The Christian Churches, the State, and Genocide in Rwanda

The churches in Rwanda have exercised considerable political influence during both the colonial and post-colonial periods. Although formally autonomous institutions subordinate to the state, in actuality they have cultivated political influence through their religious teachings and secular role as the loci of material and social resources. However, there is at least one key factor, which has contributed to their fluctuating political influence within Rwanda. During the colonial period, the dominant Catholic Church functioned within a colonial regime of indirect rule, predicated on sustaining the political authority of a Tutsi-dominated Central Court presiding over the territories roughly contiguous with the present-day republic. This threefold division of power and authority acted as a brake upon the hegemonic ambitions of the Church, the royal house and the colonial administrators. Following the abolition of the monarchy in 1961, the structure of political power and authority of the state was fundamentally transformed, clearing the way for the emergence of a ‘state church’ whose political role in the two Hutu dominated post-colonial republics would have significant historical implications. In this essay, I argue that it was this structural transformation of the Rwandan polity marking the shift from a trilateral to a dual relationship between state and Church -, which contributes to our understanding of how the Church became embroiled in the mass violence and genocide in the twentieth century Rwandan polity


Introduction
The churches first gained a foothold in the Central Kingdom of Rwanda shortly after the advent of German indirect rule in 1898 when in February 1900 Catholic missionaries of the Society of Missionaries of our Lady of Africa, or 'White Fathers', arrived in the royal capital Nyanza seeking permission from the mwami or king to begin their work (Carney 2012a:82;Longman 2010:38-39). The arrival of Europeans coincided with the violent aftermath of the 1896 Rucunshu coup d'état, which marked the accession of King Yuhi Musinga, the undesignated suc-cessor to the last independent monarch of Rwanda, king Kigeri Rwabugiri (circa 1867-1895).
During the pre-colonial period, political power in the Rwandan Central Kingdom was exercised by what Linden describes as 'a military aristocracy' which presided over a 'chain of clientship relations checked on the one side by the monarchy, the king's men, and on the other by lineages and clans' effecting 'an internal tendency towards an equilibrium' (Linden 1977:ix). Des Forges similarly points to the existence of a complex web of clan and lineage ties and associations, which gradually coalesced around 'outstanding' leaders who were instrumental in the formation of the Rwandan state (Des Forges 1999: 31). However, both Des Forges and David Newbury challenge the notion that the social and political institutions of Rwanda's precolonial state were characterized by a tendency towards stability. As David Newbury notes, the idea that the Rwandan state possessed a system of checks and balances constitutive of a 'cohesive society' with 'clear, static and standardized administrative institutions' (D. Newbury 2011:xxv) belies the fact that both the composition and institutions of the kingdom were in a constant state of flux (Des Forges 1999:31). 2 Thus although the emergence of the Rwandan kingdom as an increasingly powerful and sophisticated polity accelerated during the eighteenth and nineteenth centuries, the kingdom's expansionary drives, rather than heralding a 'centralization' of political power, was instead accompanied by internal rivalries at court 'that grew ever more ferocious over time, and because the kings were weak, they could not halt them' (Vansina 2004:163).
During the late precolonial period under the reign of Kigeli V Rwabugiri (1853-1895) there occurred a shift in political relations, which appeared to represent a centralization of power. However, as Vansina argues, this apparent concentration of power entailed a process of 'centralized anarchy' (Vansina 2004:194), which contributed towards the weakening of 'previously salient' group associations, such as lineage and clan (Longman 2010:36;see C. Newbury 1999:313). 3 The 2 Several contemporary scholars present a diametrically opposed interpretation. Thus, for example, Katongole contends that 'one could not find a more cohesive national unity than what existed in precolonial Rwanda' (Katongole 2005:71), a view shared by Gourevitch (Gourevitch 1998:47, see 73). The notion of a precolonial 'collective national identity' (Gourevitch 1998:57-58) appears curious given the fact that the concept of the 'nation' is a European construct. In any event, however one defines 'nation', Vansina argues that the 'linguistic and cultural unity of the country today did not exist in the seventeenth century and Rwanda is not a "natural" nation. … Rwanda really became a nation in the twentieth century… Formerly, neither abundance nor order flourished in the country… The reasons for the elaboration of such erroneous propositions are evident. … [T]here is the projection of a nostalgic utopia into the past, a past that contrasts with a painful present' (Vansina 2004:198, 199;see Freedman et al 2011:301-302). 3 Vansina notes that 'clans are in fact phenomena that derive from the political arena' and were as such mutable since they represented 'alliances rather than descent groups' (Vansina 2004:34).
nature of this much-vaunted centralization… did not consist in the creation or extension of a homogenous network of royal administrators, in the imposition of a network of uniform territorial sub-divisions, or in a territorial homogenization with regard to tributes or corvées. Certainly Rwabugiri acquired personal power beyond that of his predecessors, but it mostly derived from the internecine struggles among the aristocrats. And, in the end, Rucunshu proved that the king had failed to subdue them. One must therefore conclude that Rwanda as a fully centralized state is a colonial creation (Vansina 2004:194-195).
