Claiming the State: Active Citizenship and Social Welfare in Rural India 9781107199750, 9781108185899, 2018011138, 9781316649008

Table of contents :
Machine generated contents note: pt. I INTRODUCTION AND THEORY
1.Introduction: Citizenship and Social Welfare
Claiming Services, Claiming the State
Citizen and State in India
Explaining Active Citizenship
Research Setting
Research Design
Organization of the Book
2.A Theory of Active Citizenship
Claim-Making Conditions
Aspirations toward the State
Capabilities for Action
Social and Spatial Exposure
Scope Conditions
Applying the Theory
pt. II CITIZENSHIP PRACTICE IN RAJASTHAN
3.The Institutional Terrain of the State
Expansion of the Social Welfare Sector
Social Welfare Provision in Rajasthan
Rajasthan in Transition
Going Local
Setting the Stage for Claim-Making
4.Seeking the State: Claim-Making Patterns and Puzzles
The Claim-Making Landscape
Measuring Claim-Making
Disaggregating Claim-Making: What, Where, and Who?
Patterns and Puzzles
5.Encountering the State: Citizens `Social and Spatial Exposure'
Contents note continued: Social and Spatial Boundaries
Boundary Porousness
Increasing Social and Spatial Exposure
Exposure and Claim-Making Aspirations
Exposure and Claim-Making Capabilities
6.Claiming the State: Exposure as a Catalyst for Citizen Action
Testing the Relationship
Analysis and Findings
What Causes What?
Correlates of Claim-Making and Exposure
Aspirations and Capabilities
Isolation and Connectivity
pt. III CONSEQUENCES AND EXTENSIONS
7.The Consequences of Claim-Making
Material Access to Social Welfare
Claim-Making as Quotidian Citizenship Practice
Other Forms of Political Participation
The Changing Face of Clientelism
Rajasthan at a Crossroads
8.Conclusion: Active Citizenship in Rajasthan and Beyond
The Capacity to Aspire
Beyond Community and Locality
The Conditioning Effects of the State
State-Induced and Socially Produced Citizen Action
Building and Sustaining Active Citizenship
Contents note continued: Appendices
Appendix I Research Methodology
Appendix II Correlates of Claim-Making
Appendix III Correlates of Social and Spatial Exposure
Appendix IV Testing the Mechanisms: Aspirations and Capabilities
Appendix V The Consequences of Claim-Making.

Citation preview

Claiming the State

Citizens around the world look to the state for social welfare provision but often struggle to access essential services in health, education, and social security. This book investigates the everyday practices through which citizens of the world’s largest democracy make claims on the state, asking whether, how, and why they engage public officials in the pursuit of social welfare. Drawing on extensive fieldwork in rural India, Gabrielle Kruks-Wisner demonstrates that claim-making is possible in settings (poor and remote) and among people (the lower classes and castes) where much democratic theory would be unlikely to predict it. Examining the conditions that foster and inhibit citizen action, she finds that greater social and spatial exposure – made possible when individuals traverse boundaries of caste, neighborhood, or village – builds citizens’ political knowledge, expectations, and linkages to the state, and is associated with higher levels and broader repertoires of claimmaking. gabrielle kruks-wisner is Assistant Professor of Politics and Global Studies at the University of Virginia. She was previously an Academy Scholar at the Harvard Academy for International and Area Studies, and Assistant Professor of Political Science at Boston College. She holds a Ph.D. in Political Science and a Master’s in International Development and Regional Planning from the Massachusetts Institute of Technology, and a B.A. in Sociology & Anthropology from Swarthmore College.

Claiming the State Active Citizenship and Social Welfare in Rural India

GABRIELLE KRUKS-WISNER University of Virginia

University Printing House, Cambridge cb2 8bs, United Kingdom One Liberty Plaza, 20th Floor, New York, ny 10006, USA 477 Williamstown Road, Port Melbourne, vic 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06–04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781107199750 doi: 10.1017/9781108185899 © Gabrielle Kruks-Wisner 2018 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2018 Printed in The United States of America by Sheridan Books, Inc. A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data names: Kruks-Wisner, Gabrielle, 1977– author. title: Claiming the state : active citizenship and social welfare in rural India / Gabrielle Kruks-Wisner. description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2018. | Includes bibliographical references and index. identifiers: lccn 2018011138 | isbn 9781107199750 (hb) | isbn 9781316649008 (pb) subjects: lcsh: Political participation–India. | Poor–Political activity–India. | Citizenship–Social aspects–India. | Marginality, Social–Political aspects–India. | Public welfare–India. | Social service–India. classification: lcc jq281 .k78 2018 | ddc 361.954/091734–dc23 LC record available at https://lccn.loc.gov/2018011138 isbn 978-1-107-19975-0 Hardback isbn 978-1-316-64900-8 Paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Claiming the State

Citizens around the world look to the state for social welfare provision but often struggle to access essential services in health, education, and social security. This book investigates the everyday practices through which citizens of the world’s largest democracy make claims on the state, asking whether, how, and why they engage public officials in the pursuit of social welfare. Drawing on extensive fieldwork in rural India, Gabrielle Kruks-Wisner demonstrates that claim-making is possible in settings (poor and remote) and among people (the lower classes and castes) where much democratic theory would be unlikely to predict it. Examining the conditions that foster and inhibit citizen action, she finds that greater social and spatial exposure – made possible when individuals traverse boundaries of caste, neighborhood, or village – builds citizens’ political knowledge, expectations, and linkages to the state, and is associated with higher levels and broader repertoires of claimmaking. gabrielle kruks-wisner is Assistant Professor of Politics and Global Studies at the University of Virginia. She was previously an Academy Scholar at the Harvard Academy for International and Area Studies, and Assistant Professor of Political Science at Boston College. She holds a Ph.D. in Political Science and a Master’s in International Development and Regional Planning from the Massachusetts Institute of Technology, and a B.A. in Sociology & Anthropology from Swarthmore College.

Claiming the State Active Citizenship and Social Welfare in Rural India

GABRIELLE KRUKS-WISNER University of Virginia

University Printing House, Cambridge cb2 8bs, United Kingdom One Liberty Plaza, 20th Floor, New York, ny 10006, USA 477 Williamstown Road, Port Melbourne, vic 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06–04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781107199750 doi: 10.1017/9781108185899 © Gabrielle Kruks-Wisner 2018 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2018 Printed in The United States of America by Sheridan Books, Inc. A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data names: Kruks-Wisner, Gabrielle, 1977– author. title: Claiming the state : active citizenship and social welfare in rural India / Gabrielle Kruks-Wisner. description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2018. | Includes bibliographical references and index. identifiers: lccn 2018011138 | isbn 9781107199750 (hb) | isbn 9781316649008 (pb) subjects: lcsh: Political participation–India. | Poor–Political activity–India. | Citizenship–Social aspects–India. | Marginality, Social–Political aspects–India. | Public welfare–India. | Social service–India. classification: lcc jq281 .k78 2018 | ddc 361.954/091734–dc23 LC record available at https://lccn.loc.gov/2018011138 isbn 978-1-107-19975-0 Hardback isbn 978-1-316-64900-8 Paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Claiming the State

Citizens around the world look to the state for social welfare provision but often struggle to access essential services in health, education, and social security. This book investigates the everyday practices through which citizens of the world’s largest democracy make claims on the state, asking whether, how, and why they engage public officials in the pursuit of social welfare. Drawing on extensive fieldwork in rural India, Gabrielle Kruks-Wisner demonstrates that claim-making is possible in settings (poor and remote) and among people (the lower classes and castes) where much democratic theory would be unlikely to predict it. Examining the conditions that foster and inhibit citizen action, she finds that greater social and spatial exposure – made possible when individuals traverse boundaries of caste, neighborhood, or village – builds citizens’ political knowledge, expectations, and linkages to the state, and is associated with higher levels and broader repertoires of claimmaking. gabrielle kruks-wisner is Assistant Professor of Politics and Global Studies at the University of Virginia. She was previously an Academy Scholar at the Harvard Academy for International and Area Studies, and Assistant Professor of Political Science at Boston College. She holds a Ph.D. in Political Science and a Master’s in International Development and Regional Planning from the Massachusetts Institute of Technology, and a B.A. in Sociology & Anthropology from Swarthmore College.

Claiming the State Active Citizenship and Social Welfare in Rural India

GABRIELLE KRUKS-WISNER University of Virginia

University Printing House, Cambridge cb2 8bs, United Kingdom One Liberty Plaza, 20th Floor, New York, ny 10006, USA 477 Williamstown Road, Port Melbourne, vic 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06–04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781107199750 doi: 10.1017/9781108185899 © Gabrielle Kruks-Wisner 2018 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2018 Printed in The United States of America by Sheridan Books, Inc. A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data names: Kruks-Wisner, Gabrielle, 1977– author. title: Claiming the state : active citizenship and social welfare in rural India / Gabrielle Kruks-Wisner. description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2018. | Includes bibliographical references and index. identifiers: lccn 2018011138 | isbn 9781107199750 (hb) | isbn 9781316649008 (pb) subjects: lcsh: Political participation–India. | Poor–Political activity–India. | Citizenship–Social aspects–India. | Marginality, Social–Political aspects–India. | Public welfare–India. | Social service–India. classification: lcc jq281 .k78 2018 | ddc 361.954/091734–dc23 LC record available at https://lccn.loc.gov/2018011138 isbn 978-1-107-19975-0 Hardback isbn 978-1-316-64900-8 Paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

To Kiran and Asha, my rays of light and hope

Awarded the

Joseph W. Elder Prize in the Indian Social Sciences

by the American Institute of Indian Studies and published with the Institute’s generous support.

AIIS Publication Committee: Susan S. Wadley, Co-Chair Anand A. Yang, Co-Chair Deborah Hutton Ramnarayan S. Rawat Tulasi Srinivas

Contents

List of Figures List of Tables Preface Note on Caste Terminology

page xi xiii xv xxi

part i introduction and theory 1

Introduction: Citizenship and Social Welfare Claiming Services, Claiming the State Citizen and State in India Explaining Active Citizenship Research Setting Research Design Organization of the Book

3 6 12 16 20 23 26

2

A Theory of Active Citizenship Claim-Making Conditions Aspirations toward the State Capabilities for Action Social and Spatial Exposure Scope Conditions Applying the Theory

29 31 33 39 42 47 53

part ii citizenship practice in rajasthan 3

The Institutional Terrain of the State Expansion of the Social Welfare Sector Social Welfare Provision in Rajasthan

57 58 64

vii

Table of Contents

viii Rajasthan in Transition Going Local Setting the Stage for Claim-Making

69 74 82

4

Seeking the State: Claim-Making Patterns and Puzzles The Claim-Making Landscape Measuring Claim-Making Disaggregating Claim-Making: What, Where, and Who? Patterns and Puzzles

87 89 95 106 115

5

Encountering the State: Citizens’ Social and Spatial Exposure Social and Spatial Boundaries Boundary Porousness Increasing Social and Spatial Exposure Exposure and Claim-Making Aspirations Exposure and Claim-Making Capabilities

117 118 122 138 140 142

6

Claiming the State: Exposure as a Catalyst for Citizen Action Testing the Relationship Analysis and Findings What Causes What? Correlates of Claim-Making and Exposure Aspirations and Capabilities Isolation and Connectivity

148 149 154 158 162 170 174

part iii consequences and extensions 7

The Consequences of Claim-Making Material Access to Social Welfare Claim-Making as Quotidian Citizenship Practice Other Forms of Political Participation The Changing Face of Clientelism Rajasthan at a Crossroads

185 186 194 196 199 202

8

Conclusion: Active Citizenship in Rajasthan and Beyond The Capacity to Aspire Beyond Community and Locality The Conditioning Effects of the State State-Induced and Socially Produced Citizen Action Building and Sustaining Active Citizenship

207 208 211 215 222 227

Appendices Appendix I Research Methodology Appendix II Correlates of Claim-Making

231 238

Appendix III Correlates of Social and Spatial Exposure

261

Table of Contents

ix

Appendix IV Testing the Mechanisms: Aspirations and Capabilities 272 Appendix V The Consequences of Claim-Making 282 References Index

289 309

Figures

1.1 1.2 1.3 2.1 2.2 2.3 2.4 2.5 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 4.1 4.2 4.3 6.1 6.2 6.3 8.1

The Claim-Making Repertoire in Rajasthan Rajasthan, Location in India Poverty Rates in Selected Districts of Rajasthan Claim-Making Conditions Producing Aspirations Producing Capabilities Boundary Porousness, Exposure, and Claim-Making Claim-Making and the Institutional Terrain of the State Education Spending in India Health Spending in India Social Security Spending in India Literacy Rates in Rajasthan and India Social Sector Spending in Rajasthan Social Sector Spending as Percentage of State Expenditures Infant Mortality in Rajasthan and India MGNREGS Funding in Rajasthan Claim-Making Incidence by District Claim-Making Incidence by Class, Caste, and Gender Claim-Making Repertoire by Class, Caste, and Gender Index of Social and Spatial Exposure Exposure and the Likelihood of Claim-Making Exposure and the Breadth of the Claim-Making Repertoire Exposure’s Effects, Conditional on the State

page 16 21 23 32 33 40 44 52 60 60 61 66 67 68 69 80 111 112 114 153 156 157 217

xi

Tables

1.1 1.2 4.1 4.2 4.3 4.4 4.5 4.6 5.1 5.2 5.3 5.4 5.5 6.1 6.2 6.3 6.4 6.5 6.6 7.1 7.2 7.3 7.4 7.5 7.6 7.7

Selected Districts: Demographics page 24 Selected Districts: Economic and Social Indicators 24 The Claim-Making Landscape 91 Claim-Making Practice 96 Comparing Channels of Citizen-State Engagement 99 Mediated Claim-Making Channels, Conditional on Presence 102 Combining Claim-Making Practices 105 Claim-Making for Collective versus Selective Services 107 Landownership by Caste 123 Occupation and Livelihood by Caste 124 Movement Beyond the Village 127 Level of Education by Caste 130 Functional Literacy by Caste and Age Cohort 131 Indicators of Social and Spatial Exposure 150 The Effects of Exposure on Claim-Making Incidence and Repertoire 155 The Village Land-to-Labor Ratio and Claim-Making 161 Interaction Effects: Exposure and Social Standing 165 Political Knowledge, Grievance, and Efficacy 171 Testing the Mechanisms: Knowledge, Grievance, and Efficacy 173 Access to Publicly Provided Services 188 Access to Services by Landownership 190 Access to Services by Caste 190 Claim-Making and Access to Services 191 Perceived Effectiveness of Claim-Making Practice 194 Electoral, Deliberative, and Contentious Participation 197 Claim-Making and Other Political Participation 198

xiii

xiv

A1.1 A1.2 A2.1 A2.2 A2.3 A2.4 A2.5 A3.1 A3.2 A4.1 A4.2 A5.1 A5.2

List of Tables Citizen Survey: Individual and Household Characteristics Citizen Survey: Village and Panchayat Characteristics Claim-Making Practice by Class, Caste, and Gender Correlates of Claim-Making: Village and Panchayat Characteristics Correlates of Claim-Making: Individual and Household Characteristics Correlates of Claim-Making: Social and Spatial Exposure Principal Component Analysis of Indicators of Exposure Correlates of Exposure: Village and Panchayat Characteristics Correlates of Exposure: Individual and Household Characteristics Effects of Exposure on Knowledge, Grievance, and Efficacy Effects of Knowledge, Grievance, and Efficacy on Claim-Making Claim-Making and Material Access to Public Services Claim-Making and Other Forms of Participation

235 236 240 243 249 257 259 263 267 274 278 284 286

Preface

Initially, I intended to write a book about social welfare provision in rural India. I ended up writing one about the pursuit of social welfare: that is, the strategies through which people seek public goods and services central to their well-being. These subjects, while closely intertwined, represent two different dimensions (supply and demand) of the puzzle of variable social outcomes in rural India – a region in which a quarter of the population lives in poverty, where one-third cannot read or write, and almost 50 percent of children are malnourished. Governmental efforts to resolve these problems through social spending and programs in the areas of education, health, and social security are well studied. Much of this work examines the arenas in which social policy is crafted or, at a more local level, the dynamics that influence bureaucratic performance and policy implementation. I first traveled to Rajasthan, a state in northern India’s poverty belt, with the idea that these “supply side” studies were missing a critical dimension: citizen demand. Despite a global proliferation of “demand-driven” programs, the practice day-to-day of citizen claimmaking is often overlooked or taken for granted in studies of service provision. In fact, we know relatively little about how citizens’ interests vis-à-vis the state are formed and articulated, or about what the state and its service delivery apparatus look like through the eyes of ordinary people. I therefore wanted to study whether the ways in which people approached and petitioned the state influenced their access to essential services. I first became interested in Indian citizens’ pursuit of social welfare in a setting far afield from Rajasthan, during research in south India in the wake of the 2004 tsunami. There, working in affected fishing villages, I observed that different people – men, women, and members of different caste communities – sought aid through very different channels: some turned to elected local representatives, some to traditional caste leaders, and some to NGOs. Moving from the post-tsunami environment, I wanted to study similar dynamics under the xv

xvi

Preface

more quotidian (but, in many ways, also disastrous) conditions of being poor and underserved in rural India. How, I wondered, did citizens in India’s northern poverty belt navigate access to the state? Where or to whom did they turn when the water pump was broken, when the health clinic was unstaffed, or when seeking income support? I expected to find that differently placed citizens, set apart by caste or class or other features, would approach the state differently. My hunch was that the different pathways citizens pursued would have a bearing on their material access to public resources. Once on the ground in Rajasthan, however, a new set of puzzles emerged. Through conversations about service delivery, many of which took place at water sources or in local schools or other community buildings, I noted that people expressed very different opinions about what the state should deliver, if it would deliver, and whether it was worthwhile to speak up. I also noted that these differences did not appear to conform to some of the patterns that I initially expected: citizens’ approaches to the state varied within the same villages, as well as within socioeconomic and caste groupings. This variation, I realized, could not simply be a matter of rich versus poor or high versus low social standing. My research agenda thus began to shift. The central question of how citizens navigated access to the state remained the same. But I quickly realized that there was another, fundamental puzzle that needed to be addressed: namely, why citizens’ approaches to the state varied in the first place – not just across but within communities. Why did some people make claims on the state while others did not? Why did some directly seek out local officials, while others sought assistance from political intermediaries? And why did some turn again and again to the same actors or institutions, while others diversified their approaches? As time went on, I came to see these questions as uncovering a central but often neglected dimension of local citizenship practice, with important implications for our understanding of participation and representation (Who speaks up? Who speaks for or through whom?), as well as distributive politics (Who gets what from the state?). This book sets out to do three things. First, it aims to open the black box between and beyond elections, to describe the strategies through which citizens in one of India’s poorest states pursue social welfare. Far from being locked into static or singular patronage structures, I find that citizens utilize a wide array of channels, acting not just through political parties but through a range of local officials, bureaucrats, and nonstate actors. Some, though, do none of this: instead, they retreat from the state, even when faced with pressing needs. Second, the book aims to explain the varied patterns of citizen action and inaction that emerge. The central argument, broadly stated, is that claimmaking is both state induced and socially produced through direct and narrated encounters with and observations of the state. Those who traverse boundaries of community and locality (for example, of caste and class, neighborhood and village) gain information and ideas about, as well as linkages to, the state. The

Preface

xvii

core insight is that this social and spatial exposure is associated with a greater likelihood of making claims on the state and with a broader repertoire of claimmaking practices. Third, the book reflects on the implications of the observed patterns of citizen claim-making for democratic practice more broadly. Here, the story is unfinished. Rajasthan, like much of rural India, stands at a crossroads. Rapid social and economic changes are afoot: boundaries of caste, occupation, and village – once powerfully constraining – are becoming less rigid, although not uniformly so. At the same time, the reach of the state is also shifting, penetrating deeper into the lives of the rural poor through decentralization and a proliferation of social programs. The combined result is that there is “more state” visible at the local level along with more mobile citizens, leading to greater citizen–state encounters. These are conditions ripe for claim-making. Whether this is a democratic success story depends crucially on the institutionalization of claim-making over time: that is, on whether and how citizens sustain these practices and whether and how officials respond. The potential long-term effects – virtuous or vicious – are enormous. This book began as a doctoral dissertation at the Massachusetts Institute of Technology. I extend my heartfelt gratitude to the members of my committee, Richard Locke, Suzanne Berger, Lily Tsai, and Daniel Posner, who helped to shape the project since its very inception, pushing me on all aspects of the work – theoretical, methodological, and empirical. As the dissertation wound its way to book, I benefited enormously from colleagues kind enough to offer their guidance. I am particularly indebted to Thad Dunning, Patrick Heller, Anirudh Krishna, Evan Lieberman, Ashutosh Varshney, and Steven Wilkinson, all of whom provided invaluable feedback at an author’s conference hosted by Harvard Academy for International and Area Studies. John Echeverri-Gent, Rob Jenkins, Devesh Kapur, Sandip Sukhtankar, and three anonymous reviewers also commented on full drafts of the manuscript – gifts of time and intellect for which I am profoundly grateful. For their insightful reading of chapters and papers related to the book, I also thank: Matt Amengual, Adam Auerbach, Kate Baldwin, Rikhil Bhavnani, Rachel Brulé, Jennifer Bussell, Melani Cammett, Erica Dobbs, Herbert Kitschelt, Sonia Kruks, Akshay Mangla, Carol Mershon, Sara Schneiderman, Prerna Singh, Tariq Thachil, Ben Wisner, and Adam Ziegfeld. For rich discussion and comments at various conferences, workshops, and presentations, I thank Amit Ahuja, Simon Chauchard, Emily Clough, Jennifer Erickson, Kristin Fabbe, Janice Gallagher, Jason Jackson, Francesca Jensenius, Michael Levien, Lant Pritchett, Meg Rithmire, Neelanjan Sircar, Mark Schneider, Ken Sharpe, Pavithra Suranarayan, and Milan Vaishnav. Numerous scholars, practitioners, and other individuals in India helped me in the course of my research. At the heart of this book are the contributions of hundreds of unnamed people in Rajasthan. I am humbled by the gift of time that so many gave. I am also deeply grateful to Kripa Ananthpur, Nikhil Dey,

xviii

Preface

Neelima Khetan, Sanjay Lodha, Ajay Mehta, Susanne and Lloyd Rudolph, Preeti Shaktawat, Priyanka Singh, and Kavita Srivastava, all of whom in different ways provided me with support during my fieldwork. I am particularly thankful for Sanjay Lodha’s assistance in helping to organize my survey team. Neelima Khetan and her colleagues at Seva Mandir were highly generous in allowing me to draw upon the experiences and resources of their organization. Chandadevi Chipa in Udaipur provided me with a home away from home. Smita and Rajeev Sukhtankar, my mother- and father-in-law, were an enormous source of both emotional and logistical support. My fieldwork hinged upon the support of a large team. I owe an enormous debt to Vivek Kumar Shukla, who first joined me as a translator and who became a research assistant, colleague, and friend. Vivek’s skills as both an interviewer and interpreter are truly impressive, and his ability to capture local meaning and color has left a great mark on this book. A talented group of research assistants helped to oversee the survey work and assisted in carrying out interviews: Bharati Agarwal, Mrinmoy Thakur, Mahendra Singh, Jignesh Vyas, Abhishek Ranwah, Shailendra Singh, and Nidhi Jain. A group of students from Mohanlal Sukhadia University put in long hours under trying conditions to carry out the survey. They include: Shankarlal Choudhary, Ankit Jain, Rajendra Jain, Avinash Kumawat, Vishal Kumawat, Ramratan Nayak, Prakash Raiwal, Digvendra Singh, Kuldeep Singh, Dilip Singh Shekhawat, Devilal Teli, and Kanhaiyalal Teli. Ramratan Nayak and Digvendra Singh also assisted me during village case studies. I received financial support from the Fulbright Program, the Boren Fellowship, the National Science Foundation, the State Department’s Critical Language Scholarship program, and the MIT-India Program. The Harvard Academy for International and Area Studies offered me time, space, and a community of scholars in which to further develop and deepen the project. The Boston College Political Science Department provided me with a collegial home and support that enabled me to write, including the benefit of the Intersections Villa faculty writing retreat. I also received accomplished research assistance at Boston College from Robert Bohn, Emma Gleason, Varsha Ramesh, and Sean Sudol. I completed this book at the University of Virginia, where I have found happy homes in both the Politics Department and Global Studies Program. I particularly thank John Echeverri-Gent, Richard Handler, David Leblang, and Carol Mershon for their support and mentorship at UVA, as well as Daniel Davis for his excellent research assistance. At Cambridge University Press, I thank Lew Bateman for his early encouragement, Robert Dreesen for his guidance in taking the book forward, Joshua Penney and Niranjana Harikrishnan for their dedicated work on production, and Elizabeth Kelly for skilled copyediting. Portions of the book, primarily appearing in Chapters 5 and 6, draw on material previously published in World Politics 70(01), January 2018, the editors and anonymous reviewers of which I also gratefully acknowledge.

Preface

xix

Finally, my greatest debt and deepest thanks go to my family, who have lived and breathed this project with me. My parents, Sonia Kruks and Ben Wisner, have led by example – teaching me the value of scholarship that strives to make the world a more just and livable place. My husband, Sandip Sukhtankar, deserves more thanks and credit than I can express. His mark is all over this book, from helping me to think through questions of methodology and research design, to assisting in the pilot stages of my survey, to patient STATA tutorials, to acting as a constant sounding board. He is my partner in the truest sense of the word. Finally, my son Kiran Lev and daughter Asha Lily brought me the joy and balance that allowed me to finish this project. They are my human bookends: Kiran was born two months before I submitted the dissertation, and Asha arrived on the scene just a week after the book was first submitted to the press for consideration. My world is immeasurably brighter, livelier, and stickier with them in it. I thank them for their (almost) patience while I was away at my desk, and for not spilling on my keyboard.

Note on Caste Terminology

Throughout this book, I refer to members of different caste and tribal communities in India, the categorization and labeling of which are fraught tasks. I use the term “Scheduled Caste” to refer to those caste groupings that have received official designation in the Indian Constitution, recognizing their history as disadvantaged and marginalized communities for which special provisions are made for social, educational, and economic advancement (including reservation of seats in elected government, in higher education, and in certain public sector jobs). Historically, the same groups were referred to as “untouchables” – a derogatory term that denotes a sense of “pollution” should higher caste members come into contact with members of such “lower” castes. Some have instead employed the term Harijan, or “Children of God,” as coined by Mahatma Gandhi, to refer to the same lower caste groups. This term, though, has been critiqued for its paternalistic overtones. Others, including some activist groups among the lower castes, have employed the term Dalit, meaning “broken people.” I, however, do not use this term for two reasons. First, some activists and scholars rightly object to the implications of a “broken” or downtrodden status. Second, in the state of Rajasthan, where I carried out research for this book, the term Dalit was rarely employed and appeared to have little resonance among members of the so-called lower castes themselves. They instead tended to use the term “Scheduled Caste,” or “SC.” I follow suit in using the term Scheduled Caste, despite its technocratic overtones, since it is (1) the most value -neutral label of those in common usage, and (2) because it mirrors the language used by villagers themselves in Rajasthan. The “Scheduled Tribes,” also referred to as adivasis, are indigenous communities that are also recognized in the Indian Constitution as historically disadvantaged groups for which affirmative action provisions have been made. They span a diverse set of communities, most of which have distinctive cultures, xxi

xxii

Note on Caste Terminology

languages, and religions – many of which, though, also practice syncretic elements of Hinduism. In the text that follows, I refer to the Scheduled Castes and Scheduled Tribes (abbreviated as the SC and the ST) as collective groupings encompassing all caste or tribal communities that have received constitutional designation as “scheduled,” while keeping in mind the heterogeneity of castes, jatis (subcastes), and tribes that these categories encompass. I use the terms both as nouns (referring to collective categories) and as adjectives (to describe individual members of the groups, or the nature of a village or neighborhood – e.g., an SC man or an ST village). This broadly mirrors the language used on the ground in Rajasthan, where members of these communities often referred to themselves and their own communities in these same ways. I refer to non-SC and non-ST members in one of two ways. First, the “Other Backward Classes” (OBCs) are an official designation applied by the Indian states to communities that, while not included in the original “schedules” of the Indian Constitution, have also contented to varying degrees with histories of social exclusion and economic and political marginalization. They refer, broadly, to the lower-middle rankings of the Hindu varnas, or caste hierarchy. OBC categorization remains a highly contested and conceptually unclear status, which varies from state to state. In the text that follows, I follow official Government of Rajasthan designations when referring to members of the OBCs. Second, I refer to all other “unscheduled” and non-OBC caste groupings as the “General” Castes (GCs). The GCs, also sometimes referred to as the “unreserved,” typically hail from those communities in the middle to upper echelons of the Hindu varna system, reflecting their relatively higher social (and often economic) status – although this classification also encompasses a great diversity of people.

part i INTRODUCTION AND THEORY

1 Introduction Citizenship and Social Welfare

In the spring of 2009, I met a woman named Chandibai1 in Udaipur, a remote, rural district of Rajasthan in northwestern India. Udaipur is among Rajasthan’s poorest districts, and is home to one of the state’s largest adivasi (tribal or indigenous) communities – a population that is often among the poorest of the poor. Chandibai was a self-described leader in her village: a person to whom others (particularly other women) turned for help when seeking services such as the provision of drinking water, ration cards, or other welfare benefits. She wore a mobile phone around her neck and had numbers for village and district officials on speed dial. She explained: I know the system. I know who to call, and they will not ignore my call. And so everyone comes to me with their problems. Even men, who did not think a woman could do this work, know that I can assist them.2

I found Chandibai’s activism remarkable, particularly in a region known for its restrictive gender norms. As a tribal woman with little formal education, she seemed all the more unusual. She was, in fact, the kind of person – poor, female, uneducated, living in a remote, rural region – from whom we might least expect active engagement with the state. And yet, as I extended my research across Rajasthan, I came to realize that Chandibai was not simply an anomaly. While she was in some ways unusual, particularly in her boldness, she was also part of a broader pattern of active citizenship practice that extended across the state. The vast majority of citizens who I encountered actively sought and made claims on the state, taking up a wide range of strategies – both formal and informal, direct and mediated – in an effort to secure public resources. Such

1 2

Names have been changed. Quote from a focus group discussion, Gogunda block, Udaipur district, Rajasthan, April 14, 2009.

3

4

Introduction: Citizenship and Social Welfare

high rates of citizen action were striking in Rajasthan – a poor, agrarian, castedivided, and erstwhile “princely” state with strong feudal legacies.3 How could I explain citizen voice, which Hirschman reminds us can range from “faint grumbling to violent protest,”4 among people and in a region where most democratic theory would be unlikely to predict it? What, moreover, could account for varied repertoires of action and inaction among similarly situated individuals, living in the same local communities? Why did some, like Chandibai, make claims on the state – voicing needs, interests, and demands – while others did not? Scholars of India, and of democratic practice more generally, know a lot about the electoral arena – about who votes and why. The experiences of the rural residents recounted in this book, however, take us far beyond that wellstudied realm, instead highlighting the everyday practices through which citizens attempt to navigate access to the state. Around the globe, large numbers of citizens regularly turn to the state for education, health care, shelter, clean drinking water, and – for the poorest – income support, food rations, and other forms of social protection. In many places, though, it is a constant struggle to secure access to even the most basic services and entitlements. Water pumps break; health clinics are located at prohibitive distances; teachers are absent; and applications for services are ignored, lost, or subject to bribe requests. Numerous people are thus regularly denied the social rights that enable a “life of civilized being.”5 This book is about how citizens seek to overcome these problems by making claims on the state for social welfare through everyday and seemingly mundane acts such as attending a meeting, filing an application, visiting a government office, or approaching a local leader for assistance. Claim-making of this kind is an essential but often overlooked form of political participation that unfolds “beyond the romance of elections and the grandeur of social movements.”6 Quotidian acts of claim-making, while not the gripping stuff of voting booths or barricades, are no less consequential. Materially, they can mean the difference between the maintenance of a community pump and seeing that water source fall into disrepair; between securing and not securing access to a pension or employment on a government job site; between disciplining an absent teacher and sending one’s children to an empty schoolroom. They are, moreover, fundamental political acts through which citizens forge and navigate their relationship to the state. Who, then, makes claims on the state for social welfare, how, and why? Such questions are particularly salient in developing democracies across the Global

3

4 6

India’s princely states, described in greater detail later in this chapter, were nominally sovereign bodies that, while allied to the British Crown, retained partial autonomy, and which were noted for their feudal social and political relations. 5 Hirschman 1970, p. 16. Marshall 1950, p. 149. I thank Ashutosh Varshney for this phrase.

Introduction: Citizenship and Social Welfare

5

South in settings marked by high levels of clientelistic exchange, discretion, and rule bending among officials, where the gap between policy and implementation – between de jure social rights and the substantive realization of those rights – looms large.7 Access to public resources in these settings is not simply a matter of legal entitlement but rather of who can “extract them from the political system.”8 From Karl Marx to classical modernization theorists, scholars have long predicted that the poor (and the rural poor in particular) would be less engaged in politics than their wealthier urban counterparts.9 Similar predictions extend to other markers of social standing, for example race or caste.10 In rural India, echoing Barrington Moore’s famed portrayal of the “docile” Indian peasant, residents have been regularly referred to as “passive,” “fatalistic,” and “politically accepting.”11And yet, as the chapters that follow will demonstrate, active citizen–state engagement is possible – and even likely – under remote and poor conditions. In an original survey of more than 2,000 citizens across more than 100 villages of Rajasthan, I find that roughly three-quarters have personally engaged in state-targeted efforts to claim social welfare goods and services. These patterns, though, are not uniform: a significant number – almost a quarter – are notably disengaged, reporting that they have never made any contact with officials around issues of service delivery. These patterns of citizen action and inaction cannot be adequately explained, I will show, by differences in formal political institutions, by local economic development, or by individual socioeconomic standing. Claim-making practice varies both across and within localities set apart by wealth and literacy rates, as well as among individuals who share features of economic class and caste. We therefore need to look beyond boundaries of locality, class, and caste to consider the broader spaces in which citizens learn about, develop expectations of, and seek access to the state. The central argument of this book is that citizen claim-making is both state induced and socially produced, shaped by direct encounters with public officials and by narrated accounts that circulate within and beyond communities. Claim-making is shaped, first, by the institutional terrain of the state (assessed in terms of its reach, visibility, and accessibility) and, second, by an individual’s social and spatial exposure to people and places beyond the immediate community and locality (that is, beyond boundaries of class, ethnicity, neighborhood, or village). Claim-making, I argue, is most likely where the state’s terrain

7

8 9

10 11

O’Donnell 1993; Holston 2008; Bénit-Gbaffou & Oldfield 2011; Houtzager & Acharaya 2011; Jayal 2013. Banerjee 2004, p. 13. Lerner 1958; Lipset 1959; Marx 1978; Verba, Schlozman, & Brady 1995. See Chapter 4 for a broader discussion of these works. Verba, Ahmed, & Bhatt 1971; Verba, Schlozman, Brady, & Nie 1993. Moore 1966; Narain [1978], cited in Singer 2007; Hardgrave and Kochanek 2008.

6

Introduction: Citizenship and Social Welfare

is broad and uneven – marked by visible but variable social welfare provision. Institutional breadth increases citizens’ sightings of and encounters with public resources, while unevenness generates a sense of grievance (and entitlement) when the state does not deliver. Under these conditions, social and spatial exposure is a catalyst for citizen action – building knowledge of, aspirations toward, and linkages to the state.

claiming services, claiming the state Claim-making, as it is here defined, consists of citizen action in pursuit of welfare goods and services, which are broadly understood as resources intended to protect and improve the well-being of citizens and, in particular, of the poor.12 Claim-making is, by definition, state targeted.13 The state, in Weberian fashion, may be understood as an organization that exercises “legitimate” force over a designated territory and, as such, has the authority to tax, regulate, conscript, and otherwise coerce the people within that territory. It is also, in classic liberal fashion, the purveyor of public goods that would be underproduced by private actors and the market. It is, as such, charged with ensuring public security and well-being. And yet the state is not a unitary or cohesive entity; its contours are unclear and constantly shifting, appearing differently to different people. I therefore employ an intentionally loose notion of the state as an amalgam of actors, institutions, and practices through which governmental power is exercised and public resources are distributed.14 The state exists simultaneously at the national, subnational, and local levels. It is constituted and embodied by particular actors (politicians, police, bureaucrats, and street- or village-level officials), by agencies (ministries, departments, local offices), by documents (birth or death certificates, land titles, certifications), by policies (from taxation, to conscription, to social security), and by infrastructure and services (roads, schools, hospitals, water, and other amenities). Given this book’s focus, I am concerned in particular with the “developmental” arms of the state, which I define, borrowing from Corbridge et al., as “those agencies of the state and governmental practices that are charged with 12

13

14

Here I build on Cammett and MacLean (2014, p. 6), who define social welfare provision as the “direct delivery or indirect facilitation of services and programs that promote well-being and social security.” Social welfare, in this sense, encompasses a broad array of types of social spending, from public transfers – including basic income support, employment programs, and social security – to public services in the arenas of education, health, housing, and other basic infrastructure. Globally, non-state service providers are increasing in both number and in type (Cammett & MacLean 2014). While often important sources of social welfare (and thus part of the broader landscape of service provision), these actors are not the targets of claim-making as it is defined here. Instead, claim-making entails contacting or petitioning of state actors and agencies. Here I follow Corbridge, Srivastava, and Veron (2005, p. 5), who define the state as “bundles of everyday institutions and forms of rule.”

