but it strongly informed the RPF’s view of the present and the future, as well as giving it a veneer of legitimacy.

From its first days in power, the RPF imposed its view on the country’s political dispensation. Despite its formal adherence to the power-sharing formula inscribed in the 1993 Arusha peace accord, it could easily ignore such limitations, as its victory on the battlefield and the fact that it did not owe much to external forces (except Uganda) gave it a free hand to exercise power as it pleased. It did so all the more willingly since it knew that it would stand no chance in an open political contest. By August 1995, just a year after the RPF’s victory, the “government of national union” ceased to exist. The Prime Minister and other members of the cabinet (including a Hutu Minister of the RPF) resigned and left the country in protest at the closing of the political landscape, gross human rights violations and partisan appointments in the administration and the judiciary. This was just the tip of the iceberg. Many politicians, civil servants, judges and military officers who had stayed on or who returned after the RPF’s victory were threatened or disillusioned, and they fled abroad in increasing numbers from early 1995 onwards. At the same time, together with RPF-isation, a “tutsisation” drive was visible from early on. Although it officially rejected ethnic discrimination and even the notion of ethnicity, the RPF reserved access to power, wealth, and knowledge to Tutsi elites. By the end of the 1990s, about two-thirds of major positions in the state machinery were occupied by Tutsi of the RPF, and the military and intelligence services were almost exclusively in their hands.⁴⁹

As the political transition was to come to an end, the RPF set out to neutralise the opposition parties. The MDR was first divided by infiltration in the late 1990s, and banned altogether in May 2003, just before the first post-genocide national elections. At the presidential poll of 25 August 2003, President Kagame was elected with 95.05% of the vote after a campaign marred by arrests, “disappearances”, and intimidation. As all parties represented in parliament either joined the RPF list or supported Kagame during the presidential election, all the directly elected MPs were part of one political platform. An EU observer mission arrived at the paradoxical conclusion that, after the elections, “political pluralism is more limited than during the transition period”.⁵⁰ In reality, the polls returned Rwanda to de facto single party rule, and all subsequent elections were deeply flawed.⁵¹ At the end of 2015, a constitutional amendment allowed Kagame to run for a third seven-year term in 2017 (he was elected with almost 99 percent of the vote) and for two five-year terms thereafter. As he had effectively been in power since 1994, he could thus potentially lead the country for forty years. Rwanda is clearly a strong case of hegemonic authoritarianism, where under the guise of seemingly regular elections in a multi-party context the polls do not perform any meaningful function other than consolidating a dictatorship.

The RPF’s human rights record has been dismal all along. Political rights and freedoms of expression, association and the press were severely curtailed, but the RPF/RPA (Rwanda Patriotic Army, the RPF’s military wing) also killed civilians on a massive scale inside Rwanda during the period of the genocide and after, in Zaire/DRC in late 1996-early 1997, and during an insurgency in Rwanda’s north-western region in 1997–1998.⁵² More than one hundred thousand civilians were massacred by RPA “search and destroy” units in the DRC. The most comprehensive report, based on research carried out on behalf of the UN High Commission for Human Rights, concluded that the vast majority of the 617 listed incidents were to be classified as war crimes or crimes against humanity. On the issue of genocide, it noted that “[s]everal incidents listed in this report, if investigated and judicially proven, point to circumstances and facts from which a court could infer the intention to destroy the Hutu ethnic group in the DRC in part”.⁵³

The RPF has embarked on a formidable project of political, economic, social and cultural engineering, aimed at radically changing Rwanda and the Rwandans. It involved bold experiments in transitional justice, land tenure and agriculture, re-education, the spiriting away of ethnic identity, knowledge construction, spatial reorganisation (under the form of both villagisation and the redrawing/renaming of territory), and the instauration of pervasive control. The modernisation drive has been extremely fast, indeed too fast for most Rwandans: when the Rwandan government wants something, it wants it immediately, and it sets close and clear deadlines. Scott found “a pernicious combination of four elements in (…) large-scale forms of social engineering that ended in disaster”: the administrative ordering of nature and society; a high modernist ideology that believes it is possible to rationally redesign human nature and social relations; an authoritarian government that is “willing and able to use the full weight of its coercive power to bring these high-modernist designs into being”; and “a prostrate civil society that lacks the capacity to resist these plans”.⁵⁴ This is the combination of elements prevailing in post-genocide Rwanda.

