Managing Ethnicity in African Politics
Summary and Keywords
Political regulation of ethnicity has been a core dimension of state-building in Africa, and a set of different macro-political strategies was applied in African postcolonial states to deal with ethnic heterogeneity. One set of strategies consisted in attempts to completely eliminate political manifestations of ethnicity, violently through genocide (Rwanda, 1994) or mass expulsions of ethnic minorities (Uganda, 1973), consensually through secession of autonomous provinces (Eritrea, 1993; South Sudan, 2005), through legal instruments that ban the political expression of ethnic identity such as party bans, or via coercive variants of assimilation (Rwanda, 2001). An opposing option promoted the formal recognition of ethnicity through consociationalism (Burundi, 2005), ethnic federalism (Ethiopia, 1995), ethnic minority rights (Mauritius), or hegemonic control (apartheid South Africa).
Many African countries have instead opted for an informal accommodation of ethnic identity in politics, which combines the pursuit of civic nationalism and ethnic party bans with a de facto recognition of ethnic group rights through informal power-sharing, centripetal institutions, or variants of federalism which shift resources and competencies to subnational levels. The choice of strategies is, however, constrained by how interethnic relations have been shaped in the process of postcolonial state-building. Both strategies of elimination and of formal recognition are applied in ranked societies where one racial or ethnic group managed to take control of the state and in which class corresponded with ethnic affiliation. South Africa, which also belonged to this group, seems to be the only country where a liberal model of civic nation is pursued along with a strong recognition of the country’s diversity in the political and constitutional architecture.
Political regulation of ethnicity has been a core dimension of state-building in Africa. Ethnic heterogeneity was the product of colonial conquest and the formation of colonial states. Colonizers perceived ethnic groups to be building blocks of the colonial state and ethnic leaders as intermediaries within their system of domination. Postcolonial elites inherited states, which thus originated as colonial creations and did not reflect, with some very few exceptions, any existing precolonial political formations. Despite this exogeneous creation and the lack of would-be nations within the borders of nearly all of these states, the nation-state became the obvious point of reference for the rulers of new African states. The idea of self-determination, which had shaped the anti-colonial resistance and enforced the demise of the colonial empires, was inherently linked to nationalism (Davidson, 1992). As colonizers had suppressed political mobilization and social organization on political and ideological issues, ethnic identity had shaped societal and political dynamics. Postcolonial states had thus to reconcile the tensions between the existing ethnic identities, however manipulated and engineered through colonialism, and the creation of a modern nation-state.
The puzzle of how to manage ethnicity, and whether to fight, to manipulate, or to accept it, was thus a central concern of state elites, and it has remained so, as all hopes that “tribalism” would simply disappear with time had to be abandoned. Processes of political liberalization that started in many African countries since the early 1990s brought more intense competition and allowed for freedom of speech and association. Often political competition was framed in ethnic terms, fanning emotions and increasing the likelihood of outbreaks of violence among political parties representing different ethnic groups. As a return to military or single-party rule was no longer possible in the changed international environment, the thorny problem of how to manage ethnicity returned to the agenda of policymakers and constitution drafters across the continent.
Many societies worldwide faced similar challenges, and a specific body of literature, both academic and policy-oriented, emerged offering all kind of solutions for deeply divided societies, ranging from reforms of the governmental system to aspects of state structure, the electoral system, transitional justice, and party regulation (Kuperman, 2015; Reynolds, 2002). Even where this was not explicitly stated, the main assumption in this literature was that many democratizing societies were characterized by deep socio-ethnic cleavages. The conscious reform and manipulation of institutions could thus create incentives for collective actors to engage in interethnic cooperation. Interest in what “worked” also triggered large-n research about the effects of specific institutions on both peace and regime dynamics, with few comforting results (Brancati, 2006; Graham, Miller, & Strøm, 2017; Hartzell & Hoddie, 2003; Rørbæk & Knudsen, 2017; Schneider & Wiesehomeier, 2008; Selway & Templeman, 2012).
The purpose of this article is not to retrace this debate about constitutional engineering and offer a specific perspective about how ethnicity should be managed best or which institution is superior in providing peace or democracy. The emphasis will rather be on analyzing existing strategies to manage ethnicity on the African continent and to contribute to an explanation of why African elites adopted such radically different strategies. The analytical framework is thus informed by a second empirical research tradition from ethnic conflict studies (Guelke, 2012; McGarry, O’Leary, & Simeon, 2008; Wolff, 2011) that will be complemented by research on the social and political contexts of ethnic identity (Horowitz, 1985; Wimmer, 2013) The paper starts by providing a perspective on the varying structural contexts of interethnic relations on the African continent, which have represented quite different challenges and shaped the choices of policymakers. In a second step, a set of different strategies of macro-political ethnic conflict regulation is presented and their relevance in the African context discussed by making reference to case studies.1 This set of strategies is more comprehensive than the toolbox of constitutional engineers. The main criterion for discussing strategies is not what is normatively desirable but what has been practiced in postcolonial Africa.
State-Building and the Variety of Interethnic Relations
The intensity of debate about appropriate constitutional options has not been matched so far by a systematic attention to the underlying concepts of ethnicity and the varying nature of interethnic relationships.
Definitions of ethnicity typically include solidarity-based or inherited social characteristics, such as common culture, common belief systems, and common racial features: “Though culture, race, and religion are analytically distinguishable, the behaviors evoked by these forms of communal identity are so similar that they cannot readily be distinguished or disentangled” (Esman, 2004, p. 29). Within parts of the policy-oriented literature, there is a tendency to consider ethnic identity still in primordial categories, as a form of collective identity in a group that shares common inherited attributes. According to this perspective, ethnic communities become enduring historical entities along deep societal divisions, where few doubts remain about an individual’s ethnic belonging. Constructivist research, on the contrary, has shown that most individuals share a number of different identities, which can become more or less relevant in shaping behavior according to specific contexts. Constructivists would admit that violent conflict tends to reduce the contingency of ethnic affiliation. One of the substantial questions of ethnic conflict management remains, however, whether given ethnic identities need to be accommodated through fixed and formal representation of ethnic groups or ethnic identities are amenable to change via appropriate institutional mechanisms.
