Cover Image

PAPERBACK
$34.75



View/Hide Left Panel

Electoral System and Conflict in Divided Societies

Ben Reilly and Andrew Reynolds1

Introduction

This work examines whether the choice of an electoral system in a culturally plural society can affect the potential for future violent conflict. We find that it can, but that there is no single electoral system that is likely to be best for all divided societies. We distinguish four basic strategies of electoral system design. The optimal choice for peacefully managing conflict depends on several identifiable factors specific to the country, including the way and degree to which ethnicity is politicized, the intensity of conflict, and the demographic and geographic distribution of ethnic groups. In addition, the electoral system that is most appropriate for initially ending internal conflict may not be the best one for longer-term conflict management. In short, while electoral systems can be powerful levers for shaping the content and practice of politics in divided societies, their design is highly sensitive to context. Consideration of the relationship between these variables and the operation of different electoral systems enables the development of contingent generalizations that can assist policy makers in the field of electoral system design.

Several fundamental assumptions that underlie the thinking of many Western policy specialists are called into question by the evidence assembled here concerning the relationship between conflict and elections. The first assumption, derived from Western experience, is that "free and fair elections" are the most appropriate way both to avoid and to manage acute internal conflict in other countries. The second assumption, which goes hand in hand with the first, is the implicit approval of "winner take



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1
--> Electoral System and Conflict in Divided Societies Ben Reilly and Andrew Reynolds1 Introduction This work examines whether the choice of an electoral system in a culturally plural society can affect the potential for future violent conflict. We find that it can, but that there is no single electoral system that is likely to be best for all divided societies. We distinguish four basic strategies of electoral system design. The optimal choice for peacefully managing conflict depends on several identifiable factors specific to the country, including the way and degree to which ethnicity is politicized, the intensity of conflict, and the demographic and geographic distribution of ethnic groups. In addition, the electoral system that is most appropriate for initially ending internal conflict may not be the best one for longer-term conflict management. In short, while electoral systems can be powerful levers for shaping the content and practice of politics in divided societies, their design is highly sensitive to context. Consideration of the relationship between these variables and the operation of different electoral systems enables the development of contingent generalizations that can assist policy makers in the field of electoral system design. Several fundamental assumptions that underlie the thinking of many Western policy specialists are called into question by the evidence assembled here concerning the relationship between conflict and elections. The first assumption, derived from Western experience, is that "free and fair elections" are the most appropriate way both to avoid and to manage acute internal conflict in other countries. The second assumption, which goes hand in hand with the first, is the implicit approval of "winner take

OCR for page 1
--> all" models of both government and election and disapproval of arrangements that emphasize power-sharing and cooperation. The third, again derived from Western experience, is that the types of electoral systems used in the West can be successfully transplanted to the developing world. A final assumption is that stable democracies need to be based on a system of individual rights rather than group rights. This work, to varying degrees, calls all of these assumptions into question. The multi-country evidence cited here offers some insights about how to diagnose a country's situation for the purpose of selecting an electoral system that can help that country address its communal conflicts peacefully. Realistic diagnosis of key social-structural issues is a necessary pre-condition to designing a successful system. In practice, there is little evidence of such diagnosis at work in the historical record. Moreover, the choice of an electoral system involves tradeoffs among a number of desirable attributes. Thus, the role of local actors, who can draw both on international experience and on their knowledge of domestic conditions and priorities, is key. Institutions, Conflict Management, and Democracy The study of political institutions is integral to the study of democratization because institutions constitute and sustain democracies:2 as Scarritt and Mozaffar succinctly summarize, "to craft democracies is to craft institutions" (1996:3). Perhaps most important for newly democratizing countries is the way that institutions shape the choices available to political actors. Koelble notes that this emphasis on "rules, structures, codes, and organizational norms" is based upon Weber's view of organizations as constructs designed to distribute rewards and sanctions and to establish guidelines for acceptable types of behavior (1995:233). March and Olsen argue that "constitutions, laws, contracts, and customary rules of politics make many potential actions or considerations illegitimate or unnoticed; some alternatives are excluded from the agenda before politics begins, but these constraints are not imposed full-blown by an external social system; they develop within the context of political institutions" (1984:740). In his important 1991 book Democracy and the Market, Adam Przeworski develops a concept of democracy as "rule open-endedness or organized uncertainty . . . and the less the uncertainty over potential outcomes the lower the incentive for groups to organize institutionally" (1991:13). Thus his influential conclusion, central to the spirit of this paper, that was a recognition that democratic government, rather than oligarchy or authoritarianism, presented by far the best prospects for managing deep societal divisions, and that democracy itself operates as a system for managing and processing rather than resolving conflict.3

OCR for page 1
--> In their preface to Politics in Developing Countries, Larry Diamond, Juan Linz, and Seymour Martin Lipset argue that institutions influence political stability in four important respects: (i)   Because they structure behavior into stable, predictable, and recurrent patterns, institutionalized systems are less volatile and more enduring, and so are institutionalized democracies. (ii)   Regardless of how they perform economically, democracies that have more coherent and effective political institutions will be more likely to perform well politically in maintaining not only political order but also a rule of law, thus ensuring civil liberties, checking the abuse of power, and providing meaningful representation, competition, choice, and accountability. (iii)   Over the long run well-institutionalized democracies are also more likely to produce workable, sustainable, and effective economic and social policies because they have more effective and stable structures for representing interests and they are more likely to produce working congressional majorities or coalitions that can adopt and sustain policies. Lastly, (iv)   democracies that have capable, coherent democratic institutions are better able to limit military involvement in politics and assert civilian control over the military (1995:33). Institutions and Democratization in the Developing World While accepting that throughout the developing world the societal constraints on democracy are considerable, such constraints still leave room for conscious political strategies which may further or hamper successful democratization. As a result, institutions work not just at the margins, but are central to the structuring of stability, particularly in ethnically heterogeneous societies. Scarritt and Mozaffar push the critical role of institutions even further by arguing that distinct institutional arrangements not only distinguish democracies, but invest governments with different abilities to manage conflicts, and thus that the survival of third wave democracies under extremely adverse conditions often hinges on these institutional differences (1996:3). Institutional design takes on an enhanced role in newly democratizing and divided societies because, in the absence of other structures, politics becomes the primary mode of communication between divergent social forces. In any society, groups (collections of individuals who identify some sort of mutual bond) talk to each other—sometimes about resolving distributive conflicts, sometimes about planning for the national future, and often about more mundane issues of everyday concern. In the pluralist democracies of the West, there are a variety of channels of communica-