History rewriting and linking the current situation to a pre-colonial Garden of Eden are crucial aspects of the RPF’s legitimisation. The monopoly of the narrative the regime successfully promotes extends not just to Rwanda’s visions and analyses of current affairs, for instance its democratic credentials, its human rights record, or its involvement in the DRC, but also to history more generally. In summary, this official history claims that precolonial Rwanda was a unified, harmonious and peaceful society, and that ethnicity was artificially introduced by the Belgian administration and the Catholic Church in the context of a divide and rule policy. The RPF put an end to genocide that resulted from divisive politics, and restored peace and harmony.⁵⁵ However, historians find “a whole set of false propositions and assertions in this narrative”, but Vansina understands the reasons for the elaboration of such erroneous propositions: “the projection of a nostalgic utopia into the past, a past that contrasts with a painful present”.⁵⁶

Continuity … 

The period from the mid-nineteenth century to today was marked by continuity more than by change. An obvious continuity throughout the four periods is the concentration of power. The precolonial kingdom became increasingly centralised, particularly from the latter part of the eighteenth century. In a pyramid structure, regional authorities were dependents of the mwami, and below them were hill chiefs who tightly controlled the population.⁵⁷ This does not mean that the king was the sole ruler. Vansina notes that, after the 1796–1801 civil war, the king lost much of his personal power to the great families at the court who dictated policy until at least 1875.⁵⁸ Authoritarian centralisation continued in colonial days, in two ways. On the one hand, indirect rule reinforced and stabilised the power of the court and the chiefs. On the other, the Belgian administration was authoritarian and, like the indigenous one, ignored such principles as the separation of powers and the rule of law. The elective principle and checks and balances were introduced less than two years before independence, and it is not surprising that, in Rwanda like elsewhere in Africa, the new political elites practised “business as usual” and continued colonial modes of governance. In this respect, there is not much of a break between colonial rule, the de facto single-party first republic, the de jure single-party second republic and de facto single-party regime in post-genocide Rwanda.

Desrosiers and Thomson argue that the pre- and post-genocide regimes have fundamentally shared the same authoritarian concern with power and control. Both visions promoted an understanding of state-society relations that stressed respect for authority, hierarchy, and Rwandans’ place as followers of their “benevolent” leaders. Presenting themselves as harbingers of an “improved” or “new” Rwanda, both leaderships have claimed to be best able and willing to guide the country along the right path to peace, security, ethnic unity and development.⁵⁹ Likewise, looking through the lens of surveillance, Purdeková finds that “the Rwandan state of today (…) closely resembles its much maligned predecessor”.⁶⁰ The oversight structures and techniques have intensified over time, from the precolonial and colonial days, through the first and second republics, up to the present day: “striking continuities are evident across historical epochs, despite claims of decisive and even revolutionary breaks with the past”.⁶¹

Intra-regime conflict is a second continuity. The authoritarian nature of the successive regimes might suggest that they were monolithic, but they were not. As seen earlier, internal strife within the royal court and among ruling elites was common in precolonial days. Most successions to the throne were contested and led to bitter and often violent infighting, and even to civil war. Factions fought each other through false or true accusations of conspiracy, manipulation, torture and murder. Internal struggles continued during the first years of German rule, and they only came to an end when the Germans succeeded in imposing law and order the harsh way, at the same time preventing Musinga’s overthrow. Nevertheless, the Germans realised that powerful factions continued to contend within Musinga’s court, and that “beneath the veneer of absolutism was political turmoil that could lead to civil war”.⁶² It took military expeditions, the burning of entire villages and summary executions to “pacify” the country, a situation that was only achieved in 1912.⁶³ From 1916, Belgian rule no longer needed military force to prevent intrigues from destabilising the country, but they did not disappear. During the entire reign of Musinga until his removal by the Belgians in 1931, factions at the court continued jostling for power, influence and wealth.⁶⁴