In reality, therefore, the Central Court presided over a complex polity whose territorial expanse waxed and waned since it never wholly succeeded in subordinating peripheral territories in which it exercised varying degrees of power and authority over time (see e.g. Longman 2010:34). The consolidation of a clearly delineated territorial domain controlled by a central authority only began to emerge during the early years of German colonial overrule, a process, as we shall see, which coincided with the racialization of ethnic identities on the one hand, and the emergence of what I term a 'tripartite' relationship of contestation of power between the royal court, the German authorities and the churches, on the other hand.

Naming 'Ethnicity'
Longman argues that 'while the precise meaning of the terms "Hutu", "Tutsi", and "Twa" in precolonial Rwanda is a subject of considerable academic debate, most scholars agree that the terms did not refer to ethnic groups in the modern sense, as all three groups spoke the same language, shared common religious practices, 4 and lived interspersed throughout the region' 5 (Longman 2010:35). The use of the terms during the pre-colonial period nonetheless has a complex history. Catherine Newbury argues that prior to Rwabugiri's rule, there appeared to be three non-territorial ethnonyms: 'Twa', 'Hima', and 'Tutsi', which together 'established an opposition between the bulk of the population and smaller groups that did not farm. The terminology suggests that an agricultural way of life was perceived as the normal condition of the inhabitants, from whom one had to distinguish "different" people who practiced another way of life ' (C. Newbury 1999: 313). The etymology of these ethnonyms is unknown. However, Vansina contends that the terms 'Hima' and 'Tutsi' likely only coexisted in certain territories (Rwanda, Burundi, Karagwe). In these territories, 'Tutsi' seems to have referred to a social class or political elite among the herders and 'Hima' an inferior social class of 'Tutsi'. There is evidence that the term 'Hima' was abandoned in favour or 'Tutsi' following the establishment by Ndori of the Nyiginya dynasty in Central Rwanda during the seventeenth century (Vansina 2004:35, 36, 44). Originally, 'Hutu' constituted a term of 'individual disparagement' (C. Newbury 1999:314) alluding to 'rural boorishness or loutish behavior used by the elite' (Vansina 2004, 134). During the seventeenth century the term was applied to all those who were in service to the court (Vansina 2004:134). The categories 'Tutsi' and 'Hutu' could nonetheless not be distinguished on purely occupational grounds since many Tutsi were cultivators and many Hutu owned cattle (Longman 2010:35). 'Tutsi' appears to have referred to a fraction within the group of herders who formed the political elite (Vansina 2004:134).
As a governing elite began to assert its perceived superiority over commoners, The word 'Tutsi', which apparently first described the status of an individualbecame the term that referred to the elite group as a whole and the word 'Hutu' -meaning originally a subordinate or follower of a more powerful person -came to refer to the mass of the ordinary people (Des Forges 1999:32; see Vansina 2004:134-136;Longman 2010:35).
Vansina provides an account of three principal historical developments marking important shifts in the criteria defining Rwanda's three main population groups following the establishment of the kingdom. During the rule of King Cyilima II Rujugira (1675-1708) in the late seventeenth century, it became customary to refer to any combatants in the army as Tutsi, a term that 'stood in opposition to mutware, "chief", as well as to "Hutu", meaning "noncombatant"', or those who were in service to the army. It was 'in this context that the first direct and institutionalized opposition between Hutu and Tutsi arose' (Vansina 2004:135). Since noncombatants happened to be largely drawn from lineages of farmers whereas the first company was always recruited from among the pages, the elite became accustomed to referring to all cultivators as 'Hutu' as opposed to 'Tutsi', who were now associated with herders, irrespective of their origins. As the kingdom grew and its armies multiplied this practice spread across the country.
During the mid-nineteenth century, King Mutara II Rwogera (1802-1858) introduced the distinction between chief of the long grass (herders) and chief of the land (cultivators), which further consolidated and institutionalized an occupational division between Tutsi herders and Hutu farmers (Vansina 2004:135). The chiefs were 'nominated at the level of territorial provinces' (Vansina 2004:132). With the expansion of the kingdom, this practice increasingly cut across previously salient clientage, lineage and clan relationships. He notes, however, that it was only at the close of the nineteenth century that the ruling classes 'ended up labeling all herders "Tutsi" in implicit opposition to all subjects who were farmers' (Vansina 2004:37;emphasis added).