Ethnic identity might be more stable or fluid but has no meaning except in relational terms. It is the relationship to an “other,” which creates ethnic identity. Most relations among ethnic communities have been and are at present conducted peacefully. The management of ethnicity starts where interethnic relationships are perceived to be conflictive because of competing claims to land or homeland status, where members and leaders of ethnic communities harbor grievances or might have a history of ethnic-based rivalry and violence.
Which aspects of interethnic relationships could be relevant for understanding the relevance and choice of different institutional strategies? One strand of research has concentrated on the number and relative size of ethnic groups, yet ethnic heterogeneity does not automatically create deep division. It is certainly true that nearly all African societies are ethnically heterogeneous, but the political salience of ethnic differences varies greatly, both between countries and over time, and the extent to which conflicts are caused by ethnic division is highly contested (Posner, 2005). Compared to other regions of the world, African countries contain large numbers of small groups, none with sufficient numerical strength to constitute a majority in a country (Rothchild, 1986; Young, 1976). We find “many small and relatively similar groups that do not differ greatly from one another by global standards, [. . .] with multiple levels of ethnopolitical identity, [and] the absence of long histories of autonomous rule” (Mozaffar & Scarritt, 1999, p. 241).
Interethnic relations are thus not only determined by the number and relative group size or the depth of cultural differences. The postcolonial state in Africa was of critical importance for the trajectory of ethnic relationships, as control of the state was the major element of social stratification (Bayart, 1989; Wimmer, 1997). Adopting the formulation of Horowitz (1971), ethnic relationships can be ranked or unranked. Most African societies are unranked (Gisselquist, 2013). Ethnic identity does not correspond with social status, as the state has remained neutral, and language, culture, and state symbols are not tied to any particular ethnic group. The ethnic neutrality of the state does not exclude that access to the state is ethnicized, that is, ethnic identity does matter in the competition for office, jobs, and other goods. This requires groups to maintain a strong group identity but does not necessarily provide incentives to exaggerate cultural differences (Kymlicka, 2006). A minority of African states, however, does not correspond to this model of primarily unranked ethnic relationships and the ensuing competition among non-hierarchical ethnic groups. In Ethiopia, Liberia, Rwanda, Burundi, and the former apartheid states of southern Africa (Namibia, South Africa), state-building was not neutral among ethnic groups, but one racial or ethnic group managed to appropriate the state, which reflected its language, history, and culture and in which class corresponded with ethnic affiliation. These few ranked societies can thus be characterized, at least for parts of their recent history, as deeply divided societies in which ethnic or other particularistic groups are strongly polarized by overlapping political, social, and economic cleavages (Hartmann, 2013).
Access of different ethnic groups to political power has featured prominently in successive research, such as the Ethnic Power Relations (EPR) measurement of whether and to what extent ethnic groups are excluded from political power (Wimmer, Cederman, & Min, 2009). In a later book Wimmer (2013) discussed three different ethno-political configurations of power which lead to different types of ethnic conflict and ultimately different strategies of management of ethnicity. Two of these configurations are similar to Horowitz’s ranked and unranked relationships, with a third post-imperial constellation, where a legacy of indirect rule triggers secessionism as a distinct category of ethnic conflict. The sociopolitical context of ethnicity has also been covered in an index measuring the depth of divisions (Basedau, 2017), which has three main components: overlap of ethnic and religious differences, hierarchy of groups, and prior ethnic violence.
The typologies used in the literature to distinguish strategies of managing ethnicity start from the assumption that, from a comparative and global perspective, ethnic heterogeneity is the norm. Most African states are indeed inhabited by more than one ethnic group, and even where this might not be the case, migration has led to a continuous influx of ethnically diverse populations. The primary criterion to distinguish strategies is then whether ethnic heterogeneity is perceived to be a stable pattern or even a building block of the political system or whether, on the contrary, policies are pursued with the aim of reducing or erasing the open expression of ethnicity (Azarya, 2003). To recognize ethnicity within the political system is not necessarily the more peaceful or democratic option, as witnessed by the apartheid system of racial segregation in South Africa. Neither is a coercive imposition of one national identity compatible with democracy or the best recipe for peaceful social relations. The following overview of existing practice in sub-Saharan Africa thus takes up this basic distinction between strategies which aim to eliminate and those that want to manage ethnic heterogeneity (in line with McGarry & O’Leary, 1993).
The choice of specific strategies necessarily depends on a combination of different variables. Without denying the importance of prior ethnic violence, constitutional path dependencies (Schneider, 2017), regime type, or the system of government (Theuerkauf, 2012), it is our assumption that the hierarchical nature of interethnic relations is highly relevant in making sense of the radically different choices taken. In the Conclusion, this argument will be revisited in light of the empirical overview.
Eliminating Ethnic Heterogeneity
Where the main objective of constitutional dispensations and political leadership is to reduce or erase ethnic heterogeneity, the following strategies can be distinguished. Ethnic heterogeneity might be reduced through partition of territories or secession. In extreme cases state elites could be tempted to use force to homogenize their populations through genocide (i.e., killing minorities) or forced mass-population transfers (removing ethnic groups). The reduction of ethnic heterogeneity might also be achieved by less dramatic types of policymaking subsumed under the umbrella of “nation-building.” This might include more coercive variants that prohibit the use of ethnic symbols and languages and the political organization of ethnic identities (e.g., in political party names). The following section discusses the relevance of such strategies in the context of sub-Saharan Africa.
Genocide and Mass Expulsions
Genocide refers to “acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group as such” (Article 2, UN Convention on the Prevention and Punishment of Genocide ). While there is large consensus about these general definitional criteria, it is more controversial which acts are exactly covered (Hughes, 2011).
The Holocaust was clearly the reason for the enactment of the Genocide Convention and its recognition in international law. Yet there are historical episodes before World War II that would qualify as genocide, such as the displacement and attempted extermination of the Herero and Nama people in what is today Namibia by German colonial troops (1904–1907) (Zimmerer, 2007). In more recent African history the escalation of the Rwandan civil war in 1994 led to a Hutu elite mobilization for the annihilation of the minority Tutsi. The genocide was supposed to be a “terminal solution” to the long-standing conflict between the two ethnic groups. The military victory of the Tutsi-dominated Rwanda Patriotic Front in July 1994 stopped the genocide and made a mass punishment of perpetrators possible, locally based on a retrospective genocide law, and internationally with a special international criminal tribunal created by the UN Security Council (Prunier, 1995). The Rwanda genocide forced the international community, within and beyond Africa, to rethink the meaning of sovereignty and the responsibility of the international community to protect populations against gross human rights violations including genocide (as seen in the founding charter of the African Union ).