Regime infighting resumed after independence. We have seen how the gradual narrowing of the MDR-Parmehutu’s power base through the elimination of important constituencies eventually led to the downfall of the first republic. A similar phenomenon occurred under the second republic. As discussed earlier, the north took power in 1973, but Gisenyi and Ruhengeri préfectures soon fell out with each other, followed by a split between the Bushiru and Bugoyi regions inside Gisenyi. Hardly seven years after the birth of the second republic, this conflict led to major intra-regime tensions. After an episode where pamphlets were published with a virulence never seen in a country that usually exhibits a great deal of discretion and restraint, a number of regime leaders – among them the powerful director of the intelligence service – were arrested in 1980. Fearing a similar fate, others fled the country. In 1981, 47 persons suspected of plotting a coup d’état were tried. Two were given the death sentence, while 21 others were sentenced to between 2 and 25 years in jail.

The RPF has also fallen prey to intense struggle, which pitted factions against each other from the first days of the invasion, when three high ranking officers, including the military commander Gen. Fred Rwigema, were killed in an apparent internal settling of scores. Much of the infighting is largely invisible to outsiders, as debate on major issues takes place within a small inner circle. On a number of occasions, splits have however been very apparent. Already in the late 1990s, a number of RPF members left the country and turned into vocal opponents. This evolution became more pronounced after 2000, and it took a radical turn in 2010 when four leading figures who fled abroad published a document called Rwanda Briefing, which contained a long diatribe against the regime. Together with others, the four “renegades” as they were called by the regime set up an opposition party in exile under the name Rwanda National Congress. Just months later, the four were indicted and judged in absentia. Kayumba Nyamwasa and Rudasingwa were sentenced to 24 years in prison, Karegeya and Gahima to 20 years. In June 2010, Kayumba Nyamwasa was severely injured in an attempt on his life in Johannesburg, and Patrick Karegeya was found strangled in a Johannesburg hotel room on New Year’s Day 2014. These incidents had Kigali’s fingerprints all over them, Rwandan diplomats were expelled, and the relations between South Africa and Rwanda never recovered.

While I do not believe that a predominantly ethnic read of history is very instructive, there is no way of escaping the fact that a third major continuity is the salience of ethnicity, although it has had different political implications depending on the period. Political ethnicity emerged clearly in the nineteenth century. The distinction between ethnic groups that earlier referred to political positions and economic and military occupations became institutionalised. With the introduction around 1870 of the uburetwa labour tax, to which only Hutu were submitted, two hierarchical social categories came into being.⁶⁵ From indications of a situation of class or dependency or occupation, “Hutu” and “Tutsi” became absolute categories.⁶⁶ Chrétien finds a “clear discrimination between Batutsi and Bahutu, which in turn became part of everyday rural life (and not only in court functions) (…) Backed by government practices, the Hutu-Tutsi cleavage thus penetrated social life decisively”.⁶⁷ From the 1870s, the awareness of ethnic distinction spread all over the country and led to several revolts. While some of these Hutu movements targeted the court, others were aimed at the Tutsi as such. The 1897 insurrection was particularly significant as it showed that the population was conscious of a great divide between the two ethnic groups.⁶⁸

Colonial rule further institutionalised and rigidified ethnicity. Inspired by the Hamitic Hypothesis, Belgium first entrenched Tutsi rule, but switched sides in the 1950s when democratisation and independence came to the fore. Although there were underlying social, political and economic grievances, the revolution of 1959–1961 took place under an almost exclusively ethnic banner. Hutu elites dominated the first two republics, under which an overt ethnic narrative (rubanda nyamwinshi – “the majoritarian people”) and practice (such as a quota system) prevailed. Upon assuming power, the RPF set out to pursue a policy of de-ethnicisation. However, the denial of ethnicity is an essential element of the hegemonic strategies of the Tutsi elite. The claim that “there are no Hutu or Tutsi, we are all Rwandans now” allows them to hide a Tutsi ethnocracy. Collective identities were redefined in a way rarely seen elsewhere, and ethnicity was legislated away. The law reconfigured the ethnic map and entrenched the regime’s policing of relations between individuals and groups. However, de-ethnicization and reconciliation were imposed in a top-down authoritarian fashion, and all available fieldwork shows that the regime’s narrative merely reflects the public transcript, but that the hidden transcript – that of oppressed Hutu and Tutsi – is very different.⁶⁹