This practice emerged following Rwabugiri's accession in approximately 1870 following which the hated uburetwa was introduced, a form of forced labour prestation from which Tutsi were exempted. Increasingly exploitative relations of production 'aggravated' and 'poisoned' the distinction between Hutu and Tutsi: From this point on 'Hutu' and 'Tutsi' would no longer designate a relative category with respect to class or occupation but became an absolute one (Vansina 2004:135, 136;see Pottier 2002:13;Longman 2010:35). 6 The division between Hutu and Tutsi and the latter's growing status were amplified by Rwabugiri's relentless military campaigns. Widespread violence within the kingdom and 'new and more exploitative forms of land and cattle clientage' (Longman 2010:36;see Linden 1977:6) triggered several rebellions led by farmers (Vansina 2004:136), which resulted in a 'heightened awareness of ethnic differences' (C. Newbury 1988:51). Vansina characterizes Rwabugiri's rule as 'the unrelenting rise of a tide of terror that starts at court and engulfs the whole country, finally erupting into a social crisis that has now lasted for well over a century' (Vansina 2004:164). 7 6 Pottier similarly argues that it was 'through uburetwa that social relations took on a strong ethnic character before the Europeans arrived' (Pottier 2002:13) whereas Carney suggests that during the late nineteenth century Hutu-Tutsi labels developed 'ideological overtones that were missing in earlier periods of Rwandan history' (Carney 2013:13). 7 Vansina's interpretation of Rwabugiri's rule and its impact upon the transformation of social relations within the precolonial kingdom is widely regarded as authoritative. However numerous contemporary scholars contest his views. Ugirashebuja, for example, challenges the notion of an emergent 'ethnic' consciousness during the late pre-colonial period, arguing that the 'Catholic missionaries and colonizers never took into account the specific unity of the Hutu and the Tutsi peoples, their common pride which made them Rwandan, nor the fact that ninety percent of the Tutsis belonged to the masses of poor peasants' (Ugirashebuja 2004:49). It is of course true that many Tutsi were poor. However as Linden argues, 'the mass of poor Tutsi identified with the nobles and clung to their precarious superiority by despising and exploiting those of Hutu birth' (Linden 1977:227). In this sense, as Pottier argues, 'wealth, not race, was the basis of the ethnic distinction between Hutu and Tutsi' (Pottier 2002,14). Ugirashebuja, notably, does not define the term 'peoples', which appears to suggest some or other form of fixed or corporate group identification, an 'intrinsic' difference reiterated in statements such as 'each group of Rwandans certainly had its particular qualities', the 'merits' of which the colonisers failed to appreciate (Ugirashebuja 2004:50, 51). Jean-Paul Gouteux argues that the '"racialization" or "ethnicisation" of these communities is a modern, Western phenomenon. It results directly from the concept of the missionaries and colonizers, even if it was adopted by the natives, both Hutu and Tutsi' (Gouteux in Ugirashebuja 2004:49-50). Catherine Newbury's more nuanced account steers clear of essentialist categories. She argues that during the tumultuous period of Rwabugiri's rule, 'the consolidation of ethnic categories was shaped by political context. It is politics that makes ethnicity

Colonial Overrule: The Churches and the Royal Court
It was against this backdrop that King Yuhi V Musinga  acceded to the throne in 1896. The violent and contested nature of the accession was not exceptional. As David Newbury argues, carefully ordered royal successions were a myth and the transition to Musinga's rule followed in a long-standing pattern that Diverged from the ideological norms of the kingdom. Such intense competition revealed the violence at the heart of the Court and shattered the image of a peaceful society unified by custom and law and united in a powerful social coherence (David Newbury 2011:xxiv, see xxv, xxxv-xxxvi; see Reyntjens 2010:28-29; emphasis added) These circumstances, rather than the Court's ideological representations, faced the first colonial administrators, missionaries, and traders whose arrival was closely monitored by the Court. Although the initial objective of the small contingent of German administrators was limited to establishing Germany's territorial claims rather than instituting colonial rule (Des Forges 2011:24;Linden 1977:3), the German officer Captain Hans von Ramsay appeared at Court in March 1897 and 'proposed an alliance between the king and the colonial authorities that [the Queen Mother] Kajogera immediately accepted. Thus began the colonial era' (Vansina 2010:179).
The alliance marking the inception of German indirect rule had several principal features. Firstly, although the German authorities were far from passive overseers, they nonetheless adopted a policy of non-interference in the internal affairs of the royal court. A consequence of this policy was that the Rwandan state 'remained the coercive instrument of Tutsi rule' (Linden 1977:3;see C. Newbury 1999:297), which meant that the 'customary intrigues and violence' at Court persisted (Vansina 2004:179;see Linden 1977:43) until 'after the Belgian conquest in 1916' (Vansina 2004:164-165).