OCR for page 1
--> tion open through which to carry on these conversations. Individuals from different cultures and perspectives can communicate with each other through the institutions of civil society via the press, social and sporting clubs, residence associations, church groups, labor unions, and so on.4 In fledgling democracies, however, where society is more deeply divided along ethnic, regional, or religious lines, political institutions take on even greater importance. They become the most prominent, and often the only, channel of communication between disparate groups. Such societies do not yet have the mixed institutions which characterize a broad civil society. Sporting, social, and religious groups are rigidly segregated, and various peoples do not live together, play together, or really talk to each other. Similarly, many new democracies do not yet have a vigorous free press where groups can talk. This holds true in the West as well, where different media outlets speak to different social groups or classes, and where cities are often segregated along racial, ethnic, and economic lines; but divided societies in the developing world often represent the extreme of the continuum, and that is why political institutions exist as the primary channel of communication. Because political institutions fulfill this role as the preeminent method of communication, they must facilitate communication channels between groups who need to talk. If they exclude people from coming to the table, then their conflicts can only be solved through force, not through negotiation and mutual accommodation. Further, those doing the talking, the representatives, must be just that—representative. To be able to make promises and then deliver on them, each political representative needs to be accountable to his or her constituency to the highest degree possible through institutional rules. The extent to which institutional rules place a premium on the representational roles of such figures, or rather seek to break down the overall salience of ethnicity by forcing them to transcend their status as representatives of only one group or another, is central to the scholarly debate about political institutions in deeply divided societies. The Validity of Constitutional Engineering There is little dispute that institutions matter, but there is much greater dispute regarding how much one can (or would wish to) engineer political outcomes through the choice of institutional structures. In this regard there exists an important distinction between an institutional choice approach and those who seek institutional innovation through constitutional engineering to mitigate conflict within divided societies. Sisk notes that ''there has been an implicit assumption by scholars of comparative politics who specialize in divided societies that such political conflict can be potentially ameliorated if only such societies would adopt certain types of

OCR for page 1
--> democratic institutions, that is, through 'political engineering'" (1995:5). Indeed, Horowitz proposes that "whatever their preferences, it remains true that a severely divided society needs a heavy dose—on the engineering analogy, even a redundant dose—of institutions laden with incentives to accommodation" (1991:280–281). Similarly, Sartori argues that "the organization of the state requires more than any other organization to be kept on course by a structure of rewards and punishments, of 'good' inducements and scary deterrents" (1994:203). However, Sisk runs counter to Horowitz, Lijphart, Sartori, and others in arguing that constitutional engineering should not be the primary focus of research: "most scholarship about democracy in divided societies centers too much on examining the best outcomes, as opposed to looking at the ways these outcomes evolve through bargaining processes" (1995:18). Indeed, Elster supports Sisk with the view that "it is impossible to predict with certainty or even qualified probability the consequences of a major constitutional change" (1988:304). Elster and Sisk remain in the minority on this question, given that most comparative political scientists would be happy to predict with 'qualified probability' the results of a shift in electoral law or democratic system. As Sartori correctly notes, if we follow Elster's somewhat defeatist logic, then "the practical implication of the inability of predicting is the inability of reforming" (1994:200). There seems little reason to give up the potential power of institutions for conflict resolution if we are confident of some degree of predictive ability when it comes to institutional consequences. Ultimately, there is a temporal dimension to both constitutional design and the politics of institutional choice. Political actors in a fledgling democracy may choose certain structures (rationally) because they maximize their gain in the short term. Thus, negotiators may not alight upon more inclusive structures recommended by political scientists posing as constitutional engineers. However, the promise of constitutional engineering rests on the assumption that long-term sociopolitical stability is the nation's overarching goal; and the institutions needed to facilitate that goal may not be the same as those which provide maximum short-term gain to the negotiating actors in the transitional period. Thus, institutional choice and constitutional engineering are, in practice, compatible approaches. One seeks to understand what drives short-term bargains, while the other seeks to offer more long-term solutions with the benefit of comparative cross-national evidence. The task of the constitutional engineer is not only to find which institutional package will most likely ensure democratic consolidation, but to persuade those domestic politicians making the decisions that they should choose long-term stability over short-term gain.

OCR for page 1
--> Electoral Engineering and Conflict Management The set of democratic institutions a nation adopts is thus integral to the long-term prospects of any new regime as they structure the rules of the game of political competition. Within the range of democratic institutions, many scholars have argued that there is no more important choice than which electoral system is to be used. Electoral systems have long been recognized as one of the most important institutional mechanisms for shaping the nature of political competition, first, because they are, to quote one electoral authority, "the most specific manipulable instrument of politics"5—that is, they can be purposively designed to achieve particular outcomes—and second, because they structure the arena of political competition, including the party system; offer incentives to behave in certain ways; and reward those who respond to these incentives with electoral success. The great potential of electoral system design for influencing political behavior is thus that it can reward particular types of behavior and place constraints on others. This is why electoral system design has been seized upon by many scholars (Lijphart, 1977, 1994; Sartori, 1968; Taagepera and Shugart, 1989; Horowitz, 1985, 1991) as one of the chief levers of constitutional engineering to be used in mitigating conflict within divided societies. As Lijphart notes, "If one wants to change the nature of a particular democracy, the electoral system is likely to be the most suitable and effective instrument for doing so" (1995a:412). Nevertheless, the fact that electoral system design has not proved to be a panacea for the vagaries of communal conflict in many places has shed some doubt upon the primacy that electoral systems are given as "tools of conflict management." What we attempt to do in this paper is assess the cumulative evidence of the relationship between electoral systems and intrasocietal conflict, and determine under what conditions electoral systems have the most influence on outcomes. An electoral system is designed to do three main jobs. First, it translates the votes cast into seats won in a legislative chamber. The system may give more weight to proportionality between votes cast and seats won, or it may funnel the votes (however fragmented among parties) into a parliament which contains two large parties representing polarized views. Second, electoral systems act as the conduit through which the people can hold their elected representatives accountable. Third, different electoral systems serve to structure the boundaries of "acceptable" political discourse in different ways, and give incentives for those competing for power to couch their appeals to the electorate in distinct ways. In terms of deeply ethnically divided societies, for example, where ethnicity represents a fundamental political cleavage, particular electoral systems can reward candidates and parties who act in a cooperative, accommo-

