Do Ethnic Parties Promote Minority Ethnic Conflict?

This article was downloaded by: [University of North Texas], [John Ishiyama] On: 18 June 2014, At: 12:02 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Nationalism and Ethnic Politics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/fnep20

Do Ethnic Parties Promote Minority Ethnic Conflict? John Ishiyama

a

a

University of North Texas , Published online: 24 Feb 2009.

To cite this article: John Ishiyama (2009) Do Ethnic Parties Promote Minority Ethnic Conflict?, Nationalism and Ethnic Politics, 15:1, 56-83, DOI: 10.1080/13537110802672388 To link to this article: http://dx.doi.org/10.1080/13537110802672388

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions

Nationalism and Ethnic Politics, 15:56–83, 2009 Copyright © Taylor & Francis Group, LLC ISSN: 1353-7113 print / 1557-2986 online DOI: 10.1080/13537110802672388

Do Ethnic Parties Promote Minority Ethnic Conflict?

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

JOHN ISHIYAMA University of North Texas

Do ethnic parties exacerbate ethnic conflict? Many scholars have argued that the mere appearance of ethnic parties inevitably leads to a spiral of ethnic conflict and the collapse of incipient democracies. This article tests this proposition by examining how ethnic parties affect protest and communal conflict across 82 countries and 213 ethnic/communal groups from 1985–2003. Using a variety of quantitative techniques, I find that ethnic parties do mobilize minority ethnic groups to engage in protest, but there is no natural connection between the appearance of ethnic parties and the extent to which the minority group engages in communal conflict. Do ethnic parties exacerbate ethnic conflict? On the one hand, several scholars have argued that the mere appearance of ethnic parties leads to the “ethnification” of politics and a spiral towards instability and the collapse of incipient democracies.1 On the other hand, several scholars have argued that ethnic parties provide opportunities for interest articulation from groups that might normally be shut out of the political system.2 Indeed, a longstanding argument made by advocates of the consociational school is that ethnic parties actually help dampen conflict by channeling demands through legal channels, thus increasing “voice” and preventing “exit” of ethnic groups via conflict.3 The question as to the role played by ethnic parties in fomenting conflict is not simply an issue for academic debate. Indeed, many countries have explicitly adopted bans on the existence of ethnic parties as a policy “remedy” for ethnic conflict, based largely on the presumption that such parties are by their very nature destructive political entities. This was certainly the justification in Bulgaria, where constitutional bans on the existence of ethnic parties Address correspondence to John Ishiyama, Professor of Political Science, Department of Political Science, University of North Texas, 1155 Union Circle #305340, Denton, TX 762035017. E-mail: [email protected] 56

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Ethnic Parties and Minority Ethnic Conflict

57

were adopted in 1991.4 Such bans have also been employed in Turkey, Afghanistan, and Iraq.5 Further, as Anika Becher and Mathias Basedau note, ethnic party bans have been used in at least 22 sub-Saharan African states and has become one of the more popular ways in which states have attempted to contain the destructive effects of the ethnification of politics.6 Nonetheless, despite the substantial debate in the literature, there is relatively little comparative, quantitative/empirical work that has been done as to whether ethnic parties actually contribute to ethnic conflict. Most studies have either focused on single countries to explain why ethnic parties are successful7 or have concentrated on specific regions, such as Western, Eastern, and Central Europe rather than adopting a broadly comparative approach.8 In this paper, I examine the question of how ethnic parties affect ethnic protest and communal conflict by using the Minorities at Risk (MAR) database combined with data on ethnic parties across 82 new democracies or democratizing states in the developing world and 213 ethnic/communal groups. In particular I focus on the question of whether the mere appearance of ethnic parties is associated with higher degrees of ethnic group political protest and ethnic group communal conflict. I concentrate only on new democracies, in as much the extant literature that points to the dangers of ethnic parties focus on their damaging activities in newly democratizing states NOT fully consolidated democracies.9

ETHNIC PARTIES AND ETHNIC CONFLICT There has been a longstanding interest in the role played by ethnic parties in promoting ethnic conflict. Many scholars have argued that the appearance of ethnic parties is a “bad” thing for new democracies or systems in transition. From this perspective, not only does the appearance of ethnic parties deepen divisions between groups but ethnic parties serve to exacerbate conflict. Alvin Rabushka and Kenneth Shepsle offered one of the earliest explanations as to how ethnic parties promote conflict in plural societies.10 In the Rabushka and Shepsle model, a key role is played by ethnic elites and the organizations that they lead. These organizations engage in the politics of ethnic outbidding, which ultimately undermines multiethnic cooperation and inevitably leads to nondemocratic, ethnically exclusive states. Other authors have similarly argued that because ethnic parties make their political appeal specifically on ethnicity, their emergence often has a centrifugal effect on politics.11 This is especially harmful to new democracies, where democratic institutions are quite fragile. Indeed, under such conditions, ethnic competition can easily turn into ethnic conflict. This is because the competition for votes for the ethnic party involves mobilizing the ethnic group—and

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

58

J. Ishiyama

the best way to do that is to use inflammatory and confrontational rhetoric, distinguishing between “us” versus “them.”12 As Richard Gunther and Larry Diamond note, “the electoral logic of the ethnic party is to harden and mobilize its ethnic base with exclusive, often polarizing appeals to ethnic group opportunity and threat . . . the ethnic party’s particularistic, exclusivist, and often polarizing political appeals make its overall contribution to society divisive and even disintegrative.”13 In addition, ethnic parties, indirectly contribute to worsening ethnic tensions by promoting party politics along cultural lines, which often leads to the marginalization and exclusion of a cultural minority. As a consequence such minorities may feel encouraged to resort to undemocratic or even violent means in order to counter this dominance. Further, merely by promoting identity-based politics, ethnic parties can significantly raise the stakes of the political game, reinforcing group identities and thus raising the likelihood of conflict. For these reasons as well, ethnic parties increase the likelihood of intercommunal conflict and threaten the survivability of new democracies.14 Thus, from the above perspective, the mere appearance of an ethnic party should signal an increase in the rise of interethnic conflict. In short, as Donald Horowitz notes: By appealing to electorates in ethnic terms, by making ethnic demands on government, and by bolstering the influence of ethnically chauvinistic elements within each group, parties that begin by merely mirroring ethnic divisions help to deepen and extend them. Hence the oft heard remark in such states that politicians have created ethnic conflict.15

On the other hand, several scholars contend that ethnic parties can play a constructive role in promoting intergroup accommodation. Indeed, advocates of the consociational school have long argued that by promoting the emergence of ethnic parties and then representing them broadly this will facilitate the integration of as many subcultures as possible into the political game, thus creating the conditions for interethnic cooperation.16 Further, by securing representation for minority groups, openness serves to facilitate the integration of disaffected groups into the political system, which ultimately leads them to moderate their demands. Frank Cohen argues the broader the representation the more likely the ethnic group feels bound to the existing system—as he puts it “by making institutions more accessible and making ethnic cleavages more explicit, ethnic groups will engage in more frequent but less intense conflict. They will use moderate means of resistance to effect change in the status quo.”17 Others, such as Sherrill Stroschein contend that ethnic parties do not cause ethnic conflict but emerge as the result of it—that is, they reflect differences that already exist. Nonetheless, ethnic parties can channel demands into more legitimate forms of participation and thus allow conflicts to be

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Ethnic Parties and Minority Ethnic Conflict

59

resolved politically rather than through violence.18 Others have similarly attributed the preservation of ethnic peace in various settings to the representation of ethnic parties in Parliament.19 Ishiyama demonstrated that in the postcommunist world, ethnic parties have served to assist in bringing into the political process those who would have been otherwise alienated by the emerging democratic systems in the region.20 Current research, such as Johanna Birnir, Kanchan Chandra, James Fearon and David Laitin, and Daniel Posner, offers a more contingent view of the link between ethnic cleavages and conflict and stresses the importance of ethnicity and ethnic cleavages as cost-effective strategic resources for group formation, interest definition, and collective action.21 Perhaps one of the strongest and most articulate proponents of the notion that ethnic parties can have a positive effect on the stabilization of new democracies is Kanchan Chandra. Chandra directly attacks the notion of ethnic outbidding, which is so central to the argument that the mere appearance of ethnic parties sets off a chain reaction leading to a spiral of extremism that destroys democratic politics altogether. Rather, she argues that ethnic parties can help sustain democracy if these parties are institutionally encouraged to compete on multiple dimensions rather than on just the unidimensional axis of ethnicity. Indeed, political institutions that restrict “ethnic politics to a single dimension destabilize democracy, whereas institutions that foster multiple dimensions of ethnic identity can sustain it.”22 In a similar vein, Johanna Birnir, in examining patterns of ethnic politics in a broadly comparative way, contends that the ethnification of politics does not necessarily translate into violence.23 Like Chandra, she argues that ethnic identity serves as a stable but flexible information shortcut for political choices and assists in stabilizing party formations and hence the development of democracy. If violence results it is largely the result of political institutional factors, particularly restrictions on access to the executive. This exclusion is what leads to violence, not the political mobilization of ethnicity. Birnir argues that, ceteris paribus, ethnic parties (which she refers to as ethnic “attractors”) are predisposed to seek peaceful means to gain access to political power. This is because, as with all parties, ethnic attractors seek to act on the behalf of a constituency and seek leverage for that constituency. In turn voters, who use ethnic identity as a shortcut to sort through candidate preferences, prefer parties that act on behalf of the ethnic constituency (this could be a nonethnic party as well). This provides for a strong incentive for the ethnic attractor to gain access to the political executive, and this is best achieved through peaceful means.24 Why is it the case, then, that members of some ethnic groups appear to peacefully support their group in electoral politics, while others do not support their groups, exit electoral politics and even engage in protest and violence? Her answer is that if political intransigence and violence result, it