The violent conflict in Sudan’s Darfur province (2003–2009) was also initially qualified as a case of genocide by the UN Special Advisor on the Prevention of Genocide. A UN report from 2005 eventually established that the Sudanese government was responsible for a number of war crimes, with up to 300,000 people dead, most of them through hunger and disease, and more than 2 million displaced (Hughes, 2011, p. 129). The UN recommended International Criminal Court (ICC) prosecutions but did not state that the Sudanese government was pursuing a genocide. In 2008, the ICC prosecutor charged the Sudanese president and other leaders with genocide and crimes against humanity. Such claims have been rejected, not only by the African Union, but also by academics such as Mamdani (2009), according to whom the Darfur conflict was primarily a violent civil war between rival groups competing for land in a context of high environmental stress (see Prunier, 2005).
The state-directed movement of ethnic groups (McGarry, 1998) has not featured prominently as a technique of conflict regulation on the African continent, with the exception of apartheid South Africa (which will be discussed below in more detail) and the expulsion of the Indian community from Uganda in 1972. By public decree President Idi Amin gave the ca. 45,000 Ugandans with Indian descent 90 days to leave the country (Mamdani, 1993). This measure was, however, not the result of an ethnic conflict between the majority population and this minority but an attempt by the head of state, who had taken power by military coup, to gain legitimacy by mobilizing against the economically successful Indians and by confiscating their properties and distributing them among his supporters. President Museveni, soon after taking power in 1986, eventually invited the Indian community back to Uganda.2
Partition and Secession
Partition and secession refer to the creation of separate ethnically more homogeneous states as a strategy to solve ethnic conflicts. The main difference between the two is that partition refers to the division of a given state into several new constituent units, typically by some international authority, while secession describes the attempt by one ethnic group to leave an existing state by forming a new state and gain recognition for it from the international community. In contrast to well-known cases from other continents such as India/Pakistan, Palestine, or Cyprus (O’Leary, 2011), there are no cases of partition in postcolonial Africa.
Secession, on the contrary, has been a theme throughout the recent history of sub-Saharan Africa. The non-violability of international borders had been one of the guiding principles of the African state system, which emerged after political independence in the 1960s. The principle was also enshrined in the founding treaty of the Organization of African Unity (OAU) and accepted by all African states except Somalia and Morocco. Throughout the years, political independence was claimed by numerous groups across the continent in the presence or absence of violent conflict within the respective countries. Only in rare cases did these claims gained some broader relevance, as in the failed secession of Biafra during the Nigerian civil war (1967–1970) or the withdrawal of Somaliland from Somalia (Englebert & Hummel, 2005). Other center-periphery conflicts (Casamance/Senegal, Cabinda/Angola, Tuareg movements in Mali and Niger) did typically not lead to the formation of ethno-nationalist struggles. There were two successful secessions in Eritrea (1993) and South Sudan (2011), and in both cases the rump state (Ethiopia, Sudan) agreed to popular referendums, and eventually the loss of territory. One could argue, however, that in both cases the territorial separation did not end violence or the underlying conflict. In South Sudan, soon after independence in 2012, a civil war broke out, pitting the two main ethnic groups against each other. Relations with Sudan have remained strained. Only five years into independence, Eritrea fought a war with Ethiopia between 1998 and 2000 resulting in an estimated 70,000 deaths. These African cases seem to confirm the general skepticism about secessionism as a good strategy to manage ethnic conflict (Horowitz, 2003; Jenne, 2012).
Banning the Political Expression of Ethnicity
Assimilation is a strategy by dominant ethnic groups to induce individual members of minorities to give up their ethnic identity and to acculturate to a national cultural mainstream. As indicated in the introductory section, there were few potential dominant ethnic groups that could take their cultural identity as a basis for state-building. Access to higher education or full participation in government and the economy was often linked to proficiency in the national language inherited from colonizers and not one of the indigenous languages. Such policies thus promoted a rather civic and not an ethnic nationalism. Only a few countries, such as Ethiopia (during the emperor’s rule) or Liberia (before 1980), encouraged one language and culture over the rest.
One standard instrument applied by a majority of African states has been legal bans on ethnic or religious parties (Bogaards, Basedau, & Hartmann, 2010). It has been claimed that “the most overt marker of a coercively assimilationist state is the illegalization of minority nationalist, ethnic and communal parties” (McGarry et al., 2008, p. 43). One might object that the main rationale for ethnic party bans in Africa was to keep ethnicity out of politics, not assimilation into a dominant culture. In other countries, party politics was banned altogether to curb the ethnicization of politics. When Ugandan President Museveni gained power in 1986, he directly banned all political parties, allowing for political participation only in the “movement system,” an umbrella “no-party” organization. This system was justified by the role religious and ethnic parties had presumably played for provoking Uganda’s descent into many years of internal strife and war. Uganda’s general ban on the activities of political parties was upheld in the 1993 constitution, and only in 2005 was multi-party rule readmitted. (For more detail on the Ugandan system, see Carbone, 2001; Makara, Rakner, & Svåsand, 2009.) Museveni managed to sell this system as an African variant of democracy, but he clearly followed in the footsteps of military regimes in Nigeria and Ghana (Union Government of 1978), which had also prescribed no-party systems as the best way to manage or eliminate ethnicity from politics (Owusu, 1979). Blocking the public expression of ethnic identity had generally been one of the main justifications for the prolonged era of single-party rule across the continent. Although formally not a single-party state, post-genocide Rwanda provides the extreme case in this regard. With its 2001 law on “divisionism,” any speech, written statement, or action that is likely to cause conflict or can be interpreted as advocacy of ethnic groups is banned and can be declared a criminal offense (Niesen, 2010, p. 716). This would also include any public inquiry into the ethnic background of cabinet members. As ethnicity is a “divisive” colonial construct, it was officially replaced by the promotion of “Rwandan-ness” (Reyntjens, 2013; Vandeginste, 2014).