From the outset of colonial rule, the Catholic Church played a more ambiguous role since its activities were not restricted to the religious realm but overflowed into the secular political affairs of both the Royal Court and the colonial state, a feature of colonial indirect rule in Rwanda that would endure even after the abolition of the monarchy in 1961. For although the White Fathers did not initially pursue an overtly political agenda, their mission was defined by the belief of the order's founder, significant (or, indeed, insignificant), not ethnicity which invariably defines politics. The paradox is that ethnicity was simultaneously the product of politics and yet, at times, a powerful determinant of the shape of political culture'. She rejects what she terms an 'instrumentalist' view according to which 'ethnic identities are simply products of external machinations' or 'colonial intrusions', although it is clear that the notion of a collective corporate 'other' originates with the colonists (C. Newbury 1999:313).
Cardinal Charles Lavigerie, 8 that the conversion of the non-Christian commoners might best be achieved by focusing evangelistic efforts on the ruling elite (Longman 2010:39;see Linden 1977:3;Des Forges 2011:27). Towards this end, the key to winning the chiefs lay in taking their power seriously, making certain that they realized that Christian teaching would support their authority (Des Forges 2011:27).
The Church's policy of forging relations with indigenous political leaders thus complemented that of the colonial authorities, who regarded the ruling elite's internal authority as the keystone of their political strategy (Longman 2010:40). More particularly, this meant non-interference in the social and political prerogatives of the Central Court, such as ritual practices, the ritual authority of the king (Linden 1977:44), trade (Des Forges 2011:47), the appointment of chiefs, clientage relations, and the suppression of internal dissent. On the other hand, this policy entailed siding with the Central Court in its drive to consolidate its authority over formerly semi-autonomous as well as recently conquered but rebellious territories, notably the northern clanlands and Gisaka in the southeast. 9 The 'top-down vision of evangelization' thus positioned the Church 'as a close ally of the colonial state and the Tutsi-dominated royal court' (Carney 2012a:174).
The Court nevertheless viewed the missionaries with suspicion, engaging in ongoing intrigues to limit their impact on the Central Kingdom. The Court's interest in the missionaries was limited to secular learning, including teaching Musinga to read and write. Religious teachings were forbidden at Court, although they were permitted amongst Hutu and Twa commoners. This policy seems to have reflected the Court's awareness that 'full acceptance of the new faith might be inconsistent with complete loyalty to the mwami'. It also effectively, if temporarily, blocked the strategy adopted by the Fathers of evangelizing members of the royal court (Des Forges 2011:29;see Linden 1977:32-33).
Since the royal court's resistance to the Father's religious teachings did not prevent the latter from fostering direct ties with the local chiefs, the full implications of these daily interactions between the Fathers and the commoners only gradually manifested themselves. At first, Musinga and his closest associates tried to block the evangelization of Tutsi by granting the Fathers sites for mission stations in the rebellious northwestern provinces of Bugoyi and in Gisaka. Refusing this offer, the Fathers secured a site on Mara hill in the south. Accompanied by the king's brother, Cyitatire, the missionaries were instead led to the site of Save hill within his domain, a densely populated area a mere twenty kilometers south of the capital of Nyanza. These were tactical concessions by both the Court and Cyitatire. The Court viewed the southern province of Bwanamukari as troublesome and inhospitable to outsiders whereas Cyitatire feared assassination and aided the missionaries by placing his Hutu subjects at their service to construct the first mission station. This 'was the first intimation of the Fathers' later role as powerful allies in court politics' (Linden 1977:33;see Des Forges 2011:28).
By the end of 1903 five mission stations in the Central Kingdom 10 had managed to gain the trust of Hutu commoners, rapidly expanding their ties to include trade and the provision of medicine and other social services. For their part, the Hutu began to view the Fathers as their protectors and surrogate patrons (Linden 1977:42;see Lemarchand 1978, 364), a development that alarmed the Court (Des Forges 2011:30;Linden 1977:43). For what this signified was the emergence of a 'flourishing and largely theocratic peasant Church [which] grew up around isolated mission stations'. Court policy itself now posed a threat to Tutsi domination since it led to 'the rise of the Hutu Church with an educated clergy and a separate clientship network' (Linden 1977:3;see Lemarchand 1978:364).