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

60

J. Ishiyama

is not because of the ethnification of politics but rather the denial of political access to an ethnic group. It is the shutting out from the core of power that produces the kinds of violence and instability that is commonly associated with ethnic politics in the existing literature. Despite the debate on the central role played by ethnic parties in the literature, there is little in the way of a direct test as to whether or not the appearance of an ethnic party promotes conflict. Rather much of the literature (particularly the “contingent” approach) emphasizes the importance of incentives facing ethnic parties as a way to contain conflict as opposed to directly assessing whether such parties have an independent effect on communal conflict. Does the mere appearance of an ethnic party impact violence and protest (independently from other exogenous factors)?

OTHER FACTORS AFFECTING ETHNIC CONFLICT Beyond the debate as to whether ethnic parties independently promote ethnic conflict, there are a number of other control variables that need to be considered. In particular, these include political institutions, cultural and economic differences between the ethnic group and other groups in society, and contextual factors such as the level of globalization, economic growth, and resources. Much of the literature, including both those who contend that the presence of ethnic parties exacerbates conflict and those who argue that ethnic parties play a positive role in consolidating new democracies, point to the importance of political institutions. In particular scholars have long pointed to the importance of the electoral system, the structure of the executive, and federalism in affecting the course of ethnic politics. Indeed, once cultural identity is politicized, the sustained intensity of that political relevance depends on state institutions. Those institutions define the rules of political membership, representation, and resource allocation. When these institutions structure membership, representation, and resource allocation according to previously established cultural criteria, “identity politics” dominate the political game. Alternatively, if cultural identity had not been previously politicized, institutions could create that political relevance or prevent its initial emergence. Indeed, institutional rules and procedures can be structured in ways that prevent cultural identity from becoming politically relevant at all.25 Advocates of the consociational school (but others as well) have long argued that Proportional Representation (PR) electoral systems combined with parliamentary government provide the representation that allows for the incorporation of ethnic interests, thus diffusing extremist demands.26 The focus on the independent effect of political institutions is particularly true of the consociationalist approach to ethnic party behavior. Although, as

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Ethnic Parties and Minority Ethnic Conflict

61

Pippa Norris (2002) rightly points out, the electoral system, while important, really is only one component in consociational systems of democracy, the bedrock of consociational theory is the assertion that Proportional Representation (PR) electoral systems combined with parliamentary government are the fundamental institutions upon from which many other arrangements flow. PR provides the representation that allows for the incorporation of ethnic interests, thus diffusing extremist demands. On the other hand as John McGarry and Brendan O’Leary note: “a majoritarian system of liberal democratic government, designed to create strong powers for the governing party, is no guarantee of liberty for ethnic minorities. A ‘winner takes all’ system in the presence of ethnic parties ensures that ethnic competition will be regarded as a zero-sum conflict.”27 Stephen Saideman, David Lanoue, Michael Campenni, and Samuel Stanton also find that more proportional systems are better at containing ethnic conflict than are majoritarian or plurality systems.28 Others, however, question the wisdom of adopting PR systems to “remedy” the ethnic politics. George Tsebelis and Paul Brass for instance argue that longer term use of term proportional arrangements may serve to reinforce and perpetuate rigid segregation along narrow ethnic-cultural, religious, and linguistic cleavages.29 Joel Barkan warns of the dangers of PR producing extreme multipartism and fragmentation, which may promote instability in new democracies.30 Benjamin Reilly cautions against the “one size fits all” arguments in support of PR and argues that district-based systems can control and channel ethnic political demands in potentially constructive ways.31 In addition to electoral rules, there is also the argument that presidentialism promotes conflict as well. From this perspective, presidentialism promotes “zero-sum” politics in which the winner takes all and the loser receives nothing.32 In a presidential system where the executive is powerful, this essentially leads to the exclusion of minorities from the centers of power. This is particularly problematic in countries that are ethnically divided. Thus, for Mainwaring presidential systems inherently militate against “meaningful representation” and participation “in governing coalitions.”33 Although this is also possible in a parliamentary system where a majority can run roughshod over a political minority, this is unlikely; what is more likely is that parties in the parliamentary system will share power in coalitions. A corollary to this argument is posed by Johanna Birnir who contends that what really matters is whether the ethnic group is represented in the political executive.34 Thus minorities do not totally lose out and hence are provided with a larger stake in maintaining the existing system.35 On the other hand, Donald Horowitz contends that parliamentarianism is just as much to blame for political instability as presidentialism, particularly if one examines the record of postcolonial Asia and Africa. For instance, in Nigeria, where, under a parliamentary system, a “cluster of ethnic groups”

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

62

J. Ishiyama

acquired a majority of legislative seats and totally excluded all other groups from political power, parliamentarianism contributed to the Biafran civil war. On the other hand, in presidential systems, Horowitz argues that the division of powers by different parties automatically guarantees a decidedly nonwinner-take-all result that prevents the monopoly of power by one group over another. The real culprit for the “exclusionary” characteristics of a system is not presidentialism, but the electoral system, that is, the method by which presidents and legislators are elected to office. He contends that, in essence, the argument critics of presidentialism employ “boils down to an argument not against the presidency but against plurality election, not in favor of parliamentary systems but in favor of parliamentary coalitions.”36 Federalism is also an often cited structural “remedy” to ethnic conflict.37 Generally many scholars have argued that federalism provides access for representation (particularly at regional levels) for ethnic political demands.38 By doing so, this helps channel ethnic demands in more constructive ways and helps dampen the intensity of ethnic conflict. Further, by lowering the level of ethnic competition to the local or regional level, federalism helps “insulate” the political center from ethnic conflict. Others have suggested that cultural differences fuel ethnic conflict; although there has been some debate here as well. Whereas many political science scholars (see for instance Robert Kaplan and Daniel Moynihan, and to some extent Samuel Huntington)39 suggest that conflict is more likely when cultural groups are different, political psychologists like Arjun Appadurai, Julie Kristeva, and Vamik Volkan contend cultural clashes are more likely when groups share common characteristics.40 Arjun Appadurai, in particular, contends cultural clashes are more likely when groups share common characteristics. He explores the links between ethnic violence and globalization, arguing that bodily violence between social intimates is a means of trying to fix or stabilize ethnic identity amid the uncertainties of the post-Cold War world. In particular, Appadurai concerns himself with violence involving persons and groups with some prior degree of familiarity such as neighbors and kinsmen. He focuses on the connection between the indeterminacy of ethnic boundaries and brutality. The killing, torture, and rape associated with ethnic violence is not simply a means of eliminating “other” but “involves the use of the body to establish the parameters of this otherness, taking the body apart, so to speak, to divine the enemy within.”41 For Appadurai, this search for certainty through self-mutilation occurs as a result of the uncertainties of the current era: “The maiming and mutilation of ethnicized bodies is a desperate effort to restore the validity of somatic markers of otherness in the face of uncertainties posed by census labels, demographic shifts, and linguistic changes, all of which make ethnic affiliations less somatic and bodily, more social and elective.”42 Thus for Appadurai, ethnic violence is most likely to occur not between groups that are culturally different, but ironically between groups that are culturally similar.