African regimes have finally differed in the role they have attributed to traditional leaders in the politics of their countries. Most African countries maintained the institution of chieftaincy after independence but stripped the traditional authorities of their political and legal powers. Traditional leaders continued to distribute land where the state de facto granted them the autonomy to do so, while in many other cases the state enforced the right to farm to non-autochthonous groups (Boone, 2007). It is noteworthy that the second scenario, which has pushed furthest in ignoring traditional and “ethnic group” rights, has led to the politicization of ethnic identity and, as in Côte d’Ivoire, ultimately to violent conflict.
Managing Ethnic Heterogeneity
We turn now to strategies that accommodate ethnic diversity. Accommodation might lead to an explicit recognition of ethnic markers in politics (consociationalism, ethno-federalism, hegemonic control, minority rights, and reserved seats) or institutions can acknowledge ethnic identity implicitly (informal power-sharing, centripetalism, non-ethnic federalism).
Consociationalism and Ethnic Power-Sharing
The most ambitious institutional strategy to manage ethnicity is without doubt the consociational model as developed by Lijphart. His core thesis that some smaller European democracies had produced stable democratic government by institutionalizing power-sharing among the core ethnic and religious segments sounded attractive to all elites in the newly independent countries who had witnessed the failure of majoritarian Westminster-type regimes to produce stability and democracy. Consociationalism consists in the combination of four different elements: a grand coalition by the political leaders of all significant segments in a society (which might manifest itself in different institutional forms); a mutual veto to protect minority group interests; proportionality as the principal standard of political representation, civil service appointment and allocation of public funds; and finally a high degree of autonomy for each segment to run its own internal affairs (Lijphart, 1977, p. 25). More recent writing on the topic has distinguished between different types of consociationalism, including more moderate variants, in which not all ethnic groups have to be included (McGarry et al., 2008). As the different segments within a society have difficulties in living peacefully together, so the logic goes, each group should have the right to govern itself as much as possible. Interethnic cooperation is restricted to the respective elites who have the task to protect the group interests through bargaining and vetoing within coalition governments. The system also relies on a clear recognition of which groups should be accommodated.
In the African context, consociationalism has been proposed in a number of constitutional reform processes (Uzodike, 2004), but only Burundi’s post-Arusha constitution comes close to realizing a consociational model. The 2005 constitution is constructed around the objective to balance ethnic representation at all levels in the public institutions. The constitution determines the ratio between the ethnic groups, with the Tutsi overrepresented vis-á-vis the Hutu (40% vs. 60%). The government is thus composed of no more than 60% Hutu, and the constitution requires all parties to form a coalition government composed of ministers representing all parties that have obtained at least 5% of the votes in the parliamentary elections. The directly elected president has two vice-presidents who represent different ethnic communities (Vandeginste, 2014). In case of parliamentary election outcomes not respecting these constitutional quotas, the electoral commission will co-opt additional members of parliament. The same ethnic criteria and distribution ratio are applied in the security sector, in the judicial apparatus, and for appointment of public administration and state-owned companies (Sullivan, 2005). The Burundian system thus requires that all candidates and public officials identify themselves as members of a specific ethnic group (Hutu, Tutsi, or Twa).
Consociationalism should not be confused with power-sharing, which has become one of the standard models to end violent conflict or protracted constitutional or electoral crises on the African continent. The conflict management literature has indeed a wide understanding of power-sharing which might relate to any institution dividing or sharing political, economic, territorial, and military power. Political power-sharing within this research perspective covers very different situations, ranging from roughly proportionate appointment of leaders of warring groups within government to a new constitution based on an explicit pact among all main political-military actors. While consociationalism refers “to a fairly elaborate technique of constitutional engineering, power sharing can best be seen as involving ad hoc concessions intended to give opposition groups a stake in transitions to democracy. One is enshrined in a set of carefully calibrated constitutional norms; the other is more in the nature of an improvised bricolage, aimed at co-opting the bad guys” (Lemarchand, 2006, p. 3).
The concept of power-sharing has been applied to describe both peace agreements between armed movements in Liberia or Côte d’Ivoire and coalition governments between government and opposition parties in Kenya or Zimbabwe (Cheeseman & Tendi, 2010; LeVan, 2011). These power-sharing agreements have met much criticism for not bringing peace or democracy (Jarstad, 2008; Mehler, 2009; Sisk, 2013; Spears, 2012). Interestingly, none of these agreements claims to represent ethnic or religious groups as such, and while rebel movements or opposition parties might have their main support among specific ethnic groups, such power-sharing cannot be labeled as management of ethnicity.
Informal Ethnic Power-Sharing
Most students of African politics would agree that informal ethnic power-sharing was a characteristic mode of regulating political conflicts in many single-party systems across the continent. Rothchild (1986) introduced the concept of “hegemonial exchange” to describe stable patterns of inclusion of ethno-regional spokespersons within a presidential single-party regime. The system combined elements from hegemonic control systems (see below) with the horizontal exchange logic of consociationalism: “As an ideal type, it is a form of state-facilitated coordination in which a somewhat autonomous central-state actor and a number of considerably less autonomous ethno-regional interests engage, on the basis of commonly accepted procedural norms, rules, or understandings, in a process of mutual accommodations” (Rothchild, 1986, p. 72).3 In such an informal cartel of ethnic elites, the individual representatives or ethnic intermediaries could be substituted, but the proportional representation of ethnic groups in administration and government was generally guaranteed.
Typical examples were Kenya under Kenyatta and Côte d’Ivoire under the presidency of Houphouet-Boigny. Single-party rule allowed Houphouet-Boigny to select key political personnel and to co-opt potential ethno-regional challengers of his rule (and of the de facto dominance of the president’s Baoulé ethnic group) into a hegemonial coalition. Over time a system developed that was compared to the UN Security Council: the representatives of some ethnic groups would always be considered in portfolio allocation and distribution of key public posts, while all other groups rotated on a non-permanent basis. But while representation in the UN Security Council is determined by formal rules, ethnic power-sharing in Côte d’Ivoire was not. The Ivorian example also demonstrates that the stability of such informal ethnic power-sharing relies both on the skills of individual leaders and on a restricted political competition. Upon Houphouet-Boigny’s death in 1993, political liberalization and leaders with much less sensitivity and judgment spelled the end of informal power-sharing (Banégas, 2006; Dozon, 2000).
Evidence from many African states (and beyond) indicates that consideration of ethno-regional interests in the allocation of political posts remains an important mechanism in many polities, even under competitive conditions (Arriola, 2013). One example is the informal ethno-regional power-sharing in Nigeria known as “zoning” (Suberu, 2015), where top political positions rotate among the country’s main ethno-regional blocs.
