Military history

Chapter One

Defining the Crisis of Postcolonial Citizenship: Settler and Native as Political Identities

IN THE decade that followed African political independence, militant nationalist intellectuals focused on the expropriation of the native as the great crime of colonialism. Walter Rodney wrote How Europe Underdeveloped Africa. But no one wrote of how Europe ruled Africa. The great contribution of underdevelopment theorists was to historicize the construction of colonial markets and, thereby, of market-based identities.1 The popularity of political economy spread like a forest fire in the postindependence African academy precisely because it historicized colonial realities, even if in a narrowly economic way. Political economy provided a way of countering two kinds of anthropological presumptions embedded in various theories of modernization. The first was that identities in colonized societies were not grounded in historical processes. The second was that the beginning of a history for these societies was precisely colonialism, the point at which they were said to be historically animated by “cultural contact” with the West.2

The limits of political economy as a framework for political analysis became clear in the face of postcolonial political violence. For political economy could only explain violence when it resulted from a clash between market-based identities: either class or division of labor. From this point of view, political violence had to be either revolutionary or counterrevolutionary. In the face of political violence that cut across social classes rather than between them, and that was animated by distinctions crafted in colonial law rather than those sprouting from the soil of a commodity economy, explanations rooted in political economy turned arid. Animated by noneconomic distinctions, this violence was neither revolutionary nor counterrevolutionary; it was simply nonrevolutionary. It is this limit that seems to have provided an opening for a second coming of cultural explanations of political conflict.

Simply put, cultural theories claim that conflict arises from the difference between cultures. We are thus said to be in the grip of ethnic conflict locally and—more ominously—a clash of civilizations globally.3 As cultures and civilizations appear armor-plated, the colonial promise of cultural contact is said to have soured into postcolonial cultural conflict. If we are to come to analytical grips with the spread of nonrevolutionary political violence, I suggest we recognize that the process of state formation generates political identities that are distinct not only from market-based identities but also from cultural identities. Faced with a growing tendency to root causes of violence in cultural difference, however, the more pressing need is to differentiate between cultural and political identities, so as to distance oneself analytically from a growing culture-coded racism.

To focus on the construction of political identity is not to deny significant overlaps—or interrelations or even determinations—among cultural, economic, and political processes. No Chinese Wall exists between the political and other domains. Political identities may originate from the cultural or the economic domain.4 This much is clear from a reading of theorists of both social class and the nation-state. It was, after all, Marx’s prediction, and Lenin’s endeavor, that economic class would become political class through self-consciousness and self-organization. Similarly, all those who paid homage to the nation-state assumed that the cultural would become the political through the self-determination of the nation. As they heralded the “nation” in the Western world, anthropologists described “tribe” in Africa—like “caste” in India—as the cell form of social life. In both cases, they presumed a lack of historical dynamism. To understand how “tribe” and “race”—like “caste”—got animated as political identities, we need to look at how the law breathed political life into them.

Yet, no one historicized the political legacy of colonialism, of the colonial state as a legal-institutional complex that framed and set in motion particular political identities. The tendency was to discuss agency in an institutional void, by focusing on how it was harnessed to the colonial project; Marxists called the agents “compradors” and nationalists called them “collaborators.” Both bemoaned “tribe” and “tribalism” as a colonial concoction while assuming “race” and “racism” to exist as something real, in a positivist sense. Neither tried to historicize race and ethnicity as political identities undergirded and reproduced by institutions of colonial vintage—perhaps because neither had yet managed sufficient analytical distance from that legacy.


In an era when political crisis and civil war have come to be seen as ubiquitous, it is surprising how little academic writing there has been on the question of political identity.5 The tendency, rather, has been the opposite: to see political identity as derivative of either market-based or cultural identities. An earlier intellectual left tended to see political identity as derivative of market-based class identities such as worker and capitalist or landlord and tenant or merchant and peasant.6 The intellectual right had a habit of arguing that “real” identity was cultural, and that political identity was in reality an expression of cultural identity.7 The left had its verifying literature on class struggle and revolution, and the right had its counterpart literature on nationalism and tribalism.

The middle ground, where the two overlapped, was defined by the notion of the “nation-state.” Everyone—that is, everyone from Max Weber to V. I. Lenin—agreed that “self-determination” meant the right to one’s own state, and that the “self” in “the right of self-determination” was a cultural self. Even when Stalin orchestrated a Marxist consensus around the notion that the nation was not any cultural community, but one that had come to self-consciousness by virtue of a common economy, his point was to highlight the single market as a web of material relations knitting together a single cultural community.8 The bottom line was still that the nation was a common cultural community, and that self-determination made of it a political community. Cultural identity remained the bedrock of political identity.

The mainstream Marxist left assumed that market-based identities would come to self-consciousness as political identities within the context of the nation-state. That this did not happen was taken as evidence of either a false consciousness or a refracted consciousness. In the post–Cold War period, however, the focus of left analysis shifted from class movements (movements of workers, peasants) to social movements (movements of women, youth, ethnic and racial minorities). As the literature on class struggle gives way to that on social movements, it is no longer the right intelligentsia alone, but also many on the left, who now calls for self-determination for ethnicities. To that extent, too, there has been a growing tendency to presume that political identities either are or should be derivative of cultural identities.