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Ethnic Parties and Minority Ethnic Conflict

63

There has also been a growing interest in the linkage between globalization and ethnic conflict with many scholars viewing globalization as bringing about a new era of intense cultural and ethnic conflicts.43 The general argument made by advocates of this approach is that globalization has intensified nationalistic and localistic sentiments as a result of greater international economic integration. Inequities generated by globalization, or so the argument goes, generate parochial forms of resistance rooted in an imagined past that never was. The homogenization of culture brought about by globalization elicits reactions that exalt differences and local particularisms.44 Globalization forces a materialistic and superficially universalistic set of Western values on the rest of the world, and this elicits a violent reaction. The homogenizing influence of globalization creates strong opposition to wholesale adoption of the values and standards of the international community. Further, the growth of communication and electronic media (including television, but increasingly the Internet) erodes traditional values and moral restraints, by introducing the “glitz of Hollywood.” In addition, economic integration, perhaps the most seductive form of globalization, has tremendous social and political consequences, including potentially triggering ethnic and cultural conflict. The global spread of market capitalism, and its attending shrinking transport costs and growing communications facilities means that both goods and people can more easily move from one country to another.45 The movement of peoples and goods leads to competition for jobs and other economic benefits, and an expanding set of material desires and aspirations, most of which are frustrated, creating greater resentments and the desire to single out scapegoats. Many theorists thus suggest a direct connection between economic globalization and ethnic conflict. For instance, as Wright and Macmanus argued, although earlier there was considerable hope that economic interdependence would bring about “world harmony,” in fact the new intimacy brought about by increased trade and financial flows, and other forms of economic interdependence has provoked “political and cultural backlashes that were the seeds of serious conflict.”46 Other scholars have also pointed to the importance of economic disparities as a source of ethnic conflict (or economic grievances). It is argued that economic disparity, as captured by differential rates of poverty between any two ethnic, religious, or socioeconomic groups of people, would directly result in conflicts over resources.47 Ethnic conflict has also been seen as the result of economic competition between ethnically differentiated segments of the working class or between ethnically differentiated traders and customers.48 Thus, based upon this, we would expect that communal conflict would be greater under conditions of economic disparity. In addition, there are other economic factors that also stimulate conflict. These are less related to “grievances” and more to economic “greed.” As Paul Collier and James Fearon and David Laitin have noted, the availability of resources provides an incentive for “greed” that in turn provides incentives to

64

J. Ishiyama

engage in conflict.49 In particular, a most important risk factor is that countries that have a substantial share of their income (Gross Domestic Product [GDP]) coming from the export of primary commodities are radically more at risk of conflict.

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

WHAT IS AN ETHNIC PARTY? There have been several recent attempts to define an ethnic party. For instance, Gunther and Diamond contend that a “purely ethnic party seeks only to mobilize the votes of its own ethnic group [italics in original].”50 Similarly, Kanchan Chandra and Daniel Metz define an ethnic party as a “party that overtly represents itself to the voters as the champion of the interests of one ethnic group or a set of groups to the exclusion of another or others, and makes such a representation central to its mobilizing strategy.”51 Both of these definitions emphasize how the parties portray themselves and presuppose the existence of competitive elections (which makes considerable sense given that what gives life to parties is the prospect of power). In this sense both of these definitions conceive of parties as primarily electoral organizations that seek power, which is consistent with the general definition of political parties as offered by Leon Epstein, Giovanni Sartori, and Kenneth Janda.52 In a similar vein, Herbert Kitschelt argues that the defining feature of ethnic parties (which he refers to as “particularistic sociocultural parties”) is that they limit their appeal to a particular ethnic or regional constituency and “explicitly seek to draw boundaries” between ethnic “friends” and “foes.”53 For Kitschelt the ethnic party does not pursue a universalistic program but rather seeks to secure material and political benefits for the ethnic group. Thus, unlike other types of political parties, electoral mobilization is not intended to attract additional voters outside of the group to support the party.54 As a result, the potential electoral base of the party is defined and limited by ethnicity. Thus, the ethnic party, according to Brass, in essence, strives to become “that one political organization dominant in representing the demands of the ethnic group against its rivals.”55 For the purposes of this paper I employ Brass’s definition of the ethnic party as an organization that seeks to explicitly represent the interests of the ethnic group and use the operationalization of that definition as proposed by Chandra and Metz. An ethnic party is identified as the first party to either (a) proclaim itself as the primary representative of the ethnic group and only that group OR (b) is widely regarded as the first party to represent the interests of that group and only that group. Thus, this definition would include parties that identify themselves as the representative of a particular group (such as the Magyar Coalition in Slovakia or the Bodo Peoples’ Liberation Front in India) as well as parties that are widely regarded as ethnic despite

Ethnic Parties and Minority Ethnic Conflict

65

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

proclaiming themselves officially as nonethnic (such as the Movement for Rights and Freedoms in Bulgaria, which is largely a party of ethnic Turks). In this article I limit the inquiry to include only those countries that have had at least one reasonably competitive election from 1985–2003, since the ethnic outbidding argument is premised on the existence of some degree of political competition. Of the 213 ethnic groups in the sample, 89 were represented by a party that explicitly claimed to represent (and only represent) the ethnic group and met the criteria as an ethnic party as identified above.

DATA As indicated above, this article is interested in testing the relationship between the presence of an ethnic party and protest and communal conflict. Thus, the focus is not on all ethnic groups, but only groups that are potential candidates to engage in violent conflict. This includes all “groups that define themselves using ethnic criteria (who) make claims on behalf of their collective interests against the state, or against other political actors.”56 For the purposes of this paper, then, the units of analysis are the minorities at risk. A minority at risk is defined by Gurr as a group that “collectively suffers or benefits, from systematic discriminatory treatment vis-`a-vis other groups in the country or countries in which it resides” or it “is the focus of political mobilization and action in defense or promotion of the group’s self-defined interests.”57 The MAR data have been criticized extensively over the years, including the lack of conceptual clarity and the use of ambiguous terminology, the lack of demonstrable intercoder reliability, sample selection bias, and questions regarding the validity of the coding procedures.58 However despite these problems the MAR data are unquestionably the most comprehensive collection of data that records ethnic group characteristics and behaviors, both violent and nonviolent. The most recent data includes a sample of 337 individual ethnopolitical groups and are thus very well suited for broad comparative analysis.59 One of the principal problems facing the MAR is that of sample-selection bias. A solution employed by Johanna Birnir begins by first identifying an appropriate universe of cases exogenously to MAR.60 This sample is then selected from the cases that are available in the MAR data set. In this paper, I began with the cases defined exogenously by Birnir and then use the sample from the MAR she derived. However, I further limit the sample by using the following criteria: • I only include in the data set those minorities at risk in countries in which a particular, identifiable ethnic or racial group is present (thus excluding groups regional or religious groups such as those labeled as “northerners”

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

66

J. Ishiyama

or “Sunni”) for the years 1985–2003 (roughly the beginning of the end of the Cold War with the ascension of Gorbachev in the Soviet Union and the latest year in the data set). • Further, since the focus of virtually all of the literature on the effects of ethnic parties presupposes the existence of some type of competitive election, I limit the country cases to only those that have had at least one relatively competitive election (even if only a limited or flawed election) from 1985–2003. Operationally, this meant a score of at least a −3 on the Polity IV, Polity2 democracy-autocracy score. • I focus only on developing countries rather than the richest countries in the world (for example, the high income Organization for Economic Cooperation and Development countries as defined by the World Bank), largely because developing countries are at greater risk for democratic reversal as the result of ethnic conflict than are the wealthier (and more democratically consolidated high-income countries). This left a total of 213 groups (from 82 countries) selected from the MAR data base. The list of countries, minority groups, ethnic parties, and founding dates of these parties are provided in Table 1.

VARIABLES Since I am interested in how ethnic parties affect the activities of ethnic groups, for this study there are two primary dependent variables: (a) the extent to which ethnic group members engage in various forms of protest; (b) the extent to which the ethnic group has engaged in communal violence with other groups. Both of these variables were taken directly from the Minorities at Risk (MAR) Phase IV data set. Protest is measured using the annual protest scores for the period 1985–2003 from the MAR. Protest scores were coded from “0” to “5” where: 0 = No protest reported 1 = Verbal Opposition (Public letters, petitions, posters, publications, agitation, etc.) 2 = Symbolic Resistance: Scattered acts of symbolic resistance (for example, sit-ins, blockage of traffic, sabotage, symbolic destruction of property) or political organizing activity on a substantial scale. 3 = Small Demonstrations: A few demonstrations, rallies, strikes, and/or riots, total participation of less than 10,000. 4 = Medium Demonstrations: Demonstrations, rallies, strikes, and/or riots, total participation of less than 100,000. 5 = Large Demonstrations: Mass demonstrations, rallies, strikes, and/or riots, total participation greater than 100,000.