A final strong continuity lies in the nature of the state which, unlike in much of Africa, is strong and well internalised by citizens. Rwanda is not a colonial creation, and an ancient state tradition plays an undeniable role in the maintenance of an efficient pyramid-like structure. The Rwandan Leviathan is highly centralised and hierarchical, and it reaches every inch of the territory and every citizen.⁷⁰ Echoing the situation in colonial days and under the two previous republics, a mere two years after the extreme human and material destruction of 1994, the state had been rebuilt. Rwanda was again administered from top to bottom, territorial, military and security structures were in place, the judicial system was re-established, and tax revenues were collected and spent. The regime was able within a short time to establish total control over state and society. This control was seen in the maintenance of an efficient army, able to operate inside and far beyond the national borders; the establishment of “re-education”, “solidarity” and “regroupment” camps; the villagisation policy; tense relations of distrust with the UN and NGOs; and the establishment of an important intelligence capacity, with the Directorate of Military Intelligence (DMI) operating inside the country and the External Security Organization (ESO) in charge of operations abroad. Before as after the genocide, the regimes displayed a strong belief in managing, monitoring, controlling, and mobilising the population. Both showed a strong belief in using the state in projects of economic and social engineering. The RPF’s “Vision 2020” echoed the five-year development plans under MRND rule. All citizens are considered agents of development who march together under the stewardship of forward-looking and enlightened leaders.

 … and change

But change occurred also in some important respects. Discontinuities can be most prominently seen in two areas: the dominant role of the military, both as an institution and as a source of ethics, and diverging approaches to the issue of ethnicity. We have seen the centrality of the military institution for the court actors. Not just the army as an institution, but military values more generally were pervasive. The recruitment and indoctrination of intore were mentioned earlier as a strong instrument to spread these values across the entire elite and to promote the glorification of a warrior ethos. The centrality of the army and its values disappeared for a full century, during colonial days and the two pre-1994 republics. Of course, the armies lost their military function as the European powers imposed “Pax Germanica/Belgica”, but the itorero training continued, in which military skills alongside history, etiquette, poetry, dance and the use of words were still taught to young Tutsi. Ruhamiriza, one of Helen Codere’s respondents, recounts his being trained from around 1912 when he was a young teenager until he was sent to school aged around 25.⁷¹ Itorero eventually disappeared as a military institution and became limited to the teaching and practice of music and dance, until the RPF gave it new substance.

The role of the military was limited between independence and 1994. The army was small, was not involved in a single external war or in domestic keeping of order, and played no significant role in politics. Even after the army took power in 1973, civilians dominated the political systems just years after the coup. There were no references to the military institution or values in both regimes’ ideology. Rather the ethical reference was that of the hard working peasant. In this ideology of rural romanticism, only the Hutu were the real peasants, while the Tutsi belonged to a feudal class.⁷²

Militarisation returned almost overnight under RPF rule, and the continuity with the pre-colonial era was explicitly affirmed. Both the army as an institution and military values are actively promoted. Institutionally, this shows in the fact that the army and the intelligence services became the pillars of the regime and that a small circle of officers, mainly coming from the Ugandan diaspora, takes all important decisions in an opaque way, whereby the government only deals with the day-to-day management of administrative affairs and parliament is a mere rubber stamp. In Dorsey’s words, these formal institutions are the screens of the networks that are effectively in charge. The RPF has kept the shell of these institutions, but stripped them of any effective power.⁷³ Another way in which the institutional prominence of the military shows is the army’s profound involvement in the economy through a holding enterprise, a bank and an insurance company. Together with revenue from peace-keeping operations abroad, these activities give the army considerable financial clout, which is at least in part kept out of the state budget, as well as a great deal of autonomy.