If we are to understand the specificity of political crisis and the possibility of political action, we will need to distinguish political identity from both cultural and market-based identities. Political identities exist in their own right. They are a direct consequence of the history of state formation, and not of market or culture formation. If economic identities are a consequence of the history of development of markets, and cultural identities of the development of communities that share a common language and meaning, political identities need to be understood as a specific consequence of the history of state formation. When it comes to the modern state, political identities are inscribed in law. In the first instance, they are legally enforced.

If the law recognizes you as member of an ethnicity, and state institutions treat you as a member of that ethnicity, then you become an ethnic being legally and institutionally. In contrast, if the law recognizes you as a member of a racial group, then your relationship to the state, and to other legally defined groups, is mediated through the law and the state. It is a consequence of your legally inscribed identity. If your inclusion or exclusion from a regime of rights or entitlements is based on your race or ethnicity, as defined by law, then this becomes a central defining fact for you the individual and your group. From this point of view, both race and ethnicity need to be understood as political—and not cultural, or even biological—identities.

The tendency on the left has been to think of the law as individuating or disaggregating classes, and thus creating false identities. But the law does not just individuate; it also collates. It does not just treat each person as an abstract being—the owner of a commodity in the market, a potential party to a contract—it also creates group identities. Legally inscribed and legally enforced, these identities shape our relationship to the state and to one another through the state. In so doing, they also form the starting point of our struggles.

Political identities are the consequence of how power is organized. The organization of power not only defines the parameters of the political community, telling us who is included and who is left out, it also differentiates the bounded political community internally. This it does by acknowledging different kinds of identities in law. It is identities so acknowledged in law—and thus legally enforced—that form the basis of different political identities. Legal enforcement makes these identities the basis of participation in state-organized institutional and political life. By so doing, it encapsulates them, but without freezing them. Though legally enforced identities constitute the starting point of political action, they do not necessarily limit or even map the course of that political action.

It is not only political identities enforced by power, and thus defined “from above,” that are framed in legal definitions. These definitions also form a starting point for the forging of identities “from below”—for the simple reason that this forging “from below” is not in isolation but in contention with power. Even the most radical political action, such as action that accompanies the rise of insurgent identities, has to take as its starting point identities enforced by law, even if to break out of a legal straightjacket. This is why, whether officially enforced or insurgent, political identities need to be understood in relation to the process of state formation.

To sharpen the distinction between cultural and political identities, it will be useful to underline a point of contrast between cultural and political communities. More than anything else, a common cultural community signifies a common past, a common historical inheritance. In contrast, a political community testifies to the existence of a common project for the future. The distinction is often blurred because the past flows into the future, as it always does, creating a significant overlap between cultural and political communities. Yet, there are instances when there is a radical rupture between the two, as in the case of diasporic and immigrant communities. As we shall see, these instances most clearly illuminate the difference between cultural and political identities.

Even when defined as binaries in law, political identities are not always polarized. To understand the dynamic that polarizes political identities, we need to look at polarized identities as the end point of a historical dynamic, rather than positing them as its starting point. At the same time, it is when political identities do become polarized that they become most unlike cultural identities. Whereas cultural identities tend to shade into one another, with plenty of middle ground to nurture hybridity and ambiguity, there is no middle ground, no continuum, between polarized identities. Polarized identities give rise to a kind of political difference where you must be either one or the other. You cannot partake of both. The difference becomes binary, not simply in law but in political life. It sustains no ambiguity.

Every state form generates specific political identities. I shall illustrate this, first, with regard to two forms of the colonial state, characterized by direct and indirect rule. Direct rule tended to generate race-based political identities: settler and native. Indirect rule, in contrast, tended to mitigate the settler-native dialectic by fracturing the race consciousness of natives into multiple and separate ethnic consciousnesses. Once we have understood the dynamic whereby distinctive forms of the colonial state tend to generate distinctive types of political identities, we shall be in a position to understand the process of formation of Hutu and Tutsi as political identities through different periods in the history of the Rwandan state.


From the outset, modern Western colonialism presented itself as a civilizing project. In Kipling’s well-known phrase, colonialism was “the white man’s burden.” Said to signify the pinnacle of Western civilization, the modern state was considered testimony to the organizing genius of Western man.9 The architecture of the modern state was inscribed in modern law, Western law. And rule of law was in turn central to the construction of civilized society, in short, civil society.

Wherever in the non-Western world the white man carved out colonies, the civilizational project was marked by a turn-key import: Western law. At its outset, the Western colonial project was no less than to wipe clean the civilizational slate so as to introduce Western norms through Western law; modernization would have to be Westernization. Whatever the differences in practice between colonial powers, one single claim defined a shared civilizational project: whether rulers or ruled, Westerners or non-Westerners, all those subject to the power of the state would be governed through imported Western law. To be sure, there were discriminations that set the colonizers apart from the colonized, and even different groups of colonized apart from one another, within this single universe of modern law. The legal basis of group discrimination was race.

The shift from direct to indirect rule marked the first major retreat from this shared civilizational project. In contrast to the single legal universe of direct rule, indirect rule constituted separate legal universes. In addition to a racial separation in civil law between natives and nonnatives, as under direct rule, indirect rule divided natives into separate groups and governed each through a different set of “customary” laws. Every ethnic group was now said to have its own separate set of “customary” laws, to be enforced by its own separate “native authority,” administering its own “home area.” Thereby, the very category “native” was legally dismantled as different groups of natives were set apart on the basis of ethnicity. From being only a cultural community, the ethnic group was turned into a political community, too. The political project of the regime of “customary” laws was to fracture a racialized native population into different ethnicized groups. The basis of group distinction under indirect rule was both race and ethnicity.