Ethnic Parties and Minority Ethnic Conflict

67

TABLE 1 List of Countries, Ethnic/Communal Groups, Ethnic/Communal Parties, and Year Party Founded

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Country

Group

AFGHANISTAN AFGHANISTAN ALBANIA ALGERIA

HAZARAS UZBEKS GREEKS BERBERS

ANGOLA ANGOLA ANGOLA ARGENTINA AZERBAIJAN AZERBAIJAN AZERBAIJAN BANGLADESH

BAKONGO OVIMBUNDU CABINDA JEWS ARMENIANS LEZGINS RUSSIANS CHITTAGONG HILL TRIBES

BANGLADESH BANGLADESH BELARUS BELARUS BOSNIA BOSNIA BOSNIA BOTSWANA BRAZIL BRAZIL BULGARIA

HINDUS BIHARIS RUSSIANS POLES SERBS CROATS MUSLIMS SAN BUSHMEN AFRO-BRAZILIANS AMAZONIAN INDIANS TURKS

BULGARIA BURUNDI BURUNDI CAMBODIA CAMBODIA CAMEROON

ROMA HUTUS TUTSIS CHAMS VIETNAMESE KIRDIS

CAMEROON COLOMBIA COSTARICA CROATIA CROATIA CYPRUS CZECHREP

BAMILEKE BLACKS ANTILLEAN BLACKS SERBS ROMA TURKISH CYPRIOTS ROMA

CZECHREP DEM. REP. CONGO DEM. REP. CONGO DEM. REP. CONGO DEM. REP. CONGO DEM. REP. CONGO DEM. REP. CONGO DEM. REP. CONGO DEM. REP. CONGO DJIBOUTI

SLOVAKS BAKONGO LUBA LINGALA LUNDA, YEKE KIVU REGION NGBANDI HUTUS TUTSIS AFARS

Ethnic party

Year party founded

Hizb-e-Wahdat National Islamic Party Omonia Rally for Culture and Democracy

1990 1991 1992 1989

Sadval

1990

Chittagong Hill Tribal People’s Coordination Association

1972

Union of Belarusan Poles Serb Democratic Party Croatian Defense Council Party of Democratic Action First People of Kalahari Unified Black Movement

1990 1990 1991 1989 1991 1978

Movements for Rights and Freedoms Democratic Roma Union

1990 1990

Movement for the Defense for the Republic

1992

Serb People’s Party Party of Roma of Croatia Republican Turkish Party Romani Democratic Initiative

1991 1991 1970 1989

Front for the Restoration 1991 and Unity of Democracy (Continued on next page)

68

J. Ishiyama

TABLE 1 List of Countries, Ethnic/Communal Groups, Ethnic/Communal Parties, and Year Party Founded (Continued)

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Country

Group

ECUADOR EGYPT ESTONIA ETHIOPIA ETHIOPIA ETHIOPIA ETHIOPIA ETHIOPIA

BLACKS COPTS RUSSIANS AFARS ERITREANS NILO-SAHARANS OROMO SOMALIS

ETHIOPIA

TIGRAYANS

ETHIOPIA

AMHARA

FIJI GEORGIA

EAST INDIANS ABKHAZIANS

GEORGIA GEORGIA

ADZHARS OSSETIANS (SOUTH)

GEORGIA GHANA GHANA GHANA GUINEA GUINEA GUINEA GUYANA GUYANA HUNGARY INDIA INDIA INDIA INDIA INDIA INDIA INDIA INDIA

RUSSIANS ASHANTI EWE MOSSI-DAGOMBA FULANI MALINKA SUSU AFRICANS EAST INDIANS ROMA KASHMIRIS NAGAS SANTALS SIKHS MIZOS TRIPURAS ASSAMESE BODOS

INDONESIA INDONESIA

CHINESE EAST TIMORESE

INDONESIA INDONESIA IRAN

PAPUANS ACEHNESE AZERBAIJANIS

IRAN IRAN IRAN IRAN IRAN IRAN IVORY COAST

BAHA’IS BAKHTIARI BALUCHIS KURDS TURKMEN ARABS LEBANESE

Ethnic party

Year party founded

Russian Party of Estonia Afar Liberation Front

1991 1992

Oromo Liberation Front Ogaden National Liberation Front Tigrayan People’s Liberation Front All Amhara People’s Organization Fiji Labor Party AYDGYLARA (Popular Front of Abkhazia)

1973 1984

ADEMON NYKHAS (Ossetian Popular Front)

1989

Union for the New Republic Guinean People’s rally

1992 1992

People’s National Congress

1957

Roma Social Coalition National Conference Party

1991 1956

Shiromani Akali Dal Mizo National Front Tripura Upajati Juba Samiti Assam People’s Association Bodo People’s Progressive Front

1920 1956 1969 1985 1967

Timorese Democratic Movement Free Papua Movement Free Aceh Movement Azerbaijani United Islamic Front

1974

Arab Council Movement

1975 1992 1985 1988

1965 1976 2002

1979

(Continued on next page)

Ethnic Parties and Minority Ethnic Conflict

69

TABLE 1 List of Countries, Ethnic/Communal Groups, Ethnic/Communal Parties, and Year Party Founded (Continued)

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Country

Group

JORDAN KAZAKHSTAN KAZAKHSTAN KENYA KENYA KENYA KENYA KENYA KENYA KENYA KENYA KYRGYZSTAN KYRGYZSTAN

PALESTINIANS RUSSIANS GERMANS KIKUYU LUO MAASAIS SOMALIS TURKANA/POKOT KALENJINS LUHYA KISII RUSSIANS UZBEKS

LATVIA LEBANON

RUSSIANS DRUZE

LEBANON LEBANON LIBERIA LITHUANIA LITHUANIA MACEDONIA

MARONITE CHRISTIANS PALESTINIANS AMERICO-LIBERIAN POLES RUSSIANS ALBANIANS

MACEDONIA MACEDONIA

SERBS ROMA

MADAGASCAR MALAYSIA MALAYSIA MALAYSIA MALAYSIA

MERINA CHINESE DAYAKS EAST INDIANS KADAZANS

MALI MALI MEXICO MEXICO MOLDOVA MOLDOVA NAMIBIA NAMIBIA NAMIBIA NIGER NIGER NIGER NIGERIA NIGERIA NIGERIA

TUAREG MANDE MAYANS ZAPOTECS GAGAUZ SLAVS SAN BUSHMEN BASTERS EAST CAPRIVIANS DJEREMA-SONGHAI HAUSA TUAREG HAUSA-FULANI IBO OGANI

NIGERIA

YORUBA

Ethnic party

Party of National Unity and Concord Russian Citizens Party Druze Progressive Socialist Party National Liberal Party

Polish Electoral Action Union of Russians Party for Democratic Prosperity Democratic Party of Serbs Democratic Progressive Party of Roma in Macedonia

Year party founded

1993 1995 1949 1958

1994 1995 1990 1992 1992

Democratic Action Party Party Bansa Dayak Sarawak Malaysian Indian Congress United Pasok-Momogun Kadazan Organization

1966 1983 1957 1963

Gagauz Peoples Party

1989

Igbo People’s Congress 1994 Movement for the Survival of 1990 the Ogoni People Oodua People’s Congress 1995 (Continued on next page)

70

J. Ishiyama

TABLE 1 List of Countries, Ethnic/Communal Groups, Ethnic/Communal Parties, and Year Party Founded (Continued)

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Country

Group

NIGERIA

IJAW

PAKISTAN PAKISTAN

AHMADIS BALUCHIS

PAKISTAN PAKISTAN

HINDUS PASHTUNS (PUSHTUNS) SINDHIS MOHAJIRS BLACKS CHINESE BOUGANVILLEANS

PAKISTAN PAKISTAN PANAMA PANAMA PAPUA N.G. PERU PHILIPPINES

BLACKS (AFRO-PERUVIANS) IGOROTS

PHILIPPINES

MOROS

REP. OF CONGO ROMANIA ROMANIA

LARI GERMANS MAGYARS (HUNGARIANS) ROMA CHECHENS TATARS KARACHAY ROMA AVARS INGUSH LEZGINS BURYAT

ROMANIA RUSSIA RUSSIA RUSSIA RUSSIA RUSSIA RUSSIA RUSSIA RUSSIA RUSSIA RUSSIA RUSSIA SENEGAL

KUMYKS TUVINIANS YAKUT DIOLAS IN CASAMANCE

SIERRA LEONE SIERRA LEONE SIERRA LEONE SIERRA LEONE SINGAPORE SLOVAKIA

CREOLES LIMBA MENDE TEMNE MALAYS HUNGARIANS

SLOVAKIA SOUTH AFRICA SOUTH AFRICA SOUTH AFRICA SOUTH AFRICA

ROMA ASIANS COLOREDS XHOSA ZULUS

Ethnic party

Year party founded

Movement for the Survival of the Ijaw

1992

Baluchi People’s Liberation Front

2000

Mohajir Qaumi Movement

1986

Bouganvillean People’s Congress

1999

Cordilleran People’s Movement Moro National Liberation Front

1984

Democratic Alliance of Hungarians in Romania Democratic Union of Roma

1989

Ittifak

1990

Avar National Movement

1990

Sadval (Unity) Buryat-Mongolian People’s Party Tenglik-Equality

1990 1990

1971

1990

1989

Diola Movement of Democratic Forces in Casamancai

1982

Hungarian Coaltion-Coexistence Democratic Union of Roma

1990 1990

Inkatha Freedom Party 1975 (Continued on next page)

Ethnic Parties and Minority Ethnic Conflict

71

TABLE 1 List of Countries, Ethnic/Communal Groups, Ethnic/Communal Parties, and Year Party Founded (Continued)