Despite these gains, the zealotry of the missionaries rapidly formed what Linden describes as a 'theocratic tyranny' (Linden 1977:38), which triggered a largely spontaneous countrywide uprising against the White Fathers in 1904 (Linden 1977:54). Although the Court did not instigate the uprising, it nonetheless demonstrated its ability to manipulate the unrest to its own advantage (Des Forges 2011:46). 11 Conversely, as the Fathers were increasingly drawn into the political affairs of the kingdom, the distinction between 'temporal and spiritual realms' (Lin-den 1977:38) became blurred, a development which had the potential to destabilize the regime of indirect rule instituted by the German authorities.
A series of events now occurred which would prove to have a long-lasting impact on the relationship between the Court, the White Fathers and the territory's colonial rulers. Firstly, it was only with the aid of the German authorities that the White Fathers were able to end the uprising. Secondly, however, since the missionaries far outnumbered the Germans, 12 the German authorities were compelled to enlist the support of the superiors of the Mission Society to curtail the activities of the Fathers. Finally, in the face of the growing political influence of the White Fathers amongst the commoners the German authorities moved to shore up the political authority of the Central Court (Linden 1977:39, 41), thereby seeking to reassert their overall control of the territory.
The outcome of the 1904 crisis was an unwritten entente between the court, the White Fathers and the Germans. Each party recognized that any lasting alliance between the other two could render it impotent. Each was restrained by prudence, fear or formulated policy from too overt verbal or military attack on the other (Linden 1977:57;see Pottier 1989:430). 13 This marked the emergence of a fragile and mutable tripartite relationship of diverse but intertwined interests and objectives that would -considerable turbulence and recurrent crises notwithstanding -endure until the abolition of the monarchy in 1961.

The Churches and Ethnicity
A key to understanding the pivotal role played by the Catholic missionaries in framing the system of colonial rule in Rwanda revolved around their part in propagating the 'Hamitic vision of Rwandan society during the first decades of the twentieth century' (Carneya 2012a:174). John Hanning Speke's so-called 'Hamitic hypothesis' and the teachings of a range of 'minor evolutionary sociologists and physical anthropologists' (Linden 1977 (Linden 1977:52). Moreover, this ideological schema overlapped in important ways with the teachings at the Central Court itself. For the ideology of Musinga's court was founded on the conviction of Tutsi social and political superiority, an idea which had congealed into a form of 'ethnic' consciousness during the final years of Rwabugiri's violent and divisive rule. 14 Accordingly, the interests of the royal court were served by colonial ideological assumptions, which were in turn reinforced by the Catholic clergy. Des Forges argues that this accord between the 'political consciousness' of Rwandans and Europeans (Linden 1977 As Carney notes, 'colonial theorists developed their Tutsi paradigm from their experiences at the royal court' (Carney 2013:12), rather than imposing a ready-made, inflexible theoretical/ideological frame upon the kingdom.
The arrival of Protestant missionaries in the wake of the 1904 rebellion had the effect of consolidating this ideological accord (Longman 2010:41, 45). Since the new arrivals were viewed by the Catholic superiors -notably Vicar Jean-Joseph Hirth 15 and Vicar delegate Father Leon Classe 16 -as potential competitors at Court, and since they feared any German initiative to settle Protestant nationals to reinforce their claim to the territory, the missionaries were cautioned by superiors not to permit the Catholic Church to become too closely associated with Hutu (Longman 2010:46), a policy which institutionalized church support for the ruling elite.
14 Van Hoyweghen contests this interpretation, arguing that it was only during the late colonial period that 'social class… hardened into ethnicity' (Van Hoyweghen 1996:381). This interpretation is now widely rejected, as is the notion of pre-colonial Rwanda as a 'feudal state' composed of 'castes'. 15 Apostolic vicar of the Catholic region known as Nyanza Meridional (Longman 2010:39). 16 Vicar Apostolic since 1913.
Henceforth, the question of the political control of the state was bound to notions of 'ethnicity' and 'race' as the defining criteria of political rule, a principle that outlived colonial rule. 17 The advent of Belgian rule in 1916/7 did not alter the imperative of maintaining cordial relations with the royal court, which remained a linchpin of indirect rule, despite the fact that the court was deprived of some of its power by the Belgian authorities (Des Forges 2011:135). Conversely, although the Belgian authorities were fellow Catholics, they were compelled to reign in several priests who 'once again took on significant secular authority that challenged the power of local chiefs, adjudicating disputes and seeking economic advantages for the Hutu Christian converts' (Longman 2010:49). This may have reflected a contradiction in Catholicism 'between the egalitarian ideology of Christian brotherhood and the centrality to salvation of a hierarchically organized institution through which Grace flowed from the top downwards' (Linden 1977:2). It nonetheless also represented a delicate challenge for the Belgian authorities, who recognized the important role played by the Fathers in securing colonial rule, but who also realized that the mission's twofold objectives of ensuring 'the ascendancy of chiefs who supported the church' and furthering 'the cause of developing a [centralized] Christian government' (Longman 2010:50, 51) were not necessarily compatible.