OCR for page 1
--> datory manner to rival groups; or they can punish these candidates and instead reward those who appeal only to their own ethnic group. However, the "spin" which an electoral system gives to the system is ultimately contextual and will depend on the specific cleavages and divisions within any given society. That said, it is important not to overestimate the power of elections and electoral systems to resolve deep-rooted enmities and bring conflictual groups into a stable and institutionalized political system which processes conflict through democratic rather than violent means. Some analysts have argued that while established democracies have evolved structures which process disputes in ways that successfully avoid "conflict," newly democratizing states are considerably more likely to experience civil or national violence (see Mansfield and Snyder, 1995). The argument that competitive multiparty elections actually exacerbate ethnic polarism has been marshaled by a number of African leaders (for example, Yoweri Museveni in Uganda and Daniel arap Moi in Kenya) in defense of their hostility to multiparty democracy. And it is true to say that "elections, as competitions among individuals, parties, and their ideas are inherently just that: competitive. Elections are, and are meant to be polarizing; they seek to highlight social choices" (Reynolds and Sisk, 1998:18). Elections may be "the defining moment", but while some founding elections have forwarded the twin causes of democratization and conflict resolution, such as South Africa and Mozambique, others have gone seriously awry, such as Angola and Burundi. While it is important not to overemphasize the importance or influence of political institutions (and particularly of electoral systems) as factors influencing democratic transitions, it is more common when dealing with developing countries that the reverse is true: scholars and policy makers alike have typically given too much attention to social forces and not enough to the careful crafting of appropriate democratic institutions by which those forces can be expressed. As Larry Diamond has argued, "the single most important and urgent factor in the consolidation of democracy is not civil society but political institutionalization."6 To survive, democracies in developing countries need above all "robust political institutions" such as secure executives and effective legislatures composed of coherent, broadly based parties encouraged by aggregative electoral institutions.7 We thus return to the underlying premise of constitutional engineering as it relates to electoral system design: while it is true that elections are merely one cog in the wheel of a much broader framework of institutional arrangements, sociohistorical pressures, and strategic actor behaviors, at the same time electoral systems are an indispensable and integral part of this broader framework. One electoral system might nurture accommodatory tendencies which already exist, while another may

OCR for page 1
--> make it far more rational for ethnic entrepreneurs to base their appeals on exclusionary notions of ethnochauvinism. What the collective evidence from elections held in divided societies does seem to suggest is that an appropriately crafted electoral system can do some good in nurturing accommodative tendencies, but the implementation of an inappropriate system can do severe harm to the trajectory of conflict resolution and democratization in a plural state (see Reilly, 1997a; Reynolds and Sisk, 1998). Given this, is it possible to outline criteria that one might use to judge the success or failure of any given electoral system design? In light of the multicausal nature of institutions, democracy, and political behavior, it would be foolhardy to say with absolute certainty that a particular electoral system was solely, or even primarily, responsible for a change—for better or worse—in ethnic relations in a divided society. Nevertheless, with the benefit of a holistic view of a nation's democratization process, it is possible to highlight instances where the electoral system itself appears to have encouraged accommodation, and those where it played a part in exaggerating the incentives for ethnic polarization. We hope that the typologies and analytical tools introduced in this paper as part of a contingent theory of electoral system design may help future research elucidate such electoral system effects. Our Knowledge to Date To date, our academic knowledge of electoral systems and their consequences has been predominantly based upon the more generic and abstract study of electoral systems as decision-making rules, structuring games played by faceless "rational actors" in environments which are often devoid of historical, socioeconomic, and cultural context. A comprehensive body of work exists which points to the mathematical effects of various systems on party systems, proportionality, and government formation (see, for example, Farrell, 1997; Grofman and Lijphart, 1986; Lijphart and Grofman, 1984; Lijphart, 1994; Rae, 1967; and Taagepera and Shugart, 1989). This is not to deride those very important works—and the discipline as a whole—rather it reflects the fact that much less work has been carried out on the subject of electoral systems, democratization, and conflict resolution. In addition, the majority of work on electoral systems to date has exhibited both a strong bias toward the study of established democracies in the West, and has been mostly country specific. This paper seeks to give a fillip to the increasingly important and more truly comparative study of how electoral systems can be crafted to improve the lot of divided societies. Historically, Huntington has identified three periods in which each contained a "wave" of transitions of states from nondemocracy to multi-

OCR for page 1
--> party competition (Huntington, 1991). Each wave saw the crafting of new constitutions for a new order, and electoral systems were regularly the most controversial and debated aspect of the new institutions. Huntington's "first wave" takes in the period from 1828 to 1926 when the United States, Great Britain, Australia, Canada, New Zealand, and a number of smaller European states began to evolve degrees of multiparty competition and "democratic" institutional structures. The debates over electoral systems (especially in Scandinavia and continental Europe) during this first wave of democratization mirrored many of the debates that new democracies are experiencing in the 1990s—the perceived trade-offs between "accountability" and "representativeness," between a close geographical link between elector and representative and proportionality for parties in parliament (see Carstairs, 1980). Huntington's second wave encompasses the post-second world war period through the decolonization decades of the 1950s and early 1960s. This wave saw many states either inherit or receive electoral systems designed and promoted by outside powers. West Germany, Austria, Japan, and Korea are examples of such ''external imposition" by Allied powers in the postwar period, while virtually all the fleetingly democratic postcolonial nation states of Africa and Asia inherited direct transplants of the electoral and constitutional systems of their colonial masters. Finally, the "third wave" of democratization, which began with the overthrow of the Salazar dictatorship in Portugal in 1974 and continues on to this day, gives us a wealth of case study material when it comes to assessing electoral system design in new democracies and those societies divided by cultural or social hostilities. In 1997, Reynolds and Reilly found only seven countries (out of 212 independent states and related territories) which did not hold direct elections for their legislatures, and of those 98 were classified as "free" on the basis of political rights and civil liberties in the 1995–1996 Freedom House Freedom in the World (Reynolds and Reilly, 1997). Therefore, we can be confident that a considerable range of comparative material is available for a study of how electoral systems influence democratization and stability in divided societies. This is even more so if we are mindful not to ignore the important lessons of nineteenth-century emerging democracies such as the British dominions (Canada, Australia, South Africa, New Zealand), with their divisions between and within "settler" and "indigenous" groups, or the multiethnic societies of continental Europe (Belgium, Austria-Hungary, Switzerland, the Netherlands, and Luxembourg), which fulfilled many of the classic elements of "plural" or "divided" societies at the turn of the century. In both groups, electoral system design was seen as a means of dealing with divisions and, particularly in the European examples, as a tool of accom-