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Country

Group

SRI LANKA SRI LANKA

INDIAN TAMILS SRI LANKAN TAMILS

TAIWAN TAIWAN TAIWAN TAJIKISTAN TAJIKISTAN TANZANIA THAILAND THAILAND TOGO TOGO TURKEY TURKEY UGANDA UGANDA UGANDA UGANDA UGANDA UGANDA UGANDA UGANDA UGANDA UGANDA UKRAINE

ABORIGINAL TAIWANESE MAINLAND CHINESE TAIWANESE RUSSIANS UZBEKS ZANZIBARIS CHINESE MALAY-MUSLIMS EWE KABRE KURDS ROMA ACHOLI ANKOLE BAGANDA KAKWA KARAMOJONG KONJO/AMBA LANGI LUGBARA/MADI LUGBARA/MADI NYARWANDA RUSSIANS

UKRAINE UKRAINE VENEZUELA YUGOSLAVIA

CRIMEAN TARTARS CRIMEAN RUSSIANS BLACKS KOSOVO ALBANIANS

YUGOSLAVIA

CROATS A

YUGOSLAVIA YUGOSLAVIA

SLOVENES HUNGARIANS

YUGOSLAVIA

SANDZAK MUSLIMS

YUGOSLAVIA

ROMA

YUGOSLAVIA ZAMBIA ZAMBIA ZAMBIA ZIMBABWE

SERBS BEMEBE LOZI TONGA NDEBELE

Ethnic party

Year party founded

Tamil United Liberation Front

1972

Barisan National Revolution

1960

Kurdish Workers Party

1984

Russian Movement in Ukraine

2000

Russian Party of Crimea

1999

Democratic League of Kosova Democratic Alliance of Croats of Vojvodina

1989

Democratic Community of Hungarians in Vojvodina Muslim Party of Democratic Action Democratic Political Party of Roma

1991 1990 1990 1990

Sources: Arthur Banks, T. Muller, and W. E. Overstreet, Political Handbook of the World (Washington DC: CQ Press, various years); Janusz Bugajski, Political Parties of Eastern Europe: A Guide to Politics in the Post-Communist Era (Armonk, NY: ME Sharpe, 2002); Alan Day, Political Parties of the World, 6th Edition. (Farmington Hills, MI: Gale Group, 2005); Minorities at Risk Project, http://www.cidcm.umd.edu/mar/ [accessed 13 June 2007].

72

J. Ishiyama

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

The second measure the annual “communal conflict index” measures the extent to which the particular minority group is engaged in conflict with other groups. As with the protest measure, the communal conflict index is measured ordinally and is coded using the following values. Each measure was coded per year from 1985–2003. 0 1 2 3 4 5 6

= = = = = = =

No conflict manifest Acts of harassment Political agitation Sporadic violent attacks Antigroup demonstrations Communal rioting Communal warfare

The primary independent variable, presence of an ethnic party, is coded as a simple dummy variable where “1” connotes the presence of such a party and “0” indicates the absence of such a party, per country, per year. In addition, other alternative explanatory variables are included in the analysis. These are democratization, the level of constraints on the executive (which is a rough surrogate for presidential systems), a measure of the electoral system, federalism, the extent to which an ethnic group is regionally concentrated, cultural differences between the minority group and other groups in a country, and economic differences between the minority group and other groups in the country. To measure democratization, I use the “Revised Combined Polity Score” (Polity2) from the Polity IV Project for each country from 1985–2003. The POLITY score ranges from +10 (strongly democratic) to −10 (strongly autocratic). To roughly measure presidentialism I use a surrogate, the “constraints on the executive” variable from the Polity IV data base. Operationally, this variable refers to the extent to which there are institutionalized constraints on the decision-making powers of chief executives. Such limitations may be imposed by any “accountability groups.” Given that presidential systems are characterized by powerful, less constrained executives than mixed or parliamentary systems, this measure can be used as an indicator of the strength of executive authority. A seven-category scale is used. XCONST Executive Constraints Index 0 = Unlimited Authority: There are no regular limitations on the executive’s actions. 1 = Intermediate Category 2 = Slight to Moderate Limitation on Executive Authority: There are some real but limited restraints on the executive. 3 = Intermediate Category

Ethnic Parties and Minority Ethnic Conflict

73

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

4 = Substantial Limitations on Executive Authority: The executive has more effective authority than any accountability group but is subject to substantial constraints by them. 5 = Intermediate Category 6 = Executive Parity or Subordination: Accountability groups have effective authority equal to or greater than the executive in most areas of activity. To include this measure in the regression analyses below, I created a “dummy” variable by combining categories 1–4 and coding this as “0” and combining 5, 6, and 7 and scoring this as a “1.” As a measure of the electoral system I use the natural log of the average district magnitude for lower house legislative elections. This measure was advocated by Rein Taagepera and Matthew Shugart as the best measure of the electoral system in that the district magnitude (defined as the average number of seats per legislative district) was the single most important electoral system’s dimension in affecting party behavior.61 I take the measure from the World Bank’s Data on Political Institutions data base (DPI2004).62 To measure whether or not a country was federal in a given year, I refer to the list of federal states provided by the Forum of Federations organization (http://www.forumfed.org/). I code this variable simply as a dummy variable, where “1” connotes a federal system and “0” otherwise. Finally, as argued by Birnir, access to the executive is a key variable that also explains the extent to which conflict and protest occur. To measure whether the party had access to the executive, I use a very simple dummy variable to measure whether the identified ethnic party held a cabinet portfolio and was part of the governing coalition in any year between 1985–2003. Another relevant variable that impacts the behavior of ethnic group is the extent to which the group’s population is regionally or geographically concentrated.63 The greater the extent of geographic concentration, it is argued, the more likely the ethnic group will engage in political activities and the greater the temptation for separatism. Thus we would expect that the greater the level of geographic concentration the more likely the group will engage in acts of protest and communal conflict. GROUPCON Groups spatial distribution Value Label 0 = Widely dispersed 1 = Primarily urban or minority in one region 2 = Majority in one region, others dispersed 3 = Concentrated in one region This variable is also “dummied” where 0 and 1 are coded as “0” and 2 and 3 are coded as “1.”

74

J. Ishiyama

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

To measure cultural and economic differentials, I employ the ordinal measures developed by the MAR project. This cultural differentials index was based upon the assessment of differences regarding six dimensions: Different Ethnicity/Nationality, Different Language, Different Historical Traditions, Different Religion, Different Social Customs, and Different Residence. To determine the degree of difference between the target minority group and other population groups, in terms of cultural differentials, the six dimensions were combined into a composite index. CULDIFXX Cultural Differentials Index Value Label 0 = No Differences 1 = Slight Differentials 2 = Substantial Differentials 3 = Major Differentials 4 = Extreme Differentials As with the previous variables, I created another dummy variable by combining the values 0 and 1 (coded as “0”) and combining the values 2, 3, and 4 (coded as “1”). The Economic Differentials Index is coded as a seven-point scale. It is based upon six economic dimensions, including: Income, Land/Property, Higher Education, Presence in Commerce, Presence in Professions, and Presence in Official Positions. As with the political differentials index, the economic differentials index also ranges from −2 to +4, with the low score indicative of substantial economic advantages enjoyed by the minority group, and +4 indicating substantial economic disadvantages faced by the minority group. Again with this variable I create another dummy variable combining the scores −2 to 0 (coded as “0”) and 1 to 4 (coded as “1”). Three additional control variables I employ relate to the national economic context and the likelihood of protest and conflict. These include a measure for globalization (Foreign Direct Investment [FDI] as a proportion of Gross Domestic Product) and the GDP per capital annual growth rates. Further as measure of the dependence on primary commodities, I use the measure that Collier and Hoeffler use regarding primary commodity exports as a percentage of GDP. All of these measures were derived from the World Bank’s World Development Indicators.

ANALYSIS AND RESULTS Did the presence of an ethnic party exert an independent effect on the annual level of protest engaged in by the ethnic group in a given country in a given year from 1985–2003? In Table 2, I report the results of two models that regress the dependent variable, the annual protest index, against the

75

Ethnic Parties and Minority Ethnic Conflict

TABLE 2 Coefficient Estimates and Collinearity Diagnostics for Ordered Logit Models, Dependent = Annual Protest Index

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Model 1 (Ordinal Logit) (standard error) [VIF] Lag Protest Ethnic Party Federalism Dummy Variable Constraints on Executive Authority Access to Executive Dummy Polity2 Log Average District Magnitude Group Concentration Dummy Cultural Differentials Dummy Economic Differentials Dummy Foreign Direct Investment/Gross Domestic Product (GDP) GDP per Capita Annual Growth Rate in Percentages Primary Commodity Exports as percent of GDP ∗p

≤ .05, ∗∗ p ≤ .01, N = 3795.