The shift from direct to indirect rule was eventually made by every colonial power.10 To understand the sort of compulsion that made for the shift, one needs to reflect on the astonishingly arrogant claim with which modern colonialism announced itself to the world at large. Under direct rule, the colonial state justified the hierarchy of colonial society as a civilizational imperative. Within a single legal order—one based on modern law—it distinguished a political minority from a political majority. The language of the law tried to naturalize political differences in the colony by mapping these along a civilizational ladder. As the litmus of a civilizational test, the law separated the minority of civilized from the majority of those yet-to-be-civilized, incorporating the minority into a regime of rights while excluding the majority from that same regime.11 The law thus enfranchised and empowered as citizens the minority it identified as civilized, and at the same time disempowered and disenfranchised the majority it identified as yet-to-be-civilized. The unintended consequence of direct rule was to produce a bipolar identity between the colonizer and the colonized and to mark this difference by race. Its tendency was to divide colonial society into two racialized groups and thereby to turn race into the primary difference between these groups.

It is this unintended consequence that made for a fundamental crisis of direct-rule colonialism. As those excluded from the regime of rights made sense of the basis of their exclusion, they tended to organize along racial lines, more or less submerging all other differences as secondary. The more this marker of colonial civilization was turned into a standard for the anticolonial struggle, the more race turned into an anathema. This development posed a new question for colonial rule: how to dismantle and fragment race, and thus the colonized majority, into several political minorities. Indirect rule was the response to this dilemma. Alongside race, indirect rule introduced another political marker: ethnicity. Instead of treating the colonized as a single racialized mass, indirect rule sliced them over, not once but twice. The first division separated the nonindigenous—governed through civil law as nonnatives—from the indigenous, the natives. The second division sliced the natives into so many separate ethnicities. With each ethnicity governed through its own “customary law,” the plural legal order among the colonized not only produced plural political identities—as ethnicities—it also claimed that they in turn reflected just as many preexisting cultural identities.

The cultural policy of the colonial power augmented the distinctions written into colonial law. As in law, so in culture, policy in the civic sphere was based on the principle of identity and discrimination. The civic sphere was culturally assimilationist, with the culture of the colonized considered the hallmark of civilization, the destiny of all those considered capable of being civilized, and thus becoming citizens. In contrast, cultural policy in the customary sphere was based on the principle of difference: it claimed to reproduce “ethnic” difference through an ethnicized “customary” law, not to soften “racial” difference through the application of a uniform civil law. In its legal and cultural policy, the civic sphere was characteristic of direct-rule colonialism, and the customary sphere of indirect-rule colonialism. The amalgam of the two in the actual colonial experience in twentieth-century tropical Africa supplemented racial exclusion in the civic sphere with ethnic fragmentation in the customary sphere, thereby combining direct rule in the civic sphere with indirect rule in the customary sphere.

Said to spring from biology and culture, the difference between races and ethnicities involved two kinds of claims: one about hierarchy, the other concerning diversity. Claims about cultural difference and hierarchy were at the same time turned into foundation stones for building and reproducing legal and political differences. Culturally, races were said to be a civilizing influence and ethnicities in dire need of being civilized. Whereas race claimed mainly to reflect a civilizational hierarchy, ethnicity was said to be primarily about a cultural diversity. Neither claim excluded the other. Thus, the discourse on hierarchy did not preclude a statement about the diversity of unequal races. Similarly, the discussion on ethnic diversity did not mean that some ethnic groups (such as those with an internal hierarchy and a state) were not considered more civilized than others (usually those considered stateless). Yet, as representation, race was vertical but ethnicity horizontal. Legally, races were ruled through a single law: riddled with racially discriminatory provisions, civil law was an amalgamation of imported law and local initiative under colonial conditions. In contrast, each ethnicity was ruled through a separate set of customary laws. While civil law discriminated against the lower races in the civil sphere, customary law discriminated against ethnic strangers in the ethnic sphere. Politically, ethnicities lived under separate Native Authorities, each reinforcing its version of customary law in its own ethnic homeland, whereas races lived under a single civic authority. Governed under a single law, living under institutions designed for cultural assimilation, and subjugated to a single administrative authority, races were meant to have a common future—but not so ethnicities. In reality, the distinction between race and ethnicity illuminated the political difference between the nonindigenous and the indigenous.

What did it mean to be constructed as a race as opposed to an ethnicity? In the African colonies, only “natives” were said to belong to ethnic groups. “Nonnatives” were identified as races. While ethnicities were said to be indigenous, races were presumed to be nonindigenous. Ethnicity was said to mark an internal difference among those constructed by colonial law as indigenous to the land. Race marked an external difference, a difference with others, those legally constructed as nonindigenous. Through its discourse on race and ethnicity, the colonial state tried to naturalizepolitical differences, not only between the colonizer and the colonized, but also—and this is the important point here—between two kinds of colonized: those indigenous and those not.

