The Colonial Origins of Ethnic Violence in India

By Ajay Verghese

The neighboring north Indian districts of Jaipur and Ajmer are identical in language, geography, and religious and caste demography. But when the famous Babri Mosque in Ayodhya was destroyed in 1992, Jaipur burned while Ajmer remained peaceful; when the state clashed over low-caste affirmative action quotas in 2008, Ajmer's residents rioted while Jaipur's citizens stayed calm. What explains these divergent patterns of ethnic conflict across multiethnic states? Using archival research and elite interviews in five case studies spanning north, south, and east India, as well as a quantitative analysis of 589 districts, Ajay Verghese shows that the legacies of British colonialism drive contemporary conflict.

Because India served as a model for British colonial expansion into parts of Africa and Southeast Asia, this project links Indian ethnic conflict to violent outcomes across an array of multiethnic states, including cases as diverse as Nigeria and Malaysia. The Colonial Origins of Ethnic Violence in India makes important contributions to the study of Indian politics, ethnicity, conflict, and historical legacies.

• Language: English
• Publisher: Stanford University Press
• Release date: Mar 2, 2016
• ISBN: 9780804798174

Book preview

Stanford University Press

Stanford, California

© 2016 by the Board of Trustees of the Leland Stanford Junior University.

All rights reserved.

No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press.

Printed in the United States of America on acid-free, archival-quality paper

Library of Congress Cataloging-in-Publication Data

Verghese, Ajay, author.

The colonial origins of ethnic violence in India / Ajay Verghese.

pages cm—(Studies of the Walter H. Shorenstein Asia-Pacific Research Center)

Includes bibliographical references and index.

ISBN 978-0-8047-9562-3 (cloth : alk. paper)—

ISBN 978-0-8047-9813-6 (pbk. : alk. paper)—

ISBN 978-0-8047-9817-4 (electronic)

1. Ethnic conflict—India.  2. Political violence—India.  3. India—Ethnic relations.  4. India—Colonial influence.  I. Title.  II. Series: Studies of the Walter H. Shorenstein Asia-Pacific Research Center.

HN690.Z9S6284 2016

305.800954—dc23

2015028115

Typeset by Bruce Lundquist in 11/14 Adobe Garamond

The Colonial Origins of Ethnic Violence in India

Ajay Verghese

Stanford University Press

Stanford, California

THE WALTER H. SHORENSTEIN ASIA-PACIFIC RESEARCH CENTER

Studies of the Walter H. Shorenstein Asia-Pacific Research Center

Andrew G. Walder, General Editor

The Walter H. Shorenstein Asia-Pacific Research Center in the Freeman Spogli Institute for International Studies at Stanford University sponsors interdisciplinary research on the politics, economies, and societies of contemporary Asia. This monograph series features academic and policy-oriented research by Stanford faculty and other scholars associated with the Center.

ALSO PUBLISHED IN THE SHORENSTEIN ASIA-PACIFIC RESEARCH CENTER SERIES

The New Great Game: China and South and Central Asia in the Era of Reform

Edited by Thomas Fingar (2016)

Rebranding Islam: Piety, Prosperity, and a Self-Help Guru

James Bourk Hoesterey (2015)

Global Talent: Skilled Labor as Social Capital in Korea

Gi-Wook Shin and Joon Nak Choi (2015)

Failed Democratization in Prewar Japan: Breakdown of a Hybrid Regime

Harukata Takenaka (2014)

New Challenges for Maturing Democracies in Korea and Taiwan

Edited by Larry Diamond and Gi-Wook Shin (2014)

Spending Without Taxation: FILP and the Politics of Public Finance in Japan

Gene Park (2011)

The Institutional Imperative: The Politics of Equitable Development in Southeast Asia

Erik Martinez Kuhonta (2011)

One Alliance, Two Lenses: U.S.-Korea Relations in a New Era

Gi-Wook Shin (2010)

Collective Resistance in China: Why Popular Protests Succeed or Fail

Yongshun Cai (2010)

The Chinese Cultural Revolution as History

Edited by Joseph W. Esherick, Paul G. Pickowicz, and Andrew G. Walder (2006)