∗∗∗ p

1.46∗∗∗ .37∗∗∗ .20∗∗∗ .06∗∗ −.08∗∗

(.04) (.09) (.10) (.02) (.01)

[1.18] [1.21] [1.13] [1.23] [1.13]

.04 −.03 .29∗ −.08 −.01

(.03) (.09) (.12) (.09) (.01)

[1.15] [1.04] [1.12] [1.07] [1.15]

−.01 .00

Model 2 (Ordinal Logit) 1.44∗∗∗ (.04) [1.17] .38∗∗∗ (.09) [1.20] .19∗∗ (.10) [1.18] −.03∗∗ .01∗∗ .03 −.03 .31∗∗ −.09 −.02

(.01) [1.15] (.01) [1.31] (.03) [1.17] (.10) [1.09] (.12) [1.12] (.08) (1.07) (.01) [1.14]

(.01) [1.09]

−.01

(.01) [1.10]

(.00) [1.18]

.00

(.00) [1.19]

Pseudo R 2 = .55

Pseudo R 2 = .54

≤ .001.

eight independent variables (along with a lagged endogenous variable as a corrective for autocorrelation). In addition to the coefficient estimates and standard errors, I also report the Variance Inflation Factor (VIF) scores to detect problems with multicollinearity.64 Model 1 reports the results of a time series cross-sectional (TSCS) model with a lagged endogenous dependent variable. Since the dependent variable is ordinal, I ran ordinal logit models. As indicated by the Durbin Watson statistic, there is little in the way of autocorrelation after including the corrective lagged endogenous dependent variable. Further after conducting a White’s test there appeared to be no problems with heteroscedasticity—hence corrective measures (such as a White Correction) were not required. Since I found a high degree of collinearity between the Polity2 variable and the XCONS (constraints on the executive) measure (VIF = 10.91), I ran two separate models, dropping the XCONS in one model, and the Polity2 variable in the other. From Table 2, it is clear that ethnic groups that were represented by an ethnic party were significantly more likely to engage in protest than were ethnic groups that were not represented by such parties. This is consistent with the literature that suggests that ethnic parties mobilize their constituents (but for the consociationalists they channel political activities into less extreme and violent actions). In addition, the greater the degree of democracy was also associated with the greater the level of protest. This would suggest

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

76

J. Ishiyama

that the greater the openness in the political system the more channels for nonviolent interest articulation. This is also supported by the relationship between the constraints on the executive and ethnic protest. In addition, federal systems were likely to experience more ethnic group protest than nonfederal systems (again consistent with the notion that political openings in the system promote ethnic group political mobilization). However, the electoral system had no bearing on the level of ethnic group protest, contrary to much of the consociational literature. Further, economic globalization as measured by FDI/GDP and economic growth measured by annual change in GDP per capita were unrelated to protest, as was primary commodity exports as a percentage of GDP. Importantly, however, access to the political executive had a significant dampening effect on protest. Further, groups that were culturally different from the majority were more likely to engage in protest than those that were not. This did not appear to be related to the extent to which groups were geographically concentrated nor the extent to which they were economically deprived relative to the majority population. Finally, there were no problems with multicollinearity, as indicated by the low-VIF scores, which were all less than 2.65 Second, does the presence of an ethnic party promote communal conflict? As mentioned above, this is the primary issue that divides those who argue that ethnic parties promote a cycle of destruction in new democracies, and those that argue (like Kanchan Chandra and the consociational scholars) that ethnic parties can play a more positive role. Table 3 reports the results of two models, which regress the dependent variable, the annual communal conflict index, against eight independent variables and a lagged endogenous dependent variable. As indicated in Table 2, again it appears that the presence of an ethnic party exerts a significant and independent effect on communal conflict—that is, ethnic groups that are represented by an ethnic party are significantly more likely to engage in conflict. Unlike the situation with protest, democracy and limits on executive authority have little impact on communal conflict, and federalism appeared to be only marginally related to communal conflict (surprisingly the coefficient sign was also positive, indicating that federal systems were positively related to communal conflict). Thus, although these political variables appeared to affect the level of protest, they did not relate to communal conflict. The single most important structural factor is access to the political executive—it had a significant dampening effect on communal conflict. Finally, the extent to which groups were geographically concentrated was positively and significantly related to communal conflict, as was the variable economic differences, and consistent with the work of Collier, and Fearon and Laitin, primary commodity exports as a percentage of GDP, across all four models.66 It would be tempting to conclude that ethnic parties were the cause of both increased ethnic group protest and ethnic group communal conflict (which would support scholars like Rabushka and Shepsle, and Horowitz)

77

Ethnic Parties and Minority Ethnic Conflict

TABLE 3 Coefficient Estimates and Collinearity Diagnostics for Ordered Logit Models, Dependent = Annual Communal Conflict Index

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Model 1 (standard error)) [VIF] Lag Conflict Ethnic Party Federalism Dummy Variable Constraints on Executive Authority Access to Executive Dummy Polity2 Log Average District Magnitude Group Concentration Dummy Cultural Differentials Dummy Economic Differentials Dummy Foreign Direct Investment/Gross Domestic Product (GDP) GDP per Capita Annual Growth Rate in Percentages Primary Commodity Exports as percent of GDP ∗p

≤ .05, ∗∗ p ≤ .01, N = 3795.

∗∗∗ p

1.33∗∗∗ .43∗∗∗ .35∗ .04 −.03∗∗

(.05) (.14) (.15) (.04) (.01)

[1.10] [1.17] [1.15] [1.11] [1.15]

.16∗∗ 1.05∗∗∗ .45∗ .35∗ −.03

(.04) (.25) (.22) (.14) (.02)

[1.07] [1.08] [1.10] [1.06] [1.15]

.02

Model 2 1.33∗∗∗ (.05) [1.10] .49∗∗∗ (.13) [1.12] .33∗ (.15) [1.15] −.03∗∗ .02 .16∗∗ .91∗∗∗ .32 .38∗ −.03

(.01) [1.09]

.01

.02∗∗∗ (.00) [1.18]

(.01) (.01) (.05) (.23) (.21) (.14) (.02)

[1.13] [1.13] [1.10] [1.06] [1.10] [1.05] [1.15]

(.01) [1.09]

.02∗∗∗ (.00) [1.18]

Pseudo R 2 = .62

Pseudo R 2 = .62

≤ .001.

but this would be premature. Indeed, it may be the case that the presence of an ethnic party neither causes protest nor conflict, but rather the existence of an ethnic party is the result of preexisting higher levels of protest and conflict. In other words, ethnic parties do not cause anything but are preceded by protest and conflict that give rise to them. Although this may be the case, the real question is whether the level of protest and conflict increases for the individual group AFTER the appearance of the ethnic party. To examine this in Table 4 I report a simple difference of means test that compares the protest and conflict scores before and after the first appearance of an ethnic party. As indicated in Table 4, interestingly, protest levels increase significantly after the first appearance of the ethnic party. However, there is little change in the annual communal conflict index. The appearance of an ethnic party does little in the way of increasing levels TABLE 4 Difference of Means, Annual Protest Index, and Annual Communal Conflict Index, Before and After Appearance of Ethnic Party, 1985–2003+ Mean score before appearance of party

Mean score after the appearance of party

t test

.79 .67

1.53 .59

−6.54∗∗∗ .56

Annual Protest Index Annual Communal Conflict Index +Only ∗p

includes parties that appeared after 1985. ≤ .05, ∗∗ p ≤ .01, ∗∗∗ p ≤ .001.

78

J. Ishiyama

TABLE 5 Chi-Square Statistics Annual Protest Index by Presence of Ethnic Party and Annual Communal Conflict Index by Presence of Ethnic Party Annual protest index by Annual communal conflict index presence of ethnic party by presence of ethnic party Single Member District Systems Multiple Member District Systems

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

∗p

≤ .05,

∗∗

p ≤ .01,

∗∗∗ p

136.05∗∗∗ 162.88∗∗∗

167.01∗∗∗ 69.93∗∗∗

≤ .001.

of communal conflict, at least for the period 1985–2003. These results suggest that the appearance of ethnic parties do lead to political mobilization as indicated by the significantly increased protest scores—however their appearance does not lead to an increased propensity to engage in communal conflict. However, is this effect independent of other political/institutional features such as the electoral system? Indeed as noted above, according to the consociational school, ethnic parties moderate their demands especially in the face of proportional representation electoral systems. This would suggest that systems that are broadly representative (such as higher seat magnitude systems) would be more likely to lead to the moderation of ethnic parties than in countries that use small magnitude electoral systems (such as the use of single-member districts). In Table 5, I report the results of a series of chi-square statistics, cross-tabulating the appearance of an ethnic party and protest and conflict while controlling for single versus multimember district magnitudes. As indicated there is no difference in the relationship between the appearance of an ethnic party and protest and conflict when comparing single versus multimember district systems. The appearance of an ethnic party is significantly related to both protest and conflict under both electoral systems. Ethnic parties thus exert an independent effect on protest (but as indicated in Table 4, not on communal conflict).