Subject Races as Virtual Citizens

Race was about nonnatives and ethnicity about natives; yet, the legal distinction between nonnatives and natives was not quite the same as the political difference between the colonizer and the colonized. By making a distinction between two kinds of colonized groups, those indigenous and those not, the legal system blurred the colonial difference rather than illuminating it. This it did by highlighting the commonality between the colonizer and a minority among the colonized: that both were nonindigenous. It thus created a contradictory middle ground between the colonizer and the colonized. This middle ground was occupied by subject races, those from the colonized who were identified as nonindigenous.

Though all races were presumed to be a civilizing influence, some were said to be more so than others. We have seen that the hierarchical distinctions between races were expressed in civil law as so many internal discriminations. Instead of a universal citizenship, civil law gave rise to different categories of citizens. Alongside the master and colonizing race, the law constituted subject races. While members of the master race were the only full citizens in the colony, members of subject races were virtual citizens, deprived of rights of citizenship, yet considered to have the potential of becoming full citizens. Though colonized, they came to function as junior clerks in the juggernaut that was the civilizing mission. Without being part of colonial rulers, they came to be integrated into the machinery of colonial rule, as agents, whether in the state apparatus or in the marketplace. As such, they came to be seen as both instruments and beneficiaries of colonialism, however coerced the instrumentality and petty the benefits. Though part of the colonized population, the subject races received preferential treatment under the law. In contrast, subject ethnicities were set apart, and literally sat upon, legally. They were the core victims of colonial rule.

The subject race experience was marked by both petty privilege and petty discrimination. Elevated to the point that they were governed through civil law, subject races were at the same time the target of specific forms of racial discrimination under the same law. Theirs was a contradictory experience: on the one hand, they were elevated above natives and treated as virtual citizens, part of the hierarchy of civilized races; on the other hand, they were subjected to racial discrimination, which emphasized their position in the lower rungs of that hierarchy. They thus became the source of both collaborators with the colonial enterprise and nationalists who agitated against it. The legal and political distinction that cut through the colonized—native subjects and subject races, the majority ethnicized and the minority racialized—evokes the distinction that Malcolm X drew between the House Negro and the Field Negro in a different historical circumstance. As one with access to petty privilege and preferential treatment, the House Negro was prone to identifying with the master: as a clone who would mime “we sick” if the master was sick. And so did many among the subject races.

The subject races of colonial Africa were many. The best known of these were the Asians of East Africa, the Indians and “Coloureds” of South Africa, the Arabs of Zanzibar, and the Tutsi of Rwanda and Burundi. Historically and culturally, they represented a mishmash. The Asians of East Africa, like the Indians of South Africa, were obviously nonindigenous in origin. But South African “Coloureds,” Zanzibari “Arabs,” and Tutsi of Rwanda and Burundi were not at all obviously alien. The category “Arab” included both those ancestrally Arab and those culturally Arab, both immigrants and children of the soil. To the extent that one could become an Arab, Arab was more of a hybrid identity, as was obviously the identity “Coloured.” But the law brooked no hybridity, or ambiguity. At the other end of this list of subject races were the Tutsi. Though wholly indigenous to Africa, we shall see that the Tutsi were constructed by colonial ideology as well as law as nonindigenous Hamites. The claim that the Tutsi—or, for that matter, the Arabs or the “Coloureds”—were nonindigenous, part of the hierarchy of nonindigenous races, apart from the Bantu who were said to be indigenous, needs to be understood more as a legal and political construct than as representation of a historical and cultural reality.


The political legacy of indirect-rule colonialism in Africa was a bifurcated state: civic and ethnic, the former governed through civil law and the latter through customary law. But the colonial legacy went beyond legal pluralism to a set of institutionally entrenched discriminations: civil law was racialized and customary law ethnicized. At the basis of two different political identities, race and ethnicity, were two sets of discriminations. Racial discrimination in the civic sphere reproduced race as a political identity, just as ethnic discrimination in the customary sphere translated ethnicity from a cultural to a political identity. This twofold discrimination, civic and ethnic, became the basis of a distinction between two types of citizenship in the postcolonial period: civic and ethnic.

Civic and Ethnic Citizens

Civic citizenship is a consequence of membership of the central state. Both the qualifications for citizenship and the rights that are its entitlement are specified in the constitution. These rights are mainly individual and are located in the political and civil domain. In contrast, ethnic citizenship is a result of membership in the Native Authority. It is the source of a different category of rights, mainly social and economic. Further, these rights are not accessed individually but by virtue of group membership, the group being the ethnic community. The key socioeconomic right is the right to use land as a source of livelihood.

Ethnic citizenship does not just evoke a cultural difference. It has material consequences also. A civic citizen may acquire land—like any other material good—only through a market transaction, by purchasing it. For a civic citizen, a kin-based property transaction is more or less limited to inheritance. But an ethnic citizen can claim land as a “customary” right, a kin-based claim that is a consequence of membership in an ethnic group. It is also how most peasants access land in a Native Authority. The immediate practical consequence of being defined a citizen of nonindigenous origin is this: since they do not have their own Native Authority, nonindigenous citizens are denied “customary” access to land. No wonder a land-poor peasant sees the struggle for land as part of a struggle for ethnic belonging.