Ethnic Nationalism in Korea: Genealogy, Politics, and Legacy

Gi-Wook Shin (2006)

Prospects for Peace in South Asia

Edited by Rafiq Dossani and Henry S. Rowen (2005)

Contents

List of Illustrations

Acknowledgments

List of Abbreviations

Glossary of Key Terms

Introduction

Chapter 1—Colonialism, Institutions, and Ethnic Violence in India

Chapter 2—Violence in North India: Jaipur and Ajmer

Chapter 3—Violence in South India: Malabar and Travancore

Chapter 4—Explaining Violence in East India: Bastar

Chapter 5—Patterns of Ethnic Violence Across Contemporary India

Chapter 6—The Indian Model of Colonialism

Conclusion

Appendix 1: Archival Research

Appendix 2: Elite Interviews

Appendix 3: Secondary Source Research (online)*

Appendix 4: Quantitative Analysis (online)*

Notes

Bibliography

Index

*Appendix 3 and Appendix 4 can be found online at http://sites.google.com/site/ajayaverghese/home/data.

List of Illustrations

Figures

Figure 1.1: Communal Composition of Mughal Mansabdars

Figure 1.2: Ethnic Stratification in Colonial India

Figure 3.1: Evictions in Malabar, 1862 to 1880

Figure 3.2: Land Tenure System in Kerala, 1957 to 1958

Maps

Map I.1: Location of Case Studies

Map 1.1: British Provinces and the Princely States, 1909

Map 2.1: Colonial Rajputana

Map 2.2: Postcolonial Rajasthan

Map 3.1: Colonial Kerala

Map 3.2: Postcolonial Kerala

Map 4.1: Colonial Bastar

Map 4.2: Postcolonial Bastar

Map 6.1: Colonial Burma

Map 6.2: Colonial Malaya

Map 6.3: Colonial Nigeria

Tables

Table I.1: India’s Ethnic Demography, 2001

Table I.2: Book Case Studies

Table 1.1: Major Princely States in India, 1931

Table 2.1: Jaipur and Ajmer Historical Comparison

Table 2.2: Jaipur State Civil List, Mahakma Khas, 1929 to 1930

Table 2.3: Communal Representation in Ajmer Government, 1925

Table 2.4: Hindu and Muslim Education Statistics in Ajmer, 1933 to 1934

Table 2.5: Contemporary Religious Violence in Rajasthan

Table 2.6: Composition of Castes in Jaipur and Ajmer, 1901

Table 2.7: Caste Violence in Rajasthan, 2009

Table 3.1: Malabar and Travancore Historical Comparison

Table 3.2: Communal Representation in Travancore Government, 1930s

Table 3.3: Composition of Castes in Malabar and Travancore, 1901

Table 3.4: Agricultural Statistics for Kerala, 1956

Table 5.1: Descriptive Statistics

Table 5.2 Colonialism and Caste/Tribal Violence Casualties, 2005 to 2009

Table 5.3: Colonialism and Hindu-Muslim Riot Casualties, 1990 to 1995

Table 5.4: Matching Analysis, Colonialism and Caste/Tribal Violence, 2005 to 2009

Table 5.5: Instrumental Variable Analysis of Hindu-Muslim Riot Casualties, 1990 to 1995

Acknowledgments

There are so many people who assisted in the process of writing this book and I am glad to have the opportunity finally to thank them. The project began at The George Washington University (GW) under the supervision of several great teachers. Most of what I know about being a scholar I learned from Manny Teitelbaum. He was never afraid to spare the criticism, and this book wouldn’t be the same without his tough love. Henry Hale’s brilliant feedback was outdone only by his lightning-fast e-mail turnaround. Henry Farrell was always there to cite just the right book or article when I needed guidance. Dane Kennedy, Irfan Nooruddin, and Kanchan Chandra served as outside readers for the project and helped push my thinking on a number of important issues. I also owe thanks for various reasons to GW Professors Robert Adcock, Eric Lawrence, Harvey Feigenbaum, Shawn McHale, and Kimberly Morgan. In addition, I made many good friends at GW, among them Colm Fox, Enze Han, Brian Karlsson, Craig Kauffman, J. J. Mikulec, Joseph O’Mahoney, and Mike Schroeder. Colm deserves special thanks for creating several of the maps in this book.