Claiming Services, Claiming the State

7

improving or protecting the incomes, capabilities and legal rights of poorer people.”15 Citizens encounter this dimension of the state as they seek welfare goods and services in the fields of health, education, housing, employment, basic infrastructure, and social security.16 Claim-making involves efforts to navigate the state’s social welfare apparatus: that is, engaging the actors, agencies, and institutions that directly and indirectly shape the provision of such goods.17 This, by nature, entails a seeking behavior: it involves contacting, petitioning, or otherwise approaching public officials or intermediaries. The resources in question can be both collective goods with high-spillover benefits that accrue to entire groups and localities (for example, communal drinking water, roads, schools, or health clinics), as well as selective goods with lowspillover benefits to households or individuals (for example, rations, cash transfers, pensions, or public employment).18 They also encompass both legal entitlements and goods allocated in a more discretionary manner – a distinction that often becomes moot as the frequent bending of rules makes access even to entitlements highly unpredictable. It would be misleading to attempt to attribute the distribution of these resources exclusively to bottom-up processes of claim-making. A wide range of “supply-side” factors, from public-sector capacity to technocratic and political decision-making and bureaucratic norms at the national and subnational levels, inform the allocation of public resources. The linkage between citizen voice and access to services is thus seldom direct. The nature of a particular good and the structures through which it is allocated will also inevitably inform whether and how citizens seek that good. Some services, such as income support through employment on a government worksite, are by nature demand

15

16

17

18

Ibid, p. 7. This definition differs from the notion of the developmental state popularized in the 1980s and 1990s, which centered on the objectives of rapid economic and industrial growth. For a discussion of social development and the developmental state, see Evans and Heller (2015). In the rural Indian context, the social welfare sector is most often defined by the governmental budgetary categories of “social services” (health, education, housing, social security) and “rural development” (income support and employment) (Mooij & Dev 2004; World Bank 2011). Claim-making in pursuit of social welfare is one part of a broader array of practices that shape and reflect citizen–state relations. Citizens also engage (or attempt to disengage from) the state around issues of public security, dispute resolution, taxation, or conscription. These other arenas of citizen–state engagement play a critical role in conditioning the claim-making environment. Predatory state behavior with regard to taxation or property rights, for example, can create a climate of distrust that is hardly conducive to claim-making. I return to a discussion of the conditioning effects of the state in Chapter 2. On the distinction between high- and low-spillover services, see Besley, Pande, and Rao (2004). I distinguish “publicly provided” welfare goods from the more narrowly defined notion of “public goods,” which refers to services and infrastructure that are non-rival (one person’s consumption does not detract from another’s) and non-excludable (where it is impossible to prohibit access to the good). Certain welfare services may also be “public” in this narrower sense, but others can be rival, excludable, and selective in nature.

8

Introduction: Citizenship and Social Welfare

driven and thus require citizen action in the form of applications. Others – for example, large-scale infrastructure projects – may be the subjects of planning and budgetary decisions taken far from the local level. Claim-making, then, can be best understood as an often necessary but also often insufficient condition: citizen voice is rarely enough, alone, to ensure the delivery of public resources, but it can play a critical role in influencing their provision. It follows that this book is not a study of service delivery per se, nor is it primarily a study of citizens’ material receipt of resources – although these themes are taken up in later chapters.19 Rather, in studying the conditions under and strategies through which citizens seek resources from the state, I raise questions that are prior to – and so set the stage for – discussions of material distribution and access to social welfare. Claim-Making as Active Citizenship Practice Claim-making is not only materially important as a potential means to resources. As a set of practices through which citizens navigate access to the state, it is also of intrinsic value as an end (and subject of study) in itself. In making claims on the state for social welfare, citizens are negotiating the bundle of rights they are due by virtue of their membership in a given political society. They are, in other words, practicing citizenship. Citizenship is at once a legal status, an identity, and a set of obligations and rights.20 Citizenship rights, following T. H. Marshall’s foundational typology, encompass civil rights (that allow for personal freedom), political rights (to participate in the exercise of power), and social rights (to economic welfare and human well-being).21 It is these social rights – and the means through which citizens claim them – that are the primary concern of this book. Historically, in democracies in both the North and South, social rights were rarely codified in law but rather were promoted through social policies in the areas of education, health care, unemployment insurance, and social security.22 Recently, though, some states – among them Brazil, South Africa, and India – have moved to legislate social rights. Brazil’s “Citizen’s Constitution” of 1988 lists rights to “education, health, work, leisure, security, social security, protection of motherhood and childhood, and assistance to the destitute.”23 19 20

21

22

23

See, in particular, Chapter 7 on the consequences of claim-making. As Somers (2008, p. 5) puts it: citizenship is the “right to have rights” including “both de jure and de facto rights to membership in a political community.” Marshall 1950. Social rights, for Marshall, are both cultural (the right to “social heritage”) and socioeconomic (the right to a modicum level of economic and human well-being, secured by the state through “the educational system and social services”). On the evolution of social spending in Europe and the United States, see Lindert (2004). On the extension of social welfare in the Global South, see Seekings (2005). Constitution of the Federal Republic of Brazil, Article 6, Constitutional Amendment No. 26, 2000.

Claiming Services, Claiming the State

9

South Africa’s post-apartheid constitution of 1996 similarly asserts the right to basic nutrition, housing, health care, and social services.24 India’s constitution recognizes, among other fundamental rights, the “right to life” – and, in a series of directive principles, lays out mandates for India’s states to provide for adequate means of livelihood, social security, and a “decent” standard of living.25 And yet, across all of these settings, a great divide exists between the seemingly progressive rhetoric of the state at the national and legislative levels and the often regressive implementation of social policy at the local levels. There is, in other words, a profound “gap between the status and the reality of citizenship.”26 The effort of citizens to bridge this gap is an essential element of active citizenship, understood here as the quotidian practices through which citizens negotiate their social rights vis-à-vis the state.27 Citizenship in this sense is both a status (a set of rights) and an exercise (a set of practices).28 This conception draws on a long tradition that sees “citizenship as participation” and as an “expression of human agency in the political arena.”29 Claim-making is integral to active citizenship practice in this sense. Indeed, the very notion of citizenship can be understood, as Tilly asserts, as a “set of mutually enforceable claims relating categories of persons to agents of government.”30 Importantly, though, claim-making does not necessarily or consistently conform to democratic ideals of participation that rest on assumptions of political equality and effective representation. Acts of claim-making, as we will see, can be narrowly particularistic as well as dependent on powerful brokers and intermediaries. 24 25

26 27

28 29

30

Constitution of the Republic of South Africa, chapter 2, “Bill of Rights,” ss. 26–29. The Constitution of India, “Fundamental Rights,” Article 21 (21A), and “Directive Principles of State Policy,” Articles 38–39, 41–43, 46–47. Houtzager & Acharya 2011, p. 4. In emphasizing citizens’ state-targeted pursuit of social welfare services, I build, in particular, on Houtzager and Acharya’s (2011) notion of active citizenship, which they define (p. 3) as “present when individuals negotiate the terms of their access to mandated public goods and services in ways that are publicly sanctioned and protected.” My own notion of active citizenship, however, is not restricted to “sanctioned” practices; rather, citizens pursue goods and services in diverse ways – both formal and informal, sanctioned and unsanctioned. My conception is also distinct from that employed in neoliberal theory, in which active citizenship implies self-provisioning by civil society and a lesser role for the state (Cf. Miraftab & Wills 2005). Oldfield 1990; Somers 1993; Lister 1998. Lister 1998, p. 228. Also see Cornwall and Gaventa (2000), Mettler and Soss (2004,) Isin and Nielsen (2008). The notion of claim-making thus straddles two interlinked conceptions of citizenship. In the first, citizenship is a status that confers a set of civil, political, and social rights. In the second, one’s status as a citizen is derived from participation in the political community. The two conceptions are interrelated since, as Somers (1993) argues, legal frameworks granting citizenship rights are the product of social mobilization, which, in turn, responds to those rights. Tilly 1999, p. 253, emphasis added. These include both “bottom-up” claims (citizens’ demands for state-controlled resources) and the “top-down” claims (by states on citizens) for resources and compliance, through taxation, conscription, and regulation, which states make on citizens.

10

Introduction: Citizenship and Social Welfare

Claim-making practices thus often reflect the hierarchies of social exclusion in which they are embedded. Claim-making is also an inherently costly undertaking, requiring investments of time and various forms of capital (human, social, and financial), while also carrying risks of political or social reprisal. And yet, claim-making remains an important manifestation of citizen voice, made all the more salient in environments where such voices have historically been lacking. How might citizens respond when faced with material needs? Some, simply put, might not respond. Indeed, scholars have long puzzled over a lack of citizen action, particularly among the poor and marginalized who seemingly would stand to gain the most from activism. Banfield, for example, highlighted a lack of political engagement – what he called “amoral familism” – among rural Italian peasants who turned inward to insular familial networks.31 Gaventa, writing from a very different perspective, similarly sought to explain inaction in poor mining communities in rural Appalachia, asking: “Why in an oppressed community where one might intuitively expect upheaval, does one instead find, or appear to find, quiescence?”32 And yet some citizens, far from quiescent, take action. Hirschman, in his seminal work on responses to declining organizational performance, suggests that citizens have, broadly, two available responses: exit or voice. Exit involves a disengagement from the state. Some might “vote with their feet,” moving to a place that better meets their needs, or might purchase private alternatives.33 We can imagine, for example, a community in which residents buy bottled water, run their own generators, visit private hospitals, and send their children to private schools. Traditionally, these have been the retreats of the wealthy, who purchase private services rather than contend with deteriorating public systems.34 In recent years, though, non-state service providers – ranging from kinship networks, to nongovernmental organizations, to charities, and private companies – have expanded in poorer communities as well, thus increasing opportunities for a partial exit from the state by substituting private for public services. For others, exit might involve rejection of – or even outright resistance to – the state.35 In most places, though, the state – even where weak, corrupt, or capricious – remains an important provider of services, particularly for the poor. This is true despite decades of economic liberalization that sought to reduce the size and

31 32

33 34 35

Banfield 1967. Also see Putnam’s (1993) portrayal of southern Italy. Gaventa 1980, p. 3. Also see Bennett and Bennett (1986) on apathy and political indifference, and Ferguson (1990) on depoliticization. Tiebout 1956; Oates 2006. See, for example, Baviskar (2003) on Delhi, and Coy and Pöhler (2002) on Latin American cities. Such resistance is manifest in “everyday” forms of noncompliance – for example, squatting on public land, dodging fares, or siphoning off public water or electricity (Scott 1985; Chatterjee 2004; Holston 2008). At the most extreme, it can take the form of full-fledged uprisings motivated, in part, by low levels of service provision (Berman et al. 2011; Cammett 2014).

Claiming Services, Claiming the State

11

role of the public sector.36 Under these conditions, marked by economic and social dislocation, the state – even as it is constrained – is all the more critical as a source of social protection.37 Indeed, as Sandbrook et al. note, “The state remains the only entity with the legitimacy and capacity to capture and redirect the wealth that society produces, despite the immense challenges that globalized capital flows pose to its autonomy and robustness.”38 Understanding when and how citizens engage rather than exit from or rebel against the state is therefore of critical consequence. Citizen voice takes many forms. In their groundbreaking work on participation in the United States, Verba et al. drew attention beyond the voting booth, highlighting a range of non-electoral forms of political activity (campaign work, financial contributions, joining a local association, acts of organized protest, and contacting officials) that vary in the time, money, and civic skills they require.39 More recently, scholars have taken up similar studies in less developed contexts, asking not just whether but also how citizens engage the state. Collier and Handlin, for example, drawing on research from urban Latin America, distinguish between the interest and electoral arenas – defining the interest arena as “the considerably more informal locus of specific interest articulation and problem solving.”40 Within this interest arena, citizens engage in diverse practices, from direct appeals to officials, to brokerage through third parties, to collective claim-making through associations, and contentious collective action.41 In the Indian setting, a small but growing number of studies have similarly attempted to map citizen participation, variously highlighting the role of political parties, local associations, fixers, and brokers, as well as forms of direct citizen interface with local officials.42 Collectively, these works highlight the diversity of citizens’ approaches to the state. In so doing, they cast up a new set of puzzles that prompt us to ask, in addition to whether and how citizens make claims on the state, why their approaches vary.

36

37 38 39 41

42

On the paradoxical effects of state retrenchment, see MacLean (2011b) who employs data from sub-Saharan Africa to demonstrate that the rural poor are in fact more likely to rely on public services than more affluent urban residents who can purchase private alternatives. Rudra 2003; Sandbrook, Edelman, Teichman, & Heller 2007; Jayal 2013. Sandbrook, Edelman, Heller, & Teichman 2007, p. 253. 40 Verba, Schlozman, & Brady 1995. Collier & Handlin, 2009, p. 8. See, for example, Dunning (2009) and Houtzager and Acharya (2011), both of whom document variation in citizens’ problem-solving repertoires across Latin American cities. MacLean (2010, 2011a), writing of the sub-Saharan context, similarly distinguishes between electoral and nonelectoral participation, including contacting political leaders, attending community meetings, collectively raising issues, and joining demonstrations or protests. Krishna (2002, 2011) provides one of the few rigorous accounts of nonelectoral participation in rural India, highlighting the role of brokers. A somewhat larger body of work explores the urban terrain – including Harriss 2005; Jha, Rao, & Woolcock 2007; Bertorelli et al. 2014, 2017; Heller et al. 2015. Bussell (2015) offers one of the few cross-sectional studies, employing rural and urban data from across 20 states.

12

Introduction: Citizenship and Social Welfare

This book, accordingly, takes up two sets of questions related, first, to the incidence of claim-making and, second, to the repertoire of practices employed. Why, under the same formal democratic frameworks, do some make claims on the state in pursuit of social welfare while others do not? And among those who do make claims, why do different people pursue different pathways to the state? India – a consolidated democracy that is also home to some of the world’s most persistent poverty – is a particularly intriguing context in which to examine these questions.

citizen and state in india Despite decades of rapid economic growth, India lags behind in social development, ranking 130th out of 188 countries on the United Nations Human Development Index (which jointly assesses per capita income, education, and life expectancy).43 Twenty-two percent of Indians live under the national poverty line of roughly $1.90 per day, adjusted for purchasing power parity.44 The national poverty line, however, has been criticized for being too low to meet basic human needs. Other, international estimates set a higher bar and, accordingly, find a higher incidence of poverty: the World Bank, for example, reports that 58 percent live on $3.10 per day or less, while the UNDP classifies 54 percent of Indians as poor according to a multidimensional poverty index.45 Roughly a quarter of the adult population cannot read, while over 40 percent of Indian children younger than five are underweight – one of the highest rates of child malnourishment in the world.46 There is, in other words, a yawning gap between procedural democracy – marked by regular elections and commitments to civil and political rights – and its substantive outcomes, assessed in terms of human well-being. This deprivation persists, paradoxically, alongside extensive initiatives to combat poverty, including some of the world’s largest public welfare programs. While the scope of India’s welfare commitments has increased steadily since independence, social spending has accelerated most dramatically over the last twenty years – beginning in the 1990s with a wave of social rights legislation. This legislation prompted a proliferation of centrally sponsored and (subnational) state welfare schemes, among them the Mahatma Gandhi National Rural

43 44

45

46

UNDP 2015. This 2011 national poverty rate is calculated according to the Government of India’s “Tendulkar methodology,” establishing poverty lines for rural and urban areas in each state based on monthly per capita consumption expenditures for food and nonfood items. The national poverty line is roughly equivalent to the World Bank’s international poverty line of $1.90 (purchasing power parity), under which 21.3 percent of Indians were classified as poor in 2011. OPHI 2016. The index combines income alongside indicators of health, education, and standard of living. IFPRI 2013.

Citizen and State in India

13

Employment Guarantee Scheme (MGNREGS), the National Rural Health Mission (NRHM), and the “Education for All” Campaign, to name but a few. Unprecedented amounts of goods, services, and money are thus being funneled into the social welfare sector.47 And yet these resources are rarely allocated in a rule-bound or regularized fashion; rather, they are subject to the discretionary – and, at times, arbitrary – control of local officials, as well as to leakage and graft at all levels of the administrative hierarchy.48 Access to the state and its resources is thus highly uneven and, for many, elusive – even as the state grows increasingly central to the lives and livelihoods of its citizens. How do Indian citizens navigate these gaps? India is regularly celebrated as the world’s largest democracy, noted for its more than seven decades of mostly free elections, competitive contests in most states, and high rates of voter turnout – particularly among the poor and the lower castes.49 And yet voting is widely acknowledged as a “blunt” form of voice that conveys relatively little about a person’s needs and preferences.50 Protest, in contrast, is an information-rich activity focused on particular demands. Social rights campaigns in India have effectively employed contentious strategies to shape policy at the state and national levels.51 Most ordinary citizens, though, do not engage regularly in social movement mobilization. Just 17 percent of those surveyed for this book in Rajasthan, for example, reported that they had participated in a rally, strike, or other kind of protest activity, and just 11 percent were aware of an active social movement organization in their vicinity. We are left, then, with a blank space in our understanding of citizen–state relations in India, between and beyond elections and movements.

The View from Rajasthan What do citizens do on a quotidian basis when seeking the services and social protection of the state? This was the question that brought me to Rajasthan. However, when I started my research, the director of a rural development NGO

47 48 49 50

51

These trends are discussed in greater depth in Chapter 3. Chandra 2004; Gupta 2012; Sukhtankar & Vaishnav 2015. Mitra & Singh 1999; Yadav 2000; Alam 2004; Kumar 2009; Ahuja & Chhibber 2012. Riker (1982, p. xviii), for example, states that “outcomes of voting are, or may be, inaccurate or meaningless amalgamations [such that] what the people want cannot be known.” Also see Verba et al. (1995) and Collier and Handlin (2009). Landmark legislation such as the Right to Information Act, the National Rural Employment Guarantee Act, the Right to Education Act, and the National Food Security Act emerged in large part because of civil society mobilization and advocacy, including public interest litigation. These campaigns have been led by lawyers, public intellectuals, and social activists with support from high-ranking officials in the Indian Administrative Service. While these campaigns have effectively employed protest strategies including marches and road blocks, they are largely lacking in mass mobilization (Jenkins & Goetz 1999; Jayal 2013; Hertel 2015).

14

Introduction: Citizenship and Social Welfare

told me that I was asking the wrong question. The question, she said, is not how do citizens make claims on the state, but why they do not. Rural Rajasthan, in her view, was best described by a lack of citizen–state engagement. She was not alone in this opinion. Indeed, the Indian state is regularly portrayed as a distant entity beyond the reach of the aam aadmi – the common citizen.52 The state, in this view, has failed its citizens so often and so deeply that many simply stop trying to gain access to services. Rajasthan, which prior to independence was ruled as a series of socially regressive “princely” states, is often characterized in these terms. Krishna, for example, notes the prevalence of “despondent democrats,” particularly in rural Rajasthan, who, despite voting in large numbers, “expect that their voices will simply go unheard.”53 My initial fieldwork, though, began to complicate this view. I began my research in the district of Udaipur. In this area, tribal residents (primarily from the Bheel tribe, a historically disadvantaged community) tend to live in hamlets that are located at a distance from the main village. I visited a number of these hamlets, hoping to learn something about how some of Rajasthan’s poorest citizens navigate access to the state. I arrived during the dry summer months, at a time when pine ka pani (drinking water) is often the first thought on people’s minds. In one hamlet,54 where lack of access to drinking water was an acute problem, a group of men offered a rather bleak view. “We have no water, no services,” they told me. “What can we do? God willing, we survive.” When I asked if they had complained or sought assistance from the local government, one man stated: Why waste your breath? Sarkar [the state] does nothing for us. They come at elections, they eat the votes, then they go away again and we are forgotten.

These, I thought, were most certainly “despondent democrats.” And yet, just a few kilometers away in a Bheel hamlet of another village,55 I encountered a very different situation. Here water flowed from the government hand pump, and women were clustered around it holding buckets. Until recently, I was told, this pump was also running dry, and women had to walk long distances to another village to collect water. Some weeks earlier, however, a government technician had come to install a small motor that allowed the water to be drawn from a deeper level. “We made him come,” one of the men explained: When the water is not flowing, it is the responsibility of the sarkar. They bring the water, from god knows where, but they bring the water. 52 53 54

55

Yadav 1999; Fuller & Benei 2001; Corbridge et al. 2005; Krishna 2011. Krishna 2011, p. 102. Quotes from this hamlet are drawn from author interviews, Gogunda block, Udaipur district, Rajasthan, February 22, 2010. Quotes from this hamlet are drawn from author Interviews, Gogunda block, Udaipur district, Rajasthan, February 24, 2010.

Citizen and State in India

15

When I asked how they had managed to get the technician to come, I was told that residents had signed a petition, which they had presented to the local government council (the Gram Panchayat). They had done this over and over again, “countless times,” and when a district official visited the village, they took him to the hand pump and kept him there until he promised to help. They had also spoken to staff from a local NGO, in the hopes that they might put additional pressure on the officials. As one man recounted, it was a matter of “avaaz” (voice): “We raised our voices until they had to hear.” In the first hamlet, residents appeared to have given up hope of state responsiveness; they had, by their own admission, had done little to hold the sarkar to account. In the second hamlet, in contrast, the sarkar was the target of direct demands, as residents doggedly pursued local officials to call for repairs to their water source. As I traveled to other villages across the state, I came to realize that the active residents of the second hamlet were not the exception but the norm. In fact, as we will see in detail in the chapters that follow, the vast majority of those that I met (and a full 76 percent of those surveyed) were engaged in claim-making activity of some kind. More than two-thirds also reported that the claim-making practices they pursued were at least somewhat effective in solving problems. To be clear, citizen–state relations remained fraught; most were frustrated in their efforts to gain access to services, and many expressed high levels of dissatisfaction. But most, it appeared, had not given up hope (even if faint) that their voices might be heard. Instead they both expected and pursued services from the state. Most strikingly, these patterns persisted across socioeconomic groups: among the relatively rich and the poor, and the so-called upper and lower (Scheduled) castes and tribes.56 However, citizens engaged the state in different ways, both direct (involving face-to-face contact with local officials, bureaucrats, and politicians) and mediated (working through non-state actors and institutions). These various approaches to the state were not mutually exclusive but were combined into diverse repertoires, ranging from those who relied on a single approach (just over 20 percent) to those who pursued multiple pathways to the state (54 percent) (Figure 1.1). Those in despair in the first hamlet, though, were not simply outliers. Almost a quarter of those surveyed reported that they had never contacted state officials (directly or indirectly) for assistance in securing welfare services of any kind. This figure is sobering in a setting where citizens rely heavily on the state for income support, wage employment, food rations, health care, and other critical services. It is highly unlikely, given the inadequacies of service delivery in this region, that these residents simply did not need anything from 56

India’s Scheduled Castes (SCs) and Scheduled Tribes (STs) are communities that, due to long histories of social and economic marginalization, have received official designation in the Indian Constitution requiring that legal provisions are made for their social, educational, economic, and political advancement. See the note on caste terminology following the preface.

Introduction: Citizenship and Social Welfare

0

5

Percent 10 15

20

25

16

0

2

4

6

8

Claim-Making Repertoire (number of practices)

figure 1.1 Distribution of the Claim-Making Repertoire in Rajasthan Source: Citizen Survey (n = 2,210), Rajasthan, 2010–2011.

the state. Indeed, non-claim-makers include both richer residents who presumably could afford to self-provide and poorer residents who disengage from the state – even in the face of striking levels of deprivation.

explaining active citizenship My challenge thus became to explain these varied approaches to and retreats from the state. To begin, I turned for guidance to a rich literature that seeks to explain variation in citizens’ political behavior. Within this, distinct bodies of work help to illuminate different conditions and prerequisites for participation. And yet, when applied to the Rajasthani context, I found that most existing explanations cast up more questions than answers. Cross-national studies explore the role of formal political institutions, examining how differences in electoral laws, executive power, or party competition shape citizen–state linkages.57 Historical institutional accounts apply a similar cross-national lens, examining the legacies of colonial rule, state formation, and the timing and sequencing of democratization on contemporary patterns of representation and participation.58 However, this scholarship, while effective in highlighting the importance of political and historical institutions, cannot account for variation in places with shared histories and political structures. A study of varied citizenship practice within a single state and, for that matter, within the same districts, administrative blocks, and even villages, thus begged exploration of more local-level factors. 57 58

For a review of institutional theories of citizen–state linkages, see Kitschelt (2000). See, for example, Acemoglu and Robinson (2006). For a review of historical institutionalism in comparative politics, see Thelen (1999).

Explaining Active Citizenship

17

At the subnational and local levels, many of the most common explanations for divergent patterns of participation are resource based, suggesting that differences in citizen–state engagement reflect underlying levels of local economic development. The expectation, in basic terms, is that richer, more developed places will have a more engaged, active citizenry.59 A related body of work shifts the analysis to the individual, predicting limited political participation among the poor.60 Some argue that this is a matter of resources (time, money, skills, education) that are necessary for participation, but which are restricted among the poor.61 Others see it as reflective of broader power dynamics that determine whether people can participate: that is, the political spaces that are open and the rules (formal or informal) that determine who has access to those spaces. The poor, in this view, are structurally constrained in ways that inhibit political action.62 The result, as Appadurai succinctly suggests, is that the “capacity to aspire” is not evenly distributed; imagined possibilities of access and action are more expansive for some while restricted for others. For Appadurai, this is a matter of economic class; horizons of aspiration are broader for those who are “better off” and more “brittle” for the poor.63 And yet class inequality, like the preceding socioeconomic factors, is inadequate as an explanatory framework. While powerful in explaining aggregate patterns of economic and political exclusion, a class framing cannot account for variation among similarly situated (and similarly poor) groups; nor can it explain the visibly high aspirations of poor and marginalized citizens. Indeed, the data from Rajasthan (introduced in detail in Part II of this book) reveal a striking lack of difference in claim-making practice across villages set apart by different levels of wealth and literacy rates. There is also little significant difference in the likelihood of claim-making attributable to individual landownership or caste – although these features do play a role in constraining the breadth of practices employed. There is, moreover, substantial variation among similarly placed groups of individuals who share features of class and caste. Other scholars have highlighted a gender gap in participation. Early studies of political behavior noted that men were more active than women across a

59

60

61 62

63

Lipset 1959, 1960; Almond & Verba 1963; Przeworski & Limongi 1997; Przeworski, Cheibub, Alvarez, & Limongi 2000. Lerner 1958; Milbrath & Goel 1977; Verba, Nie, & Kim 1978; Rosenstone & Hansen 1993; Verba, Schlozman, & Henry Brady 1995; Dunning 2009. Brady, Scholzman, & Verba 1995; Verba et al. 1995. See, for example, Mills (1956), who argues that economic position allows the “power elite” in society to dominate the political system. Bachrach and Baratz (1963, 1970), building on this view, emphasize class-based exclusion in decision-making institutions. Also see Lukes (1974) and Gaventa (1980) on ideological barriers to participation among the poor. Appadurai 2004, p. 69.

18

Introduction: Citizenship and Social Welfare

wide spectrum of political activities.64 Over time, these trends have started to shift – particularly in the electoral arena where women in many parts of the world now vote in numbers as high as or higher than men.65 In India, though, women continue to fall behind men across a wide array of associational, civic, and political activities.66 These dynamics are most pronounced in northern India, where many women live under variants of the purdah (veil) system, under which norms of gender seclusion constrain their participation in the public sphere. Rajasthani women, in particular, have some of the lowest levels of education in India, as well as some of the lowest rates of labor force participation. Not surprisingly, my survey data show that women in Rajasthan are significantly less likely than men to engage in claim-making. And yet, as we will see, a majority of women (a full 60 percent) still do make state-targeted claims – even under these restrictive conditions. The preceding theories all seek to explain divergent patterns of participation between elite and marginalized groups. And yet, as I repeatedly found, knowing where someone stood in the local social structure – that is, where one lives, if one is rich or poor, belongs to a high or low caste, is a man or a woman – could not adequately predict whether that person would engage the state. This is not to say that such features do not matter. They do – particularly, as we shall see, when it comes to explaining how a person makes claims on the state. But as predictors of the incidence of claim-making – that is, of how likely an individual is to engage the state in pursuit of services – they explain surprisingly little. Moreover, as the accounts of the Bheel men in the preceding section remind us, practices vary among those with similar social and economic status: equally poor men from the same tribe took very different stances toward the state. My aim, then, became to develop a theory that would explain these divergent patterns of citizenship practice both across and among similarly situated individuals.

Argument in Brief Early on in my research in rural Rajasthan, I noted that residents often referenced goods and services beyond their own neighborhoods when discussing their expectations of the state. They would note, for example, that 64 65

66

Verba, Nie, & Kim 1978; Barnes & Kaase 1979; Burns, Schlozman, & Verba 2001. In the United States, for example, women have voted in greater numbers than men in every presidential election since 1980. These trends have held across a wide range of countries (Norris 2002). Norris (2002, p. 98) notes, “the [voting] gender gap has closed or even reversed in all societies except India, where women continue to turnout at markedly lower rates than men” (emphasis added). This gap, though, may be beginning to close in India as well (Yadav 2000; Chhibber 2002) – particularly following high rates of women’s turnout in the 2014 elections (Vaishnav 2015).

Explaining Active Citizenship

19

another area had new paved road, or that there was better water supply in another village. As one man in Udaipur put it: You would have to be blind not to see! There are so many [public] schemes. Over there [gesturing to another neighborhood] they have CC [paved] roads, and the panchayat [village council] brought water connections to every house. Where is our road? Where is our water?67

Seeing or hearing about service delivery and resources in other places, it seemed, generated a sense of grievance – and, by extension, sense of entitlement – among those who were not well served. People also appeared to learn about successful and unsuccessful claim-making practices through the accounts of others. Those individuals who moved around the most – traversing boundaries of caste or tribe, village or neighborhood – seemed to have greater expectations of what the state should and could provide compared to those who moved in more circumscribed spheres. They also appeared to have more and better information about public services and about how to claim them, as well as greater opportunities to do so. They had, in sum, greater aspirations toward the state, as well as greater capabilities for state-targeted action. Why, though, would citizens’ aspirations and capabilities vary? The answer to this puzzle, I argue, lies in an examination of citizens’ social and spatial exposure – that is, in how far beyond the immediate community and locality one’s networks extend. Varied degrees of exposure reflect the unevenness of social and spatial boundaries, which vary in their rigidity or porousness. Where rigid, these boundaries limit the people, places, information, and ideas to which a person is exposed and, in so doing, restrict the claim-making repertoire. More porous boundaries, in contrast, increase citizens’ exposure to information and narratives about the state, as well as to potential linkages to public officials and agencies. This increases both the likelihood and diversity of claim-making by expanding the horizon of possibility – both for what the state might provide and for the actions citizens might undertake. My central argument (developed in full in the next chapter) is that citizens who traverse local social and spatial boundaries will, all else equal, be more likely to make claims on the state, and will do so through a broader repertoire of practices, than those who remain more constrained. This prediction, however, is dependent on the underlying polices and presence of the state, which sets the stage for claimmaking. Social and spatial exposure can be catalytic – but only where the state is capable of delivering a broad array of services but uneven in actually doing so. In other words, there must be resources to be had but difficulty in obtaining them. There is little incentive to seek services if there is no hope that the they will be delivered, and little need to do so where they are readily available. The following chapters develop this theory, and test it in the rural Rajasthani setting.

67

Author interview, Bargaon block, Udaipur district, December 15, 2010.

20

Introduction: Citizenship and Social Welfare

research setting Rajasthan is India’s seventh most populous state, home at last count to almost 69 million people (Figure 1.2). Formerly twenty-two princely states grouped together within the province of Rajputana, the state of Rajasthan only emerged as a single administrative territory in 1949, two years after Indian independence.68 Prior to this, each princely state, while officially under British rule, retained considerable autonomy. The princely states were themselves further subdivided into smaller fiefdoms (jagirs) ruled by local feudal lords (jagirdars).69 The result was a patchwork of states and fiefdoms with deeply feudal traditions and allegiances, reinforced by economic dependencies – often in the form of bonded labor. Modern-day Rajasthan is shaped by the legacies princely and jagirdari rule.70 It remains in many ways a divided state, with districts that reflect historical princely boundaries and which are additionally set apart by linguistic and cultural differences. Rajasthan is home to at least nine distinct linguistic communities, and close to 400 castes and tribes that, under jagirdari rule, were grouped by occupation in service to the local rulers.71 Social structures from this era persist, albeit in modified forms. For example, in interviews, it was not uncommon to hear people refer to local “big men” (mota aadmi) who are often the descendants of the jagirdars and who remain the local landed elite – some of whom still reside in “palaces” (havelis). In the words of a staff member in one of the state’s largest development NGOs, “Rajasthan in many ways is still a feudal place. Democracy may have come, but … in the villages, the king today is still the local landowner, the local big man.”72 In her work on subnational social development, Prerna Singh argues that Rajasthan’s social fragmentation contributed to low levels of public goods provision and poor human development outcomes in the early years following independence.73 “The absence of any societal preferences in favor 68

69

70

71 72

Princely states, also sometimes referred to as “native states” were nominally sovereign bodies that were not directly governed or administered by the British but rather were indirectly ruled by local kings who, in turn, were allied to the British Crown. Narain & Mathur 1989; Schomer, Erdman, Oldrick, & Rudolph (Eds.) 1994; P. Singh 2015. Jagirs were land grants, given by the princely rulers to local elites. The British recognized these land grants, ruling indirectly in these areas through the established princely rulers and, below them, local landlords. While jagir land was technically granted for life, in practice, land holdings were hereditary, and particular families became ensconced in power over time. Jagirdars exercised almost absolute power in local affairs and had the ability to collect taxes and maintain their own standing armies. The jagirdar system was abolished in 1951 by the newly independent government of India. On the political legacies of princely rule in India, see Wood (1984). On feudal political legacies, see Harriss (2011). On the economic legacies of feudalism and colonialism, see Banerjee and Iyer (2005). On linguistic and cultural divisions in Rajasthan see Lodha (n.d., b). 73 Author interview, Udaipur, Rajasthan. January 24, 2011. P. Singh 2015.

Research Setting

21

Rajasthan

figure 1.2 Rajasthan, Location in India Source: Geospatial data from Harvard University.

of collective welfare,” she writes, provided “carte blanche for the princely states and jagirdars to virtually ignore the social sector.”74 Legacies of this neglect persist today, visible in the state’s continued low human development ranking (falling in the bottom quarter of all Indian states).75 Rajasthan, though, is not India’s poorest state. In fact, since 2005, the state has enjoyed some of the fastest income poverty reduction in the country, with rates 74 75

Singh 2008, p. 19. UNDP 2011. Rajasthan is ranked 14th out of 19 major states included in this report.

22

Introduction: Citizenship and Social Welfare

falling from 34 percent in 2004 to 15 percent in 2011.76 These poverty-line estimates, though, fail to capture the intensity of deprivation in health, educational attainment, and other indicators of human well-being. This deprivation is deepest in rural Rajasthan, which is home to 75 percent of the state’s population and 80 percent of its poor.77 At 51 deaths per 1,000 live births, Rajasthan’s rural infant mortality rate is among the highest in India, far exceeding national average of 40 per 1,000.78 Literacy rates in rural Rajasthan hover around 60 percent – well below the national rate of 74 percent.79 Just 40 percent of rural households in Rajasthan have a source of drinking water in the home, compared to 56 percent nationally; and a staggering 74 percent of rural Rajasthanis practice open defecation, compared to 44 percent across India.80 Rural Rajasthan, as seen in these indicators, is the site of deep material and human deprivation, assessed in both absolute and relative terms. And yet, as we will see in Chapter 3, social spending on education, health, and other services in Rajasthan has, in fact, increased, accelerating in the 1990s to outpace other similarly poor states.81 This upward trajectory in welfare provision, coupled with its persistent poverty, makes Rajasthan an ideal site for the study of citizen claim-making. The expansion of services and programs means, in crude terms, that there is a bigger pie to divide. The allocation of resources, however, remains uneven. Poverty is concentrated in Rajasthan’s southern districts (Figure 1.3), and among the Scheduled Castes and Tribes (19 and 40 percent, respectively, of whom live below the poverty line – compared to just four percent among the General Castes).82 Perhaps most striking is the gender gap within the state. While the literacy rate for men is over 80 percent, just 44 percent women are functionally literate.83 Rajasthan also has one of the worst child sex ratios and among the highest maternal mortality rates in the country.84 76

77

78

79 81

82

83 84

World Bank 2016, citing 2011 figures using the Government of India’s Tendulkar poverty lines (the methodology for which is discussed in note 44). These state poverty calculations, though, fall substantially below other estimates. For example, a calculation of multidimensional poverty finds a 63 percent incidence of poverty in Rajasthan (OPHI 2010, citing 2006 data). Government of Rajasthan, Directorate of Economics & Statistics, 2011–2012. Poverty rates are calculated using the Tendulkar methodology. World Bank 2016. Rajasthan’s statewide (rural and urban) infant mortality rate (at 47 deaths per 1,000 live births) is also one of the highest in India. 80 Census of India 2011. The statewide literacy rate is 66 percent. World Bank 2016. For a comparative review of social welfare expenditures and performance across Indian states, see P. Singh (2015). These trends are discussed in greater detail in Chapter 3. World Bank 2016, citing 2012 census data. Just 42 percent of the ST and 52 percent of SC are literate – compared to 79 percent among the General Castes. Census of India 2011. World Bank 2016. At 244 maternal deaths per 100,000 live births, Rajasthan significantly exceeds that all-India average of 167 deaths.