The postcolonial stranger was racial in the civic sphere, and ethnic in the customary sphere. When you said you were citizen of, say, Uganda or Congo, the claim was that you were a citizen of the central state, a civic citizen. Your rights as a civic citizen were specified in the constitution of the republic. But it did not necessarily mean that you were an ethnic citizen. To be an ethnic citizen was to belong to one of the ethnicized Native Authorities so as to claim “customary” rights. Only those considered ethnically indigenous to the republic could claim ethnic citizenship and thus a Native Authority as an ethnic home. In the indirect-rule state, citizens were divided into those indigenous and those not. The former enjoyed both civic and ethnic citizenship; the latter were only civic citizens. At the same time, whereas civic citizenship could be accessed individually, ethnic citizenship could only be exercised as a group.

Who qualified as indigenous? The colonial state considered as indigenous all those who were resident on the territory it seized at the time of colonization, and only those. Anyone who came after was treated as a stranger: if they were indigenous to Africa, they were racially branded as “native” but ethnically as strangers; if they were from outside Africa, they were considered nonnative in race but—and this is the important point—were not ascribed an ethnic identity in law. Most states in Africa continue to adhere to this claim, considering as “native” only those who were present on native soil at the time of colonization, with all others considered nonindigenous. The irony is that for a postcolonial state to make this claim is to uphold the colonial state as its true parent.

Ethnic Citizenship as a Native Prerogative

Given this legacy, any power with a democratic state project in postcolonial Africa confronted a dual task: first, simultaneously to deracialize civil power and deethnicize customary power; and then, to join the two spheres in a single authority. Without creating an undivided civic authority, it would not be possible to create a single and unified citizenship.

Given this monumental task, it is not surprising that the postindependence story is mixed. On the positive side, no independent state was content to reproduce the colonial legacy wholly unreformed. Everywhere, the world of the racialized citizen and the ethnicized native changed after independence. All postindependence regimes were determined, to one degree or another, to do away with the stigma of race they associated with colonial rule. The tendency of the postcolonial state was to deracialize civic identity. Civic citizenship ceased to recognize any difference based on race or place of origin. To deracialize civic citizenship was to end the legal prerogative of the settler. That struggle continued right up to the southern tip of the continent, to apartheid South Africa. The end of apartheid marked an African achievement because it set the future of Africa apart from that of the Americas. Whereas the Americas remain a land of settler independence, nowhere in Africa has settler power survived the end of colonialism.

That is where similarities ended, and differences sprouted among different kinds of postcolonial reform agendas. On the negative side, the conservative variant of the postcolonial state—also its mainstream—continued to reproduce the native identity as ethnic, while enforcing an ethnic version of law as “customary.” The irony was that deracialization without deethnicization continued to reproduce a bifurcated citizenship. So long as Native Authorities continued to function with an ethnically defined membership, the state continued to make a distinction between two kinds of citizens: the ethnically indigenous and ethnic strangers. Even if the civic sphere ceased to make a distinction between citizens who were indigenous and those who were not, the ethnic sphere continued to make this distinction. Indigenous citizens continued to have an ethnic home (a “Native Area”) governed by an ethnic administration (a “Native Administration”). Thus, the conservative variant of the postcolonial state accepted as “authentic” the colonial construction of the native: as an ethnic being ruled by a patriarchal authority with an authoritarian and unchanging custom that needed to be enforced officially as “customary” law. Conservative postcolonial power thus replaced the settler’s prerogative with the native’s prerogative. The outcome, however, was proof enough that you could not do away with settler identity without also doing away with native identity, for settler and native were as Siamese twins born of the same colonial parent. Even if conservative nationalism turned the world designed by the settler upside down, it did not change it.


If the anticolonial struggle was about deracializing the state, the postcolonial debate was about deracializing civil society. If the anticolonial struggle was preoccupied with the question of rights across racial boundaries, the postcolonial preoccupation was with justice and entitlement. If deracialization meant political equality between erstwhile settlers and natives, justice meant nothing less than a turning of the tables at the expense of the settler and in favor of the native. If to deracialize society was to achieve a measure of social equality between settler and native, the state would have to assert the prerogative of the native at the expense of the settler.

Defining Settlers

The struggle for justice, for redress of colonial wrongs, raised afresh the question of colonial power and the distinctions it made in society. What kinds of differences in civil society were legitimate and what kinds were not? Which differences were directly and predominantly an effect of colonial power asserting the settlers’ prerogative, and which ones an effect of market differentiation? In each case, the answer turned around a more basic question: Who is a settler? For unlike “native,” “settler” was not a legal identity. Everywhere, the law spoke of “natives” and “nonnatives,” not of settlers. Settler was an insurgent assertion, a libel hurled back by natives at the core beneficiaries of colonial rule. Settler was a political identity, of those identified with the conquest state. This is why there was never one single definition of “settler,” but as many as there were perspectives among natives. As nationalism differentiated—between the narrow and the inclusive, the cultural and the political, the reactionary and the progressive, and so on—each tendency arrived at a different understanding of “settler” as a political identity. As this process unfolded, the contradictory middle ground occupied by subject races turned into a political battlefield shaping the political future of anticolonial nationalism. What was at the heart of the settler experience: immigration or conquest? Were the settlers only those directly linked to colonial power and thus its core beneficiaries, or did they include all those legally constructed as nonnatives and thus legally entitled to preferential treatment, no matter how petty the preference? Was the key political difference between colonizer and colonized, or between indigenous and nonindigenous? If settlers were created by conquest and reproduced through a form of the state that enforced a settler’s prerogative, then the very abolition of that prerogative—and of the state that enforced it—would deflate settlers and natives as political identities.12 But if settlers were created by migration, then nothing less than repatriation would erase the settler identity.