OCR for page 1
--> modation-building between potentially hostile religious or linguistic groups. Developing an Analytical Framework for a Contingent Theory of Electoral System Design Consultants on electoral system design rightly shy away from the "one-size-fits-all" approach of recommending one system for all contexts. Indeed, when asked to identify their "favorite" or "best'' system, constitutional experts will say "it depends" and the dependents are more often than not variables such as: What does the society look like? How is it divided? Do ethnic or communal divides dovetail with voting behavior? Do different groups live geographically intermixed or segregated? What is the country's political history? Is it an established democracy, a transitional democracy, or a redemocratizing state? What are the broader constitutional arrangements that the legislature is working within? Historically, the process of electoral system design has tended to occur on a fragmented case-by-case basis, which has led to the inevitable and continual reinvention of the wheel because of limited comparative information. In this paper, we seek to develop an analytical framework upon which a contingent theory of electoral system design may be built. When assessing the appropriateness of any given electoral system for a divided society, three variables become particularly salient: (1)   knowledge of the nature of societal division is paramount (i.e., the nature of group identity, the intensity of conflict, the nature of the dispute, and the spatial distribution of conflictual groups); (2)   the nature of the political system (i.e., the nature of the state, the party system, and the overall constitutional framework); and (3)   the process which led to the adoption of the electoral system (i.e., was the system inherited from a colonial power, was it consciously designed, was it externally imposed, or did it emerge through a process of evolution and unintended consequences). In the following section we describe these three key variables and then operationalize them in the conclusion. Nature of Societal Division The Nature of Group Identity As noted earlier, appropriate constitutional design is ultimately contextual and rests on the nuances of a nation's unique social cleavages. The

OCR for page 1
--> nature of division within a society is revealed in part by the extent to which ethnicity correlates with party support and voting behavior. And that factor will often determine whether institutional engineering is able to dissipate ethnic conflicts or merely contain them. There are two dimensions to the nature of group identity: one deals with foundations (i.e., whether the society divided along racial, ethnic, ethnonationalistic, religious, regional, linguistic, etc. lines), while the second deals with how rigid and entrenched such divisions are. Scholarship on the latter subject has developed a continuum with the rigidity of received identity (i.e., primordialism) on one side and the malleability of constructed social identities (i.e., constructivist or instrumentalist) on the other (see Shils, 1957; Geertz, 1973; Young, 1976; Anderson, 1991; Newman, 1991; Esman, 1994). Clearly, if ethnic allegiances are indeed primordial, and therefore rigid, then a specific type of power sharing, based on an electoral system which primarily recognizes and accommodates interests based on ascriptive communal traits rather than individual ideological ones, is needed to manage competing claims for scarce resources. If ethnic identities and voting behaviors are fixed, then there is no space for institutional incentives aimed at promoting accommodatory strategies to work. Nevertheless, while it is true that in almost all multiethnic societies there are indeed correlations between voting behavior and ethnicity, the causation is far more complex. It is far from clear that primordial ethnicity, the knee-jerk reaction to vote for "your group's party" regardless of other factors, is the chief explanation of these correlations. More often than not ethnicity has become a proxy for other things, a semiartificial construct which has its roots in community but has been twisted out of all recognition. This is what Robert Price calls the antagonistic "politics of communalism"—ethnicity which has been politicized and exploited to serve entrepreneurial ends (1995). In practice, virtually every example of politicized ethnic conflict exhibits claims based on a combination of both "primordial" historical associations and "instrumentalized" opportunistic adaptations.8 In the case of Sierra Leone, for example, Kandeh (1992) has shown that dominant local elites, masquerading as "cultural politicians," shaped and mobilized ethnicity to serve their interests. In Uganda, President Museveni has used the ''fear of tribalism" as an excuse to avoid multipartism. Nevertheless, in both the colonial and postcolonial "one-party state" eras, strategies to control and carve up the Ugandan state were based upon the hostile mobilization of ethnic and religious identities. Malawi acts as a counter-factual to the primordial ethnicity thesis and offers an example of how political affiliations play out differently when incentive structures are altered. In the multiparty elections of 1994, a history of colonial rule, missionary activity, and Hastings Banda's "Chewa-ization" of national

OCR for page 1
--> TABLE 4 Nature of State and Electoral System Choice Nature of Democracy List PR (Consociationalism AV (Centripetalism) STV (Consensualism) Communal Rolls, Party Block Vote (Explicitism) Transitional Democracy South Africa 1994-, Bosnia 1996 Fiji 1997 Estonia 1990 Lebanon 1990-,Fiji 1997 Established democracy Belgium, Spain, Switzerland Australia, PNG 1964–1975 — Mauritius, India Democratic failure Sri Lanka 1983, Suriname 1980, Guyana 1980 — Northern Ireland 1973 Lebanon 1975, Fiji 1987

OCR for page 1
--> mocracy outside the West uses PR, although PR has been a common choice in transitional democracies in Africa in recent years. The final variable that may prove illuminating is whether breakdowns of democracy have occurred more or less under a particular system choice. As Table 4 suggests, advocates of different approaches can point to democratic successes and failures among divided societies. It is also the case, however, that countries such as Fiji, Lebanon, Northern Ireland, Sri Lanka, and others have persisted with (or reintroduced in modified form) the same electoral system design in place when democracy broke down. As discussed earlier, however, the bulk of new democracies in the post-war period simply adopted the electoral systems of their former colonists, evidencing equally unsatisfactory results across both majoritarian and proportional systems. Conclusion If the foregoing section suggests more randomness than regularity, it is still possible to isolate several factors that appear to be crucial when choosing between different models of electoral system design. First, the intensity of conflict does appear to have had an impact on the choice of electoral system for many divided societies. Specifically, the experience to date suggests that centripetal methods have been adopted in cases of more moderate conflicts and/or more fluid group identities, while list PR has tended to be adopted for transitional elections in more intensely conflictual situations. This fits with our earlier theoretical speculation that systems which require a degree of bargaining and cross-ethnic voting may be less realistic in extremely divided societies—where interethnic bargains, if any, may have to be made by elites alone-than in cases where there is a degree of fluidity to ethnic identities. This is why a system which combines elements of both approaches—such as STV—may well offer an attractive ''middle road" position, combining as it does some incentives for vote-pooling with reasonably good proportionality. Unfortunately, the use of STV in divided societies has been extremely limited and inconclusive to date. Nonetheless, there is some encouraging evidence from Northern Ireland's 1998 elections, where STV formed part of a wider prescription for power-sharing between the Catholic and Protestant populations, that STV served to advantage the pro-agreement, non-sectarian center (Wilder, 1998). The experience of systems in which ethnicity is explicitly recognized in the electoral system is somewhat contradictory. Both Lebanon and Fiji have suffered democratic breakdowns under such systems, but both have chosen to reintroduce elements of communalism in their new constitutions. It may well be that the value of such approaches lies in their ability