CONCLUSIONS The above paper examined the question of whether ethnic parties exert an independent effect on levels of protest and communal conflict. Although the results above indicated that ethnic parties do mobilize ethnic groups to engage in protest (and this effect is independent of political institutional and other contextual variables), there is no evidence that ethnic parties independently promote communal conflict. Although the appearance of ethnic parties is associated with communal conflict, the level of communal conflict was as high before the appearance of an ethnic party as it was after. These findings would also call into question whether “remedial” actions such as ethnic party bans are effective at all, given that ethnic parties do not

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Ethnic Parties and Minority Ethnic Conflict

79

independently promote conflict. In fact, they may achieve exactly the opposite than what was intended. Greater exclusion may lead to the increased attractiveness of extra legal actions on the part of the banned groups, particularly groups that are politically mobilized. The above results also suggest that better institutional “correctives” may exist rather than ethnic party bans. In particular, of the variety of purported institutional remedies to assuage conflict (electoral systems, federalism, access to the executive, and constraints on executive authority) only access to the executive by the minority at-risk group has a significant dampening effect on the group’s propensity to engage in both protest and communal conflict (a finding supportive of Johanna Birnir’s work). Access to the locus of political power thus provides a powerful incentive to dampen both ethnic political mobilization and conflict by providing groups a stake in the system. Other institutional remedies are less promising. Federalism both promotes protest AND communal conflict (contrary to the dampening effects claimed by proponents of federalism as an institutional remedy for ethnic politics). This may due to the existence of local regional governments that can (a) be targets for mobilization for groups that may be a minority nationally, but a majority regionally and (b) provide springboards for national power, thus emboldening the demands of cultural minorities.67 Presidentialism (as measured by constraints on the executive) increases protest (perhaps because as Juan Linz argues presidentialism militates against meaningful representation of minority groups) but has no independent effect on conflict. Finally, the electoral system has no discernable effect on the propensity of a group to engage in either protest OR communal conflict (contrary to much of the literature that suggests the electoral system is a key institutional variable affecting both). The above results may also suggest a “step function” in the connection between ethnic group protest and communal conflict. Although ethnic parties mobilize their primary constituencies, there is no inevitable connection between the ethnification of politics and ethnic conflict. Whether groups engage in conflict is dependent on other factors (both cultural and demographic as well as access to the executive). Although it has been popular to argue that the ethnification of politics inexorably leads to conflict and the dissolution of new democracies, the above results suggest otherwise—conflict does not follow the appearance of ethnic parties as night follows day. As suggested by Johanna Birnir, it is likely that communal conflict is caused by factors external to ethnic mobilization rather than mobilization itself. However, once conflict begins, only access to the executive appears to have any dampening effect on a group’s propensity to engage in communal conflict. This might suggest that only “grand coalitions” can provide enough inducement for groups to stop fighting once the fight begins (which may be the only solution to the current strife in places like Kenya in 2008, for example).

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

80

J. Ishiyama

Indeed, whether or not ethnic mobilization evolves into communal conflict is likely to depend heavily on state responses to ethnic mobilization. Although on the surface there appears to be a fundamental difference between the “perils of ethnification” and “contingent” approaches to the role played by ethnic parties, it should be remembered that Rabushka and Shepsle also pointed to the important role played by state responses. Perhaps it is the timing of the state response that is as important as the type of response? Perhaps access is important early in the process and other measures are required later? Although these questions are well beyond the current scope of this project (which sought only to examine the proposition that the mere appearance of ethnic parties was connected to ethnic conflict) they represent promising avenues for future inquiry.

NOTES 1. Donald Horowitz, Ethnic Groups in Conflict (Berkeley: University of California Press, 1985); Alvin Rabushka and Kenneth Shepsle, Politics in Plural Societies: A Theory of Democratic Instability (Columbus, OH: Merrill, 1972). 2. Johanna K. Birnir, Ethnicity and Electoral Politics (New York: Cambridge University Press, 2007); Kanchan Chandra, Why Ethnic Parties Succeed: Patronage and Ethnic Headcounts in India (Cambridge: Cambridge University Press, 2004). 3. Arend Lijphart, Democracy in Plural Societies: A Comparative Exploration (New Haven: Yale University Press, 1977). 4. John Ishiyama and Marijke Breuning, Ethnopolitics in the New Europe (Boulder: Lynne Rienner, 1997). 5. Matthias Basedau, Matthijs Bogaards, Christof Hartmann, and Peter Niesen, “Ethnic Party Bans in Africa: A Research Agenda,” German Law Journal, Vol. 8, No. 6 (2007), pp. 618–34, 618. 6. Anika Becher and Matthias Basedau, “Promoting Peace and Democracy Through Party Regulation? Ethnic Party Bans in Africa,” GIGA working papers 66 (Hamburg: German Institute of Global and Area Studies/Leibniz-Institut f¨ur Globale und Regionale Studien, 2008). 7. For instance, India as in Kanchan Chandra’s 2004 study. 8. Sherrill Stroschein, “Measuring Ethnic Party Success in Romania, Slovakia and Ukraine,” Problems of Post-Communism, Vol. 4, No. 1 (2001), pp. 59–69; John Ishiyama, “Institutions and Ethnopolitical Conflict In Post-Communist Politics,” Nationalism and Ethnic Politics, Vol. 6, No. 1 (2000), pp. 51–67; Ishiyama and Breuning; A notable exception is the recent work of Johanna K. Birnir, 2007. 9. Robert Hislope, “Intra-Ethnic Conflict in Croatia and Serbia: Flanking and the Consequences for Democracy,” East European Quarterly, Vol. 30, No. 4 (1997), pp. 471–94; Thomas Koelble, “Towards a Theory of Nationalism: Culture, Structure and Choice Analyses Revisited,” Nationalism and Ethnic Politics, Vol. 1, No. 1 (1995), pp. 73–89. 10. Rabushka and Shepsle, pp. 20–30. 11. Benjamin Reilly, “Political Engineering of Parties and Party Systems.” Paper presented at the Annual Meeting of the American Political Science Association, Philadelphia, PA, 2003. 12. Koeble, pp. 73–89. 13. Richard Gunther and Larry Diamond, “Types and Functions of Parties,” in Larry Diamond and Richard Gunther (Eds.), Political Parties and Democracy (Baltimore and London: Johns Hopkins University Press, 2003), pp. 23–4. 14. Becher and Basedau, p. 9. 15. Horowitz, p. 291. 16. Lijphart, pp. 1–20. Arend Lijphart, The Politics of Accommodation: Pluralism and Democracy in the Netherlands, 2nd ed. (Berkeley: University of California Press, 1974); Hans Daalder, “The Consociational Democracy Theme,” World Politics, Vol. 26, No. 4 (1974), pp. 604–21; Kenneth D. McCrae, Consociational Democracy: Political Accommodation in Segmented Societies (Toronto: McClelland and