My point is that every nationalist movement was called upon to determine its attitude to the subject races; similarly, those occupying the middle ground found themselves having to declare their preference for future membership of a political community. The variety of answers on both sides testified to a dynamic that rapidly differentiated nationalism. In the range of answers that emerged during the anticolonial movement of the late fifties, two leaders—Nyerere in Tanzania and Kayibanda in Rwanda—marked the extremes. Nyerere stood for a single unified citizenship, both deracialized and deethnicized.13 Kayibanda championed a racialized nationalism—of the Hutu—built on the very political identities institutionalized by colonialism: Hutu and Tutsi.14 Tanzania came to be a paragon of political stability in the region, the one postcolonial state that did not turn entire groups into refugees. Rwanda signified a postcolonial pursuit of justice so relentless that it turned into revenge as it targeted entire groups from the previously colonized population, groups it first victimized and turned into refugees, and later annihilated.

Political Identity and Political Violence

By politicizing indigeneity, the colonial state set in motion a process with the potential of endlessly spawning identities animated by the distinctions indigenous and nonindigenous, and polarizing them. This indeed set the context in which political violence unfolded in Africa, colonial as well as postcolonial. The starting point of that violence was colonial pacification, which took on genocidal proportions where settlers set out to appropriate native land, as in southern Africa and Congo. Settler colonies continued to be the main focus of political violence during the anticolonial struggle, though the initiative shifted from the settler to the native. While it has been widely noted that Africa’s most violent anticolonial struggles unfolded against the master race in the settler colonies, few have noted that Africa’s worst postindependence violence targeted its subject races: the Tutsi in Rwanda in 1959, the Arabs in Zanzibar in 1963, the Asians in Uganda in 1972, and finally, the Tutsi in Rwanda in 1994. And even fewer have noted the line along which the settler/native dialectic has unfolded in postcolonial Africa, as it has moved from targeting racial strangers in the civic sphere to targeting ethnic strangers in the Native Authorities.

If the master races of colonial Africa were the first group to be defined as settlers, the subject races were the second, and the ethnic strangers in Native Authorities the third, to be targeted as settlers. The more the native/settler dynamic proliferated, the more groups of settlers it created: first the master race, then the subject races—both racial strangers—and finally ethnic strangers from the erstwhile native population. The more this happened, the more the nationalist movement built up through the anticolonial struggle unraveled in the postcolonial period. It is conservative nationalism, with its nativist notions of political identity, that was intent on branding every immigrant, whether racial or ethnic, as a “settler.” Victimized in 1959–63, 1973 and 1991–94, the Tutsi of Rwanda belonged to the second category of postcolonial settlers, whereas the Banyarwanda expelled from Uganda in 1982–83 and 1990 and the Banyamulenge of Congo, targeted in 1994, belonged to the third and last category of postcolonial settlers. As we shall see, the case of the Tutsi—considered racial strangers in Rwanda and ethnic strangers in Uganda and Congo—brought together both outcomes in an explosive combination.


My analysis of Hutu and Tutsi as identities differs from the mainstream literature on Rwanda in two important ways. First, whatever other disagreements they may have, historians and political analysts of Rwanda have been preoccupied with finding a singleanswer to the question: Who is a Hutu and who is a Tutsi? In contrast, I argue that Hutu and Tutsi have changed as political identities along with the state that has enforced these identities. There cannot therefore be a single answer that pins Hutu and Tutsi as transhistorical identities. Second, unlike those preoccupied with thesearch for origins—whether biological or cultural—of Hutu and Tutsi, I argue that the clue to Hutu/Tutsi violence lies in two rather contemporary facts. The origin of the violence is connected to how Hutu and Tutsi were constructed as political identities by the colonial state, Hutu as indigenous and Tutsi as alien. The reason for continued violence between Hutu and Tutsi, I argue, is connected with the failure of Rwandan nationalism to transcend the colonial construction of Hutu and Tutsi as native and alien. Indeed, if anything, the revolutionaries of 1959 confirmed the Tutsi minority as aliens and the Hutu majority as natives—in the well-known phrase, Rwanda nyamwinshi15—finally and rightfully come to power.

A Halfway House

Colonial Rwanda was a halfway house between direct and indirect rule, combining features of both. Like elsewhere in colonial Africa, in Rwanda too, Belgian power constructed “customary law” and “Native Authorities,” alongside civic law and civic authorities. But, unlike elsewhere in Africa, neither this law nor this authority were ethnicized. After the Belgian colonial reform of 1926–36, Hutu were not ruled by their own chiefs, but by Tutsi chiefs. The same reforms constructed the Tutsi into a different race: the Hamitic race. This made for two important differences with indirect-rule colonialism. One, the bulk of the colonized population was not fragmented along ethnic lines into so many ethnically diverse identities, each with its own “customary” law and enforcing authority; instead, they were made into a single mass—the Hutu, said to be indigenous Bantu—who cut across all Native Authorities. Two, this Bantu majority was not ruled through their own chiefs but through those constructed as racially different and superior, the Hamites. Unlike indirect rule elsewhere, but like British rule in Zanzibar, the colonial state in Rwanda produced bipolar racial identities, and not plural ethnic identities, among the colonized. A single binary opposition split the colonized population into two: anativized majority opposed to several nonnative minorities. In Zanzibar, this opposition pitted African against Arab and Asian, not just British power; in Rwanda, it pitted Hutu against Tutsi, not just Belgian power.