OCR for page 1
--> to contain and manage a deep ethnic conflict until new cleavages arise to take their place. The experience of Mauritius is instructive in this regard: now that ethnicity is no longer a core political issue, the communal elements of the Mauritian electoral system, via ethnically designated "best loser" seats, are seen as a relic of times past (Mathur, 1997). In terms of the four major electoral options for managing multiethnic conflicts, all have been successfully used in some divided societies, and all have suffered democratic failure at various points in time as well. But the respective needs of transitional versus consolidated democracies are often quite different. Put simply, the most important factor for democratic transition in electoral terms is usually a system that maximizes inclusiveness, is clearly fair to all parties, and presents minimal areas for potential preelection conflicts (such as the drawing of electoral boundaries)—goals that are usually best maximized by some form of regional or national list PR and which can lead to the election of a "grand" or "oversized" coalition government. By contrast, the priorities of a consolidated democracy may be more concerned with crafting a system which gives rise to minimal winning coalition or single party governments, is accountable in both geographic and policy terms, and allows the voters to "throw out" a government if it does not perform to their satisfaction-goals that are enhanced by a system based, at least to some extent, upon small geographically defined electoral districts that does not entrench oversize coalition governments. South Africa, which successfully conducted its transitional 1994 election using a national-list PR system and a mandated "Government of National Unity," has moved away from power-sharing measures and may change to some form of constituency-based PR system for its next elections in 2004. The differences between the needs of transitional and consolidated democracies are represented diagrammatically at Table 5. Advice For Policy Makers There is no perfect electoral system, and no "right" way to approach the subject of electoral system design. The major criteria for designing electoral systems for all societies, not just divided ones, are sometimes in conflict with each other or even mutually exclusive. Devices that increase proportionality, such as increasing the number of seats to be elected in each district, may lessen other desirable characteristics, such as promoting geographic accountability between the electorate and the parliament. The electoral system designer must therefore go through a careful process of prioritizing which criteria are most important to the particular political context before moving on to assess which system will do the best job. For example, an ethnically divided state in Central Africa might want above

OCR for page 1
--> TABLE 5 Ideal Qualities of Electoral Institutions for Transitional and Consolidated Democracies Transitional Democracy Consolidated Democracy • Inclusive • Accountable • Simple for voters to understand • Enables voters to express more sophisticated range of choice • Fairness in results (proportionality) • Ability to "throw the rascals out" • Minimize areas of conflict • Responsive to electorate • Simple to run • Promote sense of "ownership" of political process among voters • Transparent • "Minimal winning" coalitions or single-party governments • "Grand" or "oversized" coalition   all to avoid excluding minority ethnic groups from representation in order to promote the legitimacy of the electoral process and avoid the perception that the electoral system is unfair. In contrast, while these issues would remain important, a fledgling democracy in a multiethnic state in Eastern Europe might have different priorities—e.g., to ensure that a government could efficiently enact legislation without fear of gridlock and that voters are able to remove discredited leaders if they so wish. How to prioritize among such competing criteria can only be the domain of the domestic actors involved in the constitutional design process. Two levels of tension exist in the choice of electoral system options for divided societies. The first concerns those systems which place a premium on representation of minority groups (list PR and ethnically defined lists) compared to those which try to emphasize minority influence (AV and STV). As Horowitz has noted, "measures that will guarantee representation to a given ethnic or racial group may not foster the inclusion of that group's interest more broadly in the political process" (1991:165). The best option, of course, is to have both: representation of all significant groups, but in such a way as to maximize their influence and involvement in the policy-making process. This goal is best achieved by building both devices to achieve proportionality and incentives for interethnic accommodation into the electoral system itself. However, these goals are not always mutually compatible. A second level of tension exists between those systems which rely on elite accommodation (especially list PR) and those which rely on the electorate at large for moderation (AV and STV). Where elites are likely to be more moderate than the electorate, then list PR enables the major parties to include candidates from various groups on their ticket. Where the electorate itself is the major engine of moderation, then AV and other

OCR for page 1
--> systems which encourage vote pooling are likely to result in the election of more moderate leaders and more accommodative policies. When neither group is likely to display moderation, then ethnically mandated lists may need to be considered, as this provides the best way of "defusing" the salience of ethnicity as an electoral issue. It should also be remembered that, although conflict-management packages based on consociationalism, centripetalism, consensualism, and explicitism do represent alternative approaches, they are not mutually exclusive. In fact, creative constitutional engineering that utilizes appropriate levers from a number of divergent approaches may well offer the optimum strategy in some cases. A good example of this is the 1997 constitutional settlement in Fiji. In 1987 Fiji experienced an armed coup on the part of the indigenous Fijian armed forces against an elected government dominated by Indo-Fijians, which resulted in the formulation of a racially weighted constitution which discriminated against Fiji's Indian population. After years of international condemnation and economic decline, a new constitution specifically designed to promote peaceful, multiethnic government was promulgated. This constitution mandated a centripetal approach to electoral competition (via the Alternative Vote), but also included provisions borrowed from consociationalism (mandated power sharing) and, more controversially, from explicitism (a partial continuation of the system of communal representation for Fijian, Indian, and "general" electors). The new constitution is thus a structure in which a high, or even a redundant, level of institutional levers for conflict management has been deliberately built into the system. While Fiji's constitution-makers saw fit to make communal representation part of this new system, in general the comparative evidence to date suggests that explicitist approaches—ethnically mandated lists, communal rolls, racial gerrymandering, and the like—may serve artificially to sustain ethnic divisions in the political process rather than mitigating them. For this reason, we would counsel against their use in all but the most extreme cases of ethnic division. We would also recommend against systems that are overtly majoritarian in their operation: namely, the block vote and the two-round system. It is remarkable to note how many fledgling democracies in Africa, Asia, and the former Soviet Union use one or the other of these systems, considering their propensity to produce undesirable results. Both tend to reduce minority representation, and are thus unsurprisingly associated with authoritarian or other "unfree" regimes (Reynolds and Reilly, 1997:22). In addition, the Block Vote typically leads to single-party domination of parliaments and the elimination of opposition elements, while Two-Round systems place considerable strain on a state's electoral apparatus by having to run elections twice within a short