Research Design

23

Jodhpur Ajmer

Kota Poverty Rates (%) 18.3–24.9 16.3–18.3 14.4–16.3 12.7–14.4 6.8–12.7

Udaipur

figure 1.3 Poverty Rates in Selected Districts of Rajasthan Sources: 2011 Poverty rates from Bhandari and Chakraborty 2015, calculated using the Tendulkar methodology. Geospatial data from Harvard University.

research design To capture these disparities and the range of conditions within Rajasthan, I focused my research in four districts – Udaipur, Kota, Ajmer, and Jodhpur.85 These districts were purposively selected to reflect differences in population, caste and tribal composition, poverty rates, and human development status (Tables 1.1 and 1.2).86 The districts also carry the historical legacies of different colonial experiences. Udaipur, Kota, and Jodhpur were each part of 85

86

Indian states are administratively divided into districts. Districts are further divided into blocks and, within them, revenue villages – which are the most local geographical units enumerated by the Indian census. The villages are nested within Gram Panchayats, which are elected local governance bodies typically encompassing several villages. For more on these divisions and sample selection, see Chapter 4. Case selection was based on data accessed in 2009, reflecting the 2001 census and a 2008 statelevel Human Development Report. For comparison, Table 1.2 presents from both the 2001 and 2011 census.

Introduction: Citizenship and Social Welfare

24

table 1.1 Selected Districts: Demographics District

Population 2001

Ajmer Kota Jodhpur Udaipur Rajasthan India

Scheduled Castes (%) Scheduled Tribes (%)

2011

2001

2011

2001

2011

million million million million

2.6 million 2 million 3.7 million 3.1 million

18.0 19.0 16.0 6.0

18.5 20.8 16.5 6.1

2.0 10.0 3.0 48.0

2.5 9.4 3.2 49.7

56.5 million 1 billion

68.5 million 1.2 billion

17.2 16.2

17.8 16.6

12.6 8.2

13.5 8.6

2.2 1.6 2.9 2.6

Sources: Rajasthan Human Development Report (UNDP 2002), Rajasthan Human Development Report Update (IDS 2008); Census of India (2001); Census of India (2011). District level census data reported by the Government of Rajasthan’s Directorate of Economics & Statistics.

table 1.2 Selected Districts: Economic and Social Indicators District

Poverty Rate (%)

HDI Rank in Rajasthan

Literacy Rate (%)

2011–2012 (national)

1999

2008

2001

2011

Ajmer Kota Jodhpur Udaipur

11.9 12.7 13.9 19.6

10 3 13 27

10 2 9 20

64.7 73.5 56.7 58.6

69.3 76.6 65.9 61.8

Rajasthan India

14.7 21.9

14 (in India) 111 (globally)

17 (in India) 113 (globally)

60.4 64.9

66.1 73.0

Sources: Rajasthan Human Development Report (UNDP 2002), Rajasthan Human Development Report Update (IDS 2008); Census of India (2001); Census of India (2011). District level census data reported by the Government of Rajasthan’s Directorate of Economics & Statistics. District level poverty rates generated by Indicus Analytics, reported by Bhandari and Chakraborty (2015). Notes: Poverty rate is the percentage of the population under the national poverty line, using the Tendulkar methodology. HDI rank in Rajasthan reflects the district’s score (1–32) on a statewide Human Development Index.

a distinct princely state, while Ajmer was the regional center of the colonial administration and was directly ruled by the British. Between 2009 and 2011, I spent over eighteen months in villages of Rajasthan, pursing iterative stages of qualitative and quantitative research.87 First, for the purposes of theory building, I carried out roughly eighty open-ended interviews in two districts: one in-sample (Udaipur) and one out-of-sample

87

The stages of research and methodology are discussed in detail in Appendix I.

Research Design

25

(Rajsamand, which lies between Udaipur and Ajmer). These open-ended interviews with local officials and village residents cast up a series of puzzles and suggested the outlines of a theory. Building on these initial questions and insights, I then designed, piloted, and ultimately supervised the administration of a citizen survey to 2,210 randomly selected individuals residing in 105 villages nested within the four districts.88 To ensure that the theory-building sample was distinct from that used for theory testing, any village visited during these initial stages was excluded from subsequent stages of both qualitative and quantitative data collection. The survey collected information on citizens’ experiences with local governance institutions, their engagement with a range of non-state actors and institutions, and a wide range of other social and economic activities. From these data, I constructed a series of metrics of claim-making, including: its incidence (whether or not a person reported contacting public officials in an effort to gain access to social welfare goods and services); the practices employed (whether involving direct, face-to-face engagement with public officials or whether mediated through non-state actors and institutions); and the repertoire (or range) of practices employed. These data enable statistical testing of the core propositions of the book: namely, that variation in both the likelihood of claim-making and in the breadth of strategies employed can be significantly explained by differences in individuals’ social and spatial exposure beyond their immediate communities and localities. For insight into the mechanisms at play, I turn to qualitative interviews, carried out in two stages: first, concurrent with the survey, in a subsample of villages across all four districts, and second (a year later), through case studies in six purposively selected villages of two districts (Udaipur and Kota) representing the lower and upper ends of human development standing in the sample. All in all, I personally conducted 250 semistructured interviews at the village level, and oversaw another 230 interviews carried out by a small team of research assistants.89 The study’s single-state research design enables nuanced comparison of localities while holding constant macro-level features that could otherwise obfuscate important sources of local variation. This approach allowed me to hone in on factors influencing claim-making practice at the level of the village – a critical site of politics where citizens live, vote, and consume and utilize public resources. It allowed me, moreover, to investigate the micro-determinants of citizen action at the level of the individual, asking why people living in the same institutional settings navigate access to the state through different means. 88

89

Sampling methodology for the citizen survey is described in greater detail in Chapter 4 and in Appendix I. Sample summary statistics are presented in Appendix I (Tables A1.1 and A1.2). I carried out interviews in a mixture of Hindi and English, working with an interpreter when necessary – particularly where local Rajasthani dialects were prevalent. The translations that appear in this book are drawn from my field notes that, wherever possible, triangulated between multiple interpreters. See the “Note on Translation” in Appendix I.

26

Introduction: Citizenship and Social Welfare

However, the same single-state approach inhibits systematic testing of the theory under different macro conditions. Lacking cross-sectional data beyond Rajasthan, I cannot know if citizens living in other parts of India – those, for example, with a thicker and more regular or thinner and more erratic state presence – are more or less likely to engage in claim-making compared to those in Rajasthan. Nor can I know if residents of different states respond to social and spatial exposure in systematically different ways. However, historical analysis within Rajasthan (presented in Part II of the book) enables me to trace how changes in the state’s reach and visibility have altered the landscape for citizen–state engagement and claim-making over time.

organization of the book The book proceeds in three parts. The next chapter introduces a general theory of active citizenship. Stepping back from the specifics of the Rajasthani case, I build on insights from social network theory and literature on policy feedback loops to argue that citizens’ knowledge of, ideas about, and linkages to the state are both state induced and socially produced. I consider the conditions – namely need, knowledge, and a sense of entitlement and of personal and political efficacy, as well as pathways of access to the state – that both motivate and enable claim-making. These conditions, I argue, are produced through a combination of bottom-up social processes and top-down action by the state. I then posit that citizens who traverse social and spatial boundaries gain exposure to information, ideas, and contacts that build their knowledge of and expectations for service delivery, thus forging linkages to the state in ways that are likely to spur action. Finally, I discuss the limits (or scope conditions) of this theory, illustrating how the effects of exposure are dependent on the actions and inaction of the state itself. Part II of the book unpacks the theory in the Rajasthani setting. Chapter 3 examines the institutional terrain upon which claim-making unfolds, proceeding in two parts. It first documents the expansion of social welfare spending and programming, both nationally and in Rajasthan in particular. It then tells the story of Rajasthan’s political transformation over time and, in so doing, describes the shifting ground upon which citizens encounter the state. Through policies of decentralization and an influx of funds to the local level, the state is becoming more institutionalized while also penetrating deeper into rural lives. This process, though, is highly uneven, marked by persistent patterns of inequality in citizens’ access to services. This, I argue, sets the stage for high but varied claim-making practice, which reflects the relative breadth but also unevenness of the state’s presence in citizens’ daily lives. Chapter 4 develops a bottom-up view of the Rajasthani state as it is seen and approached by citizens. Turning to the citizen survey data, I examine patterns of claim-making incidence, practice, and repertoire. The data highlight puzzles at both the macro and micro levels. First, considering the

Organization of the Book

27

high aggregate levels of claim-making that persist across localities and social groups, I ask why citizen action is prevalent in places (poor and remote) and among people (the lower classes and castes) where many strands of democratic theory would least predict it. Second, I ask what accounts for different courses of action at the individual level. Why, among those living in the same local communities, do some make claims on the state while others do not? What, moreover, accounts for differences in the strategies pursued? These are the puzzles that Chapters 5 and 6 together seek to explain. Chapter 5 introduces the concept of social and spatial exposure in the rural Indian and Rajasthani contexts. Drawing on literatures in anthropology and political economy, I describe the salience of caste, neighborhood, and village cleavages. I then discuss the permanence and permeability of these boundaries, examining a series of structural, political, and institutional changes that, together, have begun to enable greater connectivity (of places) and mobility (of people). The combined effect, I argue, is a slow – but pronounced – erosion of boundaries and an associated increase in citizens’ exposure beyond their immediate communities and localities. Drawing on the qualitative accounts of villagers, I illustrate how differences in social and spatial exposure influence citizens’ state-targeted aspirations and capabilities. Chapter 6 draws on the citizen survey data to empirically assess the relationship between exposure and claim-making. Controlling for a wide range of village- and individual-level features, I find that socializing across neighborhood and caste lines, working alongside people from other castes, and venturing beyond the village are all associated with both a greater likelihood and a greater diversity of claim-making practice. I then explore the direction of causality in these relationships. In the long run, I argue, social and spatial exposure is likely to be endogenous – and thus difficulty to disentangle – from processes of social and political development. In the short term, though, I show that citizens’ networks – and hence exposure – are shaped by factors that are largely antecedent to, if not fully independent from, individual acts of claimmaking. To further address concerns over endogeneity, I introduce a proxy variable – the land-to-labor ratio – by which to distinguish the short-term effects of exposure on claim-making. Calculated as a village’s cultivable land relative to its workforce, this variable reflects the likelihood that a person – driven by land scarcity – will traverse boundaries of village, caste, and occupation in search of new economic opportunities. As a product of geographic and demographic village features, the land-to-labor ratio is independent from individual acts of claim-making in the short term. This variable thus provides a lens through which to examine the independent effects of social and spatial exposure on state-targeted citizen action. I then consider a range of alternative and additional explanations, examining whether there are omitted – or hidden – variables that might confound the analysis by driving variation in both exposure and claim-making. Finally, I turn to qualitative accounts from village case studies to consider how exposure influences citizen action, examining the

28

Introduction: Citizenship and Social Welfare

mechanisms – the development of aspirations and enhancement of capabilities – through which greater exposure drives claim-making. Part III of the book examines the implications of the patterns of claimmaking observed in rural Rajasthan. Chapter 7 considers the consequences of claim-making, both material and political. The chapter begins by reflecting on what claim-making means in the Rajasthani context for access to public resources, drawing both on qualitative accounts and on survey data on citizens’ access to services and on their perception of claim-making efficacy. I then discuss claim-making as it relates to wider patterns of political participation, including citizens’ engagement in other electoral, deliberative, and contentious arenas. Claim-making, I conclude, is of material importance as a potential means to social services, and of intrinsic value as a form of voice through which citizens negotiate their relationship to the state. The long-term political consequences of claim-making, though, remain unclear. Rajasthan, I suggest, stands at a critical juncture at which both virtuous and vicious cycles of citizen voice and state responsiveness (or lack thereof ) are possible. Chapter 8 discusses the potential extensions of the theory and argument beyond Rajasthan. Drawing on comparative accounts from elsewhere in rural and urban India, I suggest that exposure is most likely to spur citizen action under a particular set of conditions marked by a broad but uneven presence of the state in social welfare provision. Under a different set of conditions where the state’s presence is thinner or more uniform, exposure is unlikely to have the same catalytic effects. Building on these insights, I reflect, drawing on research from Brazil and South Africa as well as other studies of India, on the interplay between state and societal action in shaping both citizen mobilization and welfare provision. I conclude by reflecting on the conditions that help to build active citizenship and which, in the long term, might serve to aggregate, institutionalize, and sustain such action.

2 A Theory of Active Citizenship

In the last chapter, we met two groups of men from the Bheel tribal community, one of which actively engaged the state to demand drinking water during a time of drought, and the other of which seemed to have given up any hope of government service. We return to Rajasthan in the next section of the book, where I will show that these two groups represent broader patterns of variation in both whether and how citizens make claims on the state. First, though, this chapter develops a general theory of active citizenship. Moving beyond the particulars of the Rajasthani case, I ask: what prompts and broadens claim-making among some but constrains it among others? Active citizenship practice, I argue, is state induced and socially produced: it is learned and developed through both personal and narrated encounters with and observations of the state. As Corbridge et al. remind us, “We always see the state through the eyes of others, and with close regard for past memories, accounts that circulate in the public sphere, and how we see other people getting on or being treated.”1 Claim-making, in this sense, is a profoundly social as well as political affair, reflecting citizens’ exposure to the state as well as to other citizens from whom they learn about the state. This exposure is a product of the networks in which citizens are embedded and the boundaries (of class, ethnicity, gender, neighborhood, or village) that delimit where one goes, whom one meets, and what one observes. More porous boundaries enable broader exposure to people and places beyond one’s immediate community and locality. This social and spatial exposure, in turn, builds citizens’ aspirations toward the state (their wants and expectations) and develops their capabilities for state-targeted action (their ability to navigate and access the political system). Social and spatial

1

Corbridge et al. 2005, p. 8.

29

30

A Theory of Active Citizenship

exposure is thus a catalyst for citizen action: by building knowledge, expectations, and linkages to the state, it increases both the likelihood and the diversity of claim-making practice. More rigid boundaries, in contrast, narrow citizens’ exposure, limiting the breadth of the repertoire and making claimmaking a less likely outcome. The effects of exposure, though, are not uniform, but are conditioned by the broader institutional environment in which claim-making occurs: in particular, by the reach, visibility, and accessibility (or institutional “terrain”) of the state. The state, through its on-the-ground presence (or absence), sets the stage for citizen action, while differences in social and spatial exposure shape whether and how particular individuals engage the state. Where public performance is consistently poor and resources are markedly absent, increased exposure – far from motivating claim-making – could have a chilling effect, reinforcing low expectations. On the other hand, in settings marked by more readily and regularly available resources, claim-making may appear less urgent, leaving citizens less motivated to take action. Claim-making is thus most likely, I argue, where the state is neither absent or “failed,” but nor is it uniformly or readily accessible. In such settings, public officials demonstrate at least basic levels of capacity in the administration of public programs and provision of services, but they are uneven in actual performance. These are the conditions in Rajasthan, a state marked by rising levels of social spending but by irregular service delivery and inequitable distribution of public resources. Rajasthan is far from unique in this regard: circumstances are similar across much of India and in developing democracies around the globe, from Brazil to South Africa and beyond, where expansive public mandates for social welfare provision exist alongside entrenched patterns of social exclusion.2 This chapter proceeds in three parts. First, I consider the conditions necessary for claim-making: citizens’ interests in, knowledge of, beliefs about, and access to the state. Together these build aspirations – both for services and to action – as well as capabilities for state-targeted action. The second part of the chapter asks why these aspirations and capabilities vary, arguing that they are produced by a combination of social and state-led processes. Here I develop the concept of social and spatial exposure, which is delimited by boundaries of community and of locality. I then consider the effects of boundary porousness, which facilitates greater exposure that, in turn, motivates and enables statetargeted action by citizens. The third part of the chapter explores the limits of exposure and the costs of claim-making, underscoring a set of “scope conditions” – or parameters – that serve to limit, but at the same time clarify, the theory.

2

Houtzager and Acharya (2011, p. 2) describe these conditions as those where “electoral democracy is robust but longstanding political exclusion, authoritarian political and social institutions, and clientelist networks result in particularistic and differentiated treatment by state agents.”

Claim-Making Conditions

31

claim-making conditions In seeking to explain different courses of citizen action and inaction, I begin by asking: why would a person not make claims on the state, particularly given pressing material needs? Years before traveling to India, I grappled with this puzzle in a very different setting: New York City, where I worked as a legal advocate for a nonprofit that attempted to assist low-income and homeless residents in accessing welfare entitlements. Through this work, I encountered a wide range of people struggling, in different ways, to navigate the public benefits system. Some turned directly to welfare caseworkers or others in the local bureaucracy. Others approached intermediaries (like my own agency) for support. Still others did nothing to claim entitlements. More than a decade later and more than 7,000 miles away, I faced a strikingly similar set of dynamics in rural Rajasthan.3 Consider, then, a mother in New York City struggling to access food stamps and a rural Indian woman living in a neighborhood without water supply. What might deter these women from pursuing such critical resources? While worlds apart, there is a set of common conditions that might inhibit citizen action. First, the women simply might not know about government programs and services that could meet their needs. Even if they do, they may not feel entitled to them; that is, they may not think it is the government’s responsibility to provide the services in question, or they may think they are not the intended beneficiaries. Even if they feel entitled, they may think claim-making is not worthwhile, particularly if they have little expectation that their voices will be heard, or if they perceive costs to their actions that outweigh the benefits. And even if they have some hope that their claims will, in fact, be heard, they may not be able to take action. They may not know, procedurally, where to go, whom to talk to, what forms to fill up, and so on. Last, even if they possess all this information, they may not have access (direct or mediated) to the appropriate officials. Claim-making is only possible, in sum, when someone wants or needs something; believes that it is the state’s responsibility to meet that need; feels that state-targeted action is effective and worthwhile; knows how to navigate the political system; and has access to channels through which to act. We see these requisite conditions in Figure 2.1, where citizen action rests, first, upon a set of motivating beliefs about the state that shape the horizon of imagined possibilities (chain A) and, second, upon a set of enabling conditions that make action possible given the aspiration (chain B). A person must, in

3

Scholars have grappled with variants of this puzzle in settings across the globe. Why, across Latin American cities, are some more likely than others to engage in state-targeted interest articulation (Collier & Handlin 2009; Houtzager & Acharya 2011)? What accounts for different patterns of voice and silence among the rural poor in Appalachia (Gaventa 1980); or for different approaches to local government among low-income residents of major American cities (Small 2004, 2009)?

A Theory of Active Citizenship

32 A. Motivating Conditions

B. Enabling Conditions

INTERESTS (needs or wants)

INFORMATION (governmental performance)

Sense of ENTITLEMENT

KNOW-HOW (procedural & tacit)

Sense of EFFICACY

ACCESS (linkages to the state)

ASPIRATIONS

CAPABILITIES

CITIZEN ACTION

figure 2.1 Claim-Making Conditions

other words, both aspire to make claims on the state, and have the capacity to do so. In the following sections, I unpack each concept in turn, arguing that both aspirations and capabilities are state induced and socially produced, shaped simultaneously by the social structures and institutional environments in which citizens are embedded. In building this argument, I draw on two distinct strands of literature. The first explores the social determinants of participation, examining how individuals gain resources for political action through their membership in social networks and engagement with “nonpolitical” institutions such as the family, the workplace, and the community.4 To the extent that the state enters the analysis in this body of work, it is typically as a dependent variable; state capacity and governmental performance are studied as the outcomes of social mobilization and civic participation. The second body of work reverses the causal arrow, examining how states – through their policies and actions – influence patterns of civic and political engagement.5 The theory developed here lies at the intersection of these two bodies of work, exploring how a 4 5

Putnam 1993; Burns, Schlozman, & Verba 2001. Pierson 1993, 1996; Heller 1996; Campbell 2003; MacLean 2011a.

Aspirations toward the State STATE INDUCED

Social Policy & Service Delivery

INTERESTS (needs or wants)

33 Legislation & De Jure Rights

Public Capacity & Responsiveness

Sense of ENTITLEMENT

SOCIALLY Narratives about PRODUCED Citizen–State Relations

Sense of EFFICACY

ASPIRATIONS

Beliefs about Power (or Powerlessness)

Comparative Grievance

figure 2.2 Producing Aspirations

combination of social engagement and state action shapes, from the bottom up and the top down, citizens’ experiences of and approaches to the state.

aspirations toward the state An aspiration can be defined as both the desire to have and the ambition to achieve something – thus shaping both what a person wants and what a person does. Aspirations toward the state, accordingly, encompass both citizens’ interests (what they hope to gain) and their intended actions (the strategies through which they hope to pursue those interests). Aspirations and political voice are closely linked. As Appadurai argues, the ability to exercise voice is in large part a “cultural capacity” shaped by “ideologies, doctrines and norms.”6 These, in turn, shape citizens’ “navigational capacity” (the ability to experiment, explore, and pursue opportunities) and hence their “capacity to aspire” – that is, the conceivable horizon of possible outcomes and actions.7 Neither desires nor behaviors take shape among individuals in isolation; they are, rather, “formed in interaction and the thick of social life.”8 Figure 2.2 depicts the processes through which citizens’ claim-making aspirations are socially produced through the circulation of narratives and ideas and, at the same time, how they are state induced through public policies and the action (and inaction) of public officials. Citizen Interests To make claims on the state, a citizen must first have an interest – a need or a want – related to social welfare. This is not a trivial assumption. The wealthy have little need for welfare services and may opt to purchase other services, for example in health care and education, on the private market. Even among the less affluent, a global proliferation of non-state service providers has begun to reduce (or shift) the claim-making imperative for some.9 It is not always clear, 6 9

7 Appadurai 2004, p. 66. Ibid., pp. 66–69. Cammett & MacLean 2014.

8

Ibid., p. 67.

34

A Theory of Active Citizenship

then, that a felt need – even if acute – will translate into a state-targeted demand. In most places, though, large numbers of citizens turn to the state for the provision of social welfare. This is particularly true for the poor, in contrast to more affluent citizens who can afford to purchase private alternatives.10 The result, as MacLean has noted, is a “two-tiered social service system” in which the state remains central to the lives of the rural poor even as it is diminished in the lives of the more well-to-do.11 Most states deliver services in an irregular and unequal fashion, leading to the exclusion of substantial portions of the population. Most poor citizens, as a result, desire more and better services from the state. Much of what citizens desire is so basic as to be taken as given: food in the mouths of one’s children, adequate shelter, subsistence levels of income support. Citizens’ interests, though, are not simply innate, but emerge through the circulation of narratives about the state. This latter view of interest formation clashes with that of the liberal and pluralist traditions, in which interests are seen as the natural emanation of individual preferences.12 In this view, we should be able to observe what a person desires from the state from his or her claim-making activity. This perspective, however, has been roundly critiqued by those who argue that interests are not always directly observable. Rather, through hidden dimensions of power, including agendasetting rules, procedures, and norms, certain actors and issues are excluded from the public sphere.13 A related view holds that a person’s interests are, themselves, the product of largely “invisible” power relations that influence a person’s very wants.14 Gaventa powerfully underscores this through his study

10 11 12

13

14

Bénit-Gbaffou & Oldfield 2011; Houtzager & Acharya 2011; MacLean 2011b; Jayal 2013. MacLean 2011b, p. 1155. Politics, in this liberal view, is understood as conflict between competing interest groups (or “factions”) that form shared “passions” (Madison 1787). Pluralism lacks any theory of where such interests come from; instead they are simply given (although Madison does suggest that property ownership – or lack thereof – is a defining source of interest). Bachrach and Baratz (1970, p. 44) refer to these barriers as the “second face” of power, in which there is a “mobilization of bias … in favor of the exploitation of certain kinds of conflict and the suppression of others … Some issues are organized into politics while others are organized out.” Political power, in this view, is thus observable in a lack of conflict, where “demands for change in the existing allocation of benefits and privileges in the community can be suffocated before they are voiced.” A distinct but related body of work examines the effects of institutions on the structuring of interests in society. Immergut (1998) argues that the core of this “institutionalist” theory rests on three assumptions. First, behavior cannot reveal “true” preferences or interests. Second, the ways in which individual interests are aggregated shapes and changes them. Third, institutional structures distribute power and resources unevenly among groups, favoring certain interests over others. Lukes (1974) argues that there is an “invisible” face to power that shapes the very preferences and interests upon which people act. The “supreme exercise of power,” he suggests, is to “avert conflict and grievance by influencing, shaping, and determining the perceptions and preferences of others” (Lukes [1978], cited in Isaac, 1987, p. 13).

Aspirations toward the State

35

of “quiescence and rebellion” among Appalachia’s poor, seeking to explain inaction in mining communities among those whose grievances include poverty, dire working conditions and social deprivation in a land of mineral plenty. A person’s “conceptions of necessities, possibilities, and strategies of challenge,” he argues, reflect underlying power relations that, through processes of socialization, work to produce an apparent acceptance of the status quo.15 Thus, he writes, “Not only … might grievances be prevented from entering the political process, but they might be precluded from consideration altogether.” Still other interests are shaped and even created by the state itself.16 A person’s sense of whether and how she needs the state is not static but develops along with expanding levels of service provision. A new public housing program, for example, might generate new demands for housing, just as the introduction of a public employment program could increase demands on the state for jobs. Once in place, services often create entrenched constituencies that will fight for their continuance over time. Legislation establishing the Social Security Act and Medicare in the United States, for example, sparked new interests that galvanized participation among a class of citizens (the elderly) that previously did not engage vigorously in politics.17 In sub-Saharan Africa, citizens with past experiences with public schools and health centers were more likely to express interests in those and other services, and to participate in politics, even in a climate of welfare retrenchment.18 Citizens’ interests vis-à-vis the state, in sum, are the products of the social, political, and institutional contexts in which they are mobilized, and they will vary as those contexts change.

From Interests to Entitlement Felt needs and interests, though, are rarely enough to prompt action. The mother without food stamps in an American city and the woman without drinking water in rural India are both acutely aware of their needs but may

15 16

17

18

Gaventa 1980, p. 15. On the state’s role in organizing interests, see Berger (1972, 1981). On the state’s role in citizen mobilization, see Pierson (1993), who argues that public policies create resources and incentives that compel political action among government elites, interest groups, and mass publics. Decades earlier, Schattschneider [1935] similarly observed that “new policies create a new politics” (cited in MacLean 2011a, p. 1259, FN 1). Campbell 2003. Mettler (2005) similarly describes the engagement of veterans who benefited from the G.I. Bill who came to interpret their status as citizens differently. Pierson (1996) explores how welfare programs in the U.S. and Europe, once initiated, create entrenched political constituencies that fight for their maintenance over time. MacLean 2011a.

36

A Theory of Active Citizenship

not feel or articulate them as a grievance: that is, as a wrong that is cause for complaint or action. Even where the state’s role is clear (made visible, for example, though public food distribution systems or through public water pumps), an individual may not feel that she has a “right” to services or benefits. Instead, public service delivery might appear to be an act of state largess, rather than an entitlement. Moreover, where services are allocated in an uneven fashion, particular groups of citizens (often the poorest) may come to believe that they are not intended beneficiaries. The next step in the chain from interests to action, then, is that a person conceptualizes her needs as public entitlements: something that the state should strive to fulfill. The notion of entitlement (which infers duties and obligations upon the state) is, like the conception of need, a highly malleable concept that shifts over time in response to changes in state practice. First, entitlements are conferred upon citizens by the state. In the example above of the U.S. Social Security Act, legislation established new entitlements for senior citizens, thus altering both their perception of the state’s obligations and their self-perception as active citizens. Public policy design, in other words, “sends messages to their clients about their worth as citizens, which in turn affects their orientation toward government and their political participation.”19 National legislation on social rights – as seen, for example, in Brazil, South Africa, and India – formally establishes the parameters of the state’s obligation to provide social welfare. The legal recognition of social rights might, in turn, foster a deeper sense of entitlement among citizens. In Brazil, for example, an expansion of social rights (codified in the 1988 Constitution) was driven by increases in civil society mobilization, which, in turn, further advanced the state’s engagement in social welfare provision.20 In India, in contrast, an expansion of social rights legislation since the 1990s has not been accompanied by mass civil society mobilization. The rhetoric of “rights,” while powerful in policy arenas, remains relatively thin at the grassroots.21 There are, however, other ways in which state practice can influence citizens’ sense of entitlement – beyond the formal declaration of rights. In particular, citizens develop expectations of the state by observing how it treats others around them. Scholars have long recognized the role of

19

20 21

Campbell 2003, p. 6. The result, she writes, is a policy feedback loop, as “Citizens’ relationships with government, and their experiences at the hand of government policy, help determine their participation levels and in turn, subsequent policy outcomes.” Holston 2008; Heller 2001, 2013. These dynamics are discussed in greater detail in Chapter 8. There are clear exceptions, discussed in Chapter 3, such as the movements for the Right to Food and the Right the Information, both of which resulted in public interest litigation. These movements, though, have been largely led by elite activists who mobilize periodic local support. In interviews in localities where such movements are absent, a discourse of “rights” was notably absent among ordinary citizens.

Aspirations toward the State

37

yardstick comparisons in establishing political benchmarks.22 This process is particularly pronounced in settings where the state’s presence is both visible and uneven, spurring a sense of comparative grievance (that is, a sense of exclusion, prompting citizens to ask, “Why am I being neglected while others are not?”). This articulation of grievance represents an inferred sense of entitlement. From Entitlement to Efficacy Even if a person believes that the state should act, she may not believe that it will. Here the issue is one of beliefs about the efficacy of action.23 At play are a set of underlying beliefs, first, about personal efficacy (that is, an individual’s confidence in her ability to effect change) and, second, about the efficacy of the broader political system (that is, about governmental capacity and responsiveness).24 It matters, in other words, whether a person thinks change is possible. A personal sense of efficacy reflects internalized beliefs about one’s place in a social and political hierarchy.25 Gaventa again serves to illustrate, documenting how, through processes of socialization, the working poor in Appalachia developed a sense of powerlessness, visible in “quiescence” or an apparent acceptance of the status quo.26 Beliefs about power and powerlessness often become entangled with questions of culture. A large literature has examined the effects of a so-called “culture of poverty” on citizens’ expectations, concluding that many do not act to improve their situation (or, worse, act to their own detriment) because they can see no other viable alternative. This body of work 22

23

24

25

26

Ray (2003, p. 1), for example, writes: “individuals may use the experiences of others (and perhaps their own past experiences) as the yardstick for setting goals for themselves, against which they evaluate the pleasures and pains of their immediate experience.” Through these benchmark comparisons, individuals develop new conceptions of well-being and of the possibility for change. Early work by Campbell, Gurin, and Miller (1954, p. 187) defined political efficacy as the “feeling that individual political action does, or can have, an impact on the political process, that is, that it is worthwhile to perform one’s civic duties.” For a review of subsequent scholarship on notions of political efficacy, see Kennedy (2012). Scholars of political psychology distinguish between “internal” and “external” types of efficacy, where internal efficacy refers to an individual’s perception of her or her own ability to understand and navigate the political system, and where external efficacy refers to beliefs about the responsiveness of officials to citizen demands (Balch 1974; Niemi et al. 1991; Lieberman et al. 2014). Bandura (1997, p. 3) refers to this as “self-efficacy”: that is, beliefs in one’s capabilities to organize and execute the courses of action required to produce given attainments.” Gaventa 1980, p. 15. Power and powerlessness, in this view, are expressed in neo-Gramscian terms, through influence over norms, values, and beliefs. Gramsci (1971) employed the notion of cultural hegemony to explain ideological, as opposed to simply economic and political, control by the bourgeoisie who maintained power, not only through force and coercion, but also through influence over cultural notions of the common good that were consistent with the maintenance of bourgeois interests.

38

A Theory of Active Citizenship

has been widely critiqued for its tendency to “blame the victim,” and dismissed for its failure to account for diverse practices and adaptability among the poor.27 Observing the varied responses of residents in a single neighborhood of south Boston, for example, Small notes that the challenge is to “reconcile the theory that structural conditions have such strong impacts on the poor with the fact that poor individuals living under the same conditions manifest such different outcomes.”28 Poverty, in other words, is not equally constraining. There is, nonetheless, something to be said about how social beliefs and practices might shape a person’s sense of efficacy. Rao and Walton, for example, while rejecting a culture of poverty framework, highlight the role of culturally derived “constraining preferences,” which stem from beliefs about one’s identity and relationship to others in society and which, accordingly, influence the “perceived possibility of success or failure.”29 It is important to note, however, that notions of efficacy (and the aspiration to action) are conceptually distinct from notions of entitlement (and aspiration for services). James Scott makes a similar distinction in his work on peasant resistance in Malaysia, arguing that what the poor believe is right (their conception of justice) is not the same as what they believe is possible under prevailing social and political constraints. Overt citizen action, under such conditions, may not seem worthwhile in terms of risk or time or effort expended.30 Finally, claim-making aspirations rest upon a sense of political efficacy: that is, the belief that public officials both can and will respond to citizens’ demands.31 It depends, in other words, on perceptions of public capacity and responsiveness, reflecting both subjective and objective conditions. Subjectively, citizens are responding to narrated accounts of state action and inaction. Citizens also respond to the objective conditions of their own lives, including personal encounters with public officials and agencies. In a resource-scarce 27

28 29

30

31

For critical reviews of work on the culture of poverty, see Small, Harding, and Lamont (2010), and Rao and Walton (2004). Small 2004, p. 11. Rao & Walton 2004, p. 15. They offer as example the topic of race in the United States, building on the work of Glenn Loury, who notes that racial inequality in mortality rates, incarceration, and poverty, among other factors, corresponds with patterns of behavior (such as violence, early sexual activity, and teen pregnancy). Importantly, these behaviors are not simply the result of a “culture” of poverty, but rather the products of historical processes that constrained the set of choices, beliefs, and aspirations. Scott 1985. As Scott observes, however, the poor can engage in less visible forms of “everyday” resistance – e.g., foot dragging, non-compliance, or evasion – to register their discontent. For Scott, these “everyday” forms of resistance are evidence of a profound sense of injustice (and lack of consent) among the poor, even as they are constrained in their actions. However, these forms of resistance are conceptually distinct from claim-making – which, as I define it, implies an active engagement with, rather than evasion of, the state. Niemi et al. (1991, p. 1407) describe this “external” political efficacy in terms of “beliefs about the responsiveness of government authorities and institutions to citizens’ demands.”

Capabilities for Action

39

environment marked by poor public sector performance, a person may develop low expectations with regard to both the capacity and the willingness of public officials to respond. In Mexico, for example, researchers found that increasing citizens’ knowledge of poor performance and corruption, far from spurring political action, instead prompted citizens to withdraw from the political process.32 In this instance, greater exposure to the state in fact lowered citizens’ expectations of responsiveness, leading to a diminished sense of political efficacy. In contrast, as levels of service provision expand and, importantly, as people witness this process, citizens may come to expect more. In South Africa, for example, there has been a dramatic increase in local protests since the mid2000s, primarily focused in urban areas that – in the same time period – experienced an increase in levels of service provision.33 Objectively speaking, service provision is generally higher in Mexico than in South Africa, but it is the relative expansion (or contraction) of services that informs citizens’ expectations and, in turn, motivates (or deters) citizen action. Citizens’ sense of political efficacy is thus responsive to the institutional reach, visibility, and accessibility of the state – a theme to which we return at the end of this chapter.

capabilities for action Aspirations toward the state are a necessary but insufficient condition for claim-making. Citizens must also possess a set of claim-making capabilities, including the ability to understand and navigate the local political system (information and “know-how”), and the linkages (or access) to the state in order to act upon this knowledge (recall Figure 2.1, chain B).34 These enabling conditions, like the motivating conditions discussed above, are also state induced and socially produced (Figure 2.3). Information and Knowledge Information about the state’s functions and performance provides citizens with an understanding of how to take action in a particular local context. Citizens gain information from the state itself, for example, through state-led publicity campaigns, or through civil society campaigns and community monitoring 32 33 34

Chong, de la O, Karlan, & Wantchekon 2014. Heller 2015; Alexander 2010 (discussed in Chapter 8). I employ the term “capabilities” in a narrower sense than Amartya Sen (1990) who, in his pioneering work on human development, refers to capabilities as conditions that protect and enhance human freedoms. Claim-making capabilities, as I conceptualize them here, can contribute to human capabilities in this broader Senian sense if citizens effectively demand services (for example, in health or education) that expand their ability to live healthy and full lives. By the same token, human capabilities (such as health and literacy) can enable more effective claimmaking.

A Theory of Active Citizenship

40

STATE INDUCED

Public Information & Outreach Campaigns

INFORMATION (governmental performance) SOCIALLY PRODUCED

Narratives about State Performance, Social Campaigns

Publicity about Formal Rules & Procedures

KNOW-HOW (procedural & tacit)

Formal Participatory Spaces & Outreach

ACCESS (linkages to the state)

CAPABILITIES

Social Learning about Informal Brokerage by Non-state Actors Norms & Practices

figure 2.3 Producing Capabilities

programs.35 A vast literature theorizes the impact of these kinds of information provision activities on citizen mobilization.36 The predictions of this body of work, broadly, are that citizens with better information about what their governments are (and are not) doing will be more active in making demands; and state agents, in response to this pressure, will improve their performance. There is some evidence to support this view. In a study in Delhi, for example, slum dwellers were more likely to vote (and to vote for “better” candidates) when provided with report cards on politician performance.37 Similarly, community members in rural Uganda who were provided with information about the performance of local health care centers were more active in community monitoring.38 However, a growing body of work questions the power of information alone. In rural Kenya, for example, a campaign alerting parents to the quality of education in public schools had no measurable effect.39 Similar efforts on education in the Indian context also had no impact.40 Campaigns to inform citizens about legal entitlements, for example in relation to payment under the Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS – India’s flagship rural work program), have also not resulted in corresponding increases in citizen demands.41 Formal procedural information, thus, is not enough. This is particularly true in complex governance environments, where lines of entitlement and accountability may be unclear to citizens and to officials alike. Witsoe, for example, usefully distinguishes between official information disseminated about public programs and “practical” knowledge of the state practices through which programs are actually implemented. In a study of MGNREGS in the state of Bihar, he observes that “disseminating knowledge about provisions of the 35 36

37 40 41

Jenkins & Goetz 1999; Lieberman et al. 2014. Much of this body of work builds on insights from economics concerning the asymmetric nature of information (for example, Besley 2006; Besley & Burgess 2002). 38 39 Banerjee et al. 2011. Björkman & Svensson 2009. Lieberman et al. 2014. Banerjee et al. 2010. Niehaus & Sukhtankar 2013; Dutta, Murgai, Ravallion, & van de Walle 2014.