This difference cannot be explained as either a simple carryover from the precolonial era or its inevitable consequence. It has to be seen as a specific outcome of the articulation between precolonial and colonial institutions and ideologies. We shall see that while Tutsi privilege had its genesis in the period immediately preceding colonialism—it was not unlike Hima privilege in the neighboring kingdom of Ankole that became part of colonial Uganda—the difference between Uganda and Rwanda in forms of colonial rule made for a world of difference between the postcolonial future of the Ugandan Bahima and that of the Rwandan Tutsi. Nor can it be seen as an inevitable effect of colonial racism, the ideological perversion called the “Hamitic hypothesis.” To account for the difference that Belgian colonialism made to Rwanda, as opposed to Congo, we shall need to consider not simply its ideological, but also its institutional impact.

The Hamitic hypothesis was not articulated with reference to Rwanda only. In fact, it claimed to explain all signs of civilization in Bantu Africa—from monotheism to the use of iron and other material artifacts to the development of statecraft. Not only the Tutsi, but also the Bahima and the ruling stratum in Baganda, too, for example, were considered Hamites. The important point is that only the Tutsi—and not the Bahima elite, nor the Baganda, nor any other group considered Hamitic—were constructed as a race as opposed to an ethnic group. Only in Rwanda and Burundi did the Hamitic hypothesis become the basis of a series of institutional changes that fixed the Tutsi as a race in their relationship to the colonial state.


More than any other, the Rwandan genocide brings two events into sharp focus: the 1959 “social” revolution and the 1990 RPF invasion. More than ever before, a postgenocidal analysis demands that we take a second, and a third, look at each so that we may draw from it lessons appropriate to a postgenocide politics.

I will argue that both events are indicative of a deep-seated crisis: 1959 signals a crisis of subaltern nationalism, and 1990 a crisis of postcolonial citizenship. The promise of 1959 turned sour not because 1959 was a false revolution, as proponents of “Tutsi Power” often allege. It turned sour in spite of real social gains, because 1959 repudiated only the consequences of colonial rule, but not the native/settler dynamic that was its institutional premise. Instead of pioneering a way beyond colonially shaped identities and destinies, 1959 locked Rwanda’s fate within the world of political identities constructed by colonialism. Instead of the promised first act in a revolutionary drama that would close the curtain on the colonial era, 1959 turned into a final act desperately trying to breathe life into racialized identities born of the colonial state. Indeed, 1959 ushered in a pursuit of justice so focused that it turned into revenge.

While the crisis of nationalism can be made sense of in a single-country context, the crisis of citizenship cannot be grasped fully outside of a regional context. The RPF invasion of Rwanda from Uganda was the first signal that the crisis of citizenship had indeed taken on a regional dimension. The 1990 invasion, I argue, needs also to be understood as an armed repatriation of Banyarwanda refugees from Uganda. It signified a citizenship crisis on both the Ugandan and the Rwandan sides of the border. The RPF’s crossing of the Uganda-Rwanda border in 1990 had a double significance: on the one hand, the National Resistance Movement (NRM) in Uganda exported its first political crisis since coming to power to Rwanda; on the other, the postrevolutionary Second Republic in Rwanda got its first taste of revolutionary chickens returning home. Neither the 1990 invasion nor the dynamic that led to the 1994 genocide can be understood outside of a regional context.

Specifically, a regional approach has three advantages. First, it allows us to contrast the process of identity formation in colonial Rwanda with some aspects of identity formation in Congo and Uganda. The salient political identities in Rwanda were Hutu and Tutsi; but in Congo and Uganda, Hutu and Tutsi tended to belong to a single political identity, the Banyarwanda. While Hutu and Tutsi were exclusively political identities, Banyarwanda was a cultural identity that also became political in the context of indirect-rule colonialism in Uganda and Congo. While Hutu and Tutsi in Rwanda were racialized identities—Tutsi as nonindigenous Hamites and Hutu as indigenous Bantu—Banyarwanda was an ethnicized identity in Congo and Uganda. The consequence was that while the Tutsi were considered racial strangers in colonial and postcolonial Rwanda, the Banyarwanda were considered ethnic—but not racial—strangers in colonial and postcolonial Uganda and Congo.

This does not mean that the identity Banyarwanda did not exist on the Rwandan side of the border, or that, as they crossed to the Ugandan or Congolese sides of the border, migrants and refugees somehow ceased to think of themselves as Hutu or Tutsi and miraculously began to think of themselves as Banyarwanda only. To be sure, there were always multiple identities on all sides of the border. The same persons came to see themselves as both Hutu (or Tutsi) and Banyarwanda. And yet, the important fact—the politically salient fact—is that the law considered this same person as Hutu (or Tutsi) on the Rwandan side of the border but as Munyarwanda on the Congolese or Ugandan side of the border.