OCR for page 1
--> space of time. The continuation of such systems points to the basic problem of inertia in any electoral reform. Too often, constitutional drafters simply choose the electoral system they know best (often, in new democracies, the system of the former colonial power if there was one) rather than investigating the most appropriate alternatives. This does not mean we would necessarily advocate wholesale changes to existing electoral systems. In fact, the comparative experience of electoral reform to date suggests that moderate reforms that build on those things in an existing system which work well is often a better option than jumping to a completely new and unfamiliar system. What we do know is that there are several approaches to designing electoral systems for divided societies and that there is no single choice that is likely to be best in all cases. The optimal choice depends on several identifiable factors specific to each country, including its political history, the way and degree to which ethnicity is politicized, the intensity of conflict, and the demographic and geographic distribution of ethnic groups. While the combination of such variables in a given country gives us some useful pointers about electoral system design, it also can place considerable constraints upon constitutional engineers. The choice of electoral systems is always politically sensitive and always constrained by political considerations. Constitutional engineers in practice usually have limited room for maneuver. Nonetheless, despite such constraints, appropriate (and inappropriate) electoral system choices are powerful levers of democratic engineering, which inevitably have a marked influence on the future conduct of electoral politics. References Anderson, Benedict. 1991. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Barczak, Monica. 1997. "Electoral Rules, Responsiveness, and Party Systems: Comparing Chile and Ecuador." Paper delivered at 1997 Annual Meeting of the American Political Science Association, Washington, D.C., August 28–31, 1997. Barkan, Joel D. 1995. "Elections in Agrarian Societies." Journal of Democracy 6:106–116. Birch, Sarah. 1997. "Ukraine: The Perils of Majoritarianism in a New Democracy." Pp. 48–50 in Andrew Reynolds and Ben Reilly, eds. The International IDEA Handbook of Electoral System Design, Stockholm: International Institute for Democracy and Electoral Assistance. Carstairs, Andrew M. 1980. A Short History of Electoral Systems in Western Europe. London: George, Allen and Unwin. Cliffe, Lionel, with Ray Bush, Jenny Lindsay, Brian Mokopakgosi, Donna Pankhurst, and Balefi Tsie. 1994. The Transition to Independence in Namibia. Boulder: Lynne Rienner. Connors, Michael Kelly. 1996. "The Eclipse of Consociationalism in South Africa's Democratic Transition." Democratization 3:420–434. Constitution Review Commission 1996. The Fiji Islands: Towards a United Future. Parliamentary Paper No. 34 of 1996. Suva: Parliament of Fiji.

OCR for page 1
--> Diamond, Larry. 1995. "Nigeria: The Uncivil Society and the Descent into Praetorianism." In Politics in Developing Countries: Comparing Experiences with Democracy, Larry Diamond, Juan Linz and S.M. Lipset, eds. Boulder: Lynne Rienner. Diamond, Larry, Juan Linz, and S.M. Lipset, eds. 1995. Politics in Developing Countries: Comparing Experiences with Democracy. Boulder: Lynne Rienner Diamond, Larry. 1996. "Towards Democratic Consolidation." Pp. 227–240 in The Global Resurgence of Democracy, Larry Diamond and Marc F. Plattner, eds. Baltimore and London: Johns Hopkins University Press. de Silva, K.M. 1994. Ethnic Diversity and Public Policies: Electoral Systems. Geneva: UNRISD. Elliott, Sydney. 1992. "Voting Systems and Political Parties in Northern Ireland." Pp. 76–93 in Northern Ireland: Politics and the Constitution, Brigid Hadfield, ed. Bristol, U.K.: Open University Press. Elster, Jon. 1988. "Arguments for Constitutional Choice: Reflections on the Transition to Socialism." In Constitutionalism and Democracy, Jon Elster and Rune Slagstad, eds. Cambridge: Cambridge University Press. Elster, Jon, and Rune Slagstad, eds. 1988. Constitutionalism and Democracy. Cambridge: Cambridge University Press. Esman, M. 1994. Ethnic Politics. Ithaca and London: Cornell University Press. Farrell, David M. 1997. Comparing Electoral Systems. London: Prentice Hall/Harvester Wheatsheaf. Fearon, James D., and David D. Laitin. 1996. "Explaining Interethnic Cooperation." American Political Science Review 90:715–735. Gallagher, Michael. 1997. "Ireland: The Archetypal Single Transferable Vote System." Pp. 85–87 in The International IDEA Handbook of Electoral System Design, Andrew Reynolds and Ben Reilly, eds. Stockholm: International Institute for Democracy and Electoral Assistance. Geertz, Clifford. 1973. The Interpretation of Cultures. New York: Basic Books. Grofman, Bernard, and Arend Lijphart, eds. 1986. Electoral Laws and Their Political Consequences. New York: Agathon Press. Harris, Peter, and Ben Reilly, eds. 1998. Democracy and Deep-Rooted Conflict: Options for Negotiators. Stockholm: International IDEA. Horowitz, Donald L. 1985. Ethnic Groups in Conflict. Berkeley: University of California Press . Horowitz, Donald L. 1990. "Making Moderation Pay: The Comparative Politics of Ethnic Conflict Management." Pp. 451–475 in Conflict and Peacemaking in Multiethnic Societies, Joseph V. Montville, ed. New York: Lexington Books. Horowitz, Donald L. 1991. A Democratic South Africa? Constitutional Engineering in a Divided Society. Berkeley: University of California Press. Horowitz, Donald L. 1997. "Self-Determination: Politics, Philosophy, and Law." Pp. 421–463 in Ethnicity and Group Rights, I. Shapiro and W. Kymlicka, eds. New York and London: New York University Press. Huntington, Samuel P. 1991. The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press. Inter-Parliamentary Union. 1993. Electoral Systems: A World-wide Comparative Study. Geneva: Inter-Parliamentary Union. Jenkins, Laura D. 1994. Ethnic Accommodation Through Electoral Systems. Geneva: UNRISD. Kandeh, Jimmy D. 1992. "Politicization of Ethnic Identities in Sierra Leone." African Studies Review 35:81–99. Kaspin, Deborah. 1995. "The Politics of Ethnicity in Malawi's Democratic Transition." Journal of Modern African Studies 33:595–620. Koelble, Thomas. 1995. "The New Institutionalism in Political Science and Sociology." Comparative Politics 27:231–243.