Beyond the institutional aspect, military ethos and values permeate the entire Rwandan society. Again with reference to the pre-colonial past, the ingando re-education camps and itorero training, in which all Rwandans are supposed to participate, disseminate military values, in addition to the teaching of history and the advocacy of national unity. The participants wear military uniforms, and they are taught military tactics, how to parade and to assemble/dismantle a gun, and they eat the rations of the fighters during the bush years. The pre-colonial kings’ expansionism and military victories are vaunted, and translated in the contribution values like heroism and self-sacrifice make for development and broader social transformation.⁷⁴ Of course, the past is not the only reason for this militarisation. The circumstances in which the RPF seised power made the building of strong security forces a pragmatic necessity. There is thus a mix of tradition and modernity, with tradition reinforcing the dictates of a rational reliance on military strength.⁷⁵

Militarisation has effects beyond Rwanda’s borders that can be seen in the regime’s behaviour towards the region. While the former regimes never threatened neighbouring countries and generally maintained friendly relations with them, the RPF has engaged in large-scale military and economic adventures beyond its borders, in particular in the DRC, and, acting as a regional power, has become a menace to its neighbours. In an apparent revival of Urwanda ruratera, ntiruterwa and ku-aanda, wars were waged directly with the DRC and Uganda, and indirectly with Tanzania and Burundi.

Though I have shown the salience of ethnicity throughout the four periods, the role played by it and the use made of it are another break. We have seen that the awareness of the division between Tutsi and Hutu spread rapidly during the second half of the nineteenth century. Functions in the court and the chiefly apparatus were reserved for Tutsi, and only Hutu were subjected to uburetwa. The segregation showed clearly in itorero which was almost exclusively reserved for Tutsi, while the few Hutu and Twa were placed in a separate sub-group within the corps. Newbury sees a confirmation of the hierarchical nature of the kingdom in the emergence, under Rujugira’s reign, of two strata in the military organisation: intore (elite warriors) and ingabo (commoner warriors).⁷⁶ Even under German rule, Ruhamiriza recalls: “There were one hundred and fifty of us, almost all Tutsi. There were a few Hutu, but not more than ten. In any case the Hutu children were at one side in the dances; they were in separate quarters, and we did not eat with them or drink milk together with them”.⁷⁷

Ethnicity was made explicit, and further institutionalised and rigidified under the Belgian administration whose support for the Tutsi hierarchy only ended in the late 1950s in a revolutionary context. It then became important for the Tutsi elites to stress the “centuries old” national unity and to deny the reality of ethnic discrimination. In June 1958, mwami Mutara Rudahigwa summarised the debate on the “Hutu-Tutsi question” as follows: “These are just destructive rumours propagated by a small group of guys (des types) who act under foreign influence (…) and whose intention it is to divide the country. These enemies of the country will not succeed in dividing Rwanda (…). The entire country is united in the search of the bad tree that produces these sour fruits of division.⁷⁸ When it is found, it will be cut, uprooted and burned, so that it disappears and leaves no trace”.⁷⁹ Ethnocratic rule thus attempted to hide itself under the guise of absence of ethnicity, a discourse that can be seen again after 1994.

In opposition, the ethnic card was explicitly played in the first and second republics. The historical narrative claimed that the Twa were the first inhabitants of Rwanda, supposedly followed from 1000–500 BCE by the Hutu who settled there during the great “Bantu” migration coming from Cameroon. The Tutsi were claimed to have arrived just some centuries ago. They could therefore be labelled foreign invaders. Despite the wide popular acceptance of this chronology of human settlement⁸⁰, there is no scientific evidence to support it. However, this reading of history gave rise to the claim that Rwanda was the land of the Hutu and to conflate democracy with Hutu rule. Just as the Hutu had been before the revolution, the Tutsi were the victims of discrimination and even of genocidal violence, particularly at the end of 1963, and of course in 1994. Many fled the country, and by the end of the 1980s over half a million Tutsi lived in exile. Although their situation improved after the seizure of power by Habyarimana in 1973, when regional cleavages became more prominent, they remained second class citizens. A quota system limited their access to education and jobs, and governments would typically include one single token Tutsi. The continuing notion of the “majority people” was coupled with the fear of the return of “feudal” days, a threat that in the eyes of many Hutu materialised in 1990–1994.