Second, the regional approach not only makes for a comparative understanding of the crisis of postcolonial citizenship, it also allows for a comparative insight into initiatives that tried to address this crisis. These initiatives came from both above and below. The initiatives from below came from Congo and Uganda, contexts where the Banyarwanda were branded ethnic strangers even if they were granted civic citizenship. Defined as ethnic strangers, the Banyarwanda of Congo and Uganda sought to make a political home where they were resident. But the initiatives they took differed. Those in South Kivu pressed for an ethnic home (Native Authority) by redefining their identity from Banyarwanda to Banyamulenge, from ethnic strangers to an ethnically indigenous community. In contrast, the political refugees in Uganda who were the spearhead of the National Resistence Army (NRA) guerrilla struggle went so far as to challenge the very system of Native Authorities by redefining the very basis of rights from ethnicity to residence. Theirs was indeed a radical attempt whose objective was to deethnicize citizenship.

The initiative from above came from the Habyarimana government in Rwanda and the Museveni government in Uganda, two regimes that were often publicly at loggerheads with each other. Unlike the postrevolutionary government of Kayibanda (also known as the First Republic), which continued the racial branding of Tutsi as Hamites, the Second Republic identified with Habyarimana sought to soften Hutu/Tutsi relations by redefining Tutsi from a race to an ethnicity. But while Habyarimana was willing to remove the stigma of being alien from those Tutsi living in Rwanda, thereby even letting them participate in the political sphere, he was not willing to grant the same opening to the political diaspora of exiles and refugees spread throughout the region, and indeed the globe. The limits of the Habyarimana initiative were that the Tutsi resident outside Rwanda must find a way to regularize their stay outside. That way was indeed found with the NRA reforms that redefined the basis of rights from ancestry to residence. It is indeed this reform that the Museveni government translated into law, both when it changed the legal requirement for civic citizenship from two generations to ten years of residence, and when it replaced the very notion of ethnic citizenship by residence-based rights within local councils throughout the country. But the first major political crisis of the Museveni government also brought this very reform into question. The price of surviving that crisis was to repudiate the very reform that had promised a future beyond colonially crafted political horizons: the exiles of 1959 found their new citizenship no more than a paper promise. Thrown back to an exile status, the Tutsi guerrilla fighters in the Uganda state army, the National Resistance Army, found themselves between the Rwandan devil and the Ugandan deep sea. The invasion of 1990 was their attempt to escape the closing scissors of a postcolonial citizenship crisis in Rwanda and Uganda.

Finally, a regional approach allows us to understand the regional consequences of the Rwandan genocide. It will allow us to see how the cross-border passage from racialized to ethnicized identities, and vice versa, has made the Banyarwanda diaspora—first the Tutsi and then the Hutu—the most volatile of all diasporic networks in the region. If it were the tensions of the Ugandan civil war and its aftermath that spilled over into Rwanda in October 1990, we shall see that it was the tensions of the Rwandan civil war and genocide that spilled into Kivu that same decade. Somewhat like the Zionist state born of the Holocaust, the diasporic state born of the genocide saw itself as morally accountable for the welfare of all surviving Tutsi globally. One consequence of the genocide was to dissolve the Banyarwanda ethnicity in North and South Kivu into the by-now volcanic crucible of Rwandan politics. Out of that process of dissolution emerged ultimately hostile Hutu and Tutsi partisans.

More than any other development, the Rwandan genocide is testimony to both the poisoned colonial legacy and the nativist nationalist project that failed to transcend it. Written with an eye on both developments, this book is concerned both with the nature of political identities reproduced by the colonial state and with identity formation in the postcolonial period. To understand how the colonial and the postcolonial articulated politically, I ask questions such as the following: To what extent did the identities implanted by colonial institutions get reproduced in the nationalist movement, and to what extent did the nationalist movement and the postcolonial state manage to transcend the colonial institutional/ideological legacy?

The Banyarwanda diaspora is today the largest and the most active of all diasporic networks in the region of the African Great Lakes. Infected by civil war and genocide in Rwanda, it has over the past decade split into two distinct and antagonistic networks: one Hutu, the other Tutsi. In this form, Hutu and Tutsi identities have functioned like so many carriers that have taken the virus of political conflict from one side of the border to the next. If the Hutu demand democracy, a recognition that they are the political majority, the Tutsi demand justice, a claim that the right to life must precede any recognition of a political majority. Is it possible to reconcile these seemingly conflicting demands: democracy and justice? To do so, I argue, requires going beyond the notion of victors’ justice to that of survivors’ justice. Without a notion of justice appropriate to a postgenocide situation, it will not be possible to construct a political identity other than Hutu and Tutsi in postgenocide Rwanda.

If the Nazi Holocaust was testimony to the crisis of the nation-state in Europe, the Rwandan genocide is testimony to the crisis of citizenship in postcolonial Africa. But if the Nazi Holocaust breathed life into the Zionist demand that Jews too must have a political home, a nation-state of their own, few have argued that the Rwandan genocide calls for the building of a Tutsi-land in the region. While Europe “solved” its political crisis by exporting it to the Middle East, Africa has no place to export its political crisis. That need not necessarily turn into a dilemma. If taken hold of, it can be an opportunity to marshal Africa’s resources to address the crisis of citizenship in postcolonial Africa. I aim to probe that possibility in the hope that life must be possible after death—why this book is more about politics than about history, why it is ultimately a book that wrestles with the question of political reform after political catastrophe.

If you find an error please notify us in the comments. Thank you!