This was to change following the conversion of the first Tutsi nobles in December 1917. The Catholic Church was now able to play a key role in helping to rid the kingdom of pockets of Hutu political power both in Central Rwanda as well as in newly occupied territories 'historically governed by Hutu kings'. These initiatives were complemented by a major administrative overhaul of the colony beginning in 1926, which rationalized the complex 'indigenous' system of chieftaincies. This policy removed remaining obstacles to Tutsi hegemony in Rwanda (Longman 2010:52). Moreover, in 1930 all inhabitants of Rwanda were issued with ethnic identity cards, an initiative accompanied by the codification of the principle of pat-17 It should be noted that J.J. Carney, for example, presents a somewhat more nuanced account of Church schooling during this period, arguing that prior to the transition from German to Belgian indirect rule in the late 1910s, Hutu and Tutsi had been educated together and that it was only in the 1920s that the Vicar Apostolic, Léon Classe, introduced a 'two-tiered educational system' which segregated students by ethnic group'. Still, according to Carney, even this policy did not mean that "Classe's understanding of the Hutu-Tutsi demarcation [was] as clear-cut as later commentators sometimes make it appear. Classe rejected the notion of inherent Tutsi intellectual superiority and hinted at the complexity of the Hutu-Tutsi distinction. 'I would say that the Tutsi are not, in general, more intelligent than the Hutu… Tutsi refers not to origin but to social condition'" (Carney 2012b:84-85, emphasis added). Carney's description of a more inclusive practice by the Church prior to Belgian rule would nonetheless not have altered the political equation from the point of view of the Court, which would still have had to contend with an early intervention that posed a challenge to the customary social relations within the kingdom by giving Hutu access to mission schools.
rilineality, which eliminated any possibility of social mobility between the groups (Longman 2010:65;see Hintjens 1999:249;Katongole 2005:72). The Church benefited from these changes in several respects. Firstly, education, which was largely provided by mission schools, 'became the portal which gave access to political power' (Linden 1977:152 see 283). Secondly, the establishment of parallel patrimonial networks increased the political influence of the churches within the kingdom. Finally, 'numerous institutional and personal links between church and state' integrated the Church into 'existing structures of power' (Longman 1998:56, 57).
The conversion of Tutsi gained momentum throughout the 1920s, but it was only with deposition of Musinga in 1931 and the coronation of his catechumen son Mutara Rudahigwa that the final obstacle to the evangelization of Rwanda was removed (Linden 1977:152). This heralded a period of unprecedented accord between the churches, the royal court and the colonial authorities, the 'end of the young Hutu Church', and a 'corporate recognition that the source of power within the State had shifted away from the mwami' (Linden 1977:173). It also marked a period of increased rationalization of administrative structures and a rigid approach to political education and social policies, which only nominally reflected the complex realities of Rwanda's social stratification.
As the Catholic Church in particular grew into the role of a state Church, this 'state' was as much Tutsi as it was Belgian (Linden 1977:174). This was not to change for as long as the Belgians favoured the Tutsi, and together with the Church, artificially maintained the racial stratification of Rwandan society (Linden 1977:186) whilst downplaying or ignoring 'important divisions of class, region, lineage, clan, and political faction' (Longman 2010:58;see 168, 169).
Nonetheless, there was not a straightforward symmetry between the institutions and objectives of the Church and the secular colonial authorities during this period. As Longman argues, the missionaries typically engaged in political struggles by developing an independent network of clientage relationships, frequently in competition with the Central Court and the chieftancies, on the one hand, and the colonial authorities, on the other hand. In other words, as the source of independent material benefits, the churches were transformed into quasi-political institutions and 'centers of political contestation' (Longman 2001:170). The politicization of the churches would, in turn, embroil them in the political conflicts of the late-colonial period.
Following the Second World War, a new generation of missionaries arrived in Rwanda, which included a group of progressive clergy influenced by social democratic principles, forming the bedrock of 'social Catholicism'. Troubled by the poverty and degradation of the Hutu masses they helped to foster a new Hutu 'coun-ter-elite', whose members would assume leading political offices following a peasant uprising in 1959 (Linden 1977:222;see Longman 2001:169;Van Hoyweghen 1996:381;Carney 2012a:86, 92) 18 that drove Tutsi chiefs from power, heralding the end of colonial rule as well as the abolition of the monarchy. 19 Carney qualifies this account, arguing that the political allegiance of the Church hierarchy was fractured and he contests the 'historical stereotypes' that 'portray the 1950s Catholic Church as moving fully into the pro-Hutu camp' (Carney 2012a:89, see 92;see Van Hoyweghen 1996:381). Still, shifting political terrain certainly influenced Church allegiances and by the close of the decade the Hutu-Tutsi question 'would come to dominate Rwandan ecclesial and state politics' (Carney 2012a:87, see 88).

