I Say to You: Ethnic Politics and the Kalenjin in Kenya

By Gabrielle Lynch

In 2007 a disputed election in Kenya erupted into a two-month political crisis that led to the deaths of more than a thousand people and the displacement of almost seven hundred thousand. Much of the violence fell along ethnic lines, the principal perpetrators of which were the Kalenjin, who lashed out at other communities in the Rift Valley. What makes this episode remarkable compared to many other instances of ethnic violence is that the Kalenjin community is a recent construct: the group has only existed since the mid-twentieth century. Drawing on rich archival research and vivid oral testimony, I Say to You is a timely analysis of the creation, development, political relevance, and popular appeal of the Kalenjin identity as well as its violent potential.
 
Uncovering the Kalenjin’s roots, Gabrielle Lynch examines the ways in which ethnic groups are socially constructed and renegotiated over time. She demonstrates how historical narratives of collective achievement, migration, injustice, and persecution constantly evolve. As a consequence, ethnic identities help politicians mobilize support and help ordinary people lay claim to space, power, and wealth. This kind of ethnic politics, Lynch reveals, encourages a sense of ethnic difference and competition, which can spiral into violent confrontation and retribution.

• Language: English
• Publisher: University of Chicago Press
• Release date: Sep 26, 2011
• ISBN: 9780226498096

Book preview

Gabrielle Lynch is a senior lecturer in Africa and the politics of development in the School of Politics and International Studies (POLIS) at the University of Leeds.

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 2011 by The University of Chicago

All rights reserved. Published 2011.

Printed in the United States of America

20  19  18  17  16  15  14  13  12  11     1  2  3  4  5

ISBN-13: 978-0-226-49804-1 (cloth)

ISBN-13: 978-0-226-49805-8 (paper)

ISBN-10: 0-226-49804-2 (cloth)

ISBN-10: 0-226-49805-0 (paper)

ISBN-13: 978-0-226-49809-6 (e-book)

Library of Congress Cataloging-in-Publication Data

Lynch, Gabrielle.

I say to you : ethnic politics and the Kalenjin in Kenya / Gabrielle Lynch.

p. cm.

Includes bibliographical references and index.

ISBN-13: 978-0-226-49804-1 (cloth : alk. paper)

ISBN-10: 0-226-49804-2 (cloth : alk. paper)

ISBN-13: 978-0-226-49805-8 (pbk. : alk. paper)

ISBN-10: 0-226-49805-0 (pbk. : alk. paper)   1. Kalenjin (African people)—Kenya—Ethnic identity.   2. Kalenjin (African people)—Ethnic identity—Political aspects—Kenya.   3. Kenya—Ethnic relations—Political aspects.   4. Kenya—Politics and government.   5. Regionalism—Political aspects—Kenya.   6. Moi, Daniel Arap, 1924– I. Title.

DT433.545.K35L96 2011

305.896′5—dc22

2011003350

This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

I Say to You

Ethnic Politics and the Kalenjin in Kenya

GABRIELLE LYNCH

The University of Chicago Press

Chicago and London

CONTENTS

List of Tables

Abbreviations, Swahili Terms, and Note on Ethnic Nomenclature

Acknowledgments

Map 1: Kenya: Provinces and Area of Focus

Map 2: Area of Focus: Kalenjin-Dominated Districts

INTRODUCTION / The Nature and Political Salience of Ethnic Identity

ONE / Creating a Community: From Nandi Speakers to Kalenjin

TWO / Popularizing the Kalenjin: Decolonization and the First Majimbo Debate

THREE / Moi: The Making of an African Big Man

FOUR / Harambee to Nyayo: Control and Patronage in the President’s Backyard

FIVE / Democratization and the Kalenjin Vote, 1990–2002

SIX / Multiparty Politics and the Ethnic Factor, 2002–8

Conclusion: Ethnic Politics in Modern Kenya

Appendix: Multiparty Election and Referendum Results in Kalenjin-Dominated Constituencies

Notes

Sources

Index

TABLES

4.1 Average annual rates of real economic growth, 1965–92

5.1 Presidential and parliamentary election results, December 1992

5.2 Presidential and parliamentary election results, December 1997

5.3 Presidential and parliamentary election results, December 2002

A.1 Presidential election results, December 1992

A.2 Parliamentary election results, December 1992

A.3 Presidential election results, December 1997

A.4 Parliamentary election results, December 1997

A.5 Presidential election results, December 2002

A.6 Parliamentary election results, December 2002

A.7 Referendum results, November 2005

A.8 Presidential election results, December 2007

A.9 Parliamentary election results, December 2007

A.10 Referendum results, August 2010

ABBREVIATIONS, SWAHILI TERMS, AND NOTE ON ETHNIC NOMENCLATURE

ABBREVIATIONS

SWAHILI TERMS

NOTE ON ETHNIC NOMENCLATURE

In each case the most commonly used label is listed first. These labels are used in the text except in the case of direct quotes.