The concept of centripetalism goes back to the work of Donald Horowitz. His seminal book from 1985 argued that ethnicity must be acknowledged but that political institutions should not guarantee fixed representation for ethnic groups but rather create incentives for interethnic cooperation (Horowitz, 1985). He was mainly thinking about electoral systems such as the alternative vote (AV), which is a majoritarian electoral system with ranking of preferences for all candidates (Reilly, 2012). While AV has not been introduced in Africa, other institutional innovations function according to a similar logic. In Nigeria and Kenya’s presidential elections successful candidates have not only to obtain a majority of votes but also gain a quarter of the vote in at least two-thirds of the states (Nigeria) and more than half of the counties (Kenya), respectively. As the territorial concentration of ethnic groups is high and both countries have multiple ethnic groups, such rules effectively force candidates to reach out beyond their own ethnic group or alliance. The electoral system should thus create incentive structures for interethnic cooperation and moderation. Its main disadvantage is the potential indecisiveness when no candidate fulfills the requirement (such as in the 1979 Nigerian presidential elections). The necessity to gain votes across the whole territory also did not prevent Kenyan presidential contenders in 2007 from bringing the country close to an ethnic civil war. Distribution requirements might also apply to the party system. Political parties are required to have branches throughout the country (Ghana) or to present candidates in all of the country’s constituencies (Benin) as a condition for party registration or participation in elections (Bogaards, 2007).
Minority Rights and Reserved Seats
In comparison to other world regions (such as the Arab world or Eastern Europe), protection of religious and ethnic minorities is not a major feature of African political systems. As pointed out, there are few countries with one dominant ethnic group, which would threaten the cultural identity of remaining groups. In numerical terms, most countries are inhabited only by “ethnic minorities.” Nearly all African countries certainly have legal instruments to protect the rights of different minorities, but minority rights have not been considered an instrument of the management of ethnicity. Reserved seats for minorities were a hallmark of the settler colonies where the remaining white people wanted to assure their overrepresentation in the political system (Zimbabwe, 1980–1985).
An interesting exception still practiced is the electoral system of Mauritius. This island in the Indian Ocean has a mixed population with a majority of people of Indian descent, who are Hindus or Muslims. The indigenous African population is now a minority, as are the Chinese and people of European descent. Mauritius has a parliamentary system and a lively democratic competition with repeated electoral defeats of the incumbent government coalition, but it also strived for a fair representation of all ethnic and religious groups in the political system. This is guaranteed through the so-called best-loser system, where the electoral commission compares the ethnic origin of all elected members of parliament with the percentage of each ethnic group in the total population. Should the election outcome have led to an ethnically unbalanced parliament, a maximum of eight seats is awarded to candidates of the underrepresented communities. Candidates have thus to declare before the elections to which of the ethnic groups they belong (Mathur, 1997).
We have seen that ethnic conflict can be managed through territorial “isolation” of conflicting groups, i.e., through partition or secession. Territorial solutions might, however, also be realized within the borders of existing states. Both federalism and decentralization have thus been proposed as strategies for ethnic conflict management (Ghai, 1998). They rely on the assumption that conflicts can be managed by dividing power between central and regional elites. The possibility of self-rule over political, cultural, or economic issues within the ethnic homeland should thus minimize the potentially conflictive sharing of rule at the central level (Hartmann, 2013, p. 123). The crucial question is, again, how exactly ethnicity is addressed within such territorial strategies. Should subnational states or provinces be designed in a way to create ethnically homogeneous units, which are then controlled by an ethnic community, as proposed by ethnic federalism?
Ethnic federalism gained prominence in Africa through the innovative 1995 Ethiopian constitution that transformed the centralist state into a federal system with constituent states for each of the major ethnic groups. After many years of civil war, Ethiopia’s federalism avoided further state disintegration and large-scale civil war but has been hold together by the rule of a highly integrated bureaucratic party structure and actually been governed very much like “a centralized unitary state with most powers residing at the center” (Keller, 2002, p. 46). The relative success of Ethiopia’s regime in creating some degree of political stability can thus not be attributed to the decentralized features of its constitutional system (Clapham, 2006; Fiseha, 2012). It has also led to repeated contestations of the administrative borders of these ethnic states and the very limited degree of self-rule.
Federalism has in general not been a dominant strategy for the management of ethnicity on the African continent, with Nigeria as the only other substantial case of federalism.4 In Nigeria, the federal option came with independence and in an attempt by the British to “force together” a vast and ethnic and religious heterogeneous territory. “Nigeria is arguably the sole African country in which federalism is deeply entrenched and its constituent sub-federal governments are among the most powerful sub-national units on the continent” (Suberu, 2009, p. 67). Based on the experience of Biafra’s secession attempt, Nigeria has opted for a non-ethnic federalism, but Nigeria’s record in managing violent conflict has been equally mixed. Following the Biafra civil war with its massive loss of lives, the federal reforms have been successful in preventing a recurrence of large-scale ethno-secessionist conflict, but the country remains plagued by high levels of ethnic protest, violent communal conflicts, and political and economic discrimination against so-called non-indigenes at the substate level (Ejobowah, 2010; Osaghae, 2004). Many of these conflicts have occurred not despite federalism but exactly because of the specific dynamics of Nigerian federalism with ethnically heterogeneous states.
As described by Lustick (1979), hegemonic control has been the dominant mode of managing ethnicity throughout world history, especially with regard to imperial regimes or colonial empires that controlled multiple cultures within their territories through coercive domination and elite co-optation. The control is hegemonic to the extent that an open or violent ethnic competition for power becomes “unthinkable” from the perspective of subordinated communities (McGarry & O’Leary, 1993, p. 23). Intellectuals from the hegemonic group “invent and popularize reasons, biological or cultural, to demonstrate the superiority of their people” (Esman, 2004, p. 120). Control of the security apparatus is of major importance, especially where hegemonic control is exercised not by the largest ethnic group. Dominant majoritarian groups could aspire for assimilation and the integration of minorities into their own culture and language. If minorities such as the Tutsi in Burundi or the white community in South Africa did not want to share power (as they later both did with consociational or strongly proportional systems), they needed to use coercion to maintain their hegemony over the political and economic system. Inherent to such a system was the clear recognition of ethnic and racial difference.