The Churches in Post-Independence Rwanda
During the immediate post-independence period Hutu rapidly came to dominate the hierarchies of state and church. The dramatic shift to Hutu domination was nonetheless not the only important change. For the structure of political relations in the Rwandan state was also fundamentally transformed during the transition to independence.
Firstly, the abolition of the Tutsi monarchy ended the uneasy tripartite formation of political power and authority in the Rwandan state, which had been the defining feature of colonial indirect rule. With the elimination of the Central Court's ritualistic and secular power and authority, the churches were able to cultivate more direct ties with a nominally democratic and secular, Hutu ethno-nationalist central state. However, the new Hutu political elite was largely drawn from the ranks of the churches, many of who were recruited out of Catholic seminaries (Longman 1997:4). President Grégoire Kayibanda was himself a former editor of the influen-18 Although neither the Church nor the Belgian administration envisaged changes during the post-war period, a generation of new missionaries arrived in the 1940s who 'were born into a fast secularizing Europe' (Linden 1977:222) and the 'burgeoning emancipation of the working class and the growth of trade-unionism. Hence it felt it had a moral duty to speak out on social injustice. This view was opposed to the ideas of Classe, who saw society and structure as neutral and the individual as the safeguarder of morality. These two strands within Christianity have never been at equilibrium in the Rwandese Catholic Church. While Hutu abbés found the support of an emancipating social Catholicism, the Tutsi abbés on the other hand expressed anti-Belgian and anti-White Father feelings and developed a nationalist discourse which was eventually turned against them' (Van Hoyweghen 1996:381). 19 Longman asserts that a cynical interpretation of missionary support for revolution in the 1950s as a 'calculated strategy' rather than as a commitment to social justice is not without merit (Longman 1997:4;see Lemarchand 1978:364). Whatever the exact role of the Church, or more accurately the various factions within the Church, Hutu militias attacked Tutsi in mid-1959 and order was only restored by+ Belgian military intervention in November of that year, which was accompanied by the replacement of 'Tutsi political leaders with Hutu elites, helping to precipitate a longer-term ethno-political revolution that culminated with the inauguration of the Hutu-dominated First Republic in July 1962' (Carney 2012a:90).
tial Catholic newspaper Kinyamateka and personal secretary to Father Andre Perraudin (Longman 2010:70;see Van Hoyweghen 1996:381). As Carney argues, Kayibanda 'was the most prominent example of a seminarian turned journalist turned politician' (Carney 2012a:86). 20 Collaboration Between the church and the state existed at all levels of the hierarchy: hence often bishops were part of the prefecture council of development, the curies of the communal council of development, councils that prepared the programs of development in the country (Theunis in Longman 2001:77) Moreover, by shifting their allegiance to the ascendant Hutu majority, the churches perpetuated their longstanding participation in 'ethnic politics' rather than 'challenging the central principles at the root of Rwanda's ethnic conflict' (Longman 2010:66;see C. Newbury 1999:297). Indeed, the consolidation of the revolution, which would not have been possible without the allegiance of the Catholic Church, took place 'in a climate of ethnic purification' in which thousands of Tutsi were killed (Van Hoyweghen 1996:382). Secondly, however, despite the shift in the ethnic composition of the hierarchies of the churches, Tutsi still predominated in the local Protestant and Catholic parishes: This paradoxical attitude toward ethnicity produced a tension in the churches, since Hutu credited the churches… with supporting the Hutu accession to pow-er… and yet the church that they encountered (their priests and teachers) was Tutsi (Longman 2001:171;see Van Hoyweghen 1996:382) Thus whereas the political objectives or aspirations of the church converged with those of the state and a new dual institutional formation emerged, Tutsi remained embedded in a traditionally decentralized and quasi-autonomous institutional substratum. 21 The confluence of religion and politics assumed even greater complexity in 1973 when Kayibanda fell out of favour with the churches following a series of attacks 20 Van Hoyweghen notes an additional important element in regard to Kayibanda's accession. Kayibanda drew his support from Hutu in the southern provinces, a factor which underscores the importance of region in the politics of Rwandan history, a rupture which was underscored by his successor's source of support in the historically more autonomous territories of the north (Van Hoyweghen 2012:382). 21 Longman: the 'churches are loosely organized and decentralized institutions which can be exploited for diverse purposes… In the local community… churches represent an imposing presence, frequently offering more resources to distribute than the state' (Longman 2001:172). As such, local parishes are a 'major source of economic, social and political opportunity for people in the surrounding area' (Longman 2001:173) and the clergy reflected the fissions within society. The presence of the clergy thus frequently aggravated fissures within society.