ACKNOWLEDGMENTS

This book is the product of research conducted at the University of Oxford, Keele University, and the University of Leeds. At Oxford I was fortunate enough to be supervised by David Anderson, who has provided continuous encouragement, support, academic guidance, and analytic insights. The vibrant African studies community in Oxford and Leeds provided a stimulating work environment and helped me develop a more general understanding of African history and politics. This includes Gavin Williams, who did much to cultivate my early interest in African politics. The arguments have also benefited from my attendance at a number of seminars and conferences and from feedback received for papers presented therein. In addition, I benefited from excellent feedback, suggestions, and editorial comments from David Anderson, John Lonsdale, Jocelyn Alexander, Adam Ashforth, Stephen Orvis, Susanne Mueller, Nicholas Cheeseman, Michael Molcher, Tania Edwards, Nadine Beckmann, Gerard McCann, Rob Blunt, and Raphu Mustapha.

I am also grateful for the financial assistance received over the years from the British Academy, Beit Fund (Oxford), Oxford University Press Surplus Fund, British Institute in Eastern Africa, Royal Historical Society, Department of Politics and International Relations (Oxford), and Balliol College (Oxford). Balliol College served as a continuous source of support, and the British Institute in Eastern Africa served as an excellent base in Nairobi.

As a DPhil student I lived a rather nomadic life, and I owe much to the unstinting hospitality of family and friends. Particular mention must go to Steph Wynne-Jones, Mike Monaghan, Rob Blunt, Danielle Walters and her family, Bishop Stephen Kewasis and his family, Claire Medard, and Father Patrick Baraza for their warm hospitality in Kenya and the United States, and to my parents, Daniel and Sarah Horsley, David and Lucy Smith, Geoffrey Smith, Sarah Longair, Daniel Branch, my sister Catherine and her husband Simon, Ed Hughes, Pete Murphy, and Gregory Yakovlev, who all opened their homes to me and/or my belongings in the United Kingdom.

In Kenya, Nicholas Cheeseman, Daniel Branch, Gerard McCann, Rob Blunt, Danielle Walters, Steph Wynne-Jones, Mike Monaghan, Laragh Larson, Dave Eaton, Cecilia Nalelia, Jared Nyamweya, Sarah Muhoya, Claire Medard, Diane Mwako, Bishop Kewasis and his family, Daniel Kandagor, Kipkorir Menjo, Paul Kurgat, Ngengi wa Njuguna, and many others helped provide a home away from home. Some of my fieldwork was done with Claire Medard, whose research experience was of great help (particularly in the early days), and whose knowledge and understanding of local politics had a great impact on the development of the ideas and arguments in this book. The process of conducting interviews was only made possible by people’s unending willingness to take time out of their day to answer my many questions. In particular I must thank interviewees who agreed to meet with me on several occasions and also Daniel Kandagor, Albert Mshando, Sammy Mbugua, and a litany of others who helped arrange meetings.

Final thanks must go to my partner, Michael Molcher; my family; and my friends in Leeds, Oxford, London, Keele, and Kenya, who have provided endless support and light relief.

Map 1. Kenya: Provinces and area of focus

Source: Bodleian Library, Oxford

Map 2. Area of focus: Kalenjin-dominated districts

Source: Bodleian Library, Oxford

INTRODUCTION

The Nature and Political Salience of Ethnic Identity

Ethnic identities are often recent constructs, never fixed and unchanging. Nevertheless, they enjoy a seemingly natural or primordial appeal, and their potential to unite and divide depends upon assumed commonalities and differences of history and culture. Ethnic identities also enjoy global recognition through, for example, cultural and peoples’ rights and specially designed institutional frameworks—from the right to national self-determination, the rights of indigenous peoples and ethnic minorities, to consociationalism and ethnic federalism. At the same time, there is understandable concern about the potential for a sense of ethnic difference to endorse, and even demand, violent atrocities against the other.