Capabilities for Action

41

[legal] act does nothing to change the systems of practice and control” that influence the program’s day-to-day functioning.42 In these settings, citizens require tacit – rather than procedural – knowledge of how public distribution functions, including an understanding of informal practices and channels of access: where to go, whom to contact, and how best to communicate one’s needs and interests.43 Social learning is particularly important in this regard, as people gain knowledge of the inner workings of the state that cannot be conveyed simply through public information campaigns.44 Access to the State Even with requisite knowledge – both procedural and tacit – of the political system, citizens can be stymied by an absence of channels through which to put that knowledge to work. Claim-making therefore also requires outwardreaching linkages to public officials or representatives, or to brokers and intermediaries who bridge the gap. The state itself shapes opportunities for citizen action – for example, through electoral laws or other formal institutions that determine access to the political sphere.45 At a more local level, the state can both create and limit the spaces into which citizens are invited. Creating new spaces for participation was a partial impetus behind the wave of decentralization that has engulfed many developing democracies since the 1990s, resulting in a proliferation of new participatory fora – for example, town halls, planning meetings, participatory budgeting – that have become nominally accessible to more and more people around the world. However, formal efforts to increase access, even when well intentioned, often fall short; they are victims of political or local elite capture, or they are stymied by chaotic administrative practices and limited local state capacity. Citizens therefore must often seek out other channels to the state. Access to an effective broker can extend the range of officials, agencies, and public spaces with which a person has contact – thereby expanding the opportunity structure for claim-making. An exploration of citizen–state linkages must, therefore, become at once broader and more localized – encompassing both direct and mediated pathways to the state. In the absence of any such linkage, a person will simply be unable capture the attention of public officials and claimmaking efforts will fall on deaf ears. Differences in citizens’ state-targeted 42 43

44

45

Witsoe 2014, p. 24. This notion of tacit knowledge draws on the work of Michael Polanyi (1966), in which he distinguishes codifiable knowledge that can be verbally communicated from other knowledge that can only be accrued through experience and practice. Indeed, the very notion of a “repertoire” – which can be understood as a “toolkit of habits, skills, and styles from which people construct strategies of action” – implies a process of social learning (Swidler 1986, p. 273). For a review of the literature on how formal democratic institutions shape citizen-state linkages, see Kitschelt (2000).

42

A Theory of Active Citizenship

actions thus reflect differential access to the state, as well as differences in conceptions of need, entitlement, and possibility.

social and spatial exposure Together, these two elements – aspiration and capability – take us a long way toward understanding both whether and how individuals engage the state in pursuit of social welfare. The challenge, then, becomes to explain how claimmaking aspirations and capabilities develop and why they vary across and within groups and localities. A large part of the answer to this puzzle, I argue, is found by examining the spaces in which individuals encounter and learn about public officials, agencies, and resources. Claim-making, in other words, is a reflection of citizens’ exposure to the state. The remainder of this chapter takes up the task of theorizing the effects of this exposure, which is constrained by social and spatial boundaries of community and locality. The Power of Porous Boundaries There is a growing literature on boundary politics, or the ways in which people and communities define “who and what is properly included and who and what stands outside.”46 To understand the role that these boundaries play in shaping citizens’ relationships to the state, we first need to understand “the multiplicity of ways and criteria by which human communities constitute themselves”47 or, more simply, the dimensions along which commonalities (and, by extension, differences) are defined. A rich body of work examines the ways in which social cleavages demarcate the political world, shaping the lines along which citizens mobilize48 and the dimensions along which public policies are crafted.49 The precise nature of the social divisions in question – whether based on class, ethnicity, caste, tribe, religion, gender, or other characteristics – vary from context to context. All, though, define some kind of membership and, by extension, exclusion. Spatial boundaries demarcate one locality from another 46 48

49

47 Isaac 2011, p. 779. Ibid. A substantial scholarship identifies social cleavages (in particular ethnic fragmentation) as obstacles to collective action and, by extension, to participation (Alesina & La Ferrara 2000; Putnam 2007). Others, however, have argued that heterogeneity may in fact increase levels of participation, since individuals exposed to diverse networks increase their political knowledge and skill (Huckfeldt, Mendez, & Osborn, 2004; Scheufele, Nisbet, Brossard, & Nisbet, 2004). See, for example, Lieberman (2009) and P. Singh (2015) on how differences in social boundaries (particularly along ethnic lines) shape state responsiveness, including public health responses to disease (Lieberman) and the provision of social welfare in multiethnic settings (Singh). Both emphasize the relative strength or weakness of ethnic divisions, and the extent to which public policy responses are designed to reinforce or cut across those divides. As Lieberman writes (p. 29), “boundaries themselves influence how policies are framed and interpreted by citizens and elites.”

Social and Spatial Exposure

43

and may be geographical and/or administrative. Spatial divisions can reinforce social cleavages (creating geographical entities contiguous with social identities) or can cut across them (creating mixed, or fragmented, localities). Together, social and spatial boundaries delimit an individual’s exposure, constraining both movement (where a person goes) and encounters (who he or she meets and observes). These boundaries, however, are not uniform. They are highly constraining for some but more malleable for others. The drivers of boundary porousness are context specific, reflecting a complex set of factors that enable connectivity across space and mobility among people. In the rural Indian context, as we will see in Chapter 5, some of the most salient boundaries – those of caste and of village – are becoming less constraining over time due to changes in the rural economy leading to diversification in rural employment and, by extension, increased mobility and a decoupling of caste and occupation. This, in turn, has spurred increased encounters across social (caste) and spatial (neighborhood and village) lines. At the same time, new patterns of political engagement among the poor, the lower castes, and women – prompted in part by institutional interventions to promote the participation of marginalized groups – have helped to establish mixed public spaces occupied by men and women from different caste and class backgrounds. The result is a shifting rural landscape in which boundaries of locality and community are becoming more permeable. These processes, however, are slow and uneven; change has not come equally to all places and all people. There is thus more room to maneuver for some than others, reflecting the relative porousness or rigidity of social and spatial boundaries. Where boundaries are porous, I argue that the effects on claim-making are expansive: through exposure beyond the immediate community and locality, individuals gain information and knowledge of the state; develop expectations concerning service delivery; learn about potential claimmaking strategies; and forge pathways to officials.50 This, in turn, expands both their aspirations and capabilities for state-targeted action. Where rigid, the reverse is true: citizens’ are less exposed; their information, ideas, and linkages to the state are more restricted; their aspirations and capabilities more are constrained; and their claim-making activity is more limited. These causal sequences are illustrated in Figure 2.4. Importantly, the theory depicted here is probabilistic, not deterministic: exposure itself does not guarantee that a

50

To be clear, malleability or porousness does not imply that boundaries or the identities that they delineate disappear or are fundamentally altered. Basic identities (for example, of ethnicity or village) may remain “fixed,” even as more porous boundaries enable movement and interaction across those social and spatial lines. Porousness, thus, does not alter identities per se but instead enables cross-cutting encounters. I therefore employ the notion of boundary crossing in a somewhat different manner than prevailing work in ethnic politics, in which it implies actually changing or switching identities (Cf. Lieberman 2009, p. 32).

A Theory of Active Citizenship

44 Broad

BOUNDARIES OF COMMUNITY & LOCALITY

SOCIAL & SPATIAL EXPOSURE INFORMATION, IDEAS, & LINKAGES TO THE STATE

Narrow

Expansive

High & Diverse

ASPIRATIONS

CITIZEN ACTION

CAPABILITIES

CLAIM-MAKING INCIDENCE CLAIM-MAKING REPERTOIRE

Constrained

Low & Limited

figure 2.4 Boundary Porousness, Exposure, and Claim-Making

person will engage the state, but – by generating aspirations and building capabilities – it increases the chances that he or she will do so. Social and spatial exposure influences citizens’ claim-making activity in two ways: by shaping the flow of information and ideas about the state; and by influencing the pathways or channels through which citizens encounter the state. First, as Mark Granovetter has demonstrated, information is produced, disseminated, and consumed within social networks. How much and what kind of information a citizen receives is thus a function of his or her network ties. “Weak” ties are particularly important, as they facilitate the flow of new information.51 Extending Granovetter’s insight, I argue that weak ties – enabled by porous social and spatial boundaries – facilitate exchange, not only of information, but also of narratives about public entitlement and political efficacy. As Ray notes, a person’s “window of aspiration” is constrained by who or what one perceives to be similar to oneself and one’s own conditions, and thus attainable. The notion of similarity, though, “depends on how much mobility (or perceived mobility) there is in society. The greater the extent of (perceived) mobility, the broader the aspirations window.”52 In other words, boundaries shape the horizon of possibility by delimiting the people and places to which one compares oneself (what Ray refers to as the “cognitive neighborhood”). Social polarization and segregation limit the reach of these “neighborhoods.” In contrast, in more “connected” places (marked by more porous

51

52

Granovetter’s (1973) classic work documents how individuals learn about economic opportunities (namely employment) through acquaintances (weak ties) to whom they are only loosely connected but who, because they move in different circles, can be sources of new and novel information. Ray 2003, p. 3, emphasis in the original. A person’s “window of aspirations,” he explains, consists of those who inhabit his or her “cognitive world” which, by definition, are those that are both “similar” and “attainable.” This notion of similarity and attainability, however, is malleable and so, as we shall see, will shift under different conditions.

Social and Spatial Exposure

45

boundaries), the window of aspirations expands as individuals observe and thus can imagine a broader array of possibilities.53 Second, boundaries delimit the potential pathways to the state that a given individual might pursue. Social structures of class, caste, and gender impose constraints (both real and perceived), as do spatial and geographic boundaries that demarcate localities. And yet, not all individuals are equally constrained; some regularly cross these divides in the course of their social, economic, and political lives. The more mobile a person, the broader her array of contacts, her sources of information, and her encounters with the state. Boundary porousness, in other words, increases the chances – as well as channels – of accessing the state. To illustrate, take for example these stylized descriptions of women in a rural Indian village.54 One woman lives under purdah (the veil system, under which women’s mobility is strictly curtailed) and thus rarely leaves her home or neighborhood. Her spatial and social spheres are highly circumscribed; she interacts almost exclusively with women from her same caste and with her close male relatives. The other woman is a member of her elected village council occupying a seat reserved for women, who, through her elected office, enters a sphere traditionally restricted to men. She has also worked with a local NGO that seeks to empower women in local politics and, through the NGO, has attended trainings, field trips, and exchange programs in the district headquarters and state capital. As she moves about the village and beyond, she rubs shoulders with a diverse array of people, cutting across established caste, class, and gender lines. She is also employed on a government worksite, which draws her into the labor force and beyond the village. The core difference between these two women lies in their constraints. The first occupies tightly knit, locally bound spaces; her mobility is restricted, her external contacts are few, and her sources of information are limited. The second traverses both social and spatial boundaries; she sees and learns more about the state and its service delivery apparatus. As a result of this exposure, I argue, she is more likely to make claims on the state for social welfare, and will do so through a broader array of strategies, than the other woman who remains more restricted in her mobility and connectivity. Exposure and Social Capital In emphasizing the importance of social and spatial exposure in citizens’ claimmaking, I draw on existing scholarship that examines the effects of social networks on individual and collective behaviors. Occupying a central place within this body of work are theories of social capital, understood in basic 53 54

Ibid., p. 4. These hypothetical accounts are based on an amalgam of real interviews and encounters during the course of my fieldwork, described in greater depth in Chapter 5.

46

A Theory of Active Citizenship

terms as “the ability of actors to secure benefits by virtue of membership in social networks or other social structures.”55 There are, however, two very different conceptions of social capital: one that sees it as a resource generated and utilized by groups (a “communitarian” view), and one that understands it as a private resource (an “individual” view).56 The communitarian view is most closely associated with the work of Putnam, who argues that shared norms of trust and reciprocity – developed through participation in voluntary associations – enable cooperation and, by extension, civic engagement.57 The second, individual view sees social capital as a private good developed through participation in social networks. Bourdieu, for example, defines social capital as: “the aggregate of actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance.”58 Importantly, Bourdieu emphasizes that the networks in which social capital is produced are fundamentally unequal, reflecting the underlying distribution of other economic, political, and human forms of capital. The notion of social and spatial exposure developed in this chapter builds on but departs from both of these strands of scholarship. Exposure does indeed generate social resources (“capital”) that are put to use in making claims on the state. And yet exposure does not require, nor does it necessarily engender, sustained social ties. Instead, exposure can occur through much weaker or more fleeting forms of interaction – not all of which are “social” in nature.59 Exposure also does not necessarily lead to more horizontal or equitable social relations. Far from it, encounters beyond the local community may serve to heighten awareness of inequality. Exposure is thus more likely to produce a sense of grievance than of cross-cutting solidarity. This is particularly likely in environments marked by entrenched social hierarchies. The theory of active citizenship developed here therefore takes us beyond a communitarian view of social capital, to explore how social resources flow and shape political behavior in the absence of horizontality or of shared norms of trust and reciprocity. But if the communitarian view of social capital is too quick to assume horizontality, the individual view may be too quick to assume that social capital exclusively reinforces hierarchy. Levien, for example, builds on Bourdieu to 55

56 57

58 59

Portes 1998, p. 6. He writes: “Whereas economic capital is in people’s bank accounts and human capital is inside their heads, social capital inheres in the structure of their relationships.” On the tensions between these two views of social capital, see Portes (1998) and Levien (2015). Putnam et al. 1993. On the causal mechanisms linking social capital to governmental performance, see Boix and Posner (1998). Bourdieu 1986, p. 248. Small (2009), makes a similar distinction when, in asking how poor women in American cities gain access to support resources, he urges us to think beyond social ties. The most useful ties, he argues, are not always social but instead can be brokered through local institutions, for example through a day care center or a school. Encounters with other mothers in these institutions, he notes, are not necessarily marked by sustained friendship, but are nonetheless critical sites of information exchange and referral.

Scope Conditions

47

argue that “if social capital is a form of power that allows certain people to differentially exploit new … opportunities … because of ‘who they know,’ then it will tend to reproduce or expand existing class inequalities.”60 This assumes that the forms of social capital at play solely serve to strengthen existing cleavages. However, scholars have long recognized that different kinds of social ties have different effects. “Bonding” ties foster insular engagement within a group, while “bridging” ties enable engagement beyond a narrow locality or community.61 Social and spatial exposure is bridging by nature. It is true that this cross-cutting exposure does not necessarily result in newfound social equality. To the contrary, exposure often reveals citizens’ unequal social standing as well as their unequal access to and treatment by the state. And yet, I argue, it is this very inequality – and citizens’ awareness of it – that helps to spur claim-making. Thus, to the extent that exposure drives claim-making across differently-placed social groups, it may also work to diminish existing participation gaps and, in so doing, may challenge entrenched political hierarchies.

scope conditions Thus far, the theory predicts that greater social and spatial exposure will foster more and more diverse claim-making activity. This is the case, I will show in the chapters that follow, in contemporary Rajasthan, where an increasingly mobile population traverses boundaries of caste, neighborhood, and village. The same citizens are also active claim-makers. Rajasthan, however, was not always so. Once a land of feudal princely states, it was infamous for its low levels of political engagement and for negligible levels of service provision. Under such inhospitable conditions, it is unlikely that social and spatial exposure would have spurred the same high levels of state-targeted citizen action that we see today. Far from it, exposure to the regressive princely states would have been likely to reinforce a sense of futility, and thus would have worked to curtail citizen claim-making. It is therefore necessary to consider the underlying scope conditions of the theory: that is, the parameters under which we can expect exposure to drive claim-making and, to the contrary, where we might expect to see a more chilling effect. The Limits of Exposure First, in order for social and spatial exposure to motivate citizen action, a degree of personal and local resonance is required. The theory predicts that as a person’s comparative frames of reference expand, so, too, will his or her claim-making practice. There is, at first glance, a potential affinity here with 60

Levien 2015, p. 80.

61

Gittell & Vidal 1998; Narayan 1999; Putnam 2000; Varshney 2002.

48

A Theory of Active Citizenship

strands of classical modernization theory, which emphasize the importance of urban migration in reshaping political attitudes and behaviors. By uprooting individuals from “parochial” villages and embedding them in more “cosmopolitan” settings, the theory goes, citizens gain new perspectives and develop qualitatively different interests in politics.62 However, there is an important tension that distinguishes the theory of active citizenship developed here from accounts of modernization. Exposure does, indeed, rest upon contact with people and places beyond one’s own local community. And yet, as noted, those people and places cannot be too dissimilar. In comparing oneself to people who live under vastly different conditions (for example, comparing life in rural villages to that in the capital city, or comparing the conditions of the very poor to the very rich), the conceptual distances may be too great to develop new aspirations or to spur new strategies of action. Pronounced social inequalities can thus dampen the effects of exposure by creating seemingly insurmountable gaps. Claim-making, moreover, is a largely local affair embedded in the village and its environs. For exposure to spur claim-making, then, it must resonate in a way that prompts greater engagement in, rather than exit from, local politics. It follows that the social distance between an individual and the people and places to which that person is exposed cannot be so great as to limit (instead of expand) the horizon of possibility. Second, exposure is only likely to spur claim-making when, simply put, there is something to be gained. There must, in other words, be resources to be had. This requires a modicum of investment and capacity by the state. Under conditions marked by a felt absence of the state, greater exposure might, in fact, demobilize citizens, prompting exit in pursuit of private alternatives. Similarly, if a person, looking around her, sees uniformly poor public performance, she is unlikely to develop high expectations for service delivery. Exposure, in this case, is likely to inhibit claim-making by reinforcing low expectations. This is particularly true for sectors of the population who have experienced long patterns of exclusion. A poor or lower-caste individual who, upon venturing beyond the local community, continues to encounter discriminatory practices may find her sense of efficacy (both personal and political) diminished rather than bolstered. The material resources that are the subjects of claim-making, thus, must be available, visible, and perceived to be attainable. Otherwise, claim-making may simply seem a fool’s errand. On the other hand, under conditions of relative plenty where the state delivers according to programmatic and transparent rules of entitlement, 62

Lerner 1958; Lipset 1959; Gluckman 1960; Gellner 1964. Lerner’s work on the “passing of traditional society” is a classic example; in it, he presents the story of a Turkish grocer (with economic ties beyond the village) and a local chief (who derived his authority from hereditary status within the village). For Lerner, the grocer is not only more mobile and connected, but also has broader aspirations and interests in affairs beyond the immediate locality. See Chapter 4 for broader discussion of modernization theory and its limits.

Scope Conditions

49

claim-making simply may not be as necessary – or as effective.63 Such is often the case in well-served communities of advanced welfare states, where most citizens have little reason to engage the state over basic amenities as drinking water, road repair, education, or other services that are so routine that they are taken for granted.64 Claim-making, in this sense, rests upon a degree of inequality or irregularity in public performance and resource distribution: claim-making is only worthwhile when there are resources to be had, but only necessary where their uneven allocation creates the need for citizen action. Third, the perceived benefits must outweigh the costs. Claim-making carries risks of reprisal, both political and social. These risks are particularly high under authoritarian regimes,65 as well as for communities that lack clear legal standing. For example, squatters in urban India, as in many cities around the world, may encounter the state during attempts at eviction – often at the end of a bulldozer.66 Making claims for services, in such circumstances, may be a nonstarter if the potential cost is one’s home, livelihood, or even life. Fear of organized state violence – while not a recurring theme among those interviewed for this book – is also clearly part of the narrative through which citizens see the state in settings in India and across the Global South where governments and residents are locked in pitched battles over land and resources. Even absent such extreme forms of risk, claim-making can still be costly – physically, materially, and psychologically. It may mean traveling long distances and may require paying bribes, thus imposing costs of time and money.

63

64

65

66

“Programmatic” models of resource distribution are those in which goods and services are allocated according to clear and visible rules of entitlement. Non-programmatic distribution, in contrast, is less rule-bound; entitlement criteria may be less transparent or may be subverted by political actors who are selective in their implementation. On the distinction between programmatic and non-programmatic politics, see Stokes et al. (2013); Kitschelt (2000); and Ziegfeld (2016). Such conditions, of course, are not uniform across advanced industrial states where low-income communities are often severely underserved. See, for example, “Detroit cuts off water for families – and hopes for the future,” BBC News, September 21, 2014; and “Here’s one way Baltimore teens are worse off than poor youth in Nigeria and India,” Washington Post, April 30, 2015. This is matter of degree. In many authoritarian and transitional systems with weak formal accountability, there is still considerable room to maneuver. As Tsai (2007) has demonstrated in rural China, “accountability without democracy” is possible where local officials are disciplined by informal norms that incentivize them to respond to citizens’ demands. Chatterjee (2004) notes that many urban residents live beyond the legal realm, as squatters, street traders, fare dodgers, or those who illegally siphon off water or electricity. This engagement with the state is a fraught and at times violent form of contestation over resources that the state does not view as “rights.” Claims on the state for services by squatters, though, can also be a critical step toward broader claims for legal recognition. See, for example, Holston (2008).

50

A Theory of Active Citizenship

It can also be arduous and intimidating. As Corbridge et al. describe in the rural Indian context: When a widow goes to the Block headquarters to collect her pension she makes contact with the state in the form of a lower-level official and by entering a designated building … But these encounters are rarely conducted as the rulebook says they should be. The widow will be kept waiting for hours in the sun or the rain, and she might have to call on a relative or fixer … to get her business moving. Small payments might also have to be made to the accountant and/or his peon, and sometimes the payment she receives will be several rupees short.67

Claim-making might also provoke punitive responses on the part of officials who feel threatened by citizens’ demands. In an interview in rural Rajasthan, for example, a group of women recounted rumors that a local official had slapped a woman who had come to his office to make an application. Fear of rough treatment, they explained, was a primary reason why they would not approach officials (and why their husbands would not let them go).68 Another woman in a different district similarly described her hesitation when dealing with government officials: Those offices are no place for … an old woman. You stand in long lines and say namaskar. But when they see you are alone they treat you like nothing. Like dirt. They will cheat me. Maybe they will even beat me.69

In still other cases, there is fear of political reprisal. In a high-profile court case in Rajasthan, for example, social activists have been accused of assault (despite ample evidence to the contrary) by a village official whose alleged corruption they were investigating.70 There is also evidence of “blackballing” citizens applying for programs – such as MGNREGS – if they are known to be involved with social movement organizations.71 Learning about these kinds of retaliatory actions, or hearing accounts of violence, discrimination, and other forms of ill treatment at the hands of officials, increases the perceived costs associated with claim-making. There is, moreover, the risk of social reprisals in settings where competition over public resources can often be fierce. Claims on resources and inroads by marginalized groups into social and political spaces previously dominated by elites have spurred backlash movements, visible, for example, in militialike mobilization by landowning castes in parts of rural India.72 Increased

67 68

69

70

71

Corbridge et al. 2005, p. 19. This story, which was recounted by residents in a village in Mandor block, Jodhpur district, is unconfirmed. However, whether it is true or not, the story had a chilling effect on the likelihood that these women would attempt to approach the official in question. Author interview, Gogunda block, Udaipur district, February 9, 2011. Namaskar is a form of formal greeting, in which a person touches the feet of a supposed “superior.” See, for example, “RTI activist Nikhil Dey, 4 others convicted in assault case in Rajasthan,” Hindustan Times, July 8, 2017. 72 Jenkins & Manor 2017, p. 105. See, for example, Sahay (2008).

Scope Conditions

51

interaction and exposure across social lines can thus also produce darker and more perverse effects, including violence and polarization, which in turn may have a chilling effect on claim-making among the marginalized. Bringing the State Back In Altogether, the preceding discussion underscores the fact that social and spatial exposure is not always salutary or motivating in its effects. Exposure is thus perhaps best thought of as a “neutral multiplier” that reflects the broader environment in which citizens are embedded.73 It becomes necessary, then, to bring the state back into the analysis, not simply as a backdrop for claimmaking, but as a set of actors, institutions, and policies that can foster, shape, or inhibit citizens’ interests and actions.74 These dynamics have been widely studied in advanced welfare states, where scholars have long noted a feedback loop between the expansion or retrenchment of social welfare policies and citizen mobilization.75 The feedback loop literature takes us a long way toward understanding how public policy and spending influence political behavior. It operates, however, primarily at a macro scale, examining how national policies mobilize constituencies. It is thus less well equipped to explain micro-level variation in citizens’ experiences of and engagement with the state.76 This is particularly true in contexts marked by non-programmatic rather than rulebound allocation of resources, where the gap between policy and implementation looms large. Under such conditions, citizens’ experiences of public policy are often contingent on the discretion of “street-” or village-level bureaucrats and officials, the decisions and routines of whom, as Lipsky reminds us, “effectively become the public policies they carry out.”77 It thus becomes necessary to look beyond formal policies and programs to understand the ways in which citizens actually experience the state on an everyday basis. At issue, then, is the institutional terrain of the state: that is, the spaces and channels through which citizens observe and engage public officials, agencies, and resources.78 Of particular importance is the breadth of this terrain (the scope, reach, and visibility of social programs), as well as its regularity (the uniformity of public performance and of citizen access to services). Where 73

74 75

76

77 78

I borrow this term from Berman (1997), who used it to describe the conditional effects of social capital on both democratic and authoritarian forms of political organization. Skocpol 1985. See, for example, Pierson (1993, 1996) and Campbell (2003). For similar analysis in a less developed context, see MacLean (2011a) on social welfare retrenchment and citizenship practice in sub-Saharan Africa. An exception is MacLean (Op Cit.), who focuses on the micro experiences of state service provision, rather than on welfare policy per se. Lipsky 1980, p. xii. I build here on Heller’s (2013, p. 48) discussion of “institutional surface area.” Also see Bertorelli, Heller, Swaminathan, and Varshney (2014).

A Theory of Active Citizenship

52

BREADTH (scope, reach, visibilty)

REGULARITY (uniformity of performance & equity of access) Uneven Uniform

Narrow

Broad

A. "Failed" States Uniformly sparse resources

C. "Advanced" Welfare States Programmatic & regular resources

Low & limited claim-making

Low & limited claim-making

B. "Predatory" States Scarce resources & elite capture

Low & limited claim-making

D. "Intermediate" States Variable public resources & performance

High & diverse claim-making

figure 2.5 Claim-Making and the Institutional Terrain of the State

the terrain is broad, the state is central to citizens’ lives, made visible through wide-ranging social programs. Where narrow, the state is experienced more in terms of a scarcity of public resources and programs. Where the terrain is uniform, the state’s performance is relatively even across its territory, and citizens’ access to resources is relatively equal.79 Where uneven, the state’s performance is irregular and citizens’ access to resources is unequal. Figure 2.5 depicts idealized patterns of variation in the institutional terrain of the state across these two dimensions: breadth and regularity.80 If the state’s terrain is uniformly narrow (as in quadrant A), increased exposure will be unlikely to galvanize action since citizens will consistently encounter a resource-strapped and low-capacity state that underperforms. Such is the case, to take an extreme example, in Somalia – a country often referenced as an example of a “failed” or “collapsed” state, in which weak public sector capacity compels residents to turn to non-state actors, including militant groups. In other places, the state’s resources are captured and their distribution politicized, 79

80

Note, however, that regularity does not necessarily imply breadth: public performance can be uniformly narrow if public resources are equally underprovided (see quadrant A). The categories employed in Figure 2.5 are loosely modeled on those in Evans’s (1989) article, “Predatory, Developmental, and Other Apparatuses.” The ideal types in this figure should not be read in an overly constraining fashion, but should instead be taken as benchmarks indicating a range of possible outcomes. We should expect considerable variation both across and within countries, most of which will not fit neatly in one quadrant.

Applying the Theory

53

resulting in a terrain that is both narrow and uneven (quadrant B). A classic example is Mobutu’s Zaire, noted for its extractive practices and underprovision of public goods. In such “predatory” states, citizens’ sense of efficacy as well as their access to the state will be sharply curtailed, particularly where the risk of state reprisal is high.81 Exposure to the state in both of these settings (quadrants A and B) in is likely to have a chilling effect on claim-making. At the other extreme, consider a state – classic examples being those in Scandinavia – where social services are more readily available and where public resources are delivered in a programmatic fashion by a relatively effective bureaucracy. In such settings, marked by a uniformly broad state terrain (quadrant C), the impetus for claim-making may also be limited. This is because many services – particularly in health, education, and social security – are available without the need for high levels of mobilization or contestation. Officials in these more rule-bound settings also exercise less discretion in the allocation of goods, making them less likely targets of claim-making. Greater exposure, in this context, is thus unlikely to have much of an effect on claimmaking practice, which may remain low and limited simply because it is not very often needed. State-targeted citizen action, I argue, is most likely under intermediate conditions marked by the breadth and unevenness of the state (quadrant D). In such settings there are, in blunt terms, public resources to be had – but they are unequally and, at times, unpredictably delivered. This combination of breadth and unevenness both enables and motivates claim-making, increasing citizen– state encounters while provoking a sense of grievance (and of entitlement) when the state does not deliver. It is here that social and spatial exposure is most powerful in its effects. Moving around beyond their immediate community and locality, citizens come to see the state in the form of goods, services, and programs, but are often frustrated in their attempts to access such resources. This frustration – when coupled with a sense of entitlement and of personal and political efficacy – can provoke action. At the same time, exposure builds citizens’ knowledge of and access to the state – thereby increasing the capacity for action.

applying the theory As I will demonstrate in the Rajasthani case, more mobile citizens who traverse boundaries of community and locality are more likely to engage in claimmaking and do so through broader repertoires of action. A single-state study, however, is not well positioned to systematically test the effects of exposure 81

Evans (1989, p. 562) refers to the former “klepto-patrominial” Zairian state as an example of a “predatory” state in which “those who control the state plunder without any regard for the welfare of the citizenry.” In such states, the risk of state reprisal and repression outweigh the (unlikely) benefits of claim-making.

54

A Theory of Active Citizenship

under different conditions. Absent cross-sectional data that compares Rajasthan to other settings, we cannot know if claim-making is indeed lower where the institutional terrain of the state is narrower, or higher where it is broader.82 However, a historical examination of the Rajasthani state (developed in the next chapter) is revealing. As we will see, the local terrain of the state in Rajasthan has become broader over time, as witnessed in the shift from the extractive “predatory” princely states; to the abysmally low levels of social welfare provision in the early years following Independence indicative of what was, in some senses, a “failed” state; to the “intermediate” conditions of contemporary Rajasthan marked by rising but unequal service delivery. The state, in these terms, has become more vital and visible in citizens’ lives and, at times and for some, more accessible. Part II of this book tells the story of this uneven transformation and its consequences for claim-making. Chapter 3 charts the rise in social spending and proliferation of social welfare programs nationally and in Rajasthan, noting the local effects of these trends under decentralization. Chapter 4 describes the patterns of claim-making that unfold upon this expansive yet lumpy state terrain, noting high levels of claim-making but varied repertoires of action across and within socioeconomic groupings. Chapters 5 and 6 seek to explain this variation. Chapter 5 asks why different citizens see and experience the state differently, highlighting the role of shifting social and spatial boundaries and the exposure to the state that this generates. The empirical relationship between social and spatial exposure and state-targeted action is investigated in Chapter 6, confirming exposure’s catalytic effects on claim-making.

82

Chapter 8 builds from the Rajasthani case to theorize about the state conditions under which claim-making is most and least likely elsewhere in India and beyond.

part ii CITIZENSHIP PRACTICE IN RAJASTHAN

3 The Institutional Terrain of the State

Whether and how an individual seeks resources from the state is, in large part, a function of the state itself: of its policies, programs, expenditures, and its presence and visibility in citizens’ daily lives. This chapter, accordingly, examines the spaces in which citizens encounter the state in Rajasthan. I argue that the state is becoming both more visible as well as more meaningful at the local level – although unevenly so. This is the product of two trends that took hold in the 1990s, and which accelerated through the 2000s: an expansion of the social welfare functions of the state, visible in increased expenditures and programming; and the deeper institutionalization of the local governance environment, driven by decentralization. The 1990s were a moment of dramatic political and economic change in India, marking a departure from more than four decades of Congress party rule1 and state-led planning (an era that came to be known as the “permit, license, quota Raj”). The 1989 election of V.P. Singh who, as Prime Minister, led a coalition (the National Front) of centrist and regional parties, signaled the end of an era of Congress hegemony in an increasingly competitive electoral environment. The same period also saw efforts to liberalize the Indian economy – beginning with a package of structural adjustment reforms in 1991.2 Paradoxically, in a political moment marked by the 1

2

The Indian National Congress was the leading political organization of the Indian Independence movement. Following Independence, it became the dominant political party in most states, leading the central government almost consecutively for more than forty years from 1947 to 1989. India’s economic liberalization is often given a start date of 1991 when, facing a debt crisis, the government undertook a series of economic reforms including deregulation, liberalization of trade policy and the financial sector, partial privatization of state-owned enterprise, and other actions intended to attract foreign direct investment and secure the support of the International Monetary Fund. Many of these policy changes, however, have longer roots beginning in the late

57

58

The Institutional Terrain of the State

rhetoric of liberalization, the 1990s also marked the beginning of a period of expanding social spending. At the same time, institutional reforms in local governance – most notably a constitutional amendment in 1993 mandating local elections – set in motion political, administrative, and fiscal decentralization. The consolidated effects of these trends – visible in an increased but variable local presence of the state – set the stage for citizen claim-making in the late 2000s, the time of research for this book.

expansion of the social welfare sector Social spending by both the central and state governments, as we will see in the following section, increased quite dramatically beginning in the 1990s. However, this expansion is relative, reflecting weak initial commitments to social welfare. While concerns over poverty and famine were enshrined early on following Independence in the “welfarist orientation” of the Indian Constitution,3 there is a notable divide between citizens’ civil and political rights, which are defined as “fundamental” in the Constitution, and their rights to social welfare (listed in the non-enforceable Directive Principles of State Policy appened to the Constitution.4 This division was reflected in the strong rhetoric but weak social policy of the Nehruvian state of the 1950s and 1960s, under which the imperatives of industrialization and economic growth repeatedly trumped concerns over social welfare.5 The 1970s and 1980s, under the leadership of Indira Gandhi and her sons, saw a proliferation of antipoverty programs focused on basic needs (income and calories) designed, in part, to bolster electoral support for the ruling Congress party among the poor.6 These programs, though, far from asserting universal social rights, operated on a populist and clientelistic logic. The 1990s, as noted, were a watershed moment in India’s political economy. The policy environment at the time was one of public sector retrenchment,

3 4

5

6

1970s and 1980s – a period of increased trade and foreign investment (Nayar 2006; Kohli 2006; Nayar 2009). Jayal 1999, p. 39. For a discussion of this divide between civil, political, and social rights, and of the historical debates that shaped the constitution, see Guha (2008). Jayal (2013, pp. 164–168), argues that this is in large part a product of the “modernist” vision of Prime Minister Nehru, and his belief that investments in social welfare were “unaffordable” without the prior development of an underlying material base that required rapid industrialization and growth. The primary focus of these poverty alleviation programs was: (1) the creation of wage employment and opportunities for self-employment and (2) the allocation of food through the Public Distribution System and meals for school children. As Mooij and Dev (2004, p. 99) note, these priorities “reflected the then dominant conceptualization of poverty, mainly in terms of income and calorie intake.”

Expansion of the Social Welfare Sector

59

marked by calls for a scaling back of government intervention in the economy. And yet fears of (or hopes for) a “shrinking state” under liberalization were largely unwarranted. Since the 1990s, the size of the Indian government has in fact grown, keeping pace with the rapid growth of the economy.7 The central government’s share of GDP the in mid-2000s (more than a decade after liberalization), for example, was almost three times larger than in 1960 (the heyday of state-led planning).8 Many, though, have noted the difficulties of maintaining commitments to social welfare in a climate of neoliberal reform – particularly in poorer, less developed contexts.9 Indeed, India’s social spending remains comparatively low and, by many measures, is sorely inadequate.10 In relative terms, though, India’s social welfare sector has, in fact, expanded quite dramatically. As Nayar has observed, “far from the dismantlement of the welfare role of the state, the expenditures by the state speak of a greater effort on its part at enhancing it.”11 These trends are visible in Figures 3.1–3.3, which depict real per capita spending (adjusted for inflation and population growth) by the state and central governments in the areas of education, health, and social security. Spending on education and health each more than doubled in real terms from the late 1970s to the mid-2000s, while spending on social security in the form of pensions and social assistance for the elderly, the poor, and the unemployed also rose more than fourfold in the same period. These trends intensified in the mid- to late 2000s. In just the two years between 2006 and 2008, for example, central government spending on core welfare programs increased by almost a quarter.12 Commitments to social rights thus “gained momentum in a policy environment that emphasized state withdrawal from public provisioning.”13 This is reflected, in part, in a change in the official rhetoric surrounding social welfare, which began to shift in the 1990s and 2000s from a focus on “basic needs” to a broader conceptualization of “human development,”14 and from the language of “charity” to that of economic and social rights, visible in a wave of legislation that established,

7

8 9

10 13 14

Nayar 2009. Since the 1980s, public expenditures by the Indian central and state governments have remained relatively stable as a share of GDP, accounting for 21.6 percent in 1981, 24 percent in 1991, and 23 percent in 2001. This, however, obscures a dramatic twofold increase in absolute spending in the decade of the 1990s, given the rapid growth of the economy at the time. Ibid., pp. 82–83. Spending is in constant 1993–1994 prices. See, for example, Rudra (2003); Seekings (2005); and Sandbrook et al. (2007). In advanced industrial settings, in contrast, scholars have noted expansions in social spending alongside economic liberalization (Rodrik 1997; Lindert 2004). 11 12 Drèze & Sen 2013, p. 66. Nayar 2009, p. 89. Ibid., p. 37. Jayal 2013, p. 164. Mooij and Dev 2004, p. 100. The broader notion of human development encompasses investment in the areas of education, housing, health, and basic infrastructure.