on church schools and the lower ranks of the clergy, whose ethnic character recalled the widespread anti-Tutsi violence between 1959and 1965(Van Hoyweghen 1996. 22 The ethnic composition of the lower clergy now came to the fore as a complex challenge for the Church since it presented a direct threat to the political interests and institutional stability of the Church. Hence, although the Catholic Episcopal Conference 'eventually did issue a pastoral letter… condemning the ethnic nature of the attacks' (Longman 2010:87), it stopped short of condemning the specific targeting of Tutsi, despite the fact that the Church was directly affected (Longman 1997:85, see 6, 87). The formation of the Second Republic in 1973 led by President Juvinal Habyarimana following a military coup was widely welcomed by Church leaders in Rwanda. Habyarimana created a one-party state under the National Revolutionary Movement for Development, MRND with membership compulsory for all citizens. The state employed the conventional instruments of coercive force, ideological indoctrination and a nationwide system of patron-client ties to consolidate its power. Habyarimana moved rapidly to ensure the allegiance of the churches to the new regime. The Catholic Church responded by installing a loyal Hutu hierarchy and 'the Catholic Church moved the seat of its archbishop… to Kigali in 1976 to facilitate cooperation between the church and government' (Longman 2010:89). Church leaders, too, cultivated their clientage networks. However, rather than pursuing the past practice of maintaining relatively autonomous parallel tracks, Church and state entered into a 'mutually reinforcing' relationship of power and authority, which was in part based on personal links and cross-cutting institutional ties (Longman 2010:90 see 96) most especially during the late 1980s and early 1990s.
This trend, which further closed the gap between the institutions, was a continuation of the pattern that emerged following the abolition of the Tutsi monarchy in 1962. Ethnicity played a crucial role in the convergence of interests. For although Habyarimana initially appeared to adopt a more accommodative policy towards the Tutsi minority than did his predecessor, it soon became apparent that the linkage between the patrimonial clientage networks of the Church and the state respectively reflected a shared ethnic solidarity (Longman 2010:96), influenced by an emergent strain of radical ethno-nationalism (Carney 2012a:96).

Conclusion
During the years leading up to the mass violence and genocide in 1994, the cordial relations between the hierarchies of the Church and the state as well as the grow-ing authoritarianism of the Habyarimana government were challenged on several fronts.
Firstly, Habyarimana had to contend with a resurgent civil society and a movement for democratization (Longman 1995b:474). Secondly, a faction emerged within the Church influenced by liberation theologies. This grouping was sympathetic to demands for the democratic transformation of the state. Finally, pressure within the Church arose for the reform of its patrimonial structures and its close ties to the state. It was at this juncture that the interests of the secular and Church hierarchies began to coalesce around the goal of maintaining the status quo at any price: Just as political leaders felt threatened by democratization and turned to ethnic violence as a means of reasserting their control, church leaders found their own power under threat and found a similar appeal in a policy that forcefully defended the established structures of power (Longman 2010:96;see Van Hoyweghen 1996:384, 390) In other words, as pressure mounted in both Church and state to institute fundamental internal reforms as well as transform their historical relationship, both institutions embraced 'ethnic arguments as a means of regaining popular support' (Longman 2010:28;see Longman 2001:166). Longman links this development to the Church's subsequent role in the 1994 genocide, arguing that although the Church was not 'central to the planning of the genocide' (Longman 2010:28, his emphasis) it helped to 'make the genocide possible by encouraging obedience to authority and making ethnic prejudice seem consistent with Christian teachings' (Longman 2010:28).
It is not my objective here to provide my own account of the complex factors contributing to the events leading up to the 1994 genocide. Rather, I would like to highlight the impact, which the end of colonial rule had on the historical relationship between the Church and 'the state'. As we have seen, during the colonial period a threefold relationship existed between the royal court, the Church and the German administration, each of which was restrained by an 'unwritten entente' that limited their common objective of exercising a dominant political role in the colonial state. In other words, for as long as the royal court existed, each institution was 'bound' to the other in a relationship of mutually reinforcing interests governed by a pragmatic understanding of the structural limitations upon the pursuit of self-interest. For during this period, 'the state' consisted of a combination of the royal court and chieftancies on the one hand, and the colonial administration on the other hand. During this period the Church played an important mediating role between the ambitions of the royal house and the colonial imperative of containing the power but also supporting the authority of the Tutsi court as a condition of colonial indirect rule.