OCR for page 1
--> Laitin, David. 1986. Hegemony and Culture: Politics and Religious Change Among the Yoruba. Chicago: University of Chicago Press. Lakeman, Enid. 1974. How Democracies Vote. London: Faber and Faber. Lewis, W. Arthur. 1965. Politics in West Africa. London: George Allen and Unwin. Lijphart, Arend. 1977. Democracy in Plural Societies: A Comparative Exploration. New Haven: Yale University Press. Lijphart, Arend. 1985. Power Sharing in South Africa. Policy Papers in International Affairs No. 24. Institute of International Studies, University of California, Berkeley. Lijphart, Arend. 1990. "Electoral Systems, Party Systems and Conflict Management in Segmented Societies." Pp. 2–13 in Critical Choices for South Africa: An Agenda for the 1990s, R.A. Schreirer, ed. Cape Town: Oxford University Press. Lijphart, Arend. 1991. "The Alternative Vote: A Realistic Alternative for South Africa?" Politikon 18:91–101. Lijphart, Arend. 1994. Electoral Systems and Party Systems: A Study of Twenty-Seven Democracies, 1945–1990. New York: Oxford University Press. Lijphart, Arend. 1995a. "Electoral Systems." Pp. 412–422 in The Encyclopedia of Democracy, S.M. Lipset. Washington D.C.: Congressional Quarterly Press. Lijphart, Arend. 1995b. "Multiethnic Democracy." Pp. 853–865 in The Encyclopedia of Democracy, S.M. Lipset, ed. Washington D.C.: Congressional Quarterly Press. Lijphart, Arend. 1997. "Disproportionality Under Alternative Voting: The Crucial—and Puzzling—Case of the Australian Senate Elections, 1919–1946." Acta Politica 32:9–24. Lijphart, Arend. Forthcoming. Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries. New Haven: Yale University Press. Lijphart, Arend, and Bernard Grofman, eds. 1984. Choosing an Electoral System: Issues and Alternatives. New York: Praeger. Lijphart, Arend, Rafael Lopez Pintor, and Yasunori Stone. 1986. "The Limited Vote and the Single Nontransferable Vote: Lessons from the Japanese and Spanish Examples." Pp. 154–169 in Electoral Laws and their Political Consequences, B. Grofman and A. Lijphart, eds. New York: Agathon Press. MacBeth, John. 1998. "Dawn of a New Age." Far Eastern Economic Review (17 September):24–28. Madden, A.F. 1980. "'Not for export': The Westminster Model of Government and British Colonial Practice." In The First British Commonwealth: Essays in Honour of Nicholas Mansergh, N. Hillmer and P. Wigley, eds. London: Frank Cass. Mansfield, Edward D., and Jack Snyder. 1995. "Democratization and War." Foreign Affairs 74:79–97. March, James, and Johan Olsen. 1984. "The New Institutionalism: Organizational Factors in Political Life." American Political Science Review 78:734–749. Mathur, Raj. 1997. "Parliamentary Representation of Minority Communities: The Mauritian Experience." Africa Today 44:61–77. Mattes, Robert B. 1995. The Election Book: Judgement and Choice in South Africa's 1994 Elections. Cape Town: IDASA Public Information Centre. Milne, R.S. 1982. Politics in Ethnically Bipolar States. Vancouver: University of British Columbia Press. Nagata, Judith. 1979. "Review of Lijphart's Democracy in Plural Societies." International Journal 34:505–506. Newman, Saul. 1991. "Does Modernization Breed Ethnic Political Conflict?" World Politics 43:451–478. Price, Robert M. 1995. "Civic Versus Ethnic: Ethnicity and Political Community in Post-Apartheid South Africa." Unpublished paper.

OCR for page 1
--> Przeworski, A. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. Cambridge: Cambridge University Press. Putnam, Robert D., with Robert Leonardi and Raffaella Y. Nanetti. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press. Rae, Douglas W. 1967. The Political Consequences of Electoral Laws . New Haven: Yale University Press. Reid, Ann. 1993. "Conflict Resolution in Africa: Lessons from Angola." INR Foreign Affairs Brief. Washington D.C.: US Department of State. Reilly, Ben. 1997a. "The Alternative Vote and Ethnic Accommodation: New Evidence from Papua New Guinea." Electoral Studies 16:1–11. Reilly, Ben. 1997b. "Preferential Voting and Political Engineering: A Comparative Study." Journal of Commonwealth and Comparative Studies 35:1–19. Reilly, Ben. 1998a. "Constitutional Engineering in Divided Societies: Papua New Guinea in Comparative Perspective." Ph.D. dissertation, Australian National University, Canberra. Reilly, Ben. 1998b. "With No Melting Pot, a Recipe for Failure in Bosnia." International Herald Tribune, 12–13 September, p.6. Reilly, Ben, and Michael Maley. 1999. "The Single Transferable Vote and Alternative Vote Compared." In STV in Comparative Perspective, Shaun Bowler and Bernard Grofman, eds. Ann Arbor: University of Michigan Press. Reynolds, Andrew, ed. 1994. Election '94: South Africa—An Analysis of the Results, Campaigns, and Future Prospects. New York: St. Martin's Press. Reynolds, Andrew. 1995. "The Case for Proportionality." Journal of Democracy 6:117–124. Reynolds, Andrew. 1996. "Electoral Systems and Democratic Consolidation in Southern Africa." Ph.D. dissertation, University of California at San Diego. Reynolds, Andrew, and Ben Reilly. 1997. The International IDEA Handbook of Electoral System Design. Stockholm: International Institute for Democracy and Electoral Assistance. Reynolds, Andrew, and Jïrgen Elklit. 1997. "Jordan: Electoral System Design in the Arab World." In The International IDEA Handbook of Electoral System Design, Andrew Reynolds and Ben Reilly, eds. Stockholm: International Institute for Democracy and Electoral Assistance. Reynolds, Andrew, and Timothy D. Sisk. 1998. "Elections, Electoral Systems, and Conflict Management." In Elections and Conflict Resolution in Africa, Timothy Sisk and Andrew Reynolds, eds. Washington, D.C.: United States Institute of Peace Press. Rokkan, Stein. 1970. Citizens, Elections, Parties: Approaches to the Comparative Study of the Processes of Development. Oslo: Universitetsforlaget. Rose, Richard. 1976. Northern Ireland: A Time of Choice. Washington D.C.: American Enterprise Institute. Rule, Wilma, and Joseph Zimmerman, eds. 1994. Electoral Systems in Comparative Perspective: Their Impact on Women and Minorities. Westport, Conn.: Greenwood. Sartori, Giovanni. 1968. "Political Development and Political Engineering." Public Policy 17:261–298. Sartori, Giovanni. 1994. Comparative Constitutional Engineering: An Inquiry Into Structures, Incentives, and Outcomes. New York: Columbia University Press. Scarritt, James R. 1993. "Communal Conflict and Contention for Power in Africa South of the Sahara." In Minorities at Risk, T. R. Gurr, ed. Washington, D.C.: United States Institute of Peace Press. Scarritt, James R., and Shaheen Mozaffar. 1996. "The Potential for Sustainable Democracy in Africa." Paper presented at the International Studies Association, San Diego, April 19, 1996.