The Institutional Terrain of the State 600 500 400 300 200 100 0

1978–79 1979–80 1980–81 1981–82 1982–83 1983–84 1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–97 1997–98 1998–99 1999–00 2000–01 2001–02 2002–03 2003–04

Rs. per Capita (constant prices)

60

figure 3.1 Education Spending: Real Per Capita Central and State Expenditures (constant 1993–1994 prices)

120 100 80 60 40 20 0 1978–79 1979–80 1980–81 1981–82 1982–83 1983–84 1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–97 1997–98 1998–99 1999–00 2000–01 2001–02 2002–03 2003–04

Rs. per Capita (constant prices)

Source: Nayar (2009), citing National Accounts Statistics.

figure 3.2 Health Spending: Real Per Capita Central and State Expenditures (constant 1993–1994 prices) Source: Nayar (2009), citing National Accounts Statistics.

among others, the right to work (National Rural Employment Guarantee Act, 2005); the Right to Education (2009); and the right to food (National Food Security Act, 2013).15 15

Jayal 2013, pp. 174–175. Jayal attributes this explosion of legislation in part to a national climate of economic prosperity and aspiration. She writes: “The presence of Indian billionaires who figure on the lists of the world’s richest individuals, and the aspirations of the nation as a

61

140 120 100 80 60 40 20 0

1978–79 1979–80 1980–81 1981–82 1982–83 1983–84 1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–97 1997–98 1998–99 1999–00 2000–01 2001–02 2002–03 2003–04

Rs. per Capita (constant prices)

Expansion of the Social Welfare Sector

figure 3.3 Social Security Spending: Real Per Capita Central and State Expenditures (constant 1993–1994 prices) Source: Nayar (2009), citing National Accounts Statistics.

The implementation, beginning in 2006, of the National Rural Employment Guarantee Act (NREGA, also referred to as MGNREGS16) in particular signaled a “qualitative leap” in the design of social programs “matched with massive resources.”17 The national budget for MGNREGS in 2010, for example, reached Rs. 401 billion (roughly $8.9 billion USD), accounting for 3.6 percent of all government expenditures and roughly 0.5 percent of total GDP.18 MGNREGS has also created a multiplicity of new points and modes of contact between citizens and the state, including: job sites (where workers encounter officials); written documentation (in the form of job cards and “muster rolls” that record hours worked); bank accounts (where payments are directly deposited); and – in some locations – platforms for citizen oversight in the form of social audits. MGNREGS, in other words, has broadened the terrain upon which rural citizens encounter the state.

16

17 18

global economic superpower, makes abysmal poverty aesthetically and morally offensive.” Harriss (2011, p. 128), noting similar trends, suggests the existence of a Polanyian “double movement” both toward and in resistance to economic reforms. Just as Karl Polanyi (1944) argued that the “great transformation” of Europe’s markets in the early twentieth century was met by a protective countermovement, Harriss argues that similar forces are at work in India, where the “neoliberal project … is tempered by … constitutional design and state tradition, as well as by social movements … and by popular democracy.” The full name of the program is the Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS). World Bank 2011, p. viii. Sukhtankar 2012, 2017. For point of comparison, total welfare spending for the Temporary Assistance for Needy Families (TANF) program in the United States in 2013 was 0.19 percent of GDP.

62

The Institutional Terrain of the State

The funds allocated to MGNREGS make it one of India’s biggest social programs (and, indeed, one of the largest of its kind in the world).19 And yet, MGNREGS is exceeded in both its reach and expenditures by the Targeted Public Distribution System (TPDS), which distributes food and non-food items to those living below the poverty line, covering more than 23 percent of Indian households and operating with an annual budget of $9.7 billion USD.20 MGNREGS and TPDS, while among the highest profile, are accompanied by a veritable alphabet soup of other central government programs. These include, among many others, the Sara Shiksha Abhiyan (SSA, or the “Education for All” campaign), the National Rural Health Mission (NRHM), Pradhan Mantri Gram Sadak Yojana (PMGSY, funding rural roads), the Midday Meals Scheme (MDMS, funding lunches in schools); Indira Awaas Yojana (IAY, funding rural housing); and the Indira Gandhi National Old Age Pension Scheme (IGNOAPS). Centrally sponsored welfare schemes of this kind proliferated in the 1990s, rising from 190 programs in the National Planning Commission’s Fifth Plan (1974–1979) to a peak of 360 in the Ninth Plan (1997–2002).21 The resources allocated to these schemes are substantial. The World Bank, for example, calculated that the total combined value of the central government’s major welfare programs in 2004 was equivalent to Rs. 9065 (or $135 USD) per rural household – an amount representing roughly 40 percent of the rural poverty line.22 Another estimation suggests that total spending on centrally sponsored welfare schemes, coupled with food, fertilizer, and fuel subsides, would be enough (if allocated directly to poor households) to eliminate rural income poverty.23 There is no doubt that these investments have been accompanied by real improvements in social welfare: from 1990 to late the 2000s life expectancy in India rose from 58 to 65 years, while infant mortality rates fell from 81 to

19 20

21

22

23

Jenkins & Manor 2017. World Bank 2011, pp. x–xiv, citing 2010 data. TPDS coverage has in fact expanded since the passage of the 2013 National Food Security Act, to include 75 percent of the rural and 50 percent of the urban population. Chaturvedi 2011. The share of these Centrally Sponsored Schemes as a portion of the central government’s gross budgetary support to the Central Plan increased during the same time period, from 31 percent to 42 percent. This reflects large outplays for major programs such as MGNREGS (employment), SSA (education), NRHM (health), and PMGSY (rural roads). Also see Shagal (2012) and Gurtoo and Udayaadithya (2014). World Bank 2011, p. xxi. This includes spending across ten programs in the areas of food distribution, rural employment, credit, pensions, housing, school meals, school enrollment stipends, and health, life, disability, and social insurance. Kapur, Muhkopadhyay, and Subramaniam (2008, pp. 37–38) calculate that total expenditures on centrally sponsored schemes and subsidies in 2007–2008 reached $43.5 billion USD. This amount, if it were equally divided across India’s estimated 70 million households below the poverty line, would exceed the official rural poverty line.

Expansion of the Social Welfare Sector

63

47 per 1,000 live births.24 It is critical, though, not to oversell this progress. Even causal observation quickly reveals a troubling gap between social expenditures and outcomes. Neighboring Nepal and Bangladesh have seen faster human development gains in the same period of time – catching up to or surpassing India on many social indicators, despite having lower per capita income. In fact, in an analysis of the world’s sixteen poorest countries outside of sub-Saharan Africa, India ranks tenth or worse on a wide range of social indicators from infant mortality, to literacy rates, to child malnourishment.25 This gap between social spending and welfare outcomes is widely observed, and is the subject of considerable study. Scholars have attributed it to public sector inefficiencies and limited state capacity,26 and to rampant corruption at both a macro and micro scale.27 For example, by the central government’s own estimates, over one-half of grains in the TPDS system never reach their intended beneficiaries due to poor implementation and graft.28 MGNREGS is similarly plagued by corruption, although rates of leakage are difficult to calculate.29 A substantial literature similarly documents the diversion of resources by local elites or their capture within partisan and caste-based patronage networks.30 Other studies emphasize bureaucratic norms that fail to foster accountability and responsiveness, or chaotic administrative cultures that operate in an information-constrained environment.31 Whatever the (many) culprits, it is clear that high levels of human deprivation persist – even as the India’s social welfare apparatus expands. Rural citizens’ experience of the Indian welfare state is thus an uneven one: the state is both present and absent, visible and elusive, and critical and capricious in the lives and livelihoods of the poor. The puzzle, as Gupta aptly notes, is thus “not only why government programs aimed at providing nutrition, employment, housing, healthcare, and education to poor people do not succeed in their objectives but also why, when they do succeed, they do so unevenly and erratically. Why, among the universe of beneficiaries, do some people manage to receive assistance and others not?”32

24 25

26 27 28 29

30

31

Drèze & Sen 2013, p. 55. Drèze & Sen 2013. India has the highest prevalence of undernourished children of anywhere in the world – double that in all of sub-Saharan Africa combined. Kapur & Muhkopadhyay 2007; Pritchett 2009; World Bank 2011; Joshi 2017. For a review of corruption in India, see Sukhtankar and Vaishnav (2015). World Bank 2011; Joshi 2017. See Sukhtankar 2017. In one national study, Imbert and Papp (2011) finds that survey respondents can account for between 42 to 56 percent of days of work officially recorded under MGNREGS. Another study in the state of Andhra Pradesh suggests rates of leakage of 30 percent (Muralidharan et al. 2016). Banerjee 2004; Besley, Pande, & Rao 2004; Chandra 2004; Wilkinson 2007; Dunning & Nilekani 2013. 32 Gupta 2012; Witsoe 2014; Mangla 2015. Gupta 2012, p. 24, emphasis added.

64

The Institutional Terrain of the State

social welfare provision in rajasthan Responsibility for social welfare provision lies primarily with India’s states. Aggregate levels of government spending are thus even higher when including state-level programming, which accounts for as much as 80 percent of socialsector outlays.33 The states, though, are heavily dependent on fiscal transfers from the central government.34 These take several forms, including regular transfers to support the states’ planned budget expenditures; discretionary (unrestricted) transfers for “non-Plan” expenditures; and transfers for the implementation of Centrally Sponsored Schemes (CSS), which are nationally funded programs implemented by the states. To grasp the scope and reach of India’s social welfare sector, it is therefore necessary to examine the intersection of central and state programming. The state-level trajectory of social welfare provision in Rajasthan mirrors the national trends documented above, moving upwards – in particular since the 1990s. This, though, in large part reflects Rajasthan’s very low starting point. Commitments to social welfare under princely rule in the colonial era were negligible. Each princely kingdom, as noted in Chapter 1, was divided into smaller fiefdoms ruled by landlords (jagirdars) who exercised almost absolute power within their localities (jagirs). As Sharma describes, there was little if any obligation for the local landlord to “provide any public facilities to his jagir people – so much that the jagirdar in the interior areas did not even give facilities for ‘crude’ drinking water or dispensaries.”35 Pre-Independence levels of social spending were thus minimal if not nonexistent. There was, in other words, “an almost complete absence of state or societal action in the social [welfare] sphere.”36 Princely Rajputana, consequently, had among the worst social indicators in all of India. Over 90 percent of the population remained illiterate, and almost one-third of children died before the age of two – rates significantly higher than in other parts of India at the time. From these bleak initial conditions, social spending in postcolonial Rajasthan began to slowly increase and, with it, indicators of social development also began to improve. In her comparative study of welfare provision across Indian states, Prerna Singh offers a detailed account of these changes over time, noting

33 34

35

Mooij & Dev 2004, p. 105. The fiscal architecture of Indian Centre–State relations has shifted since the time of research for this book in the late 2000s. Following the 2014 national election, Prime Minister Modi moved to abolish India’s long-standing Planning Commission, replacing it in 2015 with the NITI Aayog (National Institution for Transforming India). Part of the intention in this shift was to move away from a central planning model to one that places greater responsibility on the states. These dynamics – which imply a further devolution of social policy design and implementation (but not necessarily funding) to the states – are discussed in the concluding chapter. The bulk of this book, however, examines claim-making in the pre-2015 context, when social policy fell largely under the purview of the state and national Planning Commissions. 36 Sharma 1993, cited in P. Singh (2015, p. 176). P. Singh 2015, p. 176.

Social Welfare Provision in Rajasthan

65

that issues of health, education, and other social services began to gain traction in Rajasthan’s state budgetary debates in the 1950s.37 Social investments, consequently, began to rise, and by the 1980s, Rajasthan was outspending other similarly poor states in northern India. In fact, within the poorest (socalled “BIMARU”) states at the time, Rajasthan emerged as a leader in social welfare spending.38 This trend becomes clear when comparing Rajasthan to its neighbor Uttar Pradesh (UP), which, at the time of Independence, was in a better position economically and institutionally than Rajasthan. UP was known for its productive agriculture and developed infrastructure, as well as for its effective bureaucratic administration – so much so that it was considered a “model” for other provinces under British control. Rajasthan, in contrast, was widely considered a “backward” region, infamous for its droughts, food shortages, famines, and its highly fragmented princely rule.39 Despite these considerable disadvantages, Rajasthan pulled ahead of UP in social spending in the postcolonial period. Rajasthan’s expenditures on education as a share of state GDP (3.2 percent) was, by the end of the 1960s, double that in UP (1.5 percent). In the same time period, Rajasthan allocated two and a half times more per capita on health spending than UP.40 These relative gains in social spending continued in Rajasthan through the 1980s. Importantly, though, Rajasthan still lagged behind India’s forerunners in commitments to social welfare. Rising social expenditures, moreover, did not yield proportional improvements in welfare outcomes, and Rajasthan remained at the bottom of the pack in illiteracy and infant mortality. While literacy did improve, it paled alongside national trends – leaving Rajasthan in 1971 with the highest proportion of illiterate people (at almost 80 percent) in the country. Rajasthan remained at or near the bottom through 1991, by which point literacy had risen to just 39 percent (Figure 3.4). As in the rest of India, the 1990s were also a turning point in Rajasthan, visible in accelerated social spending and in improved social welfare outcomes. As Krishna observed in the late 1990s, “Instead of being dismantled or streamlined, … state intervention … has been intensified in the name of rural development and poverty reduction.”41 Figures 3.5 and 3.6 depict Rajasthan’s planned state expenditures from 1951 to 2012. Total planned expenditures cover all categories included in Rajasthan’s Five-Year Plans. Within this, social welfare spending is concentrated in two budgetary categories: “social services” (including health, education, housing, and poverty alleviation) and “rural development” (which encompasses basic infrastructure for services such as water and sanitation, along with rural credit and 37 38

39

Ibid., p. 177. “BIMARU,” which also means “sick” in Hindi, is an acronym referring to the states of Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh, which together comprised India’s most pronounced poverty belt. 40 41 P. Singh 2015, p. 179. P. Singh 2015. Krishna 2002, p. 43.

The Institutional Terrain of the State

66 90 80 70

Literacy Rate (%)

60 50 40 30 20 10 0

1951

1961 RJ Male

1971

1981 RJ Female

1991 RJ Total

2001

2011 India

figure 3.4 Literacy Rates in Rajasthan and India (1951–2011) Source: Census of India.

agricultural extension programs).42 Figure 3.5 portrays real per capita spending for both categories, while Figure 3.6 shows total social sector spending (for social services and rural development, combined) as a proportion of all planned state expenditures. Overall levels of state spending have risen over time. Within this, the share of funds allocated to the social sector has fluctuated, with a dramatic spike beginning in the early 1990s. This increase in social spending has had real, material effects. Between 1991 and 2001, Rajasthan enjoyed the single largest increase in literacy in post-Independence Indian history, rising from 39 to just above 60 percent (see Figure 3.4).43 Progress continued in the 2000s, although at slower rates; by 2011, statewide literacy had risen to 66 percent (still lagging, 42

43

MGNREGS expenditures, which is centrally sponsored with contributions from the states, is reported separately – as discussed later in this chapter. While Rajasthan still lagged behind the national average, the gap between Rajasthan’s and India’s average literacy rates closed from fifteen to just three percentage points between 1991 and 2001. Male literacy rates in Rajasthan in fact exceeded the national average from 1991 onwards. Rajasthan’s state average, though, has been dragged down by continued lagging performance in female literacy.

Social Welfare Provision in Rajasthan

67

400

Rs. per Capita (constant prices)

350 300 250 200 Rural Development Social Services

150 100 50

2 –1

–0 20

07

7

2 –0

02 20

7 –9

97

19

2 –9

92 19

1 –9

91 19

–9

19

90

0

5 –8

85 19

0 –8

80 19

9 –7

79 19

4 74 19

9

–7

69

19

6 19

66

–6

1 19

61

–6

–6 56

19

19

51

–5

6

0

figure 3.5 Social Sector Spending in Rajasthan: Real Per Capita Planned Expenditures for Rural Development and Social Services (1993–1994 constant prices) Source: Economic Report of Rajasthan (2016).

though, behind the national rate of 74 percent). Rajasthan also recorded advances in health, although the trajectory has been less consistent. From the early 1980s to the 1990s, the state experienced declines in infant mortality that closely mirrored national trends, dropping from 108 deaths per 1,000 live births in 1981 to 79 death per 1,000 births in 1991. The decade of the 1990s, though, saw very little progress – and Rajasthan fell behind nationally.44 The gap began to close again in the 2000s: by 2011, Rajasthan recorded 52 deaths per 1,000 live births, compared to 44 per 1,000 nationally (Figure 3.7). The picture of social development in Rajasthan that emerges, in sum, is an unsteady one. Exponential advances in education in the early 2000s have since slowed, and gains in health have been modest at best. Trajectories of social welfare provision are also uneven within the state. Poverty, as we saw in Chapter 1 (Figure 1.4), is concentrated in the more remote southern and western parts of the state. Perhaps most striking, though, is the disparity across social groups. Fifty-eight and 48 percent of the Scheduled Tribe and Scheduled Caste population, respectively, cannot read or write, 44

In all, Rajasthan’s infant mortality dropped by just 6 percent in this decade, compared to 20 percent across India. Progress may have been hampered by intense droughts in Rajasthan in the same decade.

The Institutional Terrain of the State

68

45

35 1000 30 800

25

600

20 15

% of Total Expenditure

40

1200

400 10 200

5

Total Expenditure

2

7

–1 07

20

–0

02

2

0

20

7

–0

97

19

2 –9

92

19

91 19

–9

1 –9

90 19

5

–9

85

0

–8

80

19

9

–8 79

19

4

–7 74

19

9 19

69

–7

6

–6 66

19

1 19

61

–6

6

–6

–5 51

56 19

19

0

Social Spending as % Total

0

19

Rs. per Capita (constant prices)

1400

figure 3.6 Social Sector Spending as Percentage of Total Planned State Expenditures Notes: The right-hand axis shows expenditures on social services and rural development (combined) as a percentage of overall planned expenditure. The left-hand axis shows total planned expenditure per capita (constant 1993–1994 prices). Source: Economic Report of Rajasthan (2016).

compared to just 21 percent among Rajasthan’s General Castes. A full 90 percent of ST and 68 percent of SC residents practice open defecation, compared to 28 percent of GC residents; while just 17 percent of ST households have access drinking water in their homes, compared to 74 percent of GC households.45 Rajasthan’s gender gap, as noted, is also one of the worst in the country. Female literacy rates hover around 50 percent, compared to 79 percent for men (Figure 3.4).46 Notwithstanding these glaring inequalities, the fact remains that the size and reach of Rajasthan’s social welfare apparatus has increased in relative, historical terms. There is, crudely put, “more state” at the local level, visible in an expanding array of welfare programs, an influx of funds, and a corresponding proliferation of state agents and agencies. As the array of services and programs grows then so, too, does the physical presence and visibility of the state. As Fuller and Harriss explain: “A local administrative office, a government school, a police station: to enter any of these is to cross the internal boundary into the 45

World Bank 2016.

46

Among rural women, rates drop to just 35 percent.

Rajasthan in Transition

69

120

Infant Deaths per 1000 Live Births

100

80

60

40

20

0 1975

1980

1985

1990

1995 Rajasthan

2000

2005

2010

2015

India

figure 3.7 Infant Mortality Rates in Rajasthan and India (1981–2013) Source: Economic Survey of Rajasthan (2016).

domain of the state.”47 Citizens thus see and encounter the state more regularly in the course of their daily lives, even as the spoils of Rajasthan’s social investments are unevenly shared. It is this combination of increasing breadth as well as unevenness in public performance, I will argue below, that sets the stage for claim-making.

rajasthan in transition What drove the expansion in social spending, and associated although uneven gains in social welfare in Rajasthan? Trends within the state partially reflect national-level developments: in particular, the rise of social legislation and the subsequent increase in social programing. However, to fully understand the development of Rajasthan’s social welfare sector, these national dynamics need to be viewed alongside the state’s own social and political transformation in the postcolonial (and post-princely) period. Historians of Rajasthan have referred to state’s long history of princely rule as the “Thousand Year Raj,” spanning from as early as the fifth century when 47

Fuller & Harriss 2001, p. 23.

70

The Institutional Terrain of the State

upper caste Rajput landowners (members of the warrior or Kshatriya class) began to establish what came to be known as native or “princely” kingdoms.48 Under Mughal and subsequently British rule, the native kingdoms were loosely stitched together into the single province of Rajputana. The individual states, though, retained substantial autonomy, and princely Rajputana consequently remained largely isolated from the broad political and economic change that swept the subcontinent in the precolonial and colonial periods. Even following Independence in 1947 and formal incorporation into the state of Rajasthan in 1949, the princely kingdoms retained their distinct political identities, which, as Narain and Mathur remark, “survived more than three decades of [postcolonial] socio-economic change and politico-constitutional transformations.”49 Princely Rajasthan, however, was not impervious to change. Far from it, a process of political, social, and institutional transformation has been underway from the early twentieth century onwards. The drivers of change are many. From a social welfare perspective, though, there are three interconnected sets of dynamics that merit particular attention: the rise of a statewide subnational identity; increasing electoral competition; and social movement development. First, the highly fragmented allegiances of the princely states began to give way to a broader “idea of Rajasthan.”50 The emergence of a “Rajasthani” identity, Singh has argued, promoted concern for “collective” social welfare that cut across more localized notions of “us” versus “them” associated with jagir or caste.51 This, in turn, provided the social solidarity and political grounding necessary for new policies and investments to promote social development within the state – for example, in education and in health. The development of a subnational Rajasthani identity can be traced to early challenges in the 1900s to Rajput dominance by other castes, manifest in calls for an abolition of princely rule and the jagirdari system and demands for representative government.52 “Coming together as ‘Rajasthani’,” Singh writes, “allowed the challenger elite to present themselves as the legitimate ‘modern’ leaders of a united, democratic 48 49

50

51 52

Narain & Mathur 1989. Narain & Mathur 1989, p. 23. The princely states, they write (p. 2) “maintained their political identity and politico-economic isolation from the rest of India.” Björkman and Chaturvedi (1994, p. 138), similarly observe that “Rajputana experienced very little of the political organization and mass participation in the independence movement characteristic of areas directly ruled by the British.” Schomer et al. (1994). Narain and Mathur (1989, p. 23) highlight this trend, noting a “slow but steady emergence of an overarching identity of Rajasthan … transcend[ing] earlier [princely] loyalties.” P. Singh 2015, p. 38. These castes included the Brahmins, who occupied positions of high social status, along with economically dominant Mahajans with extensive trade networks, and elite members of the Jat peasant caste (in particular those who served in the military and were deployed elsewhere in India or abroad). As members of the non-Rajput elite began to travel to other parts of India, they gained exposure to new sets of political and social ideas. Upon their return to Rajasthan, these elites began to challenge Rajput hegemony.

Rajasthan in Transition

71

Rajasthan …”53 However, these early unifying efforts remained elite dominated and were fragmented by caste and urban–rural divides.54 In the 1980s and 1990s, though, this began to shift, giving way to an emerging pan-Rajasthani identity of a more popular nature – helped by the fact that a majority of residents, by this time, were born in the independent state of Rajasthan and thus lacked the same princely or jagirdari attachments as older generations. Second, and simultaneously, intensified political competition strengthened subnational allegiances as political parties began to draw on the rhetoric of Rajasthani unity in an effort to establish broader bases of support. Following Independence, Rajasthan, like most states, was part of the “Congress Raj” system, and the Congress party remained the electorally dominant party through the 1980s. Key among the opposition to Congress was the Bharatiya Janata Party (BJP), which came to power in Rajasthan for the first time in 1989. The Rajasthani BJP ran on a platform of what Jenkins has called “Rajput Hindutva,” (referring both to the state’s Rajput heritage and to broader Hindu nationalism) which, somewhat paradoxically, helped spur a sense of crosscaste, statewide unity (while further polarizing Hindu–Muslim relations). The BJP, he notes, “managed to link region, caste, and Hindu nationalism” in a way that united support around an idealized notion of Rajput values, even among the lower castes.55 Through “grassroots developmental activities,” the BJP sought to alter its image as a party of the upper castes, reaching out to tribal and lower-class groups as well as those among the Other Backward Classes (OBCs). These activities were often carried out through “social” organizations affiliated with the Rashtriya Swayamsevak Sangh (RSS) – a Hindu nationalist movement with close ties to the BJP.56 This was particularly common in the southern, tribal regions of Rajasthan, where RSS-affiliated groups helped to secure electoral support for the BJP.57 The BJP in Rajasthan thus acted more like a regional than a national party, espousing and activating a sense of

53 55

56

57

54 P. Singh 2015, p. 104. Björkman & Chaturvedi 1994. Jenkins 1998, p. 104. Rajput history and identity, Jenkins notes, were presented as an idealized manifestation of “traditional” martial Hindu values, providing a “code of conduct which cuts across religion, class, and caste” (p. 110). Key among these values was the “preservation of a stable social order through Rajput rule” (p. 109). P. Singh (2015) similarly refers to the Rajasthani BJP as a “sub-nationalist” party that draws heavily on a sense of “Rajputized” Rajasthani identity (in contrast to the heavily “Sankritized” Hindu identity, built around Brahmin values, visible in other manifestations of Hindu nationalist ideology). The RSS is the central organization within a broader “family” of Hindu nationalist organizations referred to as the Sangh Parivar, which draws together vast networks of local volunteers who engage in service and cultural activities, but which also possess characteristics of a paramilitary organization. Jenkins 1998, p. 101. In a broader study that extends across India, Thachil (2014) similarly argues that elite parties like the BJP can secure the support of poor voter through a strategy of social service provision that is implemented through a network of affiliated grassroots social organizations.

72

The Institutional Terrain of the State

“Rajasthani subnationalism.”58 Since the 1990s, the BJP and Congress have successively traded power within the state. In this increasingly competitive environment, both parties make cross-caste appeals to broaden their base of support. Both parties, consequently, trade in the language and imagery of panRajasthani identity, and both have overseen expansions in social spending intended to promote collective Rajasthani welfare. The rise of subnationalism and intensification of party competition coincided with a third development in Rajasthan: the rise of mass social movements. Through the mid-1980s, notions of pan-Rajasthani identity “remained almost entirely an elite discourse … due to the absence of a powerful, broad-based subnational movement or association.”59 Beginning in the 1980s, though, a nascent subnationalist movement began to emerge, manifest in large-scale popular rallies to demand the formal recognition of “Rajasthani” as a language.60 At the same time, a very different kind of movement was also taking root in Rajasthan, focused on class solidarities. Interest-based social movements of this kind, which coalesced around issues of economic rights rather than identity, have powerfully reshaped the social welfare arena in India – with strong roots in Rajasthan.61 In the early 1990s, anti-corruption activists – empowered by a moment of public scrutiny in response to a series of highprofile scandals – began to organize locally across India. Their aim was to shed light on corruption in public programs, and to do so through the activism of ordinary and poor people who bore the costs of such corruption. The Rajasthan-based Mazdoor Kisan Shakti Sangathan (MKSS, or the Workers and Farmers Empowerment Organization) was at the center of this movement. MKSS describes itself as a “non-party” people’s organization that campaigns for accountability, transparency, and the realization of social and economic rights. In the early 1990s, it began to organize to demand payment of back wages owed to laborers who had worked on government worksites as part of a drought relief program. In their efforts to access local records from worksites, 58

59 60

61

P. Singh 2015, p. 108. Jenkins (1998, p. 103) similarly notes that “the Rajasthan BJP itself functions like a regional party” in which regional, Rajput identity played a prominent role (unlike the BJP in other parts of the country, where it emphasized broader notions of Hindu nationalism). P. Singh 2015, p. 107. Lodrick 1994. Rajasthan is home to at least nine linguistic groups and many more local dialects. All are related to Hindi – the official language of Rajasthan. Linguists disagree on whether the vernaculars spoken in Rajasthan are in fact dialects of Hindi (as they were classified early on in the colonial administration) or whether they together constitute a distinct language (with its own variants and dialects). Ray and Katzenstein (2005) distinguish between “identity-based” and “interest-based” movements in India, arguing that they occupy different political domains. They note that interestbased movements focused on social and economic issues and poverty alleviation have generally been less effective in penetrating the electoral arena than “identity” based movements of caste, religious, regionalist, or nationalist nature. As a result, interest-based movements have instead sought traction in the courts and bureaucracy at state and national levels.

Rajasthan in Transition

73

the MKSS pioneered new strategies, carrying out social audits and holding public hearings (jan sunwai) in which laborers themselves could examine and attest to the veracity of government records. This, in turn, paved the way for public interest litigation that resulted in new legislation, including Rajasthan’s Right to Information Act of 2000 and, subsequently, a national version of the law in 2005.62 In the early 2000s, building on this success, MKSS, along with other organizations such as the People’s Union for Civil Liberties (PUCL), began to mobilize around the rights to food and to work – seeking to both expand and institutionalize the drought and famine relief programs that they had spent the last decade monitoring. The national Right to Food campaign was born out of such activism in Rajasthan, following litigation by the PUCL in which they argued that hunger related deaths in Rajasthan were a rights violation. The Supreme Court of India recognized the right to food and issued a series of orders to expand the ruling beyond Rajasthan.63 This in turn set the stage for new nutrition-related welfare programs at the levels of the Centre and the states. The National Rural Employment Guarantee Act can similarly trace its roots to social movement activism in Rajasthan. As Jenkins and Manor write: “It was in central Rajasthan that the MKSS … and other activist groups alternately fought against, worked with, upbraided, advised, and humiliated by force of counter-example the state’s senior civil servants and leading politicians from across the political spectrum” to demand the right to work.64 Rajasthan thus emerged as an epicenter for a national movement for social and economic rights. MKSS played a particularly central role, coordinating a statewide coalition of people’s organizations and linking them to broader national networks. These organizations, while led by a relatively narrow band of activists, mobilized support across caste lines – evident in episodic mass rallies and protests organized at the state and national levels. Social movement organizations have also played a key advocacy role, reflected in their ability to capture regional and national media attention, as well as their ability to garner support among influential officials and, in particular, in the courts.65 To the extent that these movements have influenced state and national debates over social welfare, their organizing efforts have trickled back to the grassroots in Rajasthan in the form of new programs and spending. 62 63

64

65

Jenkins & Goetz 1999; Jenkins 2007. Birchfield and Corsi 2009; Hertel 2015. See also Robinson (2009) and Jayal (2013) on the rise of the activist court in India. Jenkins & Manor 2017, p. 35. On the importance of Rajasthan to the development of NREGS, also see Drèze and Khera (2010). See, for example, Jenkins and Goetz (1999, p. 619), who note that much of MKSS’ success and influence stems from “its skill in developing a network of support within the elite IAS [Indian Administrative Service], among Delhi-based intellectuals and activists, and within the regional and national media.” Also see Jenkins and Manor (2017) on the ties between MKSS and senior officials in the national-level Congress party.

74

The Institutional Terrain of the State

The 1990s, in sum, was a moment of profound but incomplete change in Rajasthan. The regional isolation and parochial allegiances that were the hallmarks of the “Thousand Year Raj” began to erode – although legacies from this period clearly remain. A subnational “Rajasthani” identity began to emerge, promoted by political parties and adopted, eventually, by large segments of the population – although this, too, is nascent. Identity- as well as interest-based social movements also took root, promoting cross-cutting class solidarities among the poor in pursuit of greater accountability in the local administration of social welfare provision – although these are relatively thin on the ground. The cumulative result of these various developments is an expansion of the size and reach of Rajasthan’s social welfare sector. This process, though, is far from complete as the state’s social welfare indicators – while vastly improved – still lag behind national averages.66

going local At the same time that social spending expanded, the arms of the state also pushed deeper into the local arena – both through a proliferation of government programs implemented at the local level and through political and administrative decentralization. In 1993, the Indian Parliament enacted the SeventyThird Amendment to the Indian Constitution, mandating local self-governance through three tiers at the district (zilla), sub-district (samiti), and village (gram) levels. The Gram Panchayat (GP) is the most local elected unit. The SeventyThird Amendment formally recognized over 200,000 of these village-level bodies nationwide, including over 9,000 in Rajasthan alone. Today there are almost 10,000 Gram Panchayats in Rajasthan, each home to an average of 5,000 people. Each panchayat has anywhere from five to twenty elected members, depending on population size. These include the panchayat president (sarpanch) and ward members (wardpanchs), all of whom are directly elected by universal franchise every five years. The Seventy-Third Amendment further required the reservation of elected seats for members of the Scheduled Castes and Tribes (in proportion to their share of the local population) and for women (at a minimum of 33 percent and up to 50 percent in some states, including

66

P. Singh (2015) compares levels of levels of social development in four states – Rajasthan, Uttar Pradesh, Kerala, and Tamil Nadu, linking them to varied levels of subnational identification within four states in the same four states. Rajasthan, for Singh, is an intermediate case: it has progressed from a deeply fragmented princely region with some of the worst social welfare indicators in the country to a state with an incipient subnational identity and much improved outcomes in collective social welfare. Rajasthan, though, lags far behind both Kerala and Tamil Nadu, in terms of levels of both subnationalism and social development. She writes (p. 176): “In both absolute terms as well as relative to other Indian states, … Rajasthani subnationalism remains low and the state is correspondingly among the most socially backward of Indian states.”

Going Local

75

Rajasthan). Millions are thus voted into local office every five years, including large numbers of women and members of the lower castes and tribes. Long before the Seventy-Third Amendment, Rajasthan was among the first states in the country to take concrete steps to institutionalize local governance, establishing local elected bodies for the first time in 1959. That Rajasthan was a first mover came as something of a surprise to many who, given the state’s princely history, considered it to be “least prepared for an experiment in democratic decentralization.”67 And yet the very localism of princely rule in fact laid the groundwork for a new system of local governance following Independence. The empowerment of the Gram Panchayat was possible given a “strong sense of historical continuity shared by former princely status” and, in particular, a tradition of local allegiances and dependencies.68 The depth of democratic decentralization, however, was curtailed by this same princely legacy. Most people did not directly approach members of the panchayat and instead, when faced with the need, were more likely to turn to the local thakur (landlord) – reflecting the continued salience of feudal, jagirdari identities. The power of the panchayat was further eroded by caste division and factionalization,69 and by intervention by state-level politicians and bureaucrats who, threatened by the emergence of a new local political class, sought to restrict the panchayats’ responsibilities and funding.70 The status of these early panchayats thus quickly declined, rendering them little more than “paper tigers” in the eyes of many observers.71 The 1990s were again a turning point, as the Seventy-Third Amendment, gave new constitutional muscle to local governance. While the core elements of the amendment were mandatory, implementation was left to the states. States, accordingly, have varied in their commitment to the spirit – if not the letter – of the law. Rajasthan was, once again, a leader in implementation, moving quickly to organize panchayat elections and to adhere to other provisions such as caste and gender reservations. The first round of local elections under the new Panchayati Raj regime were held in Rajasthan in 1995 across 9,185 Gram

67 69

70 71

68 Björkman & Chaturvedi 1994, p. 132. Ibid. Clashes over land between the Jat peasant castes (empowered by land reform) and Rajputs (who, under princely rule, owned the bulk of land) “bedeviled the workings of Panchayati Raj, which required for its functioning either cooperation among contenting groups or the ability for one … to establish hegemonic domination of over the others” (Björkman & Chaturvedi 1994, p. 145). Ibid., p. 151. Mathew 1994, p. 36. Panchayat elections were regularly postponed beginning in the mid-1960s, and no local polls were held for thirteen years. Rajasthan mirrored national trends in this regard. Central government and state politicians across the country distanced themselves from the Panchayati Raj model in the 1970s and 1980s, instead embracing increasingly centralized forms of governance focused on large-scale (state-administered) programs for rural development (including Green Revolution efforts to increase agricultural production). The same period saw a nationwide reconcentration of political power in the hands of the central government, culminating in the declaration of Emergency Rule by Indira Gandhi in 1975–1977.