After seising power, the RPF resumed the narrative of Tutsi elites in the 1950s. Ethnicity was said to have been introduced by the colonial rulers practising a policy of divide and rule, and the country now reverted to the harmony that characterised Rwanda before colonial days. There are no longer Hutu, Tutsi and Twa, but only Rwandans. National unity and reconciliation are imposed top-down and the de-ethnicisation project is supported by legislation on “divisionism” and “genocide ideology”, and by prosecutions under these laws. In 2013, the programme “Ndi Umunyarwanda” (“I am Rwandan”) was launched.⁸¹ As it is clear at the same time that Tutsi exercise a disproportionate share of power in politics, the military and intelligence services, the central and local administrations, the educational system and the (para-statal) economy, this ethnic amnesia⁸² in reality serves to hide the reality of a Tutsi ethnocracy. Thus a formal quota system under the Hutu republics is replaced by an informal one under the Tutsi republic, again a return to pre-colonial reality.

Conclusion

Clearly the continuities of this longue durée outweigh the ruptures. Except during the relatively brief period of colonial rule, Rwanda was, and is, a violent society. Wars with neighbouring countries and unruly regions were frequent, as were violent confrontations within ruling elites. Scores of people, both elite and ordinary citizens, were killed until the early days of German occupation, and again from 1959 onwards, with the 1994 genocide of the Tutsi as a tragic climax. Throughout the entire period, central political power has been almost absolute, and it was reinforced and extended during early colonial rule. Indeed, today’s Rwanda is in large part a creation of colonisation: rule of the mwami was spread to parts of the country that previously were not incorporated in the Nyiginya kingdom, political and administrative management were homogenised across entire current day Rwanda.

The tradition of a strong state is ancient and continues up to the present day. Contrary to what can be seen in many other African countries, the state is pyramidal, hierarchical, centralised, and strongly internalised by citizens. Its political culture exhibits enduring characteristics: in a context of strong respect for authority and discipline, leaders succeed in managing, monitoring, controlling and mobilising the population. In apparent contrast to this monolithic image, intra-regime elite conflict is considerable, and has been a recurrent source of conflict, often violent. Since the late eighteenth century, not a single head of state has been succeeded in a constitutional, peaceful fashion.

The salience of ethnicity has been and remains another defining factor, although it has played out in different ways depending on the political dispensation of the time. Simply put, under Tutsi rule, ethnic amnesia is used to hide ethnocracy by Tutsi elites, while under Hutu rule, ethnic belonging is highlighted to justify ethnocracy by Hutu elites claiming to represent the popular majority. In both cases, this practice has been and is used to relegate the other group to a subordinate status in the political dispensation and in social and economic life. It is important to stress that these are elite policies. Throughout the entire period studied here, the gap between the elites and the people has been and remains wide. Just like the court and chiefly culture was disconnected from the daily lives of the vast majority of Rwandans in pre-colonial days, the RPF’s leadership’s urban and cosmopolitan lifestyle is miles apart from that of ordinary citizens, both Hutu and Tutsi. Likewise, during the “Hutu Republics”, besides Hutu, Tutsi and Twa, a “fourth ethnic group”, that of a small bi-ethnic bourgeoisie accumulated wealth and privileges.⁸³

A final determining continuity is the pervasiveness of the military institution and of warrior ethics and values. What is particularly striking is the re-emergence of this characteristic in 1994, after it had virtually disappeared during colonial days and the two Hutu republics. After that century-long gap, it reappeared almost seamlessly. This is well rendered by the military historian Brig. Gen. Frank Rusagara who wrote that “the RDF [Rwanda Defence Forces] today not only ensures security for all, but provides a model of national unity and integration that continues to inform Rwanda’s socio-political and economic development”.⁸⁴ Beyond the institution, military values are disseminated throughout the entire society by the widespread use of means like ingando and itorero.