OCR for page 1
--> Shils, Edward 1957. "Primordial, Personal, Sacred, and Civil Ties". British Journal of Sociology 8:130–145. Sisk, Timothy D. 1995. Democratization in South Africa: The Elusive Social Contract. Princeton: Princeton University Press. Sisk, Timothy D. 1996. Power Sharing and International Mediation in Ethnic Conflicts. Washington, D.C.: United States Institute of Peace Press. Taagepera, Rein. 1990. "The Baltic States." Electoral Studies 9:303–311. Taagepera, Rein. 1998. "How Electoral Systems Matter for Democratization." Democratization 5(3):68–91. Taagepera, Rein, and Matthew S. Shugart. 1989. Seats and Votes: The Effects and Determinants of Electoral Systems. New Haven: Yale University Press. Vengroff, Richard. 1994. "The Impact of Electoral System on the Transition to Democracy in Africa: The Case of Mali." Electoral Studies 13:29–37. Wilder, Paul. 1998. "A Pluralist Parliament for a Pluralist People? The New Northern Ireland Assembly Elections, 25 June 1998." Representation 35:97–105. Young, Crawford. 1976. The Politics of Cultural Pluralism. Madison: University of Wisconsin Press. Notes 1   Ben Reilly is a Senior Programme Officer at the International Institute for Democracy and Electoral Assistance based in Stockholm, Sweden. Andrew Reynolds is an Assistant Professor in the Department of Government and International Studies at the University of Notre Dame, USA. 2   See March and Olsen, 1984:747; and Koelble, 1995:232. 3   See Prezworski, 1991:10–14. 4   See also Putnam, 1993. 5   Sartori, 1968:273. 6   Diamond, 1996:238. 7   Diamond, 1996:239. 8   Esman, 1994:14. 9   Crawford Young also notes that the Ghanaian elections of 1996 were another example of substantial nonethnic block voting. Only the Ewe community could be categorized as "ethnic voters." 10   Crawford Young, however, argues that "95 percent of whites voted for the NP, the IFP drew its votes heavily from Zulu, and the Colored vote was importantly shaped by the communal insecurities and concerns of that group." 11   Lijphart, 1995b:853. 12   In many African states, urbanization has led to ethnic intermixing. Mines and plantations are also more likely to have multiethnic workforces and thus communities. 13   We are indebted to Crawford Young for pointing this out. 14   For example, South Africa used a classically proportional electoral system for its first democratic elections of 1994, and with 62.65 percent of the popular vote the African National Congress (ANC) won 63 percent of the national seats. The electoral system was highly proportional, and the number of wasted votes (i.e., those which were cast for parties who did not win seats in the Assembly) was only 0.8 percent of the total (see Reynolds, 1994). However, under some circumstances nonproportional electoral systems (such as FPTP) can accidentally give rise to relatively proportional overall results. This was the case in a third Southern African country, Malawi, in 1994. In that election the leading party, the United Democratic Front won 48 percent of the seats with 46 percent of the votes, the Malawian Congress Party won 32 percent of the seats with 34 percent of the votes, and the

OCR for page 1
-->     Alliance for Democracy won 20 percent of the seats with 19 percent of the votes. The overall level of proportionality was high, but the clue that this was not inherently a proportional system, and so cannot be categorized as such, was that the wasted votes still amounted to almost one-quarter of all votes cast. 15   It must be noted however that the party system fragmentation of 1919–1933 was not a direct result of the PR system adopted from post-Great War Germany as party fragmentation was equally high and problematic under the pre-1919 two-round German electoral system. As Lakeman notes, the number of parties in the Reichstag in 1912 was 21, while during Hitler's rise to power in the early 1930s the party system had coalesced to four or five major blocks (Lakeman, 1974:209). 16   Madden, 1980:20. 17   Most notably in times of war, as in Britain, and times of internal upheaval, as in West Germany in the 1970s. 18   Lijphart, 1990:11. 19   For example, the South African National Assembly elected in 1994 was 52 percent black (11 percent Zulu, the rest of Xhosa, Sotho, Venda, Tswana, Pedi, Swazi, Shangaan, and Ndebele extraction), 32 percent white (one-third English, two-thirds Afrikaans), 7 percent Colored and 8 percent Indian. And the Namibian parliament is similarly diverse, with representatives from the Ovambo, Damara, Herero, Nama, Baster, and white (English and German speaking) communities (see Reynolds, 1995). 20   See Rule and Zimmerman, 1994 and the Inter-Parliamentary Union, 1993. 21   See Reynolds, 1996. 22   See Horowitz, 1991:140–141. 23   Connors argues that in South Africa consociationalism "rather than mitigating ethnic conflict, could only wittingly or unwittingly provide a basis for ethnic mobilization by providing segmental leaders with a permanent platform" (1996:426). 24   Horowitz, 1990:471. 25   Sisk, 1996:62. 26   Scarritt, 1993:256 27   Rose, 1976:78. 28   See Taagepera, 1990. 29   Lijphart, 1977:56. 30   See Milne, 1982.