76

The Institutional Terrain of the State

Panchayats. Four more rounds of elections in 2000, 2005, 2010, and 2015 have followed to date. Starting in 1995, one third of all seats in Rajasthan’s GPs were reserved for women, rising to 50 percent in 2010. The aim was also to reserve seats for the Scheduled Castes and Tribes in proportion to their share of the district population. Rajasthan initially fell short of this goal, reserving just 80 percent of the seats that should have been set aside for SC and ST members, although this has improved in recent years.72 Rajasthan also set aside seats for the OBCs in proportion to their population share.73 However, the substantive depth of decentralization initially remained shallow. The panchayats were particularly hampered by their dependency on higher levels of government for funding. While nominally given the power to raise its own revenue through local taxes and user fees, most panchayats are severely restricted in their ability to do so. In the mid1990s, for example, Rajasthan’s panchayats reported yearly own revenues of just four rupees per capita.74 While the flow of central and state funds to the village increased over time, the share of revenue that is generated locally – and over which the panchayat has direct control – has remained vanishingly small.75 Rajasthan’s panchayats were also hindered by elite opposition and by continued mistrust of the panchayat by the broader population. State-level politicians, fearing local challenges to their authority, moved to limit the administrative reign of the panchayats. State legislators (MLAs) were given ex officio panchayat membership thus allowing them to interfere directly in local affairs.76 State legislation also vested substantial powers in the District Collectors (centrally appointed bureaucrats), enabling them to suspend or remove an elected local official, dissolve an entire local body, or countermand the orders of the panchayat. District officials can also intervene in the financial affairs of the panchayat and are given veto power over the local budget.

72

73

74 75

76

Chaudhuri 2006, p. 171. The relative representation ratio of SC and ST members in Rajasthan has hovered around 0.8, where 1 would indicate fully proportional representation. This, however, exceeds performance in a number of other states, including its neighbors Uttar Pradesh (0.6), Gujarat (0.5), and Haryana (0.2). The Seventy-Third Amendment left decisions about whether and how to reserve seats for OBCs to the state legislatures. Chaudhuri 2006, p. 181. See, for example, Mullen (2012) and Chauchard (2017), who reports that 95 percent of panchayat resources in Rajasthan come from central and state transfers, including support from the National Finance Commission and Rajasthan State Finance Commission for the implementation of state and centrally sponsored schemes. The MLAs also administer Local Area Development funds allocated for discretionary spending on public projects within their constituencies. These funds are allocated to the same twenty-nine areas that are the mandated purview of the panchayat. Some have suggested that these funds were established to undermine the GP and to allow party politicians to retain their control over local patronage (Singh 2009; Wilkinson 2007).

Going Local

77

The panchayat, as a result, is often viewed – at least in scholarly circles – as an “arm of the state”77 that is “accountable to bureaucrats rather than to the electorate.”78 Krishna (2002), drawing on data from Rajasthan collected in 1997 and 1998 (three years after the first panchayat elections), portrayed a rather bleak picture. The Gram Panchayats, he wrote: … are equipped to provide downward communication on behalf of the government, but they are not very effective for transmitting villagers’ demands and grievances upward to public officials … Furthermore, because panchayats are not well grounded in the norms and mores of village life, and control from below is virtually non-existent for these bodies, corruption is a frequent problem.79

They were, in other words, “hampered by their institutional design.”80 Deepening Local Governance At the time of research for this book, fifteen years had passed since the first round of panchayat elections in Rajasthan. The portrayal of a circumscribed local governance environment still rang partially true. And yet fieldwork revealed that a large majority of citizens – including the poor, the lower castes, and women – did in fact, turn to the Gram Panchayat when making claims on the state (a pattern that we will see in detail in the next chapter). The panchayats, I found, had become more deeply institutionalized as well as more visible in and relevant to local communities. This, I argue, is the product of several intertwined processes, each discussed in turn below, including: (1) increased proximity (both physical and social) of elected officials to their constituents; (2) greater flows of public resources to the local level; (3) a rise in the political and social power of the panchayat; and (4) increasing competitiveness in panchayat elections. First, local government has moved “closer to the people.” Those who are elected to local office can, in fact, get no closer: they are in government, gaining firsthand experience within the panchayat. All in all, well over 400,000 individuals served in Gram Panchayats in Rajasthan between 1995 and 2010. At the time of research, Rajasthan had completed four rounds of panchayat elections, cumulatively having electing more than 36,000 sarpanches across the state.81 Among these, over 6,000 Scheduled Caste, 4,000 Scheduled Tribe, 77 79

80 81

78 Singh 2009, p. 399. Jayal 2006a, p. 18. Krishna 2011, p. 110. Crook and Manor (1998) similarly lament the highly centralized control of the panchayat by higher-level officials and politicians – a theme echoed by Wilkinson (2007) and Dunning and Nilekani (2013). Krishna Op. Cit. Data from the Rajasthan State Election Commission. The estimated 36,000 sarpanches are assumed to be unique individuals, since the rotation of caste and gender reservations (established by the Rajasthan Election Commission) imposes de facto term limits on most candidates. In calculating the number of sarpanches cumulatively elected by 2010, I use a low estimate of over

78

The Institutional Terrain of the State

and 13,000 female sarpanches were elected.82 These numbers double if we include elected deputies (up-sarpanches) and increase by an order of magnitude if we consider ward panch seats (averaging ten in number per panchayat). With each successive election, the number of individuals with direct experience serving as a member of the panchayat grows by leaps and bounds. In my citizen survey, for example, a full 15 percent of respondents had personally held or had a close family member who had held an elected position within the Gram Panchayat. For the majority of villagers who have not served in local office, the panchayat is also becoming a less distant institution. While state- and national-level politicians operate at a level removed from the village, GP members come from and live in the same villages and neighborhoods. They, in all likelihood, attended the same schools, work in the same enterprises, and move in similar social circles as their constituents. The social distance between members of the GP and nonmembers is thus diminishing rapidly. This is, in part, a matter of institutional design, given reservations of seats for women, and the lower castes and tribes. The political, material, and social effects of gender and caste reservations are hotly contested, and opportunities for elite capture of reservations abound (themes discussed in Chapter 5). Nonetheless, reservations are indisputably altering the face of local politics by bringing new people into the public sphere. This in turn, has increased the representativeness and proximity of the panchayat to the local community. Second, the panchayat wields increasing financial clout, even as its formal fiscal capabilities remain minimal. The influx of central and state funds to the local level has increased over time, reflecting the expanded spending described above. Some of these funds flow directly to the panchayat. In the period between 2005 and 2010, for example, Rajasthan’s panchayats each received a yearly average of 1.2 million rupees (roughly $22,000 USD) from state and central government financial commissions.83 Even when funds are not directly allocated to the panchayat, most finances at some point pass through the GP, where local officials play a pivotal role in their ultimate distribution. The roll out of MGNREGS in the mid-2000s, perhaps more than any other social program, has altered the local political economy – and with it, the role of the

82

83

just 9,000 Gram Panchayats. In fact, the actual number of GPs has risen by almost 1,000 from 9,184 in 1995 to 9,894 in 2015. The SC and ST share of sarpanch seats are calculated on the basis of 2011 Census data, estimating that Scheduled Castes make up 17.8 percent and Scheduled Tribes 13.5 percent of the Rajasthan’s population. As noted, however, the relative representation ratio signaling the share of SC and ST members in the panchayat is not always equal to one (and was closer to 0.8 for the first two rounds of elections). Female representation is calculated on the basis of one-third seat share in 1995, 2000, and 2005, and 50 percent seat share in 2010 (following an amendment to the Rajasthan Panchayati Raj Ordinance in 2009 increasing the share of seats reserved for women from one-third to one-half in all direct panchayat elections). Chauchard 2017.

Going Local

79

panchayat.84 MGNREGS heralded a sea change in levels of local funding, marked by a dramatic flow of central and state funds to the Gram Panchayat, which is the local body charged with program administration. MGNREGS funding in Rajasthan peaked from 2009 to 2011 (the time of research for this book) at just over Rs. 50 billion (roughly $1 billion USD) (Figure 3.8). This influx of funds has breathed critical new life into the panchayats. Funding flows from the Centre with contributions from the state governments; the Centre pays all labor costs but requires that states cover 75 percent of the cost of materials. Both types of funding are administered at the local level, and the law explicitly requires that at least 50 percent of funds must be spent on projects that are designed and implemented by the Gram Panchayats. Local officials also play a large role in managing the day-to-day implementation of the program: the GP is responsible for registering potential beneficiaries, issuing job cards, selecting from a slate of projects on which to offer employment, and ensuring that workers are paid in a timely fashion. The GP, moreover, is charged with local procurement for construction projects under MGNREGS, and is allocated funds for this purpose.85 The opportunities for GP officials to influence local program implementation are therefore substantial. Opportunities for graft and personal gain are also ample. The local standing of GP officials, and of the sarpanch in particular, has thus risen dramatically with the rollout of MGNREGS. In a study of Rajasthan and Madhya Pradesh, Jenkins and Manor observe this shift, calling the infusion of funds and increase in planning authority under MGNREGS the “most important change to panchayati raj since the passage of the 73rd Amendment.” The “increased financial resources, somewhat enhanced autonomy from higher-levels of the political system, and an expanded range of administrative responsibilities,” they note, “have combined to enliven local politics considerably. For the first time in most states, almost all local groups (including the poor) see that local panchayat decisions can make a material difference in their lives.”86 Third, the panchayat has grown, not just in fiscal, but in political and social influence, emerging as a critical space in the local political economy. In all, twenty-nine key responsibilities related to village economic and social

84

85

86

Rajasthan is widely recognized as a leader in MGNREGS implementation (Drèze & Khera 2010; World Bank 2011; Jenkins & Manor 2017). This is attributed both to a long history of drought relief and famine relief programs in the state, which put in place infrastructure and institutions well equipped for the administration of programs like MGNREGS. It also, notably, reflects the continued activism and vigilance of social movement organizations like the Mazdoor Kisan Shakti Sangathan, discussed earlier in this chapter. Chauchard (2017) estimates for the sample of panchayats in his study of Rajasthan that receipt of funds for MGNREGS – including for the purchasing of materials for the program – effectively doubled the yearly GP budget. Jenkins & Manor 2017, p. 63. Sukhtankar (2017, p. 6) similarly notes that MGNREGS implementation has “advanced the legitimization of GPs as state actors.”

The Institutional Terrain of the State

80 60,000

Rs. millions (constant prices)

50,000

40,000

30,000

20,000

10,000

0 7

–0

06

20

8

–0

07

20

9

–0

08

20

0

–1

09

20

1

–1

10

20

2

–1

11

20

3

–1

12

20

4

–1

13

20

5

–1

14

20

figure 3.8 MGNREGS Funding in Rajasthan (2006 constant prices) Source: Ministry of Rural Development.

development are entrusted under the Seventy-Third Amendment to the panchayats, including issues of water supply, housing, roads, education, child development, health services, and the administration of poverty alleviation programs. However, in Rajasthan, as in most states, the necessary personnel and funds to actually carry out these activities have not been devolved to the same extent, producing a pronounced local capacity gap. And yet, despite these limits, the stature of the panchayat in local and state politics has grown. This partly reflects a strengthening of the panchayat’s legal status. The “Rajasthan Panchayati Raj (Modification of Provisions in their Applications to the Scheduled Areas) Act” of 1999 sought to give additional teeth to the Seventy-Third Amendment by mandating biannual Gram Sabhas – planning meetings that are nominally open to any adult resident of the panchayat. Little statewide data exists on the Gram Sabha in Rajasthan, although accounts suggest that they are not regularly held.87 And yet, almost half (47 percent) of those surveyed for this book reported having attended Gram Sabha meetings. Just 26 percent, though, reported having voted on a budget or 87

My own experience suggests that Gram Sabha meetings are not always regularly held; multiple instances meetings that I had hoped to observe were cancelled or postponed.

Going Local

81

proposal, while 17 percent reported actually speaking at the meeting.88 This suggests that, as in much of rural India, Rajasthan’s Gram Sabhas are often topdown rather deliberative in nature, where residents are informed about government programs and predetermined budgets.89 Perhaps more important than their formal deliberative structure, then, is the expanding number of administrative tasks charged to the Gram Panchayats. Panchayat officials are document keepers, serving as gateways to critical papers such as birth and death certificates, caste certificates, and income certification, which, in turn, are necessary for citizens to access a wide range of government programs. GP officials, for example, are responsible for categorizing who is “below” and “above” the poverty line (thus influencing the range of services for which a household is eligible), as well as for the allocation of job cards necessary for participation in MGNREGS. They also determine which households appear on beneficiary lists for goods such as materials to build a toilet, to improve one’s home, or other such schemes. Beyond this gatekeeper role, panchayat officials negotiate with higher-level officials for the release of funds and implementation of programs. GP members are responsible for chasing down resources to ensure that they are, in the end, delivered to their intended beneficiaries. Indeed, as Simon Chauchard notes, “without the influence of the GP (and especially the sarpanch), many perceived that social benefits already due to them would not reach them.”90 The panchayat thus wields unprecedented levels of formal and informal clout. This is visible in the rising status and prestige assigned to the local office. It was not uncommon in the course of my fieldwork for the arrival of the sarpanch at an official function (or at a meeting with a foreign researcher) to be accompanied by pomp and circumstance including presentation of garlands, small gifts, preparation of food, and delivery of speeches. As one villager in Udaipur put it, “now that the Raj is gone, sarpanch is king.”91 Last, the increased status (as well as potential material and social gain) associated with local political office is reflected in the growing competitiveness of panchayat elections. The Rajasthan Panchayati Raj Department reported that an average of 6.6 candidates ran for each open (unreserved) sarpanch seat in 2010, while 4.5 candidates ran for seats reserved by caste or gender. Anecdotal accounts of election violence highlight the increasingly high stakes of panchayat elections.92 Other studies in Rajasthan similarly 88

89

90 91 92

Notably, 41 percent reported that they did not believe that officials would “hear” or listen to the voices of ordinary citizens if they spoke up at a Gram Sabha meeting. In Chaudhuri’s (2006, p. 174) assessment, most states do “little more than mechanically list a variety of nominal functions and duties to be performed by gram sabhas.” Chauchard 2017, p. 73. By “Raj” he was referring both to the princely and the British Raj. For example, in one survey village in Kota where the sarpanch’s seat was reserved for a Scheduled Caste woman, election violence broke out between two higher-caste (Jat and Rajput) communities, each of which “ran” its own female SC candidate. In a dramatic turn of events, a

82

The Institutional Terrain of the State

emphasize rising local competition, noting that it has brought new dynamism as well as diversity to the panchayats, since local elites can no longer take voter support for granted and must vigorously contest elections.93 As Jenkins and Manor note, “poor people grasp the opportunities that competitive panchayat politics offers” and are “applying pressure on candidates in local elections to deliver concrete benefits – through NREGA and other programmes – rather than empty promises.”94

setting the stage for claim-making This chapter has documented the ways in which the state in Rajasthan has, over time, moved closer to the village, broadening its reach and becoming more central in citizens’ daily lives. This, I argue, sets the stage for the high but varied patterns of claim-making that we will see in the next chapter. And yet, for much of India’s history the state has operated at a level far removed from the village: a “sovereign entity set apart from society.”95 Princely Rajputana, as we have seen, was marked by a felt absence of the state. The rural poor had very few points of direct engagement with public authorities (whether understood in terms of a local jagir, a princely kingdom, or the broader colonial administration). Where present, the state was more often than not an extractive rather than developmental force in the lives of the poor. An elderly resident of Kota, who grew up in the 1930s in the waning days of the princely kingdoms, recalled this period, remarking: In the days of the Raj, we lived our lives simply. The Thakur [landlord] sat in his house and gave his orders. If he said plow, we would plow. That was it. There was little vikas [development] in those days. No one thought to trouble with the village. No one brought anything here.96

The sixty-plus years since Independence (and the last twenty years in particular) have seen, in both relative and absolute terms, an “enormous expansion of development functions for the state” alongside an “expansion of the administrative machinery” at the local level.97 The state has thus become increasingly visible in a “pervasive, sometimes subtle bureaucratization of daily life even for

93

94 96

Jat man was killed by a Rajput in the run-up to the election. This violence signals the high stakes of local elections, even where the seat in question was reserved and therefore formally beyond the reach of either the Jat or Rajput communities. See, for example, Jenkins and Manor (2017) and Chauchard (2017). Jenkins and Manor observe that greater numbers of seats are being filled by members of the lower castes; while much of this is due to political reservations, increasing numbers are also winning “open” (or unreserved seats). 95 Jenkins & Manor 2017, p. 18. Fuller & Harriss 2001, p. 23. 97 Author interview, Sangod block, Kota district, February 2011. Gupta 2012, pp. 31–32.

Setting the Stage for Claim-Making

83

the poorest segments of the population, people who are considered the farthest removed from state bureaucracies.”98 As a resident in Udaipur remarked: It all comes down to paper. Do you have the cheinet [ration card]? Are you red, or blue [categories demarcating below and above poverty line status]? Sarkar says if we are rich or poor. Sarkar says if we will eat well or not.99

The state is also visible in physical terms, through infrastructure and programs. An elderly resident of Udaipur reflected on this change, referencing the lack of roads and medical facilities in his youth: Twenty-five years ago, there was no road here. When it was a pregnant mother’s time, if she needed to go to hospital they would strap her to a board and carry her on foot. It was 20 kilometers to Udaipur. They would walk day and night. Babies would die on the way. Finally, the road was made. It started here in the village. Villagers themselves gave their own labor and dug the road. It was not paved, just a path. But slowly things shifted … The adhikaris [officials] would come with their plans and schemes: this one for paving roads, this one for [health] facilities. Today, mothers are even paid to deliver in the hospital, and the trip to Udaipur is like that [snaps fingers].100

The same elderly resident of a Kota who recalled the “days of the Raj,” above, similarly stated: Look around today and you see sarkar [the state]. It’s there in that road, in that school over there. The face of the village has changed. Sarkar brings vikas [development].101

It is true, as these residents state, that the face of the village has changed. I was continually struck in the course of my fieldwork by the frequency with which I observed the installation of new public works. Equally striking, though, was the frequency with which these projects appeared to be stalled or incomplete. Nonetheless, they were visible to the common citizen and so spoke of a promise or possibility of services, even as that promise remained unfulfilled for many. Indeed, it is difficult to move around a village in Rajasthan without encountering some kind of physical reminder of the state, whether in the form of school buildings, health clinics, roads, or water sources. An awareness of services seemed inescapable; the walls of government buildings and other public spaces were frequently marked (on top of a signature, government-issued yellow paint) with slogans advertising programs and lists of beneficiaries for old-age pensions, school scholarships, and MGNREGS employment. As a staff member of a local NGO wryly noted: “there is a whiff of paint in the air; the buildings are dripping with promises!” The same staff member, though, also quipped, “there is water, water everywhere, but not a drop to drink,” referencing the many government-issue hand pumps that are scattered throughout the

98 100 101

99 Ibid. Author interview, Gogunda block, Udaipur district, January 28, 2011. Author interview, Bargaon block, Udaipur district, January 10, 2011. Author interview, Sangod block, Kota district, February 2011.

84

The Institutional Terrain of the State

villages and their frequent disrepair. A resident of Udaipur made a similarly sardonic observation, suggesting that he could see services in neighborhoods all around him but not in his own: Sarkar is very busy these days. They are putting in CC [paved] roads over there, and a school there. See how much they are building! I have a good view of it from here.102

As these accounts illustrate, spending is up and programs are growing, but not everyone benefits in the same way from this expansion. As I argued in Chapter 2, it is this combination of the state’s breadth and its unevenness that is most likely to foster citizen action. The greater reach and visibility of the social welfare sector increases the frequency of citizen–state encounters and, in so doing, shapes citizens’ expectations with regard to service delivery. The irregularity of the state’s performance further motivates action as underserved citizens develop a sense of grievance and, by extension, of entitlement. Claimmaking, in this sense, is conditioned by the state itself, and is most likely under the intermediate conditions of a broad (or broadening) yet irregular state presence. The data on claim-making, to which we turn in detail in the next chapter, are static – gathered through a single survey. It is therefore not possible to systematically track changes in claim-making alongside the changes in the state’s presence and visibility that this chapter has documented. However, historical sources offer persuasive evidence that citizen participation in Rajasthan has indeed intensified alongside an expansion of the institutional terrain of the state. When commitments to social welfare were negligible, as they were under princely rule and in the early decades following Independence, levels of civic and political engagement were also extremely low. As the Rudolphs, noted scholars of Rajasthan, observed: the people of the princely states “were politically irrelevant and thought that they should be.”103 Björkman and Chaturvedi similarly note, the “bulk of the local population consisted of cultivators who had limited rights to the land they tilled, and certainly no control over their local government.”104 Citizens in the post-Independence period also “tended to remain aloof from the political life of the state,”105 and public participation remained highly limited. In a 1962 village study by the Rudolphs, for example, a majority of respondents reported that they were “non-participants” in village politics.106 The Government of Rajasthan’s own assessment was equally bleak, noting low attendance in panchayat meetings due to a lack of “requisite interest and enthusiasm among the people” in the 1950s and 1960s.107 Singh, drawing on interviews with former bureaucrats who served as District Collectors from 102 103 104 106 107

Author interview, Bargaon block, Udaipur district, December 15, 2010. Rudolph & Rudolph 1962, p. 142. (Cited in P. Singh 2015, p. 174.) 105 Björkman & Chaturvedi 1994, p. 140. Singh 2015b, p. 182. Rudolph & Rudolph [1962], cited in P. Singh (2015, p. 183). Rajasthan Legislative Assembly 1950–2009, cited in P. Singh (2015, p. 183).

Setting the Stage for Claim-Making

85

the 1950s to the 1980s in Rajasthan, similarly describes the “general political apathy” of the people at that time. Only three out of 38 former District Collectors noted any instance of citizen mobilization around issues of service provision.108 A thin presence of the state, in sum, was accompanied by low levels of citizen–state engagement. As the surface area of the Rajasthani state has expanded, citizens have engaged the state with greater vigor. Elderly village residents offer powerful accounts of these changes. A Scheduled Tribe woman in Udaipur, for example, noted how patterns of mobilization have changed for some of the most marginalized: Back then, we were chup chup [quiet quiet]. No one would dream of standing up the way we do today. If there was nothing to eat, we were hungry. If there was no water, we were thirsty … Nowadays, even we ladies will march and say what it is we need. Women today are not so afraid to speak.109

An elderly Scheduled Caste man in Udaipur similarly recounted some of the changes he had observed. In those days, everyone had their place. We harijan [SC] lived here on the edges and did the work of animals, as in our fathers’ time. Then Congress came and said that all should have land. Now each has his little piece. [And so] we came into the village, to panchayat. Now if the sarpanch is falling down, drinking tea on the job, we will be the first to say.110

These shifts in participation are visible in increasing rates of voter turnout in state elections, which rose from just 51 percent in 1980, to 67 percent in 2003, to 75 percent in 2013. Voter participation in Gram Panchayat elections is also strikingly high, rising from the negligibly low levels noted by the Rudolphs and others in the 1960s (prior to the Panchayati Raj reforms and the Seventy-Third Amendment), to an average turnout of 85 percent between 2005 and 2010.111 The rise in citizen participation is also visible beyond the electoral arena, in shifting patterns of public opinion that show rising interest in politics at the state and local levels. In National Election Studies survey data, the proportion of people in Rajasthan expressing interest in “public affairs” rose substantially from just a quarter in 1967 to almost 40 percent in 2004. The proportion concerned specifically with the affairs of state-level government also more than doubled from just 11 percent in 1971 to over a quarter in 1996.112 At a more local level, former District Collectors who served in the 1990s and 2000s were much more likely to report instances of citizen mobilization than those who served from the 1950s to the 1980s; more than half of those interviewed by

108 109 110 111 112

P. Singh 2015, p. 183. Author interview, Bargaon block, Udaipur district, January 19, 2011. Author interview, Bargaon block, Udaipur district, December 21, 2010. Government of Rajasthan, Department of Panchayati Raj. National Election Survey data, cited in P. Singh (2015, p. 164).

The Institutional Terrain of the State

86

Singh who served in this later period reported that they had been “personally approached regarding issues such as the absenteeism of teachers or doctors, lack of proper teaching in the village school, problems with school infrastructure, and the lack of drugs at the primary health center.”113 The next chapter turns to the citizen survey data in Rajasthan to more systematically investigate the patterns of claim-making manifest in the late 2000s. Read alongside historical accounts, these data tell a story of rising levels of citizen–state engagement and increasingly broad strategies of action.

113

P. Singh 2015, p. 185.

4 Seeking the State Claim-Making Patterns and Puzzles

This chapter develops a bottom-up view of the state’s welfare apparatus as it is seen and sought by citizens in rural Rajasthan. Turning to the citizen survey data, I demonstrate that differences in state-targeted action (or inaction) cannot be adequately explained by level of local economic development, by individual socioeconomic standing, or by ascriptive characteristics of caste or gender. While there are notable differences in their strategies of action, citizens from all walks of life actively engage the state in pursuit of social welfare. They do so in a variety of ways, contacting both elected and non-elected public officials, and engaging a wide array of non-state intermediaries. Notably, the practices most often pursued do not conform with those most referenced in the broader literature on citizen-state relations. Political parties, brokers, and civil society associations – all often cited as key vehicles for citizen interest articulation – are underutilized relative to other channels, while the Gram Panchayat, which is often dismissed as little more than a “paper tiger,” emerges as the primary site for claim-making. At the same time, however, almost a quarter of those surveyed do not pursue any of these channels, retreating from rather than engaging with the state. This chapter maps the variance in both the incidence (or likelihood) of claimmaking and the repertoire (or range) of practices employed. In what follows, I first describe the claim-making landscape (that is, the range of existing channels of access to the state); second, document the rates at which citizens turn to these various channels; and third, explore whether and how patterns of claim-making diverge across different types of public goods and services, different places, and different groups of people (men and women from different socioeconomic backgrounds). Before delving into the data, a qualitative illustration serves to highlight the diversity of citizens’ approaches to problem solving – even among those living in the same communities. The state’s local presence is real, palpable, and of vital 87

88

Seeking the State: Claim-Making Patterns and Puzzles

importance in rural India – manifest in the allocation of infrastructure, services, income support, and other resources. The distribution of these resources, though, is rarely a regularized affair: entitlements and lines of accountability are often unclear to citizens and local officials alike. As a consequence, no two people see the state in exactly the same way. These differences in perception are reflected in different courses of action and inaction. Consider, for example, the story of a broken water pump located in a Scheduled Caste neighborhood of a village in Kota district.1 When I first visited this village, residents reported that the pump had been in disrepair for several months. At first the villagers had to make do by walking long distances to collect water on the outskirts of the village, or had to pay to take water from a private well in a farmer’s field. Recently, though, government trucks had begun to visit the village twice a week with large tankers of water. This was not a common practice; I visited other nearby villages where water was in equally short supply that did not receive tanker water. Residents of the neighborhood – all from the same Scheduled Caste – offered different accounts of the broken pump and of why local officials responded by sending trucked water. A young man, who had completed several years of secondary school, offered a detailed description of the local governance structure, extending from the Gram Panchayat to the block offices of the Public Health and Engineering Department (PHED) – the state agency charged with the maintenance of drinking water infrastructure. He explained: We have to follow the chain of command. First there is the [neighborhood-level] ward panch [an elected member of the Gram Panchayat], who brings it up in the quorum [monthly GP meeting], and things go from there. So when we needed water, all the members signed a petition and the sarpanch [GP president] brought it to Sangod [the block seat]. Then PHED had to act. They sent the water trucks, and now they are sending an Engineer to make the repairs [to the water pump].

When I asked if he personally had spoken to anyone about the broken pump, he replied: “I raised it with the panchayat, that’s enough. They are the ones who have to do the work.” I asked if he raised the issue with anyone else, but was told that he did not have the time to venture beyond the panchayat since, “My work is here in the village. I can’t go running every which way. If I spend my time running to Kota to drink, what will I eat?” Another man of about the same age offered a different account centered on his interaction with officials at an administrative camp.2 The camp, he explained, was held in another village on the road to Kota city (the district 1

2

These accounts are drawn from author interviews in Sangod Block, Kota district in January and February 2011. This camp was part of the Government of Rajasthan’s “Prashasan Gaon Ke Sang” (Governance with the Village) campaign, in which officials from various public departments are periodically made available to meet with citizens to offer “on-the-spot” solutions to their problems. The camps are visited by higher-level officials and politicians from the district and state levels, including elected parliamentarians.

The Claim-Making Landscape

89

capital). Since he worked as a manual laborer on construction sites in Kota, he frequently passed through this village and so heard about the camp and about the possibility of talking to officials there. He explained: At the end of the day nothing gets done without the higher-ups. That’s why I went to the camp some months ago, to talk to the people with ways and means to get this work done. The panchayat alone is toothless.

At the camp, he talked to a worker in the office of the local Member of Legislative Assembly (MLA) – his state-level representative. He then returned home to the village, where he spoke to his local Gram Panchayat official (a ward panch), telling her that she had to follow up. He explained: “If you complain once it doesn’t happen. We have to say it many times, to so many people. Everyone and anyone, we ask and tell them of the problem. See, even now I’m telling you!” An elderly man, in contrast, said he had not spoken to anyone about the issue. I asked who he thought was responsible for bringing the water. He replied: “Who knows … all I know is that some babu brings the water.”3 What would he do if the tankers stopped coming? He shrugged: “Why waste my time with all that? Promises are made to be broken. If God wills it, we drink or we are thirsty.” When I asked if, like some of his neighbors, he had ever contacted the panchayat, he said that he did not know what happened in panchayat meetings and had never attended one. Similarly, he had not heard of the administrative camp, and did not know if the local MLA had played any role in bringing water to the village. I was, at first, frustrated by these accounts since I could not establish a clear picture of how or why the water tankers had been sent, or of what actions community members had taken. However, this very confusion is itself illustrative as it highlights different understandings of the state and its service delivery apparatus that materialize within the same village and, strikingly, among residents of the same neighborhood and caste community. These different understandings, it appeared, were linked to different repertoires of action, ranging from those who did nothing, to those pursued action exclusively within the panchayat, to those who sought out a wider range of political actors including officials and intermediaries in the village and beyond.

the claim-making landscape To begin to unpack such different claim-making practices, one first needs to examine the channels of access to the state that citizens might pursue. To do so, I turn to the citizen survey, first introduced in Chapter 1, which was carried out over the course of three months in 2010 and 2011. The survey was administered to 2,210 randomly selected individuals across 105 villages, forty 3

Babu is used here in a dismissive fashion to refer to a petty official.

90

Seeking the State: Claim-Making Patterns and Puzzles

Gram Panchayats, 8 administrative blocks, and four districts of Rajasthan.4 The survey was organized into modules, each of which explored citizens’ interactions with different actors or institutions (state and non-state) that might facilitate access to officials and to resources. The decisions to include particular actors or institutions in the survey questionnaire were inductive and deductive, drawing both on prior rounds of qualitative research (in which I observed channels actually pursued) and on a reading of existing scholarship on Indian citizen-state relations (which suggested channels that should, in theory, matter). Table 4.1 presents descriptive statistics from the survey sample, noting, first, the formal structures of appointed and elected government that enable direct interface between citizens and the state; and second, the reported presence at the village level of a range of non-state actors and institutions that potentially mediate citizens’ approaches to the state.

Administrative and Representative Arenas for Claim-Making The most local administrative unit in rural India is the revenue village, officially defined for the purposes of tax collection and public distribution. Village boundaries were established through the mapping exercises of the British Raj or by post-Independence census enumeration. Most overlap with territorial and social units that have remained relatively stable over time, many with histories that can be traced for hundreds of years. The citizen survey drew from 105 of these villages, home to an average of 1,664 people (ranging from 212 in the smallest to 6,265 in the largest).5 Revenue villages are nested within systems of both elected and appointed government.6 The lowest tier of elected government is the Gram Panchayat (GP), the nominally non-partisan council representing a cluster of villages, introduced in the last chapter. The citizen survey included forty GPs, each with an average population of 4,749, encompassing an average of four revenue villages. In drawing the village sample, I selected up to three villages per Gram Panchayat – always including the village with the panchayat headquarters,7 4 5

6

7

Survey design and sampling methodology are presented in detail in Appendix I. Population estimates are based on 2001 census data, which was the most recent at the time of the survey design and implementation. Villages with populations of less than fifty were excluded from the sample. While the appointed state bureaucracy and system of elected Panchayati Raj Institutions are intended to work in tandem, their jurisdictional boundaries can be unclear. Appointed district and block officials are known to actively intervene in the affairs of the elected panchayats, influencing both program implementation and budget allocations. Some local officials, such as the Gram Sevak (or village secretary), moreover, are “hybrid” actors that simultaneously answer to the state administration and panchayat. Within a Gram Panchayat, one village (usually the largest) is designated the headquarter village and is home to the panchayat bhavan – an administrative building from which local officials conduct their business.

91

Local administrative and territorial unit, enumerated by the census Local elected council, spanning an average of 4 revenue villages Mid-level representative and administrative unit, spanning an average of 30 GPs Primary administrative subunit of the state, spanning an average 300 GPs and 9 blocks State (MLA) or national (MP) parliamentarians and their political parties

Customary association governing social affairs of a given caste community Village-wide association of leaders from different castes Individuals who assist others in accessing government resources Neighborhood voluntary association (developmental or cultural purposes) Village-wide voluntary association (developmental or cultural purposes) Non-governmental organization (in or around village) Social, protest, or “people’s” movement (in or around village)

0.66 0.28 0.32 0.32 0.25 0.06 0.11

Sample Mean*

–

4 out of 33

Up to 3/GP 5/Block 2/District

Selected

0.47 0.45 0.47 0.47 0.43 0.24 0.32

Std. Dev.

–

4

105 40 8

Sample Size

Source: Citizen Survey (2010–2011), n = 2,210. * The sample mean for Part B is the proportion of survey respondents who reported the presence of the given actor or institution in their locality.

Caste body Intercaste Body Individual Brokers NH Association Village Association NGO Social Movement

B. Non-State Actors and Institutions (Reported Village-Level Presence)

Politicians/Political Parties

District

Revenue Village Gram Panchayat Block

A. Administrative and Representative Institutions

table 4.1 The Claim-Making Landscape

92

Seeking the State: Claim-Making Patterns and Puzzles

while randomly selecting up to two others depending on the number of villages in the GP.8 The Gram Panchayat, in turn, rests within a block – a mid-level representative and administrative unit. Block-level panchayats (panchayat samitis) are indirectly elected councils that typically include around 30 Gram Panchayats. The officials of the panchayat samiti are selected by the members of the surrounding GPs. These block-level councils serve as a link between the local villages and the district, and are the primary units through which funds for rural development flow. The panchayat samitis sit alongside administrative blocks – referred to as tehsils – that are part of the state’s bureaucratic apparatus. The most senior official at this level is the Block Development Officer, who oversees the many extension agents who administer state and central government programs. The survey covered eight blocks, sampling two panchayat samitis per district (each from a distinct tehsil), having sorted them by literacy rate before randomly selecting one above and one below the district’s mean.9 The district is the primary administrative sub-unit of the state (spanning roughly nine blocks), and is the level at which revenue collection and allocation are decided. The senior bureaucrat is the District Collector, who is a centrally appointed officer of the Indian Administrative Service (a national-level civil service body). Appointed district-level bureaucrats serve alongside district-level panchayats (zilla parishads), the officials of which are indirectly elected by the members of the panchayat samitis and who are charged with defining development priorities for the district. The survey drew from four purposively selected districts – Kota, Ajmer, Jodhpur, and Udaipur – which, as noted in Chapter 1, represent distinct socioeconomic and geographic areas of Rajasthan. Last, while not part of the survey sampling frame, citizens also reside within state and national electoral constituencies within India’s federal parliamentary system. Members of Parliament (MPs) from 543 single-member constituencies (of which there are 25 in Rajasthan) are directly elected via first-past-the-post voting to the Lok Sabha (the lower house of Parliament).10 At the state level, Members of Legislative Assembly (MLAs) are also elected from single member constituencies, of which there are 200 in Rajasthan. These politicians and the political parties with which they are affiliated are the primary channels for citizen interest representation in the electoral arena.

8 9

10

Some GPs contain only one or two large villages, and in these cases all villages were selected. The boundaries of the indirectly elected panchayat samitis and state-appointed tehsils are often but not always concurrent. In drawing the survey sample, I opted to use panchayat samiti boundaries, since these are the units in which the Gram Panchayats are formally embedded. Each selected panchayat samiti was located in a separate tehsil, thus ensuring that the sample pulled from distinct representative as well as administrative areas. Members of the Raja Sabha (the upper house of Parliament) are indirectly elected by the statelevel Legislative Assemblies.

The Claim-Making Landscape

93

Non-State Intermediaries Together, the revenue village, the Gram Panchayat, the administrative block and district, and state and national parliamentary constituencies encompass the formally recognized arenas in which claim-making unfolds. These, though, comprise only one slice of the claim-making environment and are best understood alongside non-state actors and institutions that present alternative or additional channels for citizen action and access. As Corbridge et al. have observed: “When poorer people meet local state officials, they do so with reference to their non-state networks … The state is met through … a broad range of caste leaders, brokers (dalaals), and political fixers (pryaveerkars).”11 Survey respondents were asked to report whether they were aware of such mediating actors and institutions in their locality (described in Part B of Table 4.1). First among these are customary caste bodies. Traditionally, access to the princely, colonial, and ultimately independent Indian state was mediated through caste leaders who exercised high levels of social control in a village setting. Caste leaders are still prominent today, and most villages are home to associations that are led in hereditary fashion by the elders of a given caste community. Two-thirds of survey respondents reported the presence of one or more such caste organization in their villages. These associations, sometimes referred to as caste panchayats,12 have no formal jurisdiction or relationship to the elected Gram Panchayat. Rather, they govern (in extra-judicial fashion) the internal affairs of a caste community, and are called upon to weigh in on cases of child marriage, divorce, land disputes, or other areas of social contention. Some villages are also home to intercaste bodies, variously referred to as “traditional” or “gaon” (village) panchayats, which draw together leaders from different caste communities within the village. These bodies (reported as locally present by 28 percent of the survey sample) play a role in dispute resolution where conflicts spill over communal lines – although they are de facto often controlled by the leaders of the locally dominant caste. Conceptually, villagers draw a sharp line between caste bodies (of both the single and mixed variety) and the elected Gram Panchayat. Caste panchayats, in the words of many that I interviewed, deal with “communal” affairs but not with village “development” or “politics,” which are the purview of the “sarkari” (governmental) panchayat. In practice, though, caste institutions exert considerable influence over the Gram Panchayat. In some cases, the leadership of the gram, caste, and traditional panchayat is one and the same. In other cases, caste leaders attempt to influence the nomination and voting process 11 12

Corbridge et al. 2005, p. 108. The term panchayat is a general term for “council,” borrowed from pre-colonial local governance structures in which village communities were internally organized a council of “five” (“panch”) elders.