In today’s Rwanda, constant references to history, whether factually true or not, are used as a tool of legitimation. The idealised glorification of the precolonial era supports the political objectives and strategies of the current rulers. Therefore, the longue durée is not just a historical and epistemological issue, but very much a concrete contemporary political stake, hence the efforts of the RPF to impose and tightly police its narrative. The problem is that the public and the hidden transcripts often do not tally. Jessee and Watkins show this when confronting the “tutsi” and “hutu” readings of the monarchy. For the former, the kings were benevolent; for the latter they were bloody tyrants. Divergent versions on this and other themes align with people’s own experiences, received knowledge and political preferences. History in Rwanda is highly politicised and polarised, and considered a dangerous subject by most ordinary Rwandans.⁸⁵ In a political context where the hidden transcript cannot be publicly expressed, alternative narratives encode subtle resistance to the official history.⁸⁶

Acknowledgments

René Lemarchand, Scott Straus and two anonymous referees offered very useful comments on earlier drafts of this article. The usual disclaimer applies.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. Newbury and Newbury, “Bringing the Peasants,” 833.

2. In a recent article, Mathys shows how the past and its representation shapes the present, and how contemporary political needs and strategies influence historical narratives (Mathys, “Bringing History”).

3. Des Forges, Defeat.

4. Newbury, The Land, 5.

5. Idem, 15.

6. Idem, 281–339.

7. Newbury, “Writing Royal History,” 142.

8. Vansina, Antecedents, 52; Chrétien, The Great Lakes, 159.

9. Rusagara, Resilience, 209; Kagame, Un abrégé de l’ethno-histoire, 39.

10. Vansina, Antecedents, 44.

11. Idem, 217.

12. Idem, 196.

13. Chrétien, The Great Lakes, 160–1.

14. Vansina, Antecedents, 61–62. However, as said earlier, most people in the territory now defined as “Rwanda” were either outside the domain of court power or only intermittently subjected to military requisitions. Many regions resisted court intrusions until well into the colonial period. I thank one of the reviewers for this caveat.

15. Idem, 78.

16. Rusagara, Resilience, back cover. I must make clear that I don't quote Rusagara on a par with scientific historians, but only because he articulates the historical narrative of the current regime.

17. Vansina, Antecedents, 62. Vansina adds in a footnote: “Today this literary glorification of violence persists, and it is a particularly nefarious legacy of the ancestral heritage” (246).

18. Sebasoni, Les Origines, 59–60.

19. Vansina, Antecedents, 123.

20. Rusagara, Resilience, xvi.

21. Ibid., 1.

22. Ibid., 208. In November 1996, at the beginning of the first Congo war, Rwandan president Bizimungu showed a map to the media. On it, “Greater Rwanda” included large parts of Eastern DRC. However, at the end of the 19th century, Rwanda was smaller rather than larger than it is today (see e.g. Mathys, “Bringing History,” 470–5).

23. Kagame, Un abrégé de l’ethno-histoire, 137. A former holder of high office told this author that this saying is regularly recited in conversations between President Kagame and top military officers.

24. Ibid., passim.

25. Vansina, Antecedents, 75.

26. Kagame, Un abrégé de l’histoire, 108.

27. Vansina, Antecedents, 182–4. An overview of Rwabugiri's military campaigns is offered in Newbury, The Land, 129–141.

28. Vansina, Antecedents, 195.

29. Newbury, The Land, 130.

30. Ibid., 333.

31. The succession did not obey essential prescriptions of the esoteric code ubwiru.

32. Vansina, Antecedents, 107.

33. Ibid., 109.

34. Louis, Ruanda-Urundi, 145.

35. Des Forges, Defeat, 99–129.

36. Louis, Ruanda-Urundi, 157.

37. Mainly efficiency and reduction of costs (as territories could be managed with a minimum of European personnel), and the legitimacy of indigenous institutions (better accepted than outside rulers).

38. Louis, Ruanda-Urundi, 150–1.

39. Quoted by ibid., 145.

40. A full survey can be found in Reyntjens, Pouvoir et droit, 69–177.

41. “Everything of value found in Africa was brought there by the Hamites, allegedly a branch of the Caucasian race” (Sanders, “The Hamitic Hypothesis,” 521). In Rwanda, the Tutsi were seen as Hamites.

42. Lemarchand, “Political Instability,” 318.

43. Reyntjens, “Chiefs and Burgomasters”.

44. These descriptions can be found in Le Mois, which was the MDR-Parmehutu's monthly from March 1971 to June 1973.

45. Apter, The Politics of Modernization, 307.

46. Lemarchand, “Political Instability,” 318.

47. Dozens perished in the infamous “special section” of Ruhengeri prison.

48. The sclerosis of the second Republic is analysed, among others, in Bézy, Rwanda; Reyntjens, L’Afrique des grands lacs and Sibomana, Hope for Rwanda.

49. More details can be found in Reyntjens, “Rwanda”.

50. Mission d’observation électorale, 12.

51. For details, see Reyntjens, Political Governance, 26–56.

52. More details and numerous sources can be found in Reyntjens, Political Governance, 98-123. More particularly for crimes committed in Zaire/DRC see Reyntjens, The Great African War, 80–101, 287–90.

53. United Nations, para. 31.

54. Scott, Seeing Like a State, 4–5.

55. This presentation can be found in many official statements and documents, see e.g. Republic of Rwanda, Office of the President, The unity of Rwandans; Republic of Rwanda, Office of the President of the Republic, Report on the Reflection Meetings; Republic of Rwanda, National Unity and Reconciliation Commission, The Rwandan Conflict; Courses taught in ingando reproduced in Penal Reform International, From camp to hill, 83–112. For a summary of the government's reading of history, see Buckley-Zistel, “Nation, narration, unification?”, 33–38.

56. Vansina, Antecedents, 197–198. For further analysis of the post-genocide government's attempts to rewrite history, see Pottier, Re-Imagining Rwanda; Newbury, “Canonical Conventions” and Thomson, Whispering.

57. Chrétien, The Great Lakes, 175. It must again be stressed that this applied only to areas under central court control.

58. Vansina, Antecedents, 126.

59. Desrosiers and Thomson, “Rhetorical Legacies”.

60. Purdeková, “Mundane Sights,” 61.

61. Ibid., 80.

62. Louis, Ruanda-Urundi, 126.

63. Ibid., 145–56.

64. Des Forges, Defeat.

65. Vansina, Antecedents, 135.

66. Ibid., 136.

67. Chrétien, The Great Lakes, 187.

68. Vansina, Antecedents, 136–9.

69. The reference to hidden and public transcripts is borrowed from Scott, Domination and the Arts of Resistance. The differences between the two transcripts have often been highlighted with regard to post-genocide Rwanda, see e.g. Burnet, Genocide and Thomson, Whispering.

70. The metaphor of a Leviathan comes from Straus, The Order of Genocide, 201–223 who details how and why the Rwandan state is qualitatively different from most of the rest of Africa.

71. Codere, Biography, 50–59.

72. Verwimp, Peasants.

73. Dorsey, “Violence,” 314.

74. For a more detailed treatment of militarisation under the RPF and its historical references, see Purdeková, Reyntjens and Wilén, “Militarisation of Governance”.

75. I thank René Lemarchand for having drawn my attention to this hybridity.

76. Newbury, The Land, 323.

77. Codere, The Biography, 53.

78. “Divisionism” was made a criminal offence under the RPF.

79. Rwanda politique, 37.

80. Interestingly, even the Tutsi court historian adhered to this narrative, albeit in a book written after the revolution, i.e. under “Hutu” rule: Kagame, Un abrégé de l’ethno-histoire, 22–28.

81. Strangely enough in a context of fighting ethnicity, under the same programme, President Kagame asked all Hutu to implore forgiveness for a genocide committed in their name.

82. Term coined by Lemarchand, “Burundi”.

83. Vidal, Sociologie, 28–35.

84. Rusagara, Resilience, back cover.

85. Jessee and Watkins, “Good Kings”.

86. Jessee, “The Danger,” 148; Thomson, Whispering.