The South African apartheid system implemented many features of such hegemonic control. “Of all the hegemonic systems in existence, none was more hegemonic in terms of its multidimensionality, the depth of the control exercised, and the very fact that it was carried out by a minority than South Africa” (Peleg, 2007, p. 158). It relied first on racial stratification (even poor white people enjoying more status and opportunities than black people), the physical separation of ethnic communities, the denigration of African cultures as inferior, and a massive security and judicial apparatus, which should render any hope of violent revolution unthinkable for the black populations. In a second phase, the apartheid system was able to co-opt traditional leaders and make them rulers of quasi-autonomous homelands. For many years, benefiting from the Cold War, this system was quite stable before it began to break down in the late 1980s. It remains a miracle that such a coercive mode of managing ethnicity could be replaced by a more peaceful and democratic regulation of ethnic identity (Friedman, 1995).
South Africa as a “rainbow nation” has maintained a mode of managing ethnicity, which does not deny the existence of ethnic diversity but stops short of formalizing ethnic representation within the political system (Henrard, 2005). Ethnic diversity is foremost recognized in the constitution with 11 official languages (2 European and 9 African) and entitlement to education and other public services in each of these languages. Rights of traditional leaders are entrenched in the constitution. A proportional electoral system guarantees the representation of all social and ethnic interests in parliament. A more consociational formula in the transitional constitution (1994–1996), which gave all political parties with a minimum of 5% of popular votes representation in the cabinet (and thus some veto rights), was not maintained in the 1996 final constitution. South Africa also opted for a weak form of non-ethnic federalism. With South Africa’s trajectory toward non-racial democracy, there are no hegemonic control regimes left on the African continent.
Different macro-political strategies of ethnic conflict management have been applied at some point in some countries within sub-Saharan Africa, but the broader trend within Africa has been to maintain a management strategy based on informal recognition of ethnic difference. There is generally not an assumption to deny the very existence of ethnic identity or any ambition to formalize its manifestation in the public realm. Formal institutions impose strict barriers on ethnic parties and put emphasis on individual rights, but informal institutions continue to exist in many parts of the continent that acknowledge the relevance of de facto ethnic group rights. Civic nationalism as a commitment to a common set of political institutions and political values beyond ethnic identity remains the norm, and this seems, in light of ethnic geography and multiple levels of ethno-political identity, to have worked well.
Both the various strategies that want to abolish particularistic identities and those that openly recognize ethnicity as an inherent feature of the institutional architecture remain the exception. They occur not by chance in the few ranked societies that cannot afford a strategy of informal accommodation, as ethnic identity is structurally related to social status and political power. Ethiopia and Burundi have selected the strategy of recognizing ethnicity, thereby guaranteeing some degree of shared power. Rwanda has opted for the opposing strategy of coercive assimilation. Ethno-political configurations of power, which can often be traced back to colonial or imperial politics, thus continue to shape the management of ethnicity. The escalation of violent conflict in Kenya or Côte d’Ivoire, formerly prime examples of successful informal power-sharing in unranked societies, serve, however, as a reminder that these structural constraints are not deterministic.
Among the predominantly ranked societies, South Africa seem to have managed to avoid both extremes through a careful design of minority rights and proportional representation but without ethnic federalism and grand coalition governments. The South African example also seems to show that the ranked or unranked character of interethnic relationship might change over time.
This article has presented evidence about the occurrence of different types of ethnic management and presented a tentative hypothesis about the choice of specific strategies. Other important questions such as a more systematic analysis of these various strategies’ effects on political stability, ending violent conflicts, or democratization were beyond the scope of this article, and thus remain an agenda for further research.
Arriola, L. R. (2013). Capital and opposition in Africa: Coalition building in multiethnic societies. World Politics, 65(2), 233–272.Find this resource:
Azarya, V. (2003). Ethnicity and conflict management in post-colonial Africa. Nationalism and Ethnic Politics, 9(3), 1–24.Find this resource:
Banégas, R. (2006). Côte d’Ivoire: Patriotism, ethnonationalism and other African modes of self-writing. African Affairs, 105(421), 535–552.Find this resource:
Basedau, M. (2017). Does the success of institutional reform depend on the depth of divisions? A pilot study on 34 African countries. In N. Ansorg & S. Kurtenbach (Eds.), Institutional reforms and peacebuilding. Change, path-dependency and societal divisions in post-war communities (pp. 21–54). New York, NY: Routledge.Find this resource:
Bayart, J.-F. (1989). L’état en Afrique. La politique du ventre. Paris: Fayard.Find this resource:
Bogaards, M. (2007). Electoral systems, party systems and ethnicity in Africa. In M. Basedau, G. Erdmann, & A. Mehler (Eds.), Votes, money and violence. Political parties and elections in sub-Saharan Africa (pp. 168–193). Uppsala, Sweden: Nordiska Afrikainstitutet.Find this resource:
Bogaards, M. (2014). Democracy and social peace in divided societies: Exploring consociational parties. New York, NY: Palgrave Macmillan.Find this resource:
Bogaards, M., Basedau, M., & Hartmann, C. (2010). Ethnic party bans in Africa: An introduction. Democratization, 17(4), 599–617.Find this resource:
Boone, C. (2007). Property and constitutional order: Land tenure reform and the future of the African state. African Affairs, 106(425), 557–586.Find this resource:
Brancati, D. (2006). Decentralization: Fueling the fire or dampening the flames of ethnic conflict and secessionism? International Organization, 60(3), 651–685.Find this resource:
Carbone, G. M. (2001). Constitutional alternatives for the regulation of ethnic politics? Institution-building principles in Uganda and South Africa’s transition. Journal of Contemporary African Studies, 19(2), 229–252.Find this resource:
Cheeseman, N., & Tendi, B.-M. (2010). Power-sharing in comparative perspective: The dynamics of “unity government” in Kenya and Zimbabwe. Journal of Modern African Studies, 48(2), 203–229.Find this resource:
Clapham, C. (2006). Afterword. In D. Turton (Ed.), Ethnic federalism. The Ethiopian experience in comparative perspective (pp. 231–240). Oxford, U.K.: James Currey.Find this resource:
Davidson, B. (1992). The black man’s burden. Africa and the curse of the nation-state. London, U.K.: James Currey.Find this resource:
Dozon, J.-P. (2000). La Côte d’Ivoire entre démocratie, nationalisme et ethnonationalisme. Politique Africaine, 78, 45–62.Find this resource:
Ejobowah, J. B. (2010). Territorial pluralism: Assessing the ethnofederal variant in Nigeria. Regional & Federal Studies, 20(2), 251–274.Find this resource:
Englebert, P., & Hummel, R. (2005). Let’s stick together: Understanding Africa’s secessionist deficit. African Affairs, 104(416), 399–427.Find this resource:
Esman, M. J. (2004). An introduction to ethnic conflict. Cambridge, U.K.: Polity.Find this resource:
Fiseha, A. (2012). Ethiopia’s experiment in accommodating diversity: 20 years’ balance sheet. Regional & Federal Studies, 22(4), 435–473.Find this resource:
Friedman, S. (1995). South Africa: Divided in a special way. In L. Diamond (Ed.), Politics in developing countries. Comparing experiences with democracy (pp. 531–581). Boulder, CO: Lynne Rienner.Find this resource:
Ghai, Y. (1998). Decentralization and the accommodation of ethnic diversity. In C. Young (Ed.), Ethnic Diversity and Public Policy: A Comparative Inquiry (pp. 31–71). London, U.K.: Palgrave Macmillan.Find this resource:
Gisselquist, R. M. (2013). Ethnic politics in ranked and unranked systems: An exploratory analysis. Nationalism and Ethnic Politics, 19(4), 381–402.Find this resource:
Graham, B. A. T., Miller, M. K., & Strøm, K. W. (2017). Safeguarding democracy: Powersharing and democratic survival. American Political Science Review, 111(4), 686–704.Find this resource:
Guelke, A. (2012). Politics in deeply divided societies. Cambridge, U.K.: Polity Press.Find this resource:
Hartmann, C. (2013). Territorial power-sharing and the regulation of conflict in Africa. Civil Wars, 15(Suppl. 1), 125–143.Find this resource:
Hartzell, C., & Hoddie, M. (2003). Institutionalizing peace: Power sharing and post-civil war conflict management. American Journal of Political Science, 47(2), 318–332.Find this resource:
Henrard, K. (2005). Power sharing and rights protection in the prevention and management of ethnic conflict: The case of post apartheid South Africa. In S. Noel (Ed.), From power sharing to democracy: Post conflict institutions in ethnically divided societies (pp. 140–163). Montreal, CA: McGill & Queen’s University Press.Find this resource:
Horowitz, D. L. (1971). Three dimensions of ethnic politics. World Politics, 23(2), 232–244.Find this resource:
Horowitz, D. L. (1985). Ethnic groups in conflict. Berkeley, CA: University of California Press.Find this resource:
Horowitz, D. L. (2003). The cracked foundations of the right to secede. Journal of Democracy, 14(2), 5–17.Find this resource:
Hughes, J. (2011). Genocide. In K. Cordell & S. Wolff (Eds.), Routledge handbook of ethnic conflict (pp. 122–139). London, U.K.: Routledge.Find this resource:
Jarstad, A. K. (2008). Power-sharing. Former enemies in joint government. In A. K. Jarstad & T. Sisk (Eds.), From war to democracy. Dilemmas of peacebuilding (pp. 105–133). Cambridge, U.K.: Cambridge University Press.Find this resource:
Jenne, E. K. (2012). When will we part with partition theory? Flawed premises and improbable longevity of the theory of ethnic partition. Ethnopolitics, 11(3), 255–267.Find this resource:
Keller, E. J. (2002). Ethnic federalism, fiscal reform, development and democracy. African Journal of Political Science, 7(1), 21–50.Find this resource:
Kuperman, A. J. (Ed.). (2015). Constitutions and conflict management in Africa. Philadelphia, PA: University of Pennsylvania Press.Find this resource:
Kymlicka, W. (2006). Emerging Western models of multination federalism: Are they relevant for Africa? In D. Turton (Ed.), Ethnic federalism: The Ethiopian experience in comparative perspective (pp. 32–64). Oxford, U.K.: James Currey.Find this resource:
Lemarchand, R. (2006). Consociationalism and power sharing in Africa: Rwanda, Burundi, and the Democratic Republic of the Congo. African Affairs, 106(422), 1–20.Find this resource:
LeVan, A. C. (2011). Power-sharing in Africa’s uncertain democracies. Governance, 24(1), 31–53.Find this resource:
Lijphart, A. (1977). Democracy in plural societies. New Haven, CT: Yale University Press.Find this resource:
Lustick, I. S. (1979). Stability in divided societies: Consociationalism versus control. World Politics, 31(3), 325–344.Find this resource:
Makara, S., Rakner, L., & Svåsand, L. (2009). Turnaround: The national resistance movement and the reintroduction of a multiparty system in Uganda. International Political Science Review, 30(2), 185–204.Find this resource:
Mamdani, M. (1993). The Ugandan Asian expulsion. Journal of Refugee Studies, 6(3), 265–273.Find this resource:
Mamdani, M. (2009). Saviors and survivors: Darfur, politics and the war on terror. London, U.K.: Verso.Find this resource:
Mathur, R. (1997). Parliamentary representation of minority communities: The Mauritian experience. Africa Today, 44(1), 61–82.Find this resource:
McGarry, J. (1998). “Demographic engineering”: The state-directed movemnt of ethnic groups as a technique of conflict regulation. Ethnic and Racial Studies, 21(4), 613–638.Find this resource:
McGarry, J., & O’Leary, B. (1993). Introduction: The macro-political regulation of ethnic conflict. In J. McGarry & B. O’Leary (Eds.), The politics of ethnic conflict regulation (pp. 1–40). London, U.K.: Routledge.Find this resource:
McGarry, J., O’Leary, B., & Simeon, R. (2008). Integration and accommodation? The enduring debate in conflict regulation. In S. Choudry (Ed.), Constitutional design for divided societies: Integration or accommodation? (pp. 41–88). Oxford, U.K.: Oxford University Press.Find this resource:
Mehler, A. (2009). Peace and power sharing in Africa: A not so obvious relationship. African Affairs, 108(432), 453–473.Find this resource:
Mozaffar, S., & Scarritt, J. R. (1999). Why territorial autonomy is not a viable option for managing ethnic conflict in African plural societies. Nationalism and Ethnic Politics, 5/3–4, 230–253.Find this resource:
Niesen, P. (2010). Political party bans in Rwanda 1994–2003: Three narratives of justification. Democratization, 17(4), 709–729.Find this resource:
O’Leary, B. (2011). Debating partition. Evaluating the standard justifications. In K. Cordell & S. Wolff (Eds.), The Routledge handbook of ethnic conflict (pp. 140–157). London, U.K.: Routledge.Find this resource:
Osaghae, E. E. (2004). Federalism and the management of diversity in Africa. Identity, Culture and Politics, 5(1–2), 162–178.Find this resource:
Owusu, M. (1979). Politics without parties: Reflections on the union government proposals in Ghana. African Studies Review, 22(1), 89–108.Find this resource:
Peleg, I. (2007). Democratizing the hegemonic state. Cambridge, U.K.: Cambridge University Press.Find this resource:
Posner, D. N. (2005). Institutions and ethnic politics in Africa. Cambridge, U.K.: University Press.Find this resource:
Prunier, G. (1995). The Rwanda crisis. History of a genocide, 1959–1994. New York, NY: Columbia University Press.Find this resource:
Prunier, G. (2005). Darfur. The ambiguous genocide. London, U.K.: Hurst.Find this resource:
Reilly, B. (2012). Institutional designs for diverse democracies: Consociationalism, centripetalism and communalism compared. European Political Science, 11(2), 259–270.Find this resource:
Reynolds, A. (Ed.). (2002). The architecture of Democracy. Oxford, U.K.: Oxford University Press.Find this resource:
Reyntjens, F. (2013). Political governance in post-genocide Rwanda. New York, NY: Cambridge University Press.Find this resource:
Rørbæk, L. L., & Knudsen, A. T. (2017). Maintaining ethnic dominance: Diversity, power, and violent repression. Conflict Management and Peace Science, 34(6), 640–659.Find this resource:
Rothchild, D. (1986). Hegemonial exchange: An alternative model for managing conflict in Middle Africa. In D. L. Thompson & D. Ronen (Eds.), Ethnicity, Politics and Development (pp. 65–104). Boulder, CO: Lynne Rienner.Find this resource:
Schneider, G. (2017). Matches and misfits: divided societies and the adoption of power-sharing. In N. Ansorg & S. Kurtenbach (Eds.), Institutional reforms and peacebuilding. Change, path-dependency and societal divisions in post-war communities (pp. 55–70). New York, NY: Routledge.Find this resource:
Schneider, G., & Wiesehomeier, N. (2008). Rules that matter: Political institutions and the diversity–conflict nexus. Journal of Peace Research, 45(2), 183–203.Find this resource:
Selway, J., & Templeman, K. (2012). The myth of consociationalism? Conflict reduction in divided societies. Comparative Political Studies, 45(12), 1542–1571.Find this resource:
Sisk, T. (2013). Power-sharing in civil war: Puzzles of peacemaking and peacebuilding. Civil Wars, 15(Suppl. 1), 7–20.Find this resource:
Spears, I. S. (2012). Africa’s informal power-sharing and the prospects for peace. Civil Wars, 15(1), 37–53.Find this resource:
Suberu, R. (2009). Federalism in Africa: The Nigerian experience in comparative perspective. Ethnopolitics, 8(1), 67–86.Find this resource:
Suberu, R. (2015). Managing constitutional change in the Nigerian Federation. Publius: The Journal of Federalism, 45(4), 552–579.Find this resource:
Sullivan, D. P. (2005). The missing pillars: A look at the failure of peace in Burundi through the lens of Arend Lijphart’s theory of consociational democracy. Journal of Modern African Studies, 43(1), 75–95.Find this resource:
Theuerkauf, U. G. (2012). Presidentialism and the risk of ethnic violence. Ethnopolitics, 12(1), 72–81.Find this resource:
Uzodike, U. O. (2004). Cultural diversity, societal conflict, and sustainable peace in Africa: Consociational democracy revisited. In J. M. Mbaku & S. C. Saxena (Eds.), Africa at the crossroads. Between regionalism and globalization (pp. 287–313). Westport, CT: Praeger.Find this resource:
Vandeginste, S. (2014). Governing ethnicity after genocide: ethnic amnesia in Rwanda versus ethnic power-sharing in Burundi. Journal of Eastern African Studies, 8(2), 263–277.Find this resource:
Whitaker, B. E. (2005). Citizens and foreigners: Democratization and the politics of exclusion in Africa. African Studies Review, 48(1), 109–126.Find this resource:
Wimmer, A. (1997). Who owns the state? Understanding ethnic conflict in post-colonial societies. Nations and Nationalism, 3(4), 631–665.Find this resource:
Wimmer, A. (2013). Waves of war. Nationalism, state formation, and ethnic exclusion in the modern world. Cambridge, U.K.: Cambridge University Press.Find this resource:
Wimmer, A., Cederman, L.-E., & Min, B. (2009). Ethnic politics and armed conflict: A configurational analysis of a new global data set. American Sociological Review, 74(2), 316–337.Find this resource:
Wolff, S. (2011). Managing ethno-national conflict: towards an analytical framework. Commonwealth and Comparative Politics, 49(2), 162–195.Find this resource:
Young, C. (1976). The politics of cultural pluralism. Madison, WI: University of Wisconsin Press.Find this resource:
Zimmerer, J. (2007). Colonial genocide: The Herero and Nama War (1904–1908) in German South Africa and its significance. In D. Stone (Ed.), The historiography of genocide (pp. 323–343). London, U.K.: Palgrave.Find this resource:
(1.) While interethnic relations can be conflictual at the micro-level, for example in resource-related conflicts in rural Africa between pastoralists and farmers or between autochthones and migrants, for lack of space this article will concentrate on macro-political strategies.
(2.) One could add under this heading attempts to redefine citizenship and disenfranchise large segments of the population, as pursued in Côte d’Ivoire with the ivoirité policies. Such exclusionary nation-building mechanisms did not include the forced removal of ethnic groups from the territory (Whitaker, 2005).
(3.) Bogaards (2014) has pointed to the fact that in authoritarian regimes or dominant party systems consociationalism could also be implemented within a single party, citing Kenya (under Kenyatta) and South Africa’s African National Congress as relevant cases.
(4.) One could add here the Comoros, as well as the quasi-federal systems of South Africa and Tanzania.