94

Seeking the State: Claim-Making Patterns and Puzzles

through which Gram Panchayat candidates are selected.13 They may also play behind-the-scenes roles in beneficiary selection and implementation of programs, and raise issues of communal concern with local officials. However, as we will see in the next chapter, the nature of caste relations is shifting and the political salience of caste – while still poignant – has begun to decline, particularly as parties increasingly compete across caste lines to secure votes. Other kinds of non-caste mediating institutions have thus emerged at the local level, many of which intentionally operate across caste lines. Key among these is a diverse group of individuals variously labeled “brokers” (dalaals), “fixers” (pryaveerkars), or informal “leaders” (netas).14 These actors do not hold any formal position, but practice “the art of approaching officials for favors and making the wheels of administration move in support of such favors.”15 In the Rajasthani setting, Krishna has highlighted the importance of village-level intermediaries whom he refers to as naya netas (“new leaders” – thereby distinguishing them from hereditary, caste-based leaders).16 These new leaders, he observes, typically hailed from the lower castes (emerging from a cohort of newly educated youth) and worked across caste lines within the village and beyond. For the purposes of my citizen survey, brokers were defined in broad terms (avoiding particular labels that might take on distinct meanings in different contexts), as “people who know how to get things done both inside and outside the village, and who can help others to … access government schemes and benefits.” Just under one-third (32 percent) of respondents reported that these kinds of individuals were active in their locality. Voluntary associations have also been widely celebrated as sites for interest mobilization and collective action. Some scholars, though, have cast doubt on the degree to which models of associationalism developed in the West can be applied in the Indian context, where formal associations remain thin on the ground.17 And yet, India’s voluntary sector – broadly defined to encompass NGOs, social movements, advocacy groups, and social campaigns – has grown both nationally at the local level.18 Roughly one-third (32 percent) of survey respondents

13

14 15

16

17 18

In several villages, for example, residents reported that caste elders sat together to “decide” which candidates should be given the support of their community. However, with a closed ballot, there is no monitoring mechanism to ensure that caste members in fact follow these informal directives. Oldenburg 1987; Manor 2000; Corbridge et al. 2005; Krishna 2002, 2011. Reddy and Haragopal, cited in Manor 2000, p. 817. Manor (p. 817) describes fixers as “crucial political intermediaries between the localities and powerful figures (bureaucrats and, especially, politicians) at higher levels.” Krishna 2002, 2011. Many of these new leaders gained invaluable skills, experience, and connections through their engagement with large-scale, state-sponsored rural development and poverty alleviation programs in the 1990s, in which they were incorporated through government training programs. Chhibber 1999; Krishna 2011; Harriss 2005. Kudva 2005; Ray & Katzenstein (Eds.) 2005.

Measuring Claim-Making

95

reported the presence of an active voluntary association in their neighborhood, while another quarter reported a village-wide association. These local associations vary in their degree of formality (some are membership based, while others are looser agglomerations), and coalesce around a wide range of issues – cultural and recreational (for example, religious groups, film societies, or sports leagues) and developmental (for example, village improvement societies promoting village cleanliness, or which attempt to raise funds for improved infrastructure). Still other associations operate at a regional, state, or even national level, but maintain a local presence through meetings, chapters, or organizers that operate in or around the village. Non-governmental organizations are active in a wide array of social and developmental spheres. While their local presence remains sparse (just six percent of survey respondents reported an active NGO in or around their locality), these organizations can reshape the claim-making environment in important ways. For example, NGOs have developed different relationships to the Gram Panchayat, in some cases working to strengthen its capacity and in others seeking to create alternative spaces for informal governance.19 Social movement organizations are also thin on the ground: just eleven percent of the sample reported being aware of movement activity in or around their locality. And yet, as we began see in the previous chapter’s discussion of the Mazdoor Kisan Shakti Sangathan (MKSS), social movement organizations have played a powerful role in shaping the claim-making arena by influencing state and national social welfare legislation.

measuring claim-making Which, if any, of these many channels – direct and mediated, state and nonstate – do citizens utilize when making claims on the state for social welfare? Do different kinds of people employ different strategies, particularly when seeking different kinds of resources? To answer these questions, I examine the rates at which survey respondents reported turning to the various actors and institutions described in the preceding section. For each, respondents were asked: “In your own experience, have you approached [the actor or institution in question] for help concerning [a list of welfare goods and services]?”20 They were then presented with a bundle of collective village services benefiting groups and localities and, separately, a bundle of selective benefits allocated to households and individuals. Collective village services were described as “amenities, facilities, and public works, such as schools, health clinics, roads, drainage, lighting, or drinking water,” while selective benefits were “welfare schemes and 19 20

See Chapter 5. The questions were not time-bound, in part because of difficulties of recall when specifying a particular time period in a pilot survey, and in part because I wanted to investigate cumulative practice over time.

96

Seeking the State: Claim-Making Patterns and Puzzles table 4.2 Claim-Making Practice Mean Std. Dev. Claim-Making Incidence Claim-Making Repertoire (No. practices, 0–10)

0.76 2.00

0.43 1.73

Direct Contacting: Administrative and Representative Institutions Gram Panchayat Members Bureaucrats: Block and District Levels Politicians/Parties

0.65

0.48

0.62 0.21 0.22

0.48 0.41 0.41

Mediated Approaches: Non-State Actors and Institutions Caste Body Intercaste Body Individual Broker NH Association Village Association NGO Social Movement

0.54

0.50

0.23 0.14 0.17 0.22 0.15 0.03 0.09

0.42 0.34 0.37 0.41 0.36 0.16 0.28

Source: Citizen Survey (2010–2011), n = 2,210. Note: Claim-making incidence and practices are assessed in binary terms where Yes = 1 and No = 0. Claim-making repertoire is an index of the number of practices employed.

programs such as pensions, cash assistance, subsidies or food rations, employment on government worksites, or other schemes.”21 Table 4.2 presents sample means for the reported rates of each kind of practice. I define the incidence of claim-making as the overall rate at which respondents reported any kind of state-targeted practice (direct or mediated) related to any kind of welfare goods or services (collective or selective) – characteristics that are subsequently disaggregated in the analysis that follows. Seventy-six percent reported that they have personally engaged some kind of claim-making in these terms.22 I define the claim-making repertoire as the set of all practices employed (ranging from zero to ten).

21

22

Concerns over survey length prohibited me from asking about each service or program individually. On the distinction between collective (high-spillover) and selective (low-spillover) goods see Chapter 1. The data from the large-n citizen survey are supplemented by a small-n, rapid survey (n = 232), carried out a year later in six case study villages in Kota and Udaipur districts (districts at the high and low ends of human development standing in the sample). In this small-n sample, 62 percent reported engaging in claim-making. This small-n survey utilized a semi-structured questionnaire allowing for open-ended answers. Two research assistants were employed to code the responses, working independently to ensure inter-coder reliability. In employing this different

Measuring Claim-Making

97

A glass-half-full reading might view large numbers of rural citizens pursuing public resources in one of India’s poorest regions as evidence of robust political participation. The puzzle, seen in these terms, is to explain why so many engage the state in a land of former princely states where residents have long been described as “docile” and “despondent.”23 In contrast, a glass-half-empty reading might highlight the almost one-quarter that reported having never contacted or sought out the resources of the state. In a region where residents depend heavily on government programs for income support, education, health care, subsidized food, and other services, this absence of claim-making is striking – and sobering. The puzzle, thus framed, is one of “why the dog doesn’t bark”24 – that is, of why citizens remain silent even when faced with critical material needs. No matter how it is framed, though, the central question remains: why are some citizens more likely than others to make claims on the state? Of equal concern to whether a person makes claims is how one does so. The incidence of each kind of practice is presented in Table 4.2, grouped by whether it involves direct, face-to-face contact with public officials or representatives, or whether it is mediated through non-state actors or institutions. The primary difference in these groupings lies in the qualitative nature of the interaction. Direct contacting involves personal interaction between a citizen and an agent of the state: with an elected member of the panchayat, a bureaucrat, or an officebearing state or national politician. Through these practices, a citizen inserts him or herself directly into the public sphere. Mediated practices, in contrast, involve a circuitous approach – variously brokered through caste bodies, brokers, local (neighborhood and village) associations, non-governmental organizations, or social movements.25 It is important, though, not to over-emphasize the distinction between “direct” and “mediated” approaches to the state, just as it is important not to idealize one set of practices over another.26 The boundaries between direct and mediated forms of contact are, in fact, blurry, reflecting a complex landscape in

23 24

25 26

methodology, the small-n survey offers a robustness check for the measurement of claim-making. See Appendix I for more on this methodology. Moore 1966; Krishna 2011. Sherlock Holmes asks this question to suggest that the absence of an event might require as much investigation as its occurrence. Doyle [1967], cited in Hochschild (1981). Note that all mediated claim-making practices are still state-targeted. Some have suggested that direct engagement of the state through institutionalized channels is preferable to brokered forms of contacting that are typically associated with patron-client networks marked by dependency and lack of accountability (Harriss 2005; Houtzager & Acharya 2011; Bertorelli et al. 2014). Concerns over the power of unelected intermediaries are certainly warranted. And yet, direct channels of access to the state are also easily captured by local elites or manipulated by parties or higher levels of government. Many in fact turn to brokers precisely because formal approaches to the state are captured and inaccessible – particularly for the poor. It is thus unclear that direct approaches are any more equitable than mediated approaches. Instead, I follow the lead of citizens themselves, who consider both direct and mediated practices to be necessary means of accessing the state.

98

Seeking the State: Claim-Making Patterns and Puzzles

which leaders wear multiple hats and may move into and out of formal administration at different points in time. A broker or caste leader, for example, might be subsequently be elected to the Gram Panchayat. Contacting the Gram Panchayat, in turn, can be a conduit to higher levels of government or to political parties and thus might also be considered a mediated approach.27 Moreover, as we will see in the discussion of the claim-making repertoire below, direct and mediated strategies are not mutually exclusive, but rather complement one another. Claim-Making through the Gram Panchayat Almost two-thirds of the sample (65 percent) reported directly engaging a public official or elected representative in their claim-making efforts. The bulk of this occurred through the Gram Panchayat: a full 62 percent stated that they had personally contacted an elected panchayat member for assistance when seeking access to government programs and services.28 This, though, was not always the case. While differences in methodology prohibit precise comparison over time, Anirudh Krishna’s study of Rajasthan in the late 1990s offers an important benchmark by which to assess how engagement with the panchayat has shifted.29 In a 1997–1998 survey that asked “Who helps gain access to the concerned agency of the state in the following situations (dealing with land administration or the police; getting a bank loan; replacing a school teacher; or getting wage employment),” no more than 18 percent mentioned Gram Panchayat officials.30 A little over a decade later, the role of the Gram Panchayat has shifted quite dramatically. In my 2010–2011 survey data, the GP is the first port of call for claim-making, standing head and shoulders above all other practices (Table 4.2). This reflects the deeper institutionalization of the panchayat over time: these local elected bodies were in their infancy in 1997, and have since grown in their capacity, standing, and visibility (processes described in Chapter 3).31 27 28

29

30

31

Bussell (2018). Approaching the Gram Panchayat was broadly defined to include contacting the council president (sarpanch), the vice-president (up sarpanch), or ward councilors (ward panchs). The survey data do not allow for greater disaggregation of which actors within the panchayat were contacted. Qualitative interviews, though, suggest that the preponderance of citizen-panchayat contact occurs through the elected sarpanch. Krishna 2002, 2011. In these works, Krishna draws on surveys carried out in 1997 and 1998 across 60 villages in five districts of Rajasthan (n = 1,898), as well as an additional nine villages in an adjoining district of Madhya Pradesh (n = 334). Two of the districts Krishna sampled in Rajasthan (Udaipur and Ajmer) overlap with those that I sampled for this book. Krishna’s other districts (Bhilwara, Rajasmand, and Dungarpur) have higher poverty rates on average than the remaining districts (Kota and Jodhpur) in my sample. Krishna 2011, p. 110. Responses ranged from 5 percent (for land administration and the police); to 7 percent (for getting a bank loan); 11 percent (for wage employment); and 18 percent (for replacing a non-performing school teacher). The first round of panchayat elections in Rajasthan were held in 1995. Krishna also suggests that change is likely over time. He noted (2011, p. 105) that the panchayats have been strengthened

Measuring Claim-Making

99

table 4.3 Comparing Channels of Citizen-State Engagement Gram Panchayat

Politician/ Party

Individual Brokers

Who will help a person trying to apply for a government scheme like a pension, work under MGNREGS, or some other benefit? Who will help when there is village work to be done, such as repairing the school, fixing the roads, or repairing water supply?

0.73

0.06

0.08

0.76

0.12

0.06

In your own experience, who have you approached for problems/issues related to government schemes (like pensions, rations, MGNREGS)? In your own experience, who have you approached for problems/issues related to village services (like roads, schools, water, health center)?

0.53

0.15

0.12

0.57

0.21

0.14

Source: Citizen Survey (2010–2011), n = 2,210. Note: Reports the proportion who cite a given channel (GP, party, broker) in response to each question. Channels are not mutually exclusive.

The centrality of the Gram Panchayat to citizens’ strategies of action is further highlighted in Table 4.3, which compares the rates at which citizens report turning to GP members, political parties, and individual brokers for help across a range of different issues. The first two questions are hypothetical, asking who would be most likely to assist someone seeking help with problems, first, related to “government schemes” of an individual or household nature and, second, related to collective “village services,” and so are akin to those asked by Krishna in his 1997–1998 survey. Large majorities (73 percent and 76 percent, respectively) cited the Gram Panchayat in response, compared to just 18 percent and 11 percent answering similar questions in Krishna’s study.32 The next two questions asked respondents to reflect on their personal experience, asking whether they themselves had contacted members of the GP,

32

over the fifteen years since the Seventy-Third Amendment and are “widely expected to grow into a larger role, crucially linking state actors with ordinary rural citizens.” For “selective” services, the closest comparison question in Krishna’s survey is: “Who helps gain access to the concerned agency of the state when [dealing with] replacing a non-performing teacher.” Eighteen percent highlighted the role of the Gram Panchayat in this capacity in 1997–1998. For “collective” services, the closest point of comparison in Krishna’s study is the question: “Who helps gain access to the concerned agency of the state when [dealing with] getting wage employment?”, which refers to employment on public works projects under programs such as the Jawahar Rozgar Yojana (a predecessor of MGNREGS). Here, just eleven percent of Krishna’s respondents referenced the Gram Panchayat.

100

Seeking the State: Claim-Making Patterns and Puzzles

political parties, or brokers. Fifty-three percent reported approaching the Gram Panchayat for assistance concerning selective services, while 57 percent did so for problems concerning collective village services.33 My 2010–2011 citizen survey data are consistent with those from other recent research in Rajasthan and elsewhere in India. Another survey in Rajasthan, Bihar, and Karnataka, for example, found that 73 percent of respondents asked panchayat members for assistance – most frequently with regard to “access to a government welfare scheme.”34 A qualitative study in Rajasthan and Madhya Pradesh also found that “poor villagers were interacting almost entirely with leaders and members of local [panchayat] councils …,” particularly in their efforts to access government programs such as [MG]NREGS.”35 Contacting Bureaucrats and Politicians Levels of claim-making activity are significantly lower when it comes to other forms of “direct” contacting, namely through bureaucrats and higher-level politicians. Just 21 percent of my citizen survey respondents reported directly approaching block or district level officials (Table 4.2), despite their formal jurisdiction over a substantial number of programs and services. This reflects the fact that such officials typically do not spend prolonged periods of time in the village, but rather occupy offices in the district capital or secondary cities that house administrative block seats. Their visits to the village are thus infrequent: in fact, 57 percent of respondents reported that they were not aware of a single visit from a block or district official in the past year. For most citizens, contacting bureaucrats requires travel beyond the village – a process that can be time consuming and costly. This disconnect between citizens and officials is not new. In Krishna’s earlier study, he also noted that very few directly approached the state bureaucracy.36 Similarly small numbers of citizen survey respondents (22 percent) reported turning to elected representatives at the state (MLA) or national (MP) levels. This, in fact, represents a substantial increase when compared to Krishna’s 1997–1998 survey, in which just four percent referenced party representatives when considering who would assist someone seeking access to wage 33

34

35

Taking both selective and collective services together, a full 62 percent reported personally contacting members of the Gram Panchayat (as reported in Table 4.2). An additional question, designed to test the robustness of this measure of citizen-panchayat contact, asked (in a separate module of the survey) whether one had “ever received help from Gram Panchayat officials in obtaining government services and facilities.” Fifty-seven percent responded affirmatively – a rate very much in line with responses concerning citizen contacting of the panchayat. Dunning and Nilekani 2013, p. 38. Another study drawing on survey data from Bihar similarly finds a large role for the Gram Panchayat: a fact that is “unsurprising from the perspective that the [sarpanch] is the most powerful individual closest to village residents” (Bussell 2012b, p. 20). On the centrality of the Gram Panchayat, also see Bholken (2016); and Chauchard (2017). 36 Jenkins & Manor 2017, p. 75. Krishna 2011, p. 109.

Measuring Claim-Making

101

employment on public works programs or seeking to replace an underperforming teacher.37 Even so, the rates at which citizens’ approach politicians still pale alongside rates of contacting the Gram Panchayat.38 This in large part reflects the physical distance of political parties which, like bureaucrats, also largely operate at a level beyond the village.39 Party headquarters and politicians’ offices are located in towns and cities, and visits to villages by politicians are typically restricted to the campaign season. Indeed, seventy-seven percent of those that I surveyed reported that politicians visited their village just once in the past year – usually for the purposes of campaigning. The frequency of direct contact between citizens and their elected representatives is further hampered by sheer constituency size, which averages 226,000 people per MLA in Rajasthan.40 This is not to say that political parties do not play a critical role in village politics; they do, influencing everything from panchayat elections to the budgetary process.41 Parties thus shape the claim-making landscape – even if higherlevel politicians themselves are largely absent from the village. Moreover, as we will see, access to parties and their resources is often itself a mediated affair, brokered through the Gram Panchayat or through non-state actors. Parties are therefore more central to the claim-making arena than the relatively low rates of citizen-politician engagement might at first suggest. Turning to Non-State Intermediaries More than half (54 percent) of the citizen survey respondents reported engaging with some kind of non-state actor or institution when seeking access to public resources. There are, as noted, a wide range of potential intermediaries, but no more than a quarter of the sample reported contacting any given one (Table 4.2). Caste bodies and neighborhood associations top the list at just 37

38

39

40

41

Ibid., p. 110. Across all four of Krishna’s issues areas, an average of five percent referenced party representatives, with the highest response (six percent) for “dealing with land administration or the police.” In response to similar hypothetical questions (Table 4.3), I find that just six and twelve percent cite parties as likely to assist. This is, in part, a function of the nominally non-partisan nature of the Gram Panchayats which, while often informally linked to parties, are prohibited from running on a party platform. There are thus institutional factors that work to diminish the centrality of parties in Rajasthan that are absent in other states (for example West Bengal) where parties are allowed to compete in local elections and so, by design, take on a much greater local role. The weak local institutionalization of political parties in rural India has been widely noted (Weiner 1989; Kohli 1990; Krishna 2002). Bholken 2016, p. 60. See, however, Bussell (2015, 2018), who notes the preponderance of time that MLAs spend on “constituency service” – responding to citizens’ claims and grievances. The relatively high levels of effort by politicians to engage citizens in this manner are diluted by sheer numbers in each constituency. Bussell (2018) also notes that those who are “blocked” at the local level by the Gram Panchayat are most likely to turn directly to higher-level politicians. Dunning & Nilekani 2013; Wilkinson 2007.

102

Seeking the State: Claim-Making Patterns and Puzzles

table 4.4 Mediated Claim-Making Channels, Conditional on Presence Proportion Reporting Claim-Making through Caste Body Intercaste Body Individual Broker NH Association Village Association NGO Social Movement

Presence Reported by

Contact

Full Sample

>50% Village

Where Present

0.66 0.28 0.32 0.32 0.25 0.06 0.11

0.79 0.20 0.14 0.18 0.12 0.04 0.03

0.25 0.25 0.34 0.43 0.37 0.36 0.33

Source: Citizen Survey (2010), n = 2,210. Note: Mediated channel is “present” where reported by >50 % of respondents in a village.

under one quarter each. Substantially smaller numbers turn to other mediated channels: 17 percent employ individual brokers, 15 percent seek assistance through village associations, 14 percent seek out intercaste bodies, while just nine and three percent, respectively, turn to social movement organizations and to NGOs. However, assessing citizens’ mediated approaches to the state is complicated by the fact that not all intermediaries exist in all places. Table 4.4 thus presents the incidence of each mediated channel of claim-making conditional on its reported presence in a given village. The table first depicts the proportion of the full survey sample that reported that a given intermediary was active in their locality, followed by the share of sample villages in which a majority said that the intermediary was present.42 Last, the table shows the rates at which survey respondents contacted the actor or institution in question conditional on its being reported as present by a majority in their village. With this adjustment, rates of mediated claim-making increase substantially: a quarter reported turning to single and intercaste bodies alike; roughly forty percent sought assistance from neighborhood and village associations; while one third or more reported seeking out NGOs, social movement organizations, and individual brokers. Together, the data in Table 4.4 (particularly when read alongside Table 4.2) underscore a simple but important point: claim-making practice reflects the local opportunity structure. Where present, non-state intermediaries play a significant role in facilitating claim-making, but these actors and institutions are unevenly distributed. In their study of Rajasthan, for example, Jenkins and 42

This is the share of villages where more than one half of survey respondents report a given mediated channel. The 50 percent mark sets an artificial cut off: certain intermediaries may be present but only accessible to small groups within a village. It does, however, provide a good sense of how visible a given actor or institution is within a locality.

Measuring Claim-Making

103

Manor find a large role for social movement organizations – but only in “movement-endowed” blocks with a known presence of MKSS or other affiliated organizations.43 In these blocks, they find that social movements have played a critical role in facilitating citizens’ access to local officials and agencies charged with program implementation. In my own survey, which did not purposively select on the criteria of movement presence,44 just nine percent reported turning to social movement organizations. However, in villages where a majority of respondents were aware that an organization of this kind was active in the vicinity, rates of claim-making through movements rose to 33 percent. Social movements thus do provide a critical point of citizen-state engagement in “movement-endowed” areas; most villages, though, are not so endowed.

The Changing Role of Brokers The relatively limited role of individual brokers merits further discussion, in large part because of the attention that these actors have received in accounts of Indian citizen-state relations. In his study in the late 1990s, Krishna noted large numbers turning to informal brokers (naya netas): an average of 65 percent of his survey sample referenced these “new leaders” when reflecting on who they thought would assist citizens with a range of service delivery issues. In 2010–2011, in contrast, just 17 percent of those surveyed reported having personally contacted individual brokers (Table 4.2). Adjusted for their reported presence, the rate of contacting brokers doubles but – at 34 percent – still remains limited (Table 4.4).45 Jenkins and Manor find a similarly reduced role for brokers in Rajasthan and neighboring Madhya Pradesh. They attribute this to the rising authority of the Gram Panchayat’s sarpanch who has come to play a large role in the administration of public programs like MGNREGS. Brokers, they suggest, play a smaller role since higher-level bureaucrats and political parties (those to whom brokers typically make connections) are becoming 43 44

45

Jenkins & Manor 2017. MKSS activists are concentrated around its headquarters in Rajasmand district, and the organization’s local presence does not appear to spill far beyond the district boundaries. In fact, in the more than 500 interviews carried out for this book, only one person (in Ajmer district, adjacent to Rajsamand) mentioned MKSS by name. Again, direct comparison of my data to Krishna’s is inadvisable given differences in survey design and question wording. For example, my definition of “brokers” (“people who … know how to get things done both inside and outside the village and can assist people in accessing government schemes”) differs from Krishna’s notion of the “naya neta” that he defines as those without a formal position who are “experienced in dealing with the government,” who help connect villagers with one another and to government agencies (Krishna 2011, pp. 105–106). Moreover, respondents in Krishna’s survey answered hypothetical questions. In my own hypothetical questions (Table 4.3), even smaller numbers (eight and six percent) state that brokers are likely to assist in problem solving.

104

Seeking the State: Claim-Making Patterns and Puzzles

relatively less important in the administration of programs compared to local panchayat officials, who are taking on larger roles. The highly localized nature of program implementation in effect renders brokers less necessary.46 The role of brokers thus appears to be becoming less pronounced over time.47 This is not to suggest that such actors have simply disappeared: rather, their roles have evolved. While there is, to date, little systematic data on the career trajectories of brokers, there is anecdotal evidence to suggest that many have over time joined the ranks of the Gram Panchayat.48 A “new” (naya) neta in 1997 may, by 2010, have become simply a neta (a politician).49 The odds of this transition increase with each successive round of panchayat election in which caste reservations are imposed; when political office is set aside for members of the Scheduled Castes and Tribes, naya netas from those communities are natural candidates to fill these seats. As Chauchard writes of his encounters with sarpanches in Rajasthan: “While there appears to be no clear requirement in terms of wealth, education, or political experience in order to become sarpanch, each of the officials I interviewed had been a community leader before they managed to win a local-level election.”50 Moreover, while brokers initially appeared to substitute for citizens’ direct engagement with the state (visible in the low rates of approach to the panchayat and high rates of contact with naya netas in Krishna’s data), their roles today appear to be more complementary. Indeed, as we will see in the next section, of those who contacted the Gram Panchayat, over two-thirds reported that they also turn to some kind of non-state intermediary.

46

47

48

49

50

Jenkins & Manor 2017, p. 75. They also note (p. 76), however, that there is an important distinction between the “retail” function of brokers who assist particular individuals, and the “wholesale” function of those who facilitate the exchange of political support for services for entire groups and localities. Wholesale brokers, they suggest, still play an important role in shaping the electoral calculus and broad patterns of distributive politics. Bussell (2015) highlights similar trends in an all-India context, noting relatively low rates of brokerage compared to citizen’s direct contacting of local, state, and national politicians. Krishna (2011, p. 114) anticipated the likelihood of change among the naya netas, writing: “The roles that these intermediaries play are neither stable nor predictably recurring, so there is little that is ‘institutional’ about naya netas … Their numbers should surely grow in the aggregate, but at the local level, the behavior of individual naya netas is fickle. So it is hard to predict how any given citizen, especially a poorer one, will be gaining access to welfare a year from now.” Neta in Hindi means “leader,” and typically connotes a politician – although it can also refer to a caste leader (jati neta), community leader (samajic neta) or, in the case of Krishna, an “new” leader (naya neta) playing the role of broker. Chauchard 2017, p. 81. Not all of these leaders precisely fit the description of a naya neta: some had been caste leaders, others labor organizers, and still others had campaigned for political parties during state elections. All, though, played some prior intermediary role, and so were “a resource [to which] villagers … turned to when they needed assistance with various bureaucratic transactions.” All drew upon this prior experience and standing in the community when contesting panchayat elections.

Measuring Claim-Making

105

table 4.5 Combining Claim-Making Practices Proportion Reporting Claim-Making Through

Contact Mean

Obs.

Std. Dev.

Multiple Channels Direct and Mediated Channels Direct Channels Only Mediated Channels Only

0.54 0.43 0.22 0.10

2,210 2,210 2,210 2,210

0.50 0.50 0.41 0.31

Among Those Who Contact the Gram Panchayat GP Only Plus Political Party Plus Bureaucrat Plus Intermediary Plus Broker Plus NH Assoc. Plus Village Assoc. Plus Caste Body Plus Intercaste Body Plus NGO Plus Social Movement

0.20 0.32 0.31 0.67 0.22 0.29 0.21 0.30 0.19 0.03 0.10

1,375 1,375 1,375 1,375 1,375 1,375 1,375 1,375 1,375 1,375 1,375

0.4 0.47 0.46 0.47 0.41 0.45 0.41 0.46 0.39 0.18 0.31

Among Those Who Contact Politicians/Parties Politicians/Parties Only Plus GP Plus Bureaucrat Plus Intermediary Plus Broker Plus NH Assoc. Plus Village Assoc. Plus Caste Body Plus Intercaste Body Plus NGO Plus Social Movement

0.04 0.91 0.47 0.74 0.28 0.29 0.24 0.35 0.25 0.04 0.14

480 480 480 480 480 480 480 480 480 480 480

0.20 0.29 0.50 0.44 0.45 0.45 0.43 0.48 0.44 0.20 0.35

Source: Citizen Survey (2010–2011), n = 2,210.

The Claim-Making Repertoire The strategies of action described above are combined into repertoires of action of varied content and breadth, ranging from those who do nothing at all, to those who depend on a single point of contact, to those who pursue multiple channels of access to the state. A majority (54 percent) reported engaging in more than one kind of practice (Table 4.5), employing an average of two on an index ranging from 0–10 (Table 4.2). Most, in other words, are not locked into singular modes of approach. There is, instead, a degree of room to maneuver at the local level.

106

Seeking the State: Claim-Making Patterns and Puzzles

Most citizens, moreover, combine direct and mediated approaches to the state (Table 4.5). For example, of those who reported contacting the Gram Panchayat, just 20 percent reported doing so singularly, without also contacting another channel. Almost one-third contacted bureaucrats or parties in addition to the GP, while more than two-thirds also engaged a range of nonstate intermediaries. Perhaps the most striking dynamic in these combined practices is the interplay between political parties and the Gram Panchayat. Panchayat members are nominally non-partisan and are prohibited from contesting an election on a party ticket. In practice, though, parties hold considerable sway over the GP, influencing both the nomination of candidates and the electoral process. GP candidates often run on the basis of promised links to political parties, and parties are known to help finance local campaigns. This increases the political salience of the GP, which often acts as a gatekeeper to parties and their resources. Of those who reported turning to parties for the purposes of claim-making, 90 percent also contacted the Gram Panchayat.51 This again underscores, as highlighted in Chapter 3, the growing political salience of the Gram Panchayat.

disaggregating claim-making: what, where, and who? Thus far, this chapter has presented aggregate data on the incidence of claimmaking, the practices employed, and the breadth of the claim-making repertoire. The remainder of the chapter takes a disaggregated view, considering whether patterns vary, first, according to the nature of the goods and services in question; second, by locality; and third, by individual characteristics such as socioeconomic class, caste, and gender. What Is Being Claimed? Claim-making happens in reference to particular kinds of goods and services, which are produced by different agencies and delivered according to different procedures. Is there, then, something about the nature of the goods themselves that shapes claim-making activity? Table 4.6 depicts the rates at which citizens turn to different claim-making channels, first for collective goods of a highspillover nature (schools, health clinics, roads, or drinking water); and second, for low-spillover selective benefits (pensions; food rations; MGNREGS employment; or other schemes targeted to poor and lower caste households). Across all channels, rates of claim-making are significantly higher for collective goods than selective benefits. For all kinds of goods and services, though, there is 51

Bussell (2015) finds similar patterns in a survey in Bihar, Jharkhand, and Uttar Pradesh where, among respondents who said that individuals would go to a senior [MLA] politician for assistance, only 6 percent said they would do so without also turning to the local panchayat president.

Disaggregating Claim-Making: What, Where, and Who?

107

table 4.6 Claim-Making for Collective versus Selective Services Proportion Reporting Claim-Making Through Direct Channels Gram Panchayat Bureaucrats Politicians/Parties Mediated Channels Caste Body Intercaste Body Individual Brokers NH Association Village Association NGO Social Movement Claim-Making Incidence

Collective Goods Mean

Selective Benefits Mean

Diff. in Means

0.60 (0.010) 0.57 (0.011) 0.20 (0.008) 0.21 (0.009)

0.56 (0.011) 0.53 (0.011) 0.14 (0.007) 0.15 (0.008)

0.04***

0.51 (0.011) 0.22 (0.009) 0.13 (0.007) 0.14 (0.007) 0.21 (0.009) 0.15 (0.008) 0.03 (0.003) 0.08 (0.006)

0.39 (0.010) 0.17 (0.008) 0.10 (0.006) 0.12 (0.007) 0.15 (0.008) 0.11 (0.007) 0.01 (0.002) 0.03 (0.003)

0.12***

0.72 (0.010)

0.64 (0.010)

0.04*** 0.06*** 0.06***

0.06*** 0.03*** 0.02** 0.06*** 0.03*** 0.01*** 0.06*** 0.08***

Source: Citizen Survey (2010–2011), n = 2,210. Note: Results from two-sample tests of proportion. Standard errors are shown in parentheses. Standard levels of significance apply, where * = p-value < 0.10; ** = p-value < 0.05; *** = p-value < 0.01.

considerable variation, and the overall mix of strategies is quite similar for both categories. Beyond the broad categories of “collective” and “selective” goods, though, the survey data are limited in their power to unpack whether particular resources are associated with particular kinds of claim-making. It is indeed likely that some of the choices that citizens make reflect the nature of the goods being claimed and the different agencies under whose jurisdiction those fall. Certain things, like gaining access to employment on a MGNREGS worksite, by definition fall under the mandate of the Gram Panchayat, which allocates job cards and oversees payment. It is thus unsurprising that large numbers of

108

Seeking the State: Claim-Making Patterns and Puzzles

citizens would approach the GP when concerned with accessing programs like MGNREGS. In contrast, other services – for example, water supply – do not fall under the panchayat’s jurisdiction; water, instead, is the purview of the state’s Public Health and Engineering Department (PHED). Complaints about water, then, ought to be targeted to district or block officials. As noted, though, contacting bureaucrats is often a difficult affair, and many citizens instead turn to the GP or non-state intermediaries for assistance in approaching these higher-level officials. The extent to which we should expect formal jurisdictional boundaries to guide claim-making practice is thus unclear. These dynamics were visible in village-level interviews, many of which focused on MGNREGS (a critical source of income) and on drinking water (a critical resource in a drought prone region). As the story of the broken water pump in the beginning of this chapter highlighted, different residents of the same village (and even the same caste community) pursued varied strategies in their efforts to get the PHED to deliver water. Practices were similarly mixed when it came to securing employment and payment on a government worksite. MGNREGS is a universal program (any rural household can request a job card entitling them to 100 days of work each year). It is also theoretically the case that one household’s employment should not limit another households’ opportunities. In this sense, MGNREGS might be considered a “public” program (in the classic sense of a good that is both non-rival and non-excludable). In practice, though, gaining access to job cards and to days of work can be a tricky affair, due in large part to high levels of local graft.52 In this sense, the program is selective in its benefits as well as “low-spillover” as it targets households. It is therefore unclear precisely where to place MGNREGS employment on the spectrum from “collective” to “selective.” This ambiguity is reflected in citizens’ varied claim-making strategies. In a single village in Udaipur, for example, residents gave very different accounts of their efforts to obtain work. Some reported directly and individually approaching the Gram Panchayat. A young man stated: Getting rozgar [work] is simple; if I want to work, I go to the sarpanch, and he will tell the “mate” [the MGNREGS supervisor] to put me on the list.53

Others, though, had less straightforward experiences. A woman described her efforts: There was the issue of getting documentation so that we could get work. We went again and again to get our names on the list [for job cards]. So it became necessary to do more. Our whole [women’s self-help] group went at once. There were so many of us that we crowded the office, and the officials there finally gave up the papers.54 52

53 54

On forms of corruption in the administration of MGNREGs, see Niehaus and Sukhtankar (2013) and Witsoe (2014). Author interview, Bargaon Block, Udaipur District, January 12, 2011. Author interview, Bargaon Block, Udaipur District, January 11, 2011.

Disaggregating Claim-Making: What, Where, and Who?

109

In yet another instance, residents solicited the help of a local NGO: There was a problem of payment. The money was locked up there in the panchayat for weeks. When they did give, it was not even the full amount. And so we went to the sanstha [a NGO], and they accompanied us to the panchayat bhavan.55

As seen in these accounts, as well as in those related to the broken water pump earlier in the chapter, citizens pursued varied strategies (individual and collective, direct and mediated) in their pursuit of both MGNREGS and communal drinking water. Thus, while supply-side issues related to the production and allocation of goods clearly do play a role, it is too simple to say that claimmaking practices directly reflect the nature of the goods sought. In a service delivery environment marked by high levels of uncertainty, citizens pursue multiple strategies that reach beyond agencies with formal jurisdiction over the supply of the resources in question. We therefore need to look more closely at the characteristics not just of the goods involved but of the individuals engaged in claim-making. Two closelyrelated bodies of literature emphasize resources for participation. The first predicts higher levels of civic and political engagement among those living in more developed settings marked by relative levels of affluence, economic modernization, and urbanization. The second observes a relationship between an individual’s socioeconomic class and political activity, predicting higher levels of participation among citizens with higher social standing. The survey data allow us to test each of these sets of predictions in turn. Where Is Claim-Making Most Likely? Modernization theorists predicted that rising levels of affluence, spurred by industrialization and urbanization, would prompt higher levels of participation. As incomes rise, the theory suggests, citizens become more engaged in politics, developing both the preferences and capabilities to demand more effective and accountable government.56 The same body of work also expects greater participation in more urbanized settings, where people overcome the “parochial” tendencies of rural life to form a broader, more “civic” political community.57 Could disparate patterns of claim-making within in Rajasthan thus be a result of varied levels economic modernization and development? 55 56

57

Author interview, Bargaon Block, Udaipur District, January 10, 2011. Lipset 1959; 1960. More recent scholarship on modernization includes Przeworski and Limongi (1997) and Przeworski, Cheibub, Alvarez and Limongi (2000), who argue that economic development affects the stability of political regimes, thus enabling the consolidation of existing democracies and sustaining spaces for participation. For reviews and critiques of this body of work, see Boix and Stokes (2003); and Krishna (Ed.) (2008). See Lerner (1958), Deutsch (1961), and Almond and Verba (1963) on the socially and culturally transformative effects of modernization and, in particular, urbanization, which they argue enables the development of more “cosmopolitan” values. For a review and critique of this school

110

Seeking the State: Claim-Making Patterns and Puzzles

As noted, the four districts were selected as research sites in part because of their differences in economic and human development standing. Figure 4.1 portrays district-wide claim-making averages, revealing some divergence. Residents of Jodhpur and Udaipur are, respectively, seven and five percent less likely than residents of the other districts to make claims on the state, while residents of Kota (the wealthiest district) are eleven percent more likely to do so.58 That Kota stands out as a center of claim-making is broadly consistent the prediction that, as incomes rise, so too will levels of participation. The story is complicated, though, by the fact that claim-making levels in Jodhpur and Ajmer (middleranking districts in terms of their poverty rates and human development ranking) are almost indistinguishable from those in Udaipur (which is the poorest district in the sample, with a human development score that is roughly two times lower than that of Jodhpur or Ajmer). This suggests, at the very least, that the relationship between district-level development and claim-making is a non-linear one. The data also reveal important variation in claim-making practice within districts, prompting the need to consider more local-level characteristics. Multivariate analysis (described in Appendix II, and developed in Chapter 6) confirms the lack of significant association between a range of indicators of wealth (from land and asset ownership to literacy rates) and levels of claim-making in a village. Demographic and geographic features similarly lack predictive power: neither village population, density, nor proximity to a town is significantly correlated with claim-making.59 In sum, it is simply not the case that people residing in richer or more “urbanized” places demonstrate a greater propensity to engage the state. Who Makes Claims, and How? A closely related body of work shifts the focus to the individual level, predicting that people of higher social standing will participate with greater vigor.60 As Milbrath and Goel put it: “No matter how class is measured … higher class persons are more likely to participate in politics than lower class persons.”61 In their seminal study of participation in the United States, Verba et al. (1995) document a positive relationship between socioeconomic

58

59 60

61

of thought, see Varshney (1998a), who cites India as an exception since democratization preceded India’s industrial revolution. Results are from two-sample tests of proportion in which district dummies are used to compare average claim-making incidence in each district to the average across all four districts. Differences in mean are significant for Jodhpur (p = 0.016), Udaipur (p = 0.065); and for Kota (p < 0.000), but not for Ajmer (p = 0.787). Results in Chapter 6, reported in full in Appendix II, Table A2.2. Most of this literature builds on studies of advanced industrial countries. See, for example, Milbrath and Goel (1977); Bennett and Bennett (1986); Rosenstone and Hansen (1993); Verba et al. (1995). Similar predictions extend beyond income to other markers of social standing, such as ethnicity (Brady, Verba, & Schlozman 1995). Milbrath & Goel 1977, p. 92 (emphasis in the original).

111

Claim-Making Incidence (district mean) 0 0.2 0.6 0.8 0.4

Disaggregating Claim-Making: What, Where, and Who?

Jodhpur

Ajmer

Udaipur

Kota

figure 4.1 Claim-Making Incidence by District Note: The horizontal line represents the full sample mean. Source: Citizen Survey (2010–2011), n = 2,210.

status and a wide range of political activities that extend beyond voting, including campaign work, contacting officials, protest, and other forms of collective action.62 Studies from around the world, but particularly in the West, have documented similar patterns.63 And yet, a widely noted “paradox” of Indian politics is that poor and lower-caste citizens vote at rates as high or higher than elites – despite their persistent social and economic exclusion.64 These patterns hold beyond the voting booth. In the rural Indian setting, for example, Krishna finds that there is no significant relationship between levels of wealth and participation across a range of activities including voting, campaign activity, contacting officials, and protest.65 In an urban context, Harriss finds that lower income groups in Delhi are more likely to engage in a range of problem-solving activities, including contacting officials, than higher-income groups.66 My citizen survey data similarly underscore a lack of consistent relationship between markers of socioeconomic status and the likelihood of claim-making. High rates persist across different social groups. These patterns – visible in Figure 4.2 – are confirmed by difference in means testing that compares rates of claim-making incidence across different quintiles of landownership; across different caste 62

63

64 66

Some argue that “particularized” contacting is distinct from other kinds of more “public” political activity (Verba, Nie, & Kim 1978). Here, though, the empirical record is mixed. While Verba et al. find a lack of correlation between particularized contacting and socioeconomic status, others find a positive relationship (Sharp 1982). Dunning (2009), for example, documents similar class distortions in rates of participation across Latin American cities, where the wealthy are more engaged in both direct and brokered political action. 65 Mitra & Singh 1999; Yadav 2000; Alam 2004; Ahuja & Chhibber 2012. Krishna 2008. Harriss 2005.

(a) -0.002

0.048

0.010 -0.045**

0.106***

Gender and Caste

0.098**

Caste Categories

0.036

0.013

0.022

Note: Depicts mean claim-making incidence sorted by (a) quintiles of landownership, (b) caste category, (c) gender, and (d) the interaction of caste and gender. Values on top of the bars are the difference in means for each category compared to the remaining sample. * = p-value < 0.10; ** = p-value < 0.05; *** = p-value < 0.01. The horizontal lines represent the full-sample mean. Source: Citizen Survey (2010–2011), n = 2,210.

figure 4.2 Claim-Making Incidence by Class, Caste, and Gender

Gender

-0.279***

0.099***

Quintiles of Landownership

0.007

(d)

-0.059***

-0.008

(b)

(c)

Claim-Making Incidence (Mean)

Claim-Making Incidence (Mean)

Claim-Making Incidence (Mean)

Claim-Making Incidence (Mean)

112

Disaggregating Claim-Making: What, Where, and Who?

113

categories; between men and women; and, among women, across castes.67 Across all groups, rates of claim-making exceed fifty percent. The data fail to show a linear relationship between socioeconomic status and the likelihood of claim-making. When it comes to landownership, for example, those with the most land (the fifth quintile) do not significantly out-claim the rest – including those with the least. Instead, those in the fourth quintile, are the most likely to make claims – exceeding all others by 13 percent. Broad categorizations of caste, moreover, do not play a large role in determining who makes claims on the state. While the middle-ranking OBCs are six percent less likely than all others to engage in claim-making, there are no significant differences between the SC, the ST, or the GC compared to the full sample. There is a pronounced gender gap. In a setting where many women (both Hindu and Muslim) live under variants of the purdah (veil) system,68 we might expect women’s participation to be curtailed. This is indeed the case: women are a full 32 percent less likely to engage in claim-making than men. Notably, though, tribal women are the most likely among women to engage in claimmaking: they are 16 percent more likely to do so than all other women – including women from the upper castes.69 OBC women, in contrast, are the least likely to make claims – lagging behind all other women by 18 percent.70 There is significant variation in how different groups engage the state – visible in Figure 4.3 (again confirmed through difference in means testing).71 Those with lower social standing undertake significantly narrower repertoires of action. Those who own the least land, for example, are the least likely to engage in almost all practices – with the notable exception of contacting NGOs, which they are a full 73 percent more likely to engage.72 This is reflected in a 21 percent contraction of the claim-making repertoire compared to all others quintiles of landownership. Those in the fourth and fifth quintiles, in contrast, engage in a wider range of practices than the rest of the sample – reflected in a 23 and an 9 percent broadening of repertoire, respectively. The land-rich are more likely 67

68 69

70

71 72

Results derived through two-sample tests of proportion comparing each sub-group to the remaining sample. All results presented in this chapter are significant at the 10% level or lower. Full results are presented in Table A2.1. These dynamics are described in greater detail in Chapter 5. Results from two sample t-tests, as described above, but not shown in Table A2.1. The difference in mean claim-making incidence between ST women and all other women is 10 percentage points (p = 0.016). SC women are also eight percent more likely than other women to make claims, although this result is not significant at conventional levels (p = 0.247). This likely reflects the fact that tribal and lower caste women are more likely to work outside the home – thus gaining knowledge of and access to channels for claim-making that are obstructed for other women. The difference in mean claim-making incidence between OBC women and all other women is 10.5 percentage points (p = 0.001). GC women do not differ significantly from other women. Full results are presented in Table A2.1. This likely reflects the active targeting of the poor by NGOs. The base levels at which the land poor contact NGOs still remains low (at just 3.9%). This though, represents a 1.6 percentage point increase over the mean incidence of contacting NGOs in the rest of the sample (2.2 percent).

(a)

-1.292 ***

0.166 *

SC Women

0.044

-0.212**

OBC Women

-0.198**

Gender and Caste

ST Women

0.149

Caste Categories

-0.096

-0.012

GC Women

0.110

0.304 ***

Note: Depicts mean breadth of claim-making repertoire sorted by quintiles of (a) landownership, (b) caste category, (c) gender, and (d) the interaction of caste and gender. Values on top of the bars are the difference in means for each category compared to the remaining sample. * = p-value < 0.10; ** = p-value < 0.05; *** = p-value < 0.01. The horizontal lines represent the full-sample mean. Source: Citizen Survey (2010–2011), n = 2,210.

figure 4.3 Claim-Making Repertoire by Class, Caste, and Gender

Gender

Quintiles of Landownership

0.008

(d)

-0.417 ***

0.021

0.448***

(b)

(c)

Claim-Making Repertoire (Mean)

Claim-Making Repertoire (Mean)

Claim-Making Repertoire (Mean) Claim-Making Repertoire (Mean)

114

Patterns and Puzzles

115

to engage in most practices – with the exceptions of contacting NGOs (which they are less likely to do) and contacting members of the Gram Panchayat and fixers (for which there are no significant class-based differences). Repertoires also contract for the lower castes and tribes, but expand for the upper castes. The ST, for example, engage in 11 percent fewer practices while the GC employ 15 percent more, compared to the full sample. Directly comparing those at the high and low ends of the caste hierarchy, moreover, we find that both the SC and ST engage in significantly narrower repertoires than the GC – visible in 14 and 18 percent contractions, respectively. Women also pursue 5 narrower repertories than men, given that they are less likely to engage in all kinds of claim-making practice. In sum, while class and caste standing tell us relatively little about how likely a given individual is to make claims on the state, these characteristics tell us much more about the breadth of the repertoire of practices that will be employed. However, within all repertoires, the Gram Panchayat stands out as the first port of call. It is also, along with contacting fixers, one of only two channels for which there is no significant class or caste gap in rates of contacting. Those in the upper-middle landowning class are the most likely to turn to the GP. However, the richest villagers (those with the most land) and the poorest (with the least) are no different from each other in the rates at which they contact the panchayat. In fact, majorities from all economic groups reported engaging elected members of the GP. The same is true across caste: large majorities (over 60 percent) from all castes groups approach the GP. Women, we have seen, lag significantly behind men in almost all claim-making activities, and contacting the GP is no exception. Despite this, the GP remains the most common channel of approach among women, utilized by 44 percent. Other forms of direct claim-making, though, exhibit strong class and caste biases: the land-rich and the upper castes are significantly more likely to contact both bureaucrats and parties compared to the land-poor and the lower castes and tribes. The picture is more mixed when it comes to mediated forms of claim-making. The GC are the least likely to contact neighborhood associations but the most likely to contact caste and intercaste bodies (perhaps reflecting their dominance in these customary institutions). The SC are the most likely to contact neighborhood associations, but the least likely to contact broader village associations or inter-caste bodies. The ST are the most likely group to contact NGOs (many of which target tribal communities in Rajasthan), but the Scheduled Castes are the least. The precise content of the claim-making repertoire thus varies significantly from one caste grouping to another – even as its breadth remains most restricted among the lower castes and tribes.

patterns and puzzles Taken altogether, the data underscore both the centrality of the state to rural citizens’ lives, and the complex strategies of action employed when seeking its

116

Seeking the State: Claim-Making Patterns and Puzzles

resources. Three broad patterns emerge. First, the overall incidence of claimmaking is high: the vast majority reported that they have personally engaged in state-targeted action in efforts to secure social services, infrastructure, and welfare goods. Second, the practices employed are diverse, ranging from direct (face-to-face) contacting of village officials, local bureaucrats, and state and national politicians, to mediated contact through individual brokers, a variety of local associations, NGOs, and social movement organizations. The rates at which citizens turn to these various actors and institutions vary, with the largest numbers turning directly to local elected representatives and smaller numbers turning to intermediaries. Within these repertoires of action, direct and mediated strategies appear to complement, rather than substitute one another. Third, repertoires are more restricted for women and the lower castes and tribes, compared to men from the upper castes. Lower income groups and those from the lower castes and tribes, while just as likely to take action overall, have less room to maneuver in doing so. Women are the least likely to engage the state and pursue the narrowest set of practices. And yet even among this socially and politically marginalized group, a majority still engages in claimmaking – including even higher numbers among SC and ST women. The high aggregate incidence of claim-making is in large part a reflection of the increasingly broad, but also uneven and complex, presence of the state in citizens’ daily lives – as described in Chapter 3. Rising budgets and a proliferation of social programs have, simply put, increased the size and the visibility of the public pie. There are resources to be had, and citizens are increasingly aware of this fact. They are, moreover, resources that matter. Accessing the state is, for many, an issue – if not of life or death – of whether and what one will eat, where one will work, and whether one has access to basic protection and care. The motivation for claim-making is therefore high. And yet, as we have observed, claim-making incidence and practice vary at the individual level, both across but also within local communities and socioeconomic groupings. What accounts for such different courses of action and inaction? Why do some make claims on the state while others do not? And what, moreover, accounts for differences in the strategies pursued? These are the puzzles that the next two chapters explore. As Chapter 5 will demonstrate, the state is not uniformly visible and accessible: different individuals see and encounter the state in different ways. This, I will show, reflects differences in the degree to which a person traverses boundaries of caste, neighborhood, and village, and the social and spatial exposure he or she gains beyond the immediate community and locality. As I go on to show empirically in Chapter 6, greater exposure of this kind increases both the likelihood and the breadth of claim-making practice by building citizens aspirations and capabilities for state-targeted action.

5 Encountering the State Citizens’ Social and Spatial Exposure

This chapter investigates the spaces in which citizens encounter the state in rural Rajasthan, and the implications of those encounters for claim-making practice. Chapter 3 described the changing terrain of the state in rural India and its deeper but uneven penetration into local lives and livelihoods. However, as we saw in Chapter 4, different people navigate this terrain in different ways, as they are constrained by different sets of social and spatial boundaries. What are these boundaries in the rural Indian context and what drives movement across them? In the first part of this chapter, I draw on literatures in anthropology and political economy to describe the salience of class, caste, gender, neighborhood, and village cleavages.1 The second part of the chapter discusses the permanence and permeability of these boundaries, examining a series of structural, political, and institutional changes that together have begun to enable greater connectivity and mobility. The declining centrality of agriculture and diversification of rural livelihoods; increased connectivity and mobility; rising literacy rates; increasing electoral competition; and institutional interventions targeting the lower castes, tribes, and women have, altogether, profoundly altered rural life. The combined effect of these changes, I argue, is a slow but pronounced increase in the porousness of social and spatial boundaries – particularly of caste and of village – and an associated increase in exposure to people and places beyond immediate communities and localities. This exposure, in turn, increases the frequency with which individuals encounter and learn about the state, both through personal experience and through the narrated accounts of other citizens. However, this process is not 1

Religion, of course, is another salient cleavage in much of India, and tensions across the Hindu– Muslim divide have received much attention in the ethnic politics literature. The rural Rajasthani sample considered here, though, is almost fully (98 percent) Hindu. Religion is therefore not a primary marker or cleavage within the study population.

117

118

Encountering the State: Citizens’ Social and Spatial Exposure

uniform: boundaries that remain highly restrictive for some are more malleable for others. The third part of the chapter illustrates this unevenness and its consequences for citizen action. Drawing on village-level accounts from across Rajasthan, I explore how differences in social and spatial exposure shape citizens’ aspirations and capabilities for claim-making, influencing both their linkages to the state and their expectations regarding social welfare provision.

social and spatial boundaries Boundaries, as discussed in Chapter 2, refer to the ways in which people and communities define “who and what is properly included and who and what stands outside.”2 In the rural Indian context, these boundaries are both social (centered on notions of community and kinship) and spatial (centered on place of origin and residence). Perhaps the most obvious set of social divisions in rural India revolve around caste.3 In a village setting, caste is often highly visible – identifiable through one’s family, surname, occupation, or where one lives. Caste-based norms restrict social interactions, with prohibitions extending to marriage, to religious rituals, and to more day-to-day forms of social interaction such as visiting others’ homes or sharing food.4 As such, caste demarcates communities, although the strength of demarcation (as well as rigidity of caste hierarchies) is variable.5 Historically, the caste system provided the basis for division of labor in an agrarian economy. Each caste and subcaste was associated with a particular occupation – although occupational divides were never completely rigid.6 Caste and occupation remain conceptually tied in much of rural India, where certain tasks are considered “traditional” to a given caste community – whether or not all members of that community actually engage in them. Caste also continues to 2 3

4

5

6

Isaac 2011, p. 779. Caste is an ascriptive characteristic: membership is given at birth. There is no consensus on how to define India’s caste system. Typically, however, its features include the hierarchical structuring of hereditary groups, traditionally based around occupation, with norms restricting social interactions – and, in particular, marriage – between groups (Ghurye 1969; Beteille 2011). Such prohibitions on social interactions are based, broadly, on hierarchical rankings of castes and subcastes where those closer to the “top” are associated with “purity” and those toward the “bottom” with “pollution.” “Untouchability” is an extreme form of caste-based discrimination, manifest in physical avoidance of those who do not fall within the four primary social groupings (or varnas) within Hinduism, and so are considered “outcastes.” Untouchability, which also encompasses physical humiliation and social exploitation, was banned under the Indian Constitution following Independence, but its practices persist – particularly in rural India. The permanence of caste structures has long been contested. Some, like Dumont (1970), have argued that caste hierarchies are deeply internalized among members of the upper and lower castes alike, thus restricting the possibilities of mobility, as well as challenge, from those at the bottom. Others, in contrast, have argued that caste boundaries are fluid, best understood as a set of shifting social demarcations rather than a rigid hierarchy (Srinivas 1966; Dirks 2001). Srinivas 1987; Dirks 2001; Beteille 2011; Gang et al. 2017.

Social and Spatial Boundaries

119

roughly predict many dimensions of socioeconomic status, including income, consumption, landownership, education, and human development standing.7 It follows that caste boundaries continue to represent – albeit imperfectly – socioeconomic divisions within a village. These divisions are reflected in local citizen–state relations, visible in the “persistence of hierarchical values associated with caste as a factor in the way in which government works.”8 However, as we will see throughout this chapter, the role of caste in contemporary social, economic, and political life is rapidly shifting. Other boundaries are spatial. Sociological and anthropological studies of rural India have long focused on the idea of the village as a social unit. For some scholars, a village’s identity is subsumed by a “dominant” caste in a given locality.9 Others, though, have argued that the village is an important marker of social identity in its own right (as are ideas, more broadly, of “locality, place, territory”).10 Whether village boundaries connote an identity of their own above and beyond caste depends, in part, on the ethnic composition of the village. In a single caste village, the boundaries of caste and locality are one and the same; the village is both a territorial and a social unit. Where mixed caste, villages are often divided into distinct neighborhoods (or mohallas), where each mohalla is home to a different caste community. But even where divided along caste lines, the village often retains its own social resonance. As Lambert notes, “In Rajasthan … the first question to be asked is not ‘What caste are you?’, or ‘What is your name?’, or ‘What do you do?’, but ‘Where are you from?’”11

7

8

9

10 11

For overviews of contemporary patterns of caste disparities and differentials, see Deshpande (2011); Desai and Dubey (2012); and Thorat and Neuman (2012). Harriss [2012], cited in Singh 2012, p. 22. It is important to note, though, that caste structures in Rajasthan were historically less rigid than in other parts of India. “High”-caste Brahmins in Rajasthan maintained social and intellectual prominence but were economically and politically inferior to the Rajputs (members of the Kshatriya [warrior] Hindu varna or class) who enjoyed political dominance despite their “lower”-caste status relative to Brahmins. This created what Narain and Mathur (1989) have referred to as a “de-Sanskritized” culture in which Rajput norms prevailed over Brahminical culture. Today, most Rajputs in Rajasthan are considered “upper” caste, falling within the “general” (GC) category. Srinivas (1951) popularized the notion of local caste “dominance,” defined in terms of the numerical preponderance of a particular caste coupled with their exercise of economic and political power. Dumont and Pocock (1957, p. 27) build on this notion to argue that any notion of village “solidarity” that appears to transcend caste in fact represents the power of the dominant caste. Lambert 1996, p. 93. Ibid., p. 95, emphasis added. Lambert notes that the question translated from Rajasthani dialect is literally, “Which is your village?” This can refer either to place of long-term residence or place of origin. For men (even among migrants), this is usually one and the same. For women, village identification is more complex since most adult women have moved after marriage to another village. A woman thus identifies with both a “native” place that typically refers to the natal village in which she was born and with her sasural, or mother-in-law’s home.

120

Encountering the State: Citizens’ Social and Spatial Exposure

Within villages, mohallas are also important markers of social and spatial identity. Mohallas are not always officially recognized, but their boundaries are apparent to most villagers – particularly since they often map onto caste cleavages within a village. In rapid mapping exercises before administering the citizen survey, I found that residents could name and locate different caste-based mohallas with remarkable ease and consistency. Not all mohallas, though, are drawn along caste lines. Some are mixed, with households from different castes living shoulder to shoulder. These mixed-caste mohallas are most often comprised of middle- to upper-class and caste families (drawn from the relatively more affluent among the Other Backward Classes and General Castes). The lower (“Scheduled”) castes and tribes typically reside in spatially distinct areas of the village, often along the periphery or, as is case for many tribal communities, in hamlets set at some distance from the central village. It is not uncommon to find that mohalla boundaries also demarcate differences of economic class within a village. So-called pucca (“proper”) homes built from concrete or other durable materials are typically concentrated in the center of the village where the middle- and upper-caste households reside – an area commonly referred to as the “main” village, or mukhya gaon. Government buildings (such as panchayat offices, schools, and health clinics) are most often located in the mukhya gaon, which is also much more likely to have paved roads and reliable sources of drinking water than more peripheral areas of the village. Housing on the outskirts of the village and in its outlying hamlets is very often of much poorer quality, made out of kaccha (rough or makeshift) materials. Access and interior roads in these areas are more likely to be unpaved and are often impassable by vehicle. Another set of divisions is based on gender. In most of rural India, women’s social and spatial mobility is constrained relative to men’s. This is particularly true in northern India, where many women live under purdah, a system of physical and social seclusion that prohibits participation in mixed-gender spaces or institutions.12 These norms of seclusion limit women’s engagement in the public sphere and restrict economic opportunities by curtailing the occupations that women may pursue.13 They also physically curtail a woman’s movement beyond her home and immediate neighborhood. These dynamics are particularly pronounced in Rajasthan, as compared to other parts of the country. Rajasthan has one of the worst records of gender inequality

12

13

Purdah literally means “curtain,” referring to the veil that many women who practice purdah wear, which physically and symbolically secludes them from the public sphere. Purdah is a central part of familial and social structure for women across much of South Asia, both Muslim and Hindu. Muslim and Hindu norms for seclusion vary, but, in all cases, purdah is marked by practices that restrict interactions between women and men. See Papanek (1973), Miller (1982, 1993), Mandelbaum (1986, 1993), Channa (2013). See, for example, Field, Jayachandran, and Pande (2010).

Social and Spatial Boundaries

121

nationally, and the gap between women’s and men’s attainment on a wide range of human development indicators remains staggering.14 However, not all women are equally constrained. Within the Hindu hierarchy, the enforcement of purdah is most prevalent among the “upper” castes – primarily those from GC or OBC communities.15 Women from the “lower” castes, in contrast, are more likely to work outside the home, often as agricultural laborers or on MGNREGS sites. Even where purdah is not strictly applied, restrictive gender norms continue to limit the spheres that women occupy. These limitations are felt, to varying degrees, among women from all caste and economic backgrounds. The result is that barriers of caste, neighborhood, and village are most pronounced for women. Together, divisions of caste, class, neighborhood, village, and gender play important roles in structuring social, economic, and political life. These boundaries are closely intertwined: spatial demarcations of village and neighborhood simultaneously delineate caste and socioeconomic distinctions within and across localities. Gender barriers also have a strong spatial dimension since women’s mobility is constrained both within and beyond the village. Examining these interrelated social and spatial cleavages thus casts light on how “differently placed men and women see the state in rural India.”16 Scholarship on Indian citizen–state relations regularly highlights these differences, disaggregating patterns of political participation and material distribution by caste, economic standing, gender, and locality.17 This rich literature illuminates the ways that different groups of people encounter the state. And yet, as we saw in Chapter 4, citizen–state engagement also varies among similarly placed people. This variance, I argue, can be significantly explained by differences in citizens’ social and spatial exposure. Boundaries of caste, class, gender, neighborhood, and village are not uniformly constraining; those who traverse them gain exposure to information, ideas, practices, and networks that, together, build their claim-making aspirations and capabilities.

14

15

16 17

Joshi 2008; IDS 2008; P. Singh 2015. Rajasthan is also among the states with the worst records of violence against women, including the practice of sati (bride burning), and has some of the highest rates of child marriage and female infanticide in the country. Joshi, Kocchar, & Rao 2017. In Rajasthan, purdah practices are particularly common among the high-caste Rajput communities in Rajasthan. Aspirations for upward social mobility, marked by an emulation of “upper”-caste norms, may also lead to “lower”-caste women living under variants of purdah or other restrictive practices. See, for example, Bhattacharjee (2016). Corbridge et al. 2005, p. 15 (emphasis added). Corbridge et al.’s pioneering work Seeing the State (2005) is a prime example that attempts to map experiences or “sightings” of the Indian state through the eyes of men and women from different caste and tribal backgrounds. Numerous other works, both ethnographic and quantitative in nature, parse patterns of citizen–state relations in similar ways. See, for example, Banerjee (2004); Ahuja and Chhibber (2012); Bussell (2015); Harriss (2005); Krishna (2011); Kruks-Wisner (2011).

122

Encountering the State: Citizens’ Social and Spatial Exposure

boundary porousness Barriers of caste, neighborhood, village, and, to an extent, gender are becoming more porous over time. This, I argue, is the product of a series of structural, political, and institutional transformations that are unfolding in rural Rajasthan and rural India more broadly. Linkages between caste, occupation, and socioeconomic status are beginning to fray in the context of changes to the rural economy, accompanied by livelihood diversification, increased rural–urban mobility, and rising rural literacy rates. The political salience of caste is also changing. In an increasingly competitive electoral environment, politicians can no longer count on caste-based vote banks and must instead attempt to woo mixed-caste coalitions of voters. Institutional interventions such as political reservation of seats in the Gram Panchayats, accompanied by targeted NGO programming, have also helped to establish mixed public spaces occupied by men and women from different backgrounds. In what follows, I examine each of these interrelated changes in turn.

The Changing Rural Economy Before Independence, Rajputana’s princely states were built upon feudal structures in which generations of those from the lower castes worked on the land of upper caste families. The local jagirdar (or landlord, a figure introduced in Chapter 3) controlled most of the land in a village – in turn paying taxes and tributes to the princely state in which their jagir (fiefdom) was embedded. Under this system, caste and occupation were closely tied: “each person had a fixed place … that he or she knew at birth.”18 People were dependent on a patron–client (jajmani) system in which laborers were tied to their village and were not free to travel in search of higher wages.19 Bonded labor was formally outlawed in 1976, and over time has become increasingly rare. Even so, agricultural labor remains one of India’s most caste- and place-bound occupations, marked by social hierarchies and restricted mobility.20 These dynamics, however, are shifting, following changes in land tenure and agricultural production. Across India, the lower castes are increasingly likely to cultivate their own small plots of land. This is due, in part, to land reforms following Independence that, while widely acknowledged as weak and incomplete, have nonetheless altered the rural economy in 18 19

20

Krishna 2002, p. 36. The jajmani system was marked by a caste division of labor, restricting mobility of the lower castes. See Lewis and Barnouw (1956), Gould (1958); and Commander (1983). See, for example, Jayaraman and Lanjouw (1999); Ramachandran (2008); Lanjouw and Murgai (2009).

Boundary Porousness

123

table 5.1 Landownership by Caste Land Ownership No land (proportion) Average landholding

All Sample

ST

SC

OBC

GC

0.09 13.23

0.10 6.98

0.15 7.91

0.07 15.52

0.09 18.98

Source: Citizen survey 2010–2011 (n = 2,210). Note: Landholdings are measured in bighas, where 1 bigha is equal to 0.25 hectares.

important ways. In Rajasthan in the 1950s, land reform brought an end to the jagirdari system as large swaths of jagir land were distributed to smallholders and the landless. Within a decade of these reforms, 3.4 million hectares of land were brought under the auspices of the newly formed Rajasthani state, and were subsequently parceled out to individuals – many of whom received legal title to land for the first time.21 Peasants once tied to the land became landowners in their own right, even as the plots of land remained very small.22 Landownership in Rajasthan, thus, “is no longer the monopoly of the upper castes.”23 The vast majority of citizen survey respondents own some land, including 90 percent of members of the Scheduled Tribes and 85 percent of all Scheduled Caste members (Table 5.1). This does not denote newfound equality: the upper castes (GCs) on average still own almost three and 2.5 times as much land as the STs and SCs, respectively, while the OBCs own roughly double the amounts as either the SCs or STs. Nonetheless, a majority among the lower castes and tribes do possess at least small parcels of land. These land tenure patterns are reflected in the respondents’ reported occupations and sources of livelihood (Table 5.2): a full three-quarters farm their own land, while just 5 percent work as laborers on someone else’s land.24 These patterns hold for the lower castes and tribes: 70 percent of the SC and 77 percent of the ST farm their own land.25 21

22

23 24

25

On the impact of these land reforms in Rajasthan on land use patterns, see Jodha (1985) and Joshi (2007). Land reform in Rajasthan, as in elsewhere in India, has not resulted in an equalization of ownership. The middle classes and castes (the OBCs) benefited the most, while the lower castes and poorest have lagged behind (Krishna 2002; Rathore 2007; Acharya & Sagar 2007). Krishna 2002, p. 38. Even where the lower castes continue to work on upper-caste land, they are today more likely to do so on a contractual basis as sharecropper – rather than as day laborers. Kapur et al.’s (2010) study of Dalit (SC) occupations in neighboring Uttar Pradesh documents similar trends. With increasing mechanization of agriculture, there is less use of bullocks to plough fields. Dalits (who traditionally cared for those bullocks) are therefore much less likely today to work directly for landlords. Rather, they cultivate upper-caste lands while keeping a share of the profits. The relatively high rates of OBC landownership compared to the upper castes reflect the traditional dominance of the OBC Jat community in Rajasthani agriculture. The high rates of SC and ST ownership, in contrast, reflect the land reforms of the 1950s.

124

Encountering the State: Citizens’ Social and Spatial Exposure

table 5.2 Occupation and Livelihood by Caste Occupation: proportion reporting income from

All Sample

ST

SC

OBC

GC

Agriculture – Own Land Agricultural Labor – Others’ Land Nonfarm Labor MGNREGS Employment Small Enterprise Salaried Employment

0.75 0.05 0.52 0.66 0.06 0.03

0.77 0.06 0.63 0.68 0.01 0.02

0.70 0.08 0.65 0.80 0.04 0.03

0.79 0.03 0.51 0.73 0.06 0.02

0.66 0.03 0.31 0.39 0.10 0.06

Reports More Than One Income Source Does Not Work in Agriculture

0.88 0.24

0.94 0.90 0.20 0.26

0.90 0.19

0.78 0.33

Source: Citizen survey 2010–2011 (n = 2,210).

These patterns mark a departure from the landlordism that prevailed historically, thereby shaking up established patterns of interaction across caste lines. Small-scale SC or ST farmers, for example, now often negotiate with uppercaste landowners for access to irrigation water. They might also contract labor from upper-caste farmers who own tractors or other mechanized farm equipment.26 Relationships once marked by outright dependency are thus increasingly contractual in nature. At the same time, changes in land tenure have inserted the lower castes into broader markets, propelling many beyond the village in search of new economic opportunities. A tribal man in Udaipur, reflecting on the effects of land reform, noted: In my childhood, the Rajputs grew maize and we Gamiti [ST] used to work for them. The “mota log” [big people] didn’t allow us to have any land. Slowly, the Congress government came, and they said: “they should be given land, let them also harvest.” That is when people started to venture outside, beyond the fields, to the markets, to the towns. People got aware.27

As noted, however, the parcels of land distributed remained small and in many cases were of marginal quality – enabling, at best, subsistence levels of farming.28 As family size has grown over time, the subdivision and parceling of land for inheritances has made it even less viable for new generations to maintain themselves through agriculture alone.29 Local land-to-labor ratios

26

27 28

29

Kapur et al. (2010, p. 47) note this practice in Uttar Pradesh. I observed similar transactions in Rajasthan. Author interview, Gogunda block, Udaipur district, January 30, 2011. The amount of land under cultivation increased by 50 percent following land reforms. This meant that more marginal and arid lands were incorporated into agriculture, contributing to increased land degradation (Jodha 1985). Jodha 1985; Joshi 2007; Rathore 2007.

Boundary Porousness

125

are declining, reflecting the growing scarcity of cultivable land relative to the agricultural workforce in a village. As an elderly resident of Udaipur noted: There used to be much land and such good quality land. Each had his own. Now, the population is growing and with each generation the land is divided into smaller and smaller pieces – a little for each son. Now there is so much crowding, so little space.30

The rural sector, consequently, has witnessed a decline in the centrality of agriculture.31 Statewide, nonfarm activities have increased as a share of employment, from 32 percent in 1994, to 39 percent in 2005, to 50 percent in 2013 – marking one of the fastest rates of growth in nonfarm jobs in India.32 My citizen survey data reflect these statewide patterns (Table 5.2). A vast majority of survey respondents (88 percent) reported more than one occupation, with over 50 percent working as laborers in a nonfarm setting. This is true for almost two-thirds of the SC and ST, over half of all OBC, and almost one-third of all upper caste respondents. Notably, the Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS – introduced in Chapter 3) has created new opportunities for nonfarm employment: a full two-thirds of the sample reported having a household member who has worked under MGNREGS, including 68 percent of STs and 80 percent of SCs.33 While many turn to MGNREGS or other forms of manual labor to supplement their agricultural incomes, almost one-quarter reported that they do engage at all in farming or farm labor at all – including 20 percent of ST and 26 percent of SC. As agriculture becomes less central to rural livelihoods, the lower castes and tribes have begun to diversify their livelihood strategies. As Krishna has noted of Rajasthan, “People are no longer bound to follow the occupation of their caste. Barbers have become telephone operators, sweepers have become patwaris [government record keepers], and skinners and tanners have become tractor repairmen.”34 In my own fieldwork, I encountered members of the lower castes and tribes working in crafts (such as masonry and construction), in services (for example, as drivers or guards), or as small entrepreneurs (as shopkeepers, paan-wallas selling tobacco, and so on). An elderly upper-caste

30 31

32

33

34

Author interview, Gogunda, Udaipur, February 25, 2010. Krishna 2002, 2003; Lanjouw & Murgai 2009; Kapur, Prasad, Pritchett, & Babu 2010; Gang et al. 2017. World Bank 2016. Nonfarm employment rates in Rajasthan remain below the national average, but are higher than in most other low-income states. Much of this nonfarm employment remains in the informal sector, and most are self-employed. Casual wage labor also increased across Rajasthan in the same period, from 16 to 24 percent. MGNREGS, which offers states’ minimum wages, was designed to supplement agricultural labor – particularly in the dry and lean seasons. Krishna 2002, p. 38.

126

Encountering the State: Citizens’ Social and Spatial Exposure

man described these dynamics in Udaipur, expressing a degree of nostalgia for a time in which occupations were more caste-bound: In the old times both Gamiti [ST] and Meghwal [SC] had their given place in the village. The Gamiti were supposed to protect the village boundaries, and so lived on the outskirts. The Meghwal worked for the service of the village, removing household waste or dead animals, and so were attached to the village. Now, though, their work is changing. Now everyone has his own little piece of land. They go to Udaipur to work in the factories, to work as laborers, or they take up this or that job … It used to be that the SC and ST depended on the village for their well-being. Today, every man is more independent and will make his own way.35

A longitudinal study tracking economic and social practices over ten years in the neighboring state of Uttar Pradesh documents similar changes, noting a “rapid shift out of traditional Dalit [SC] economic relationships into local occupations and professions, migration and changed agricultural practices.”36

Rural Mobility and Connectivity A related set of changes is visible in increasing movement among rural residents, many of who seek economic opportunities in nearby towns and cities or farther afield. A ST woman in Udaipur, for example, described how the need to diversify sources of income has propelled many to travel beyond her village: Everybody works the land. But it is not enough to live. If you have five family members and you don’t do some other labor, how will you live? People go to Gogunda