A number of scholars provided feedback on various incarnations of this project over the years, including Victoria Farmer, Michael Fisher, Ron Herring, Michael Hechter, Karen Leonard, James Mahoney, and Thomas Rosin. Roberto Foa read the entire manuscript at a very late stage and offered some terrific feedback and much-needed encouragement.

I owe thanks to more people in India than I can possibly recall. My greatest debt is to Asha Sarangi at Jawaharlal Nehru University, who helped me navigate the Indian bureaucracy. I thank in Rajasthan the many fine teachers at the American Institute of Indian Studies’ Hindi Language Program, Panjak Sharma and staff at the Jaipur City Palace Archives, the staff of the Institute of Development Studies, and Anupam Sharma. I must thank George and Rekha Varghese, Ronn Mathew, and George Varkey in New Delhi for opening their home to me countless times over the years. My work was also assisted by Radhakrishna Nair, Jaya Prabha Ravindran, and the staff of the National Archives of India. I owe thanks to Thomas Punnose, Jacob Punnose, Premu Philips, Zareena Parveen, and V. Rangaraj at the Andhra Pradesh State Archives and Research Institute, and to the staff of the Nehru Centenary Museum, both in Hyderabad. I must thank Prabha Taunk and the staff of the Deshbandhu Press Library in Raipur, as well as Philip and Mary Cherian, Cecil George, and the staff of the Directorate of Archives in Kerala, especially Reji Kumar, Prakash B., and Pauvabhy. In London, the staff of the British Library were very helpful while I conducted research, and I must thank Frank Verano for generously giving me a place to stay. I also thank Adam Auerbach, Peter Samuels, Julia Kowalski, Patton Burchett, and Lion König for help and camaraderie while I was conducting my fieldwork.

Several institutions provided generous funding for my research, including the Sigur Center for Asian Studies, the Loughran Foundation Endowment, the Konosuke Matsushita Memorial Foundation, and the Cosmos Club. Parts of the book were written while I was a postdoctoral fellow at the Shorenstein Asia-Pacific Research Center at Stanford University, to which I owe a great debt of gratitude. I completed the manuscript while a professor at the University of South Florida St. Petersburg, so I must thank Ray Arsenault, Erica Heinsen-Roach, Hugh LaFollette, Larissa Kopytoff, Adrian O’Connor, Felipe Mantilla, and Thomas Smith for their tireless support. I must also thank the editorial team at Stanford University Press, including Geoffrey Burn, Andrew Walder, James Holt, Anne Fuzellier Jain, Jenny Gavacs, Alice Rowan, and two anonymous reviewers.

I have too many people close to me who deserve my gratitude to list all their names here, but I need to single out two. Travis Valentine has put up with me as a friend for twenty-five years and counting. And without Jenn Ortegren’s constant love, care, and support, I am certain I would have gone crazy while writing this book. Last but not least, I need to thank my family. My ammachi, Mary Verghese, was very supportive of my research—reminding me before I went to India in 2008 that New Delhi is the capital—and I wish she was here to see the final product. My brother, Ashwin, deserves special thanks for all his support over the years. Finally, my father, Abraham, filled the house with books and gave me a great love of history and politics while I was growing up; and my mother, Anila, always made sure that I did my homework. This book is dedicated to the two of them.

List of Abbreviations

Glossary of Key Terms

adivasi: original inhabitant, member of an indigenous tribe

dalit: member of an untouchable caste

darbar: the princely court, used to refer to the government of a princely state

dargah: a shrine, used in this book to refer to the tomb of a Sufi saint in Ajmer

diwan: Chief Minister of a princely state, also spelled dewan

istimradars: landlords of Ajmer

jagirdari: an area of land controlled by a Rajput nobleman

jenmi: Hindu landlords in Malabar, also spelled janmi

kanamdars: land supervisors in Malabar

khalsa: an area of land controlled directly by the state (British or princely)