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

Ethnic Parties and Minority Ethnic Conflict

81

Stewart, 1974); Eric A. Nordlinger, Conflict Regulation in Divided Societies (Cambridge, MA: Center for International Affairs Harvard University, 1972); Val Lorwin, “Segmented Pluralism,” Comparative Politics, Vol. 3, No. 2 (1971), pp. 141–75. 17. Frank S. Cohen, “Proportional versus Majoritarian Ethnic Conflict Management in Democracies,” Comparative Political Studies, Vol. 30, No. 5 (1997), pp. 607–30, p. 613. 18. Stroschein, p. 61. 19. Lilia Petkova, “The Ethnic Turks in Bulgaria: Social Integration and Impact on Bulgarian—Turkish Relations, 1947–2000,” Global Review of Ethnopolitics, Vol. 1, No. 1 (2002), pp. 36–55, 52. 20. Ishiyama, pp. 51–67. 21. Birnir; Chandra; James D. Fearon and David D. Laitin, “Ethnicity, Insurgency, and Civil War,” American Political Science Review, Vol. 97, No. 1 (2003), pp. 75–90; James D. Fearon and David D. Laitin, “Explaining Interethnic Cooperation,” American Political Science Review, Vol. 90, No. 4 (1996), pp. 715–735; Daniel N. Posner, “Measuring Ethnic Fractionalization in Africa,” American Journal of Political Science, Vol. 48, No. 4 (2004), pp. 849–63. 22. Kanchan Chandra, “Ethnic Parties and Democratic Stability,” Perspectives on Politics, Vol. 3, No. 3 (2005), pp. 235–52, 236. 23. Birnir, pp. 59–69. 24. Kanchan Chandra and Sherill Stroschein make very similar arguments. See Chandra; Stroschein. 25. Ishiyama, pp. 51–67. 26. Pippa Norris, “Ballots not Bullets: Electoral systems, Ethnic minorities and Democratization,” in Andrew Reynolds and Scot Mainwaring (Eds.), The Architecture of Democracy (Oxford: Oxford University Press, 2002); Cohen, pp. 607–630; Lijphart; Daalder; McRae; Nordlinger; Lorwin. 27. John McGarry and Brendan O’ Leary, The Politics of Ethnic Conflict Regulation Case Studies of Protracted Ethnic Conflicts (London: Routledge, 1993), p. 25. 28. Stephen M. Saideman, David J. Lanoue, Michael Campenni, and Samuel Stanton, “Democratization, Political Institutions, and Ethnic Conflict: A Pooled, Cross-Sectional Time Series Analysis from 1985–1998,” Comparative Political Studies, Vol. 35, No. 1 (2002), pp. 103–29. 29. George Tsebelis, Nested Games: Rational Choice in Comparative Politics (Berkeley, CA: University of California Press, 1990); Paul Brass, Ethnicity and Nationalism (New Delhi: Sage Press, 1991). 30. Joel Barkan, “Rethinking the Applicability of Proportional Representation for Africa,” in T. D. Sisk and Andrew Reynolds (Eds.), Elections and Conflict Management in Africa (Washington, DC: US Institute of Peace Press, 1998), pp. 25–45. 31. Benjamin Reilly, “Electoral Systems for Divided Societies,” Journal of Democracy, Vol. 13, No. 2 (2002), pp. 156–70; Benjamin Reilly, Democracy in Divided Societies: Electoral Engineering for Conflict Management (Cambridge, MA: Cambridge University Press, 2001). 32. Juan J. Linz, “The Perils of Presidentialism,” Journal of Democracy, Vol. 1, No. 1 (1990), pp. 51–69, 56; J. A. Cheibub, Presidentialism, Parliamentarism, and Democracy (Cambridge: Cambridge University Press, 2006), p. 18; Timothy Power and Mark J. Gasiorowski, “Institutional Design and Democratic Consolidation in the Third World,” Comparative Political Studies, Vol. 30, No. 2 (1997), pp. 123–155; Matthew Shugart and John M. Carey, Presidents and Assemblies: Constitutional Design and Electoral Dynamic (Cambridge: Cambridge University Press, 1992). 33. Scott Mainwaring, “Presidentialism, Multipartism and Democracy: The Difficult Combination,” Comparative Political Studies, Vol. 26, No. 3 (1993), pp. 198–228, 223. 34. Birnir. 35. Alfred Stepan and Cynthia Skach, “Constitutional Frameworks and Democratic Consolidation,” World Politics, Vol. 46, No. 1 (1993), pp. 1–22; Linz, pp. 51–69. 36. Donald Horowitz, “Comparing Democratic Systems,” Journal of Democracy, Vol. 1, No. 4 (1990), pp. 783–94, 784 (italics in original). 37. Horowitz; Lijphart. 38. Will Kymlicka, “Emerging Western Models of Multination Federalism: Are they Relevant for Africa?,” in David Turton (Ed.), Ethnic Federalism: The Ethiopian Experience in Comparative Perspective (Athens, OH: Ohio University Press, 2006), pp. 1–25; K. Adeney, Federalism and Ethnic Conflict Regulation in India and Pakistan (New York: Palgrave, 2006); R. Suberu, Federalism and Ethnic Conflict in Nigeria (Washington, DC: United States Institute of Peace Press, 2001).

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

82

J. Ishiyama

39. Robert Kaplan, Balkan Ghosts (New York: St. Martin’s Press, 2003); Daniel Patrick Moynihan, Pandemonium: Ethnicity in International Politics (Oxford: Oxford University Press, 1993); Samuel Huntington, “The Clash of Civilizations?,” Foreign Affairs, Vol. 72 (1993), pp. 22–49. 40. Arjun Appadurai, “Dead Certainty: Ethnic Violence in the Era of Globalization,” Public Culture, Vol. 10, No. 3 (1998), pp. 225–40; Julie Kristeva, Powers of Horror: An Essay on Abjection (New York: Columbia University Press, 1982); Vamik Volkan, Cyprus—War and Adaptation: A Psychoanalytic History of Two Ethnic Groups in Conflict (Charlottesville, VA: University Press of Virginia, 1979). 41. Appadurai, p. 247. 42. Ibid., p. 246. 43. Arjun Appadurai, “Dead Certainty: Ethnic Violence in the Era of Globalization,” Public Culture, Vol. 10, No. 3 (1998), pp. 225–24; Robert Kaplan, Balkan Ghosts: A Journey through History. (New York: Vintage Books, 1993); Samuel Huntington, “The Clash of Civilizations?”, Foreign Affairs Vol. 72, No. 3 (1993) pp. 22–49; Daniel Patrick Moynihan, Pandemonium: Ethnicity in International Politics (Oxford: Oxford University Press, 1993). 44. Yahya Sadowski, The Myth of Global Chaos (Washington, DC: Brookings Institution Press, 1998), p. 95. 45. Robin Wright and Doyle Macmanus, Flashpoints: Promise and Peril in a New World (New York: Alfred Knopf, 1991). 46. Ibid., pp. 20–21. 47. S. K. Bhaumik, Gang, and M. Yun, “Ethnic Conflict and Economic Disparity: Serbian and Albanians in Kosovo,” Journal of Comparative Economics, Vol. 34, No. 4 (2006), pp. 754–73. 48. M. D. Toft, The Geography of Ethnic Violence: Identity, Interests, and Territory (Princeton: Princeton University Press, 2003); E. Bonacich, “A Theory of Middleman Minorities,” American Sociological Review, Vol. 38, No. 4 (1973), pp. 583–94; E. Bonacich, “A Theory of Ethnic Antagonism: The Split Labor Market,” American Sociological Review, Vol. 37, No. 4 (1972), pp. 547–59; William Easterly and R. Levine, “Africa’s Growth Tragedy: Policies and Ethnic Divisions,” Quarterly Journal of Economics, Vol. 112, No. 4 (1997), pp. 1203–50. 49. Paul Collier, “Economic Causes of Civil Conflict and Their Implications for Policy,” in C. A. Crocker, F. O. Hampson, and P. Aall (Eds.), Leashing the Dogs of War (Washington, DC: United States Institute of Peace Press, 2007); Fearon and Laitin, pp. 715–734. 50. Gunther and Diamond, p. 183. 51. Kanchan Chandra and Daniel Metz, “A New Cross-national Database on Ethnic Parties,” Paper Presented at the Annual Meeting of the Midwestern Political Science Association, Chicago, Illinois, 2002, p. 5. 52. Leon Epstein defines the political party as “any group of individuals, however loosely organized, whose avowed purpose is winning elections.” Leon Epstein, Political Parties in Western Democracies (New York: Praeger, 1966), p. 1; Sartori defined a party as “any political group identified by an official label that presents at elections, and is capable of placing through elections (free or nonfree), candidates for public office.” Giovanni Sartori, Parties and Party Systems: A Framework for Analysis (Cambridge: Cambridge University Press, 1976), p. 63; Still more broadly, a party can be defined as an organization that pursues a goal of placing its avowed representatives in government positions. Kenneth Janda, Political Parties (New York: Free Press, 1990), p. 5. 53. Herbert Kitschelt, “Divergent Paths of Postcommunist Democracies,” in Larry Diamond and Richard Gunther (Eds.), Political Parties and Democracy (Baltimore, MD: Johns Hopkins University Press, 2001). 54. See Horowitz, pp. 294–7. 55. Paul Brass, p. 106. 56. Ted Robert Gurr, “People Against States: Ethnopolitical Conflict and the Changing World System,” International Studies Quarterly, Vol. 38, No. 3 (1994), pp. 347–77, 349. 57. Ibid., p. 349. 58. For a recent criticism see Vladimir Tishkov, “Status in the World System and Economic Mobilization,” Journal of Peace Research, Vol. 36, No. 4 (1999), pp. 571–91; also William J. Foltz, “Minorities at Risk: A Global View of Ethnopolitical Conflicts,” The American Political Science Review, Vol. 88, No. 2 (1994), pp. 513–14. 59. For a defense of the use of the MAR data, see Birnir, pp. 168–9. 60. Birnir, pp. 168–169.

Ethnic Parties and Minority Ethnic Conflict

83

Downloaded by [University of North Texas], [John Ishiyama] at 12:02 18 June 2014

61. Rein Taagepera and Matthew Shugart, Seats and Votes: The Effects and Determinants of Electoral Systems (New Haven: Yale University Press, 1989). 62. Phillip Keefer, DPI2004: Database of Political Institutions (Washington, DC: World Bank, 2005). 63. Shaheen Mozzafar, James Scaritt, and Greg Galaich, “Electoral Institutions, Ethnopolitical Cleavages, and Party Systems in Africa’s Emerging Democracies,” American Political Science Review, Vol. 97, No. 3 (2003), pp. 379–90; Toft. 64. John Fox, Regression Diagnostics (Newbury Park, CA: Sage, 1991). 65. Ibid. 66. Toft. 67. See Brass, 1991, for this point.

John Ishiyama is Professor of Political Science at the University of North Texas. He is author or editor of numerous books and articles on Democratization, Postcommunist Politics, and Ethnic Politics in Russia, Eastern Europe, Central Eurasia, and Africa.