This begs a host of questions regarding the origins and salience of ethnic identities: How are ethnic groups formed? How might a collective compartmentalization of us and them on the basis of an assumed shared past lead to a situation in which politics is cast as ethnic? How and when does a sense of ethnic distinction lead to a reality of ethnic competition and violent conflict? In seeking to address these questions, this study analyzes the construction, development, political relevance, and appeal of one ethnic identity over time—the Kalenjin of western Kenya.

This choice of approach stems from the idea that a specifically historical interpretation can shed light on the origins and continuing appeal of ethnic identities, since the content of the ethnic message itself will continue to vary from people to people, as the culture brokers craft messages that will resonate with their own clienteles (Vail 1989, 7, 17). Some—but not all—of the findings are case specific. However, more generally, this book proposes that since ethnic groups are socially constructed and renegotiated over time, historical narratives of collective achievement, migration, injustice, persecution, and associated moral claims are in constant evolution, producing complex and contested groupings that enjoy greater relevance to local actors. As a consequence, ethnic identities provide politicians with a means to mobilize support and for ordinary people to lay claims and assert rights to space, power, and wealth. The problem is that such ethnic politics encourages a sense of difference and competition between communities, which—in the presence of resentments, elite encouragement, and the absence of institutional brakes or barriers—can erupt into violent confrontation that is legitimized by notions of defense, the settling of old scores, and group rights, with some ethnic narratives being more emotive and divisive than others.

To simplify further, this book argues that the constructed nature of ethnic identities is the source of ethnicity’s attraction and danger, as selective and interpreted histories are used to unite some and differentiate others in ways that are meaningful, contested, and unstable.

The Case Study: Kenya and the Kalenjin

Political parties in Kenya tend to be associated with particular ethnic groups, while competitive elections have displayed strong ethnic voting patterns. Kenya’s most recent general election was held in December 2007, when contested results sparked a postelection crisis that left over 1,000 people dead and almost 700,000 displaced within two months (Lynch 2009, 604). Of 1,133 recorded casualties, an estimated 405 were shot by the police, the majority killed by their fellow citizens (Kenya 2008a, 305). In a number of foreign media reports, poll violence was portrayed as a battle between members of President Kibaki’s Kikuyu community and those of his opponent Raila Odinga’s Luo community—or Kenya’s two main tribes (Somerville 2009). Yet the epicenter was in Rift Valley Province, western Kenya, where neither the Kikuyu nor the Luo predominate. It was here that over half the deaths (KNCHR 2008, 341), the majority of displacement (Lynch 2009), and iconic moments of the conflict occurred (D. Anderson and E. Lochery 2008, 328), as Kalenjin participated in targeted attacks against Kikuyu and other selected communities, most notably Kisii and Luhya. Many people were killed in these attacks, and many more were forced to take shelter with friends and relatives or in camps for the internally displaced.

The crisis was unexpected. Kenya is widely viewed as a bastion of peace and stability in a volatile region, and campaigns and voting were relatively peaceful, with high voter turnout a testament to democratic commitment. However, while shocking, the high-handed state security response was compatible with the origins and evolution of local policing (Africa Watch 1991; Hills 2007, 2008; Katumanga 2010) and escalation of extrajudicial killings (KNCHR 2008). In turn, this book will argue that, intercommunal attacks were consistent with ethnic readings of the past and present, mutually incompatible interpretations of social justice, an assumed critical juncture that offered opportunities and heightened fears, high levels of popular political skepticism, institutional decay, a culture of impunity, elite opportunism, and related strategies of action.

Most important, despite frequent media references to Kenya’s former existence as a peaceful idyll, this was not the first time that intercommunal conflict had rocked the country. The most significant precursor was the ethnic clashes of 1991 to 1993, when an estimated fifteen hundred people were killed and three hundred thousand were displaced in western Kenya (Africa Watch 1993, 1). Most of the violence occurred on cosmopolitan farms in and on the borders of Rift Valley Province, where Kalenjin and Kikuyu emerged as principal perpetrators and victims, respectively. Ethnic clashes also erupted in parts of the Rift Valley and at the coast in 1997, while western Kenya has suffered from numerous localized outbursts of intercommunal tension and conflict. At independence the colonial government feared a possible war over land and majimboism (or regionalism) between Kalenjin warriors and former Mau Mau adherents and sympathizers, and local Kikuyu residents. Two common themes emerge: (1) Sides were ethnically delineated, with the Kalenjin cast as principal perpetrators of attacks on Kikuyu neighbors in an ethnically cosmopolitan and agriculturally rich area beset by conflicting claims to land and authority. (2) The eruption of large-scale violence correlated with electoral competition and has been a strategy used and endorsed by political elites to secure control of the center, and by ordinary citizens to gain access to land and other resources and/or as a form of preemptive defense.

The label of Kalenjin is a recent construct, dating from the mid-twentieth century, when it came to embrace a number of subgroups administered as separate tribes by the colonial authorities. Groups spanned district and provincial boundaries, the racial settlement zones of the White Highlands, and the Kenya-Uganda border. Concentrated in western Kenya, the Kalenjin constitute a decided majority in Bomet, Kericho, Baringo, Koibatek, Keiyo, Marakwet, and West Pokot Districts in Rift Valley Province and Mount Elgon District in Western Province (see map 2). They also constitute a majority in the more ethnically mixed Uasin Gishu District and have a significant presence in Trans Nzoia and Nakuru Districts, Rift Valley Province.¹

The term Kalenjin literally means I say to you—a direct reference to the linguistic similarity of its members—although significant differences of dialect lead to talk of Kalenjin language clusters (Huntingford 1944, 19–20; Sutton 1976, 25). The catalog of subgroups is also a matter of debate. Those commonly listed are the Nandi, Kipsigis, Tugen, Keiyo, Marakwet, Sabaot, Pokot (although the latter hold a somewhat peripheral position),² and Terik (who are sometimes bracketed together with the Nandi [Fedders and Salvadori 1979, 167]). More problematic is the position of Ogiek or Ndorobo—labels used interchangeably for dispersed groups of forest-dwelling hunter-gatherer communities. Since, while most Ogiek speak a Kalenjin-related dialect, some use Maasai-related dialects (Blackburn 1976, 54–56), which, together with their ethnic history of permanent residence (rather than migration) and forest livelihoods, renders their position within the larger Kalenjin debatable and contested (Lynch 2006b).

To confuse the ethnic landscape further, a number of authors suggest a relationship between the Kalenjin and Tatoga in Tanzania (Ehret 1968, 122–23; Huntingford 1953b, 9; Langley 1979, 3; Sutton 1976, 34). Benjamin Kipkorir also lists the Bongom (now largely absorbed by the Luhyia) as one of the Kalenjin subgroups (2009, 392) (or alternatively as a subgroup of the Sabaot [1975, 64]), while leaders from several small communities (such as the Enderois, Sengwer, and Mount Elgon Ogiek)—which are commonly regarded as subgroups of Kalenjin subgroups—have come to assert their difference (Lynch 2006b). Some even deny the existence of a Kalenjin community and talk, for example, of a motley confederation of some eleven Nilotic groups with separate languages and cultural practices (Ajulu 2002, 266). Well-known Nandi politician Jackson Kibor went further in his declaration that the Kalenjin groupings were hatched by an individual who wanted to use the groups to ascend to power and gain wealth.³

The question of who is Kalenjin is thus subjective, and differences of opinion are indicative of the complex, ambiguous, and contested nature of ethnic identities. However, for most people within and outside Kenya, the Kalenjin constitute a recognized ethnic group. Moreover, despite their recent progeny, complex makeup, and internal divisions, the community has become one of Kenya’s most united in terms of electoral and referendum voting patterns, while members have acted as key participants in occasions of ethnic violence. Their political significance is further enhanced by the incumbency of Daniel arap Moi (a Tugen from Baringo District) as vice president of Kenya from 1967 to 1978, and president from 1978 to 2002, and by his unsuccessful call for Kalenjin to reelect President Kibaki in 2007. One point of interest is the way in which Moi’s leadership helped strengthen Kalenjin unity, at the same time as it fostered internal divisions and subgroup nationalisms. Finally, Kalenjin often self-identify as one of Kenya’s smaller tribes. Yet, at the time of the 1989 population census, they accounted for approximately 11 percent of the Kenyan population, as compared to the Kikuyu at 21 percent, Luhya at 14 percent, Luo at 12 percent, and Kamba at 11 percent (Kenya 1994, 6–2).⁴ Moreover, while the 1999 census omitted ethnic data (Kenya 2001), census results from 2009 reveal that the Kalenjin now constitute 14 percent of the population, as compared to the Kikuyu at 17 percent, Luhya at 14 percent, Luo at 10 percent, and Kamba at 10 percent (Kenya 2010). This numerical strength, together with their concentration in and around the cosmopolitan farming lands of the former White Highlands (and overrepresentation of these areas in parliament) means that the community enjoys a strong presence in the country’s politics and can act as an important swing community.

Notwithstanding their recent provenance, historical prominence, and political significance, the Kalenjin have attracted little academic attention, and most of the existing literature consists of ethnographic (for example. Hollis 1909; Huntingford 1953a, 1953b; Massam 1972; Orchardson 1961; Peristiany 1939) and historical studies (for example, D. Anderson 1993, 2004; Ellis 1976; Gold 1978; Kipkorir 1972, 1973; Matson 1970, 1980; Mwanzi 1977) of particular subgroups. In light of this relative silence and recent political developments, this book analyzes how a sense of being Kalenjin was constructed, self-consciously popularized, and adopted, how it coexists with internal divisions and debate but has nevertheless been used as a fulcrum around which to mobilize support, and the implications of ethnonationalist historical narratives and memories for a popular sense of intercommunal difference and competition and potential for violent conflict.

Since the aim is to look at how presentations of history and interconnected interpretations of identity inform a sense of difference and competition, and associated processes of political mobilization and support, the history of the Kalenjin is outlined for the ways in which it has shaped political institutions, cultures, and strategies rather than for its historical veracity per se. In turn, the rationality of ethnic construction, mobilization, and support is investigated without the constraints of a strict rational choice theory where individuals maximize expected payoffs in a context where preferences are consistent—in that they can be ranked, and are connected and transitive (D. Green and I. Shapiro 1994). Finally, the book stems from an assumption that—in any context—people are motivated by rational considerations (for example, of loss and gain) but also by more irrational feelings, such as affection, frustration, anger, and hatred.

Unlike much recent political science literature (for example, Chua 2003; Kaufman 2001; Mann 2005; Mansfield and Snyder 2005; Petersen 2002; Snyder 2000), this study does not seek to offer a comprehensive analysis of the relationship between democratization and intercommunal violence or provide a model of ethnic violence—although it does have important implications for such an endeavor and occasions of intercommunal violence are discussed and analyzed. Instead, the study focuses on a prior question: namely, the collective narratives and group perceptions that render a particular understanding of belonging and citizenship (and thus difference and competition) meaningful, appealing, and politically significant.

The arguments are based on a triangulation of recorded and nonrecorded interviews, participant observations of political meetings and rallies, newspaper articles, government and nongovernment reports, and secondary literature. Over 250 interviews were conducted between September 2004 and August 2009 with a range of actors from current and former cabinet ministers, MPs, top-level civil servants, local councillors, civil society activists, clergy, academics, lawyers, students, businessmen, and farmers in Nairobi, Rift Valley, and Western Provinces. Interviewees ranged in age from their early twenties to almost one hundred. However, while interviews provided critical insight and only a few respondents requested that their comments remain anonymous, direct citation has largely been avoided. This is due to the sensitive and emotive nature of the interview content and current political context, which demands that interview materials be handled with caution.

Local Conclusions and General Propositions

As noted, this study aims to further our understanding of the processes of ethnic construction and identification, and of political mobilization and support across Kenya and beyond through an analysis of the interaction between presentations of history and interpretations of identity, and local perceptions and elite behavior. The central argument is that—while processes of ethnic construction and negotiation are limited by the need for ethnicity to be rooted in primordial discourses of cultural similarity and shared pasts (cf. Lentz and Nugent 2000)—the main motivation for the construction and politicization of a Kalenjin alliance was (and continues to be) a nexus of fear of loss and potential for gain. More important, ethnic notions of morality and assistance, and of shared pasts and justice, provide a channel for state patronage and a discursive framework, which locals can use to question the legitimacy of social and spatial inequalities, and suggest corrective initiatives. The language of identity—and notion of I say to you—thus provides a means for ordinary citizens and politicians to influence each other and to express political arguments to ethnic others and external agents.

From this analysis, a number of interrelated themes emerge that are of particular import for those interested in the nature of ethnic identities, the reasons for their continuing political salience, and common correlations between ethnically delineated processes of political mobilization and support and periods of heightened intercommunal tension and violence.

First, ethnic groups are revealed to be socially constructed imagined communities (B. Anderson 1983), which are subject to ongoing processes of negotiation and renegotiation. These processes can include migration of individuals from one community to another, assertions of difference, amalgamation, and rebranding (as, for example, a nation or indigenous people). These processes of social construction and negotiation require a sense of cultural and linguistic similarity and difference; are shaped by available historical transcripts and presentations, and perceptions of fear and opportunity; and are initiated by culture brokers and community leaders in response to dynamic institutional and socioeconomic contexts. The implication is that (a) primordial appeals and instrumental motivation are central to processes of ethnic construction; (b) culture brokers and leaders play a key role in processes of ethnic construction and negotiation but cannot invent or conjure them, since histories, popular perceptions, and institutions also matter; and (c) ethnic groups are not created as fixed and unchanging, and thus their perpetuation, content, allies, and foes (as well as their initial construction) must always be explained.

Following these initial conclusions, chapters 1 and 2 show how the Kalenjin grouping emerged in a colonial context of indirect rule and rapid change, as local leaders drew upon cultural and linguistic similarities to amalgamate a number of colonial tribes in the face of burgeoning (if fluctuating) fears of marginalization and exclusion, and attendant opportunities to gain access to the ownership and control of geographic space and state resources. But that—since the new Kalenjin identity embraced rather than replaced diverse, geographically disparate, and complex ethnic groups—Kalenjin subgroups and their subgroupings can still differentiate themselves from one another and assert their difference to the corporate whole, allowing for ongoing debates about the meaning, relevance, and stretch of ethnic inclusion.

The suggestion is that ethnic identities are historic and moral communities, which are subject to constant negotiation, contest, and internal division due to the need for identities to make sense given interpreted histories, perceptions of the present, and forecasts of the future. In turn, the myths and symbolism of peoplehood rely on perceptions of shared pasts and interrelated futures, which individuals help shape and inform but by which they are also constrained and curtailed. These arguments support the following general proposition: Ethnic identities are socially constructed historic and moral communities that are subject to ongoing processes of negotiation and contestation. Elites help shape but cannot dictate these processes, and their role is often to respond to, articulate, and/or focus more popular sentiments of justice and injustice, morality and immorality, and inclusion and exclusion.

From the outset, Kalenjin ethnic narratives of territoriality, threat, and opportunity were intertwined with a portrait of the community’s marginality and vulnerability relative to ethnic others, and with a collective claim to the Rift Valley on the basis of prior residence and an assumed right of locals to own and control homelands. Chapters 3 through 6 reveal how this notion of belonging as sons of the soil or autochthony (cf. Geschiere 2009)—and the founding myths of danger and opportunity—have never disappeared but have evolved and gained new content over time. More important, these narratives continue to shape postcolonial memories and discursive postures regarding citizenship, rights, social and economic justice, and geographical space and are available to politicians as reason for group compensation, restitution, and/or particular assistance, and thus to mobilize and incense local constituents.

At a more general level, the interconnectedness of identity, historical presentation, and moral relations ensures that ethnic narratives provide a discursive framework that helps shape popular perceptions of group status and thus evaluations of self-worth (cf. Horowitz 2000) while also providing a moral and historical basis for a collective sense of social justice, claims, and rights. As a consequence, shared and interpreted histories—for example, of achievement, marginalization, injustice, past land use, and/or divine will—provide a discursive framework that leaders can use to mobilize support but that ordinary people can also use to assert rights to the ownership and control of space and/or the redistribution of resources. In other words, if politics is understood as who gets what, when, how (Lasswell 1950), and why, then a sense of belonging to a moral and historic community can support demands for redistribution within and among ethnic groups on the basis of virtue, rights, and/or just deserts.

One implication is that the appeal of ethnic narratives and support for ethnic spokesmen is less about rewards for patronage received and more about access to patrons in the future, protection of group interests, support for group claims, redress for past wrongs, and fear of others. In turn, when political parties become associated with particular communities or ethnic spokesmen, electoral competition can prompt reactive ethnic voting as voters choose to minimize any collective disadvantage (Horowitz 2000, 323). In contemporary Kenya, this has prompted a two-pronged logic of exclusionary ethnicity, which focuses on who should not get power and control the state’s resources (Mueller 2008, 201), and speculative loyalty, or calculation of the potential advantages of electing community spokesmen. This multifaceted understanding is important, as it differs from common analyses of African politics that tie political support along ethnic lines directly to ethnically delineated channels of political patronage and allows for support to rely, for example, on local interpretations of social and economic justice and contributions to group status as well as financial or development benefits promised and/or received, or on assistance in the broadest sense.

Most important, since the language of ethnic association is rooted in a sense of shared experience and social justice, it helps color perceptions of what can rightfully be claimed and on what grounds. As a consequence, it provides a range of strategies of acquisition and protection, which are often mutually exclusive and can easily come into conflict. At a basic level, demands by one group may foster fears of exclusion and marginalization among others, especially if demands are for redistribution from one region or community to another. At another level, historic and moral narratives can influence perceptions of citizenship in ways that are mutually incompatible. In this way, a naturalized sense of belonging to a particular soil can provide a basis on which to assert rights to the ownership and control of an area and associated resources as indigenes or autochthonous residents, but also reason to relegate others to the status of guests, foreigners, or dangerous infestation, while electoral competition and talk of benefits for locals can raise endless debates about who really constitutes a local. Similarly, while a sense of divine will, collective achievement, and/or a communal work ethic can help justify ownership, control, and wealth on the basis of toil, efficiency, and destiny, it can also provide reason to relegate others to the status of charitable causes, layabouts, or inferior beings. This complexity helps us understand how the Kikuyu became, and have remained, the principal other for the Kalenjin, due to their being cast as more advanced and favored, and as outsiders and thus guests in the Rift Valley.

The construction and negotiation of ethnic identities and the associated narratives and interpretations of shared histories and conceptions of morality and social and economic justice are thus critical for understanding intra- and intercommunal relations. Moreover, some discursive repertoires are clearly more defensive, aggressive, and exclusive than others. In this way, a defensive politics of ethnic belonging that casts others as guests—who can be ejected if they fail to follow household rules (cf. Horowitz 2000, 199)—has the potential to fuel particularly violent outcomes. This is due—as other analysts have shown in eastern Congo, Côte d’Ivoire, South Africa, and beyond (for example, Chua 2003; Comaroff and Comaroff 2001; Dunn 2009; Geschiere 2009; Manby 2009; Marshall-Fratani 2006)—to the positioning of others as undeserving of their relative success, but also from the fact that belonging is more elusive [than ethnicity] and more easily subject to political manipulation (Geschiere and Nyamnjoh 2000, 424). The term’s inherent relativity and nervousness means that notions of the other can be renegotiated at ever closer range (Ceuppens and Geschiere 2005, 403; Geschiere and Jackson 2006, 1; Jackson 2006), with the possibility, for example, of debates between and within Kalenjin subgroups about who is who and who really belongs where (cf. Jackson 2006; Marshall-Fratani 2006).

Consequently, while many analysts distinguish between good and bad ethnicity as ethnicity from below versus ethnicity from above, this case study reveals how bottom-up and top-down pressures can both be inherently exclusive and competitive, and encourage intercommunal fear, tension, and violence. For example, while people may participate in violence because they are paid or coerced, the organization and incitement of mass participation in intercommunal violence also relies on the threat that others seemingly (or actually do) pose, and the opportunity that their exclusion (or extermination) may create. Unfortunately, such sentiments can actually encourage or even force politicians to take an ethnically chauvinistic and confrontational approach if they want to retain local support.

The general propositions that emerge are that: The historical presentations and moral arguments associated with ethnic groupings—and their construction, development, and negotiation—help explain the utility of ethnic narratives, and thus the basis of support for ethnic leaders and popular sense of intercommunal difference and competition. As a result, the morality of ethnicity from below and above is ambiguous and context dependent. And that: the politics of ethnic belonging in cosmopolitan contexts where outsiders are deemed to be more economically advanced and/or politically successful can encourage a particularly defensive, pernicious, and violent politics of difference, where it is possible for the other to be reinterpreted at ever closer range.

As a result, ethnic readings of the past and associated interpretations of group rights, justice, and deserts help explain not only a