Khilafat: an Islamic movement in India to preserve the caliphate of the Ottoman Empire

mansab: ranking in the Mughal political system

Mappillas: Muslims of Malabar

Marathas: Hindu warrior caste in Maharashtra

nawab: a Mughal regional governor

Naxalites: various Maoist guerrilla groups in India

pandaravaka: land controlled by the Travancore princely government

puravaka: private land in Travancore

raj: rule, or government; India was known as the British Raj

raja, maharaja: king or great king; Hindu ruler of a princely state

rajpramukh: governor of an Indian state between 1948 and 1956

Rajputs: Hindu warrior caste in Rajasthan

reservations: affirmative action quotas

ryotwari: land in which the tribute was collected directly from the peasant

verumpattamdars: cultivators in Malabar

zamindari: a landlord area, usually in British India; a zamindar is a landlord

Introduction

Men make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly encountered, given and transmitted from the past. The tradition of all the dead generations weighs like a nightmare on the brain of the living.

—Karl Marx¹

In the north Indian state of Rajasthan there are two neighboring districts named Jaipur and Ajmer. To a traveler they would seem almost indistinguishable. Both districts have desert terrain. The people in both areas speak the same language, share a common culture, and work in the same kinds of jobs. The demography of both territories is also similar: they have roughly the same percentage of Hindus and Muslims, as well as members of high castes, low castes, and indigenous tribal groups.² Yet these two identical districts responded very differently to a pair of notable political events that occurred in India over the past several decades.

In 1992, a huge mob of Hindu extremists destroyed the Babri Masjid (Mosque) in the northern city of Ayodhya. For years this holy site had attracted the ire of various Hindu nationalist groups who believed it had been built by Muslim invaders on the site of an ancient temple marking the birthplace of the god Ram, the hero of the epic poem Ramayana. The destruction of the Babri Masjid triggered massive Hindu-Muslim riots throughout India. In Jaipur, serious riots gripped the city and led to twenty-eight deaths.³ Right next door in Ajmer, however, there was no religious rioting.

About a decade and a half later, in 2008, these two districts became sites of another political controversy when huge clashes broke out over the Indian government’s policy of reservations, or affirmative action quotas. In India, certain low castes and tribal groups are guaranteed a number of reserved spots in higher education and government jobs. A dispute over the specific allotment in 2008 led to major protests throughout Rajasthan. This time Ajmer was the city embroiled in violence—the government’s Rapid Action Force was dispatched there and the entire city was briefly shut down⁴—whereas Jaipur largely remained peaceful.

Contemporary social science research tells us that all individuals have multiple, often crosscutting (or overlapping) ethnic identities, and they can adopt different identities within different societal contexts. Why, therefore, does ethnic conflict in multiethnic states revolve around one particular identity rather than another? Why do people in Jaipur tend to fight over religion whereas people in Ajmer tend to fight over caste and tribal identities? What explains patterns of ethnic violence in multiethnic states?

Answering these questions has important implications for India, a violent country wracked by endless Hindu-Muslim riots, caste atrocities, and tribal uprisings (Iyer 2009). But it is also a central question for a number of plural states in Europe, Africa, Asia, and elsewhere. In the broadest sense, the twentieth century witnessed a remarkable amount of political violence based on ethnic identities. The Rwandan genocide, breakup of Yugoslavia, and long-running civil war in Sri Lanka are all emblematic of the brutal ethnic bloodshed that occurs around the world. Beyond this larger point, there are also three additional reasons to understand the genesis of patterns of ethnic conflict. First, not all forms of ethnic violence are equal. Many social scientists argue, for example, that religious identities may be uniquely salient (Wald, Silverman, and Fridy 2005; Grzymala-Busse 2012), and that the fusion of politics and religion is inherently more dangerous than other kinds of ethnic politics (Stark 2001; Vanaik 2007; Wilkinson 2008). A second and related point is that most scholars believe that states have some ability to manipulate the salience of ethnic identities. Policymakers might therefore benefit from de-emphasizing a salient identity—such as religion—in order to lessen the chances of conflict. Finally, understanding how patterns of violence arise can also provide insight into the construction and persistence of ethnic identities over time.

India is an ideal site for this research because it is one of the most ethnically diverse countries in the world—the birthplace of several of the world’s major religions and home to thousands of distinct castes, indigenous tribes, and more than twenty official language groups. As Lloyd and Susanne Rudolph (1987: 64) put it, India is the state Europe would have become had the Holy Roman Empire embodied itself in a modern polity. Table I.1 highlights some of the main ethnic groups studied in this book. This striking pluralism within India has also been implicated in a variety of conflicts throughout the country.

TABLE I.1   India’s Ethnic Demography, 2001

SOURCE: Data from 2001 Census.

Different regions of multiethnic states tend to experience different kinds of ethnic violence.⁵ Northern Ireland has a long history of conflict between Protestants and Catholics, but the rest of the country has been relatively immune from this sectarian violence. In some parts of Nigeria, ethnic conflict centers on religious identity, but in other areas it centers on ancestral city membership. Across Iraq, violence between tribes has engulfed some regions, but tribal politics is conspicuously absent in others. India is no exception: scholars have noted that specific parts of the country contain a master narrative that dictates how ethnic conflicts unfold.⁶ In the city of Hyderabad, for example, Hindu-Muslim riots have become endemic, but in northern Kerala, caste conflict has long defined the region’s politics. These varied conflicts are puzzling because they do not seem to be based on the sizes of the ethnic groups in a particular region (Posner 2005). Almost every district in India is Hindu-majority, yet the Hindu-Muslim divide forms the axis of conflict only in specific parts of the country. In other areas, caste, tribal, or linguistic violence constitutes the dominant narrative of conflict.⁷

Understanding patterns of ethnic violence in India begins with recognizing that the country is one of many postcolonial multiethnic states around the world. The era of European colonialism strongly influenced the long-term development of ethnic politics in a number of countries. The British experience in India, however, was unique due to a key fact: only three-fourths of India’s population were ruled by British administrators. They lived in territories known as provinces. A massive rebellion in 1857, however, prevented the rest of the country from coming under direct colonial rule. The remainder of the population—according to the 1901 census, more than sixty million people—lived under the control of largely autonomous native kings in territories known as princely states.⁸ This dichotomy in colonial rule resulted in the creation of two very different political cultures on the subcontinent, and ultimately forms the basis of patterns of ethnic violence today.

Explaining the Argument

This book advances a theory of ethnic conflict centered on history, culture, and institutions. It argues that historical legacies create cultures of conflict or cooperation that, reinforced over time through institutions, drive patterns of ethnic violence in multiethnic states. In India, the era of British colonialism structured long-term ethnic conflict outcomes. There are three components to this argument. First, the British and princely rulers had markedly different understandings of ethnicity. After the Rebellion of 1857, which many colonial administrators interpreted as a religious uprising, British administrators came to believe that caste should be promoted as the central organizing principle of a new Indian society. Princely rulers, on the other hand, had always emphasized religion, and they were encouraged to continue to do so by the British when they took a laissez-faire approach to princely areas after the Rebellion. Two different political cultures subsequently emerged across the provinces and princely states (Putnam 1993).

Second, given these differential understandings of ethnicity, colonial rulers created disparate policies of ethnic stratification (Hechter 1975; Horowitz 1985). Certain ethnic groups were privileged whereas others were subjugated, often for the most capricious of reasons. In the provinces, British administrators implemented policies that benefited high castes, discriminated against low castes and tribals, and protected religious minorities. In the princely states, native kings did the opposite: their policies benefited their coreligionists, discriminated against non-coreligionists, and protected low castes and tribals. Bifurcated colonial rule⁹ led to the creation of different fault lines of ethnic conflict across the provinces and princely states.

Third, these differential political cultures and social fissures were reinforced through institutions. After independence arrived in 1947 the princely states were unified with the provinces, and although on paper most political organizations became identical, they operated in different environments; institutions thus reproduced cultural divides across the provinces and princely states. In addition, India’s independence did not disrupt existing patterns of violence because the new government did not reform most of the vestiges of its colonial past. In the former provinces, on the one hand, low-caste and tribal groups continue to suffer under the weight of historic discrimination, and reform efforts have failed to minimize the violence. In the former princely states, on the other hand, it is mainly minority religious groups that suffer from discriminatory legacies. Once patterns of ethnic violence were created, both formal and informal institutions embedded them in society.

This book shows that the former British provinces in India experience higher levels of caste and tribal violence whereas the former princely states experience more religious violence. The reason religion forms the basis of ethnic conflict in Jaipur is that the state was controlled by a Hindu dynasty during colonialism; the reason Ajmer experiences caste and tribal violence is that it was governed instead by British administrators.

This argument relies on certain key terms that should be clarified at the outset. I follow Donald Horowitz (1985: 53) in describing the term ethnic as encompassing groups differentiated by color, language, and religion; it covers ‘tribes,’ ‘races,’ ‘nationalities,’ and castes. The three kinds of ethnic identities examined in this book are those based on caste, tribe, and religion.¹⁰ Relatedly, individuals have multiple ethnic identities. Kanchan Chandra (2012) carefully distinguishes between nominal and activated categories. The former is the set of ethnic identities to which a person may belong; the latter is the set of identities to which an individual professes membership or is assigned by others. These ethnic identities may also be crosscutting: members of a tribal group may be Hindu, and Muslims may have caste identities. The rich ethnic diversity of India means that in every region there are several potential fault lines of conflict, but violence tends to revolve around one dominant axis.

The term ethnicity, on the other hand, is about uncertainty reduction. Ethnic markers are merely one way of reducing the complexity of the social world, but they are uniquely powerful because ethnic groups share myths of a common origin, a sense of a common fate, a common culture and symbols, and physical similarities, and they face reduced barriers to communication (Brubaker, Loveman, and Stamatov 2004; Hale 2004, 2008).¹¹ Ethnicity is not simply instrumental; it also connotes a way of interpreting and making sense of the social world.

Ethnic violence is violence that occurs largely along ethnic lines.¹² Rogers Brubaker and David Laitin (1998: 428) emphasize two further components of this definition: at least one party must not be the state,¹³ and ethnic identity must be integral rather than incidental to the conflict. In regard to this last point, many scholars have argued that what seem like ethnicity-based conflicts are often not explicitly ethnic but rather interpreted or officially stylized that way after the fact (Brass 1997; Mueller 2000; Kalyvas 2003). While remaining sensitive to this argument, this book shows that the historic construction and stratification of ethnic groups created distinct patterns of violence in contemporary India.

Finally, I use a broad definition of institutions: not just formal rules, procedures or norms, but the symbol systems, cognitive scripts, and moral templates that provide the ‘frames of meaning’ guiding human action (Hall and Taylor 1996: 947). Specifically, a number of insights are drawn from the literature on sociological institutionalism (Fligstein 1990; Dobbin 1994). Most important, according to this school, institutions are a means by which culture is transmitted over time. In India, both formal institutions like political parties and informal institutions like the collective memories of ethnic groups reinforced long-term cultural divides between the provinces and princely states. The argument here can be placed alongside a number of impressive recent works that are beginning to assess the institutional bases of ethnic conflict around the world (Cederman, Wimmer, and Min 2010; Lieberman and Singh 2012; Jha 2013).

Many of the puzzles examined in this book have been the focus of scholars across the social sciences working on the politics of a variety of regions and countries. Potential answers to several important research questions—about the causes of Hindu-Muslim riots in India, the nature of ethnic salience, and the relationship between colonial rule and contemporary ethnic violence—are offered through this study.

A number of scholars have made major advancements in the study of ethnic violence in India, specifically Hindu-Muslim riots. This literature has rightly been called one of the most striking examples in recent years of the development of a cumulative research program in political science (Chandra 2006b: 207). Recently, three scholars have put forward theories to explain these riots.¹⁴ Paul Brass (1997, 2003) has argued that communal¹⁵ riots in India are not random acts of violence but rather are driven by riot specialists, many of whom belong to militant Hindu nationalist organizations such as the Vishva Hindu Parishad (World Hindu Council), the Rashtriya Swayamsevak Sangh (National Volunteer Association), and the Bharatiya Janata Party (Indian People’s Party). Ashutosh Varshney (2002) argues that violence occurs when intercommunal links—those that cross religious divides—do not exist in society. These links may be formal (business organizations, neighborhood peace committees) or informal (Hindu and Muslim families eating together), although formal links are stronger. Steven Wilkinson (2004) advances an electoral theory of Hindu-Muslim violence, arguing that politicians have the ability to prevent riots through their control of the police force. Local politicians will do this when Muslims form an important voting bloc, or when the party system is so competitive that the governing parties will need future coalition partners.

These works have illuminated fascinating and undoubtedly important aspects of Hindu-Muslim violence in India, but this book aims to correct two underlying problems from this literature. The first is a historical problem: all three theories focus largely on the twentieth century, especially the postindependence period. But communalism represents a significantly deeper dilemma for India. Marc Gaborieau (1985: 12) notes that Hindu-Muslim riots occurred routinely in the past, even long before the British arrived on the subcontinent. He recounts:

The oldest evidence I came across is from Ibn Battuta, the 14th century Moroccan traveller. He speaks of the South Indian town of Mangalore which was still under Hindu rule. There was there a community of 4,000 Muslim merchants who lived in a suburb near the Hindu town (note the spatial segregation). Then, Ibn Battuta continues, war frequently breaks out between them (the Muslims) and the (Hindu) inhabitants of the town; but the Sultan (the Hindu King) keeps them at peace because he needs the merchants.

This fact creates complications for existing arguments. For example, why did Hindu-Muslim riots occur for hundreds of years prior to the introduction of an electoral system? Though elites undoubtedly manipulated ethnic identities in the past, large-scale violence occurred despite the lack of formal electoral competition. Similarly, as responsible as Hindu nationalists may be for contemporary riots, they are not the explanation for medieval conflict. This book advances the existing literature by considering the underlying causes of ethnic violence in India. It analyzes conflict within a longer historical time frame, one spanning the precolonial, colonial, and postcolonial periods.

A second problem of existing research is its narrow focus on religion. The brutality of Hindu-Muslim riots is evident in India, but many kinds of ethnic conflict are plaguing the country. Violence involving secessionist linguistic groups in the northeast, rural low-caste uprisings in the so-called Hindi belt (a large region in north India where Hindi is the primary language), and tribal rebellions in the jungles of eastern India have been comparatively understudied in recent political science research.¹⁶ Consider the impact of caste violence: in 2010 a total of 32,712 crimes were committed against members of the Scheduled Castes.¹⁷ This book advances the existing literature by looking at religious violence in tandem with caste and tribal violence, thereby providing a more comprehensive portrait of ethnic conflict in contemporary India. Naturally one risk with such a broad scope is that different types of ethnic violence may have different causes. This is probably true—for instance, caste and tribal violence are largely driven by land issues. However, all of these conflicts may be called ethnic conflicts,¹⁸ and patterns of violence are largely driven by legacies inherited from the past.

Another area of social science research has tried to unravel the puzzle of ethnic salience: explaining why some kinds of identities become heightened or prioritized over others. There are several studies of ethnic salience,¹⁹ but two major arguments can perhaps simplify the debate. An historical account comes from David Laitin’s (1986) work on ethnic cleavages and conflict in Nigeria. Laitin set out to explain why politics in Yorubaland revolved around ancestral city membership rather than religion, even though there was a clear Muslim-Christian divide in the region. He found that British colonial administrators had expunged (154) religion as a legitimate form of ethnic classification, thereby creating a hegemony of ancestral city membership, which then became the main axis of conflict. A competing rationalist account of ethnic salience comes from the work of Daniel Posner (2005), who studied why ethnic competition in Zambia focused on either tribal or linguistic identities. He posits that individuals consider the sizes of ethnic groups in a region, then form minimum winning coalitions—that is, they pick which identities most benefit them given specific institutional rules. In this account, ethnic identities have no major purpose other than their instrumental value. According to Posner, the salience of a particular identity can be inferred from the ethnic demography of a region.

These explanations of ethnic salience are focused on African politics, but this book advances research on this question by examining the comparative implications of these theories. In India, an historical account of ethnic salience has more explanatory value than a rationalist account in that patterns of ethnic conflict are based on colonial history, not instrumental motivations. However, building on Laitin’s work, this book places special focus on the role of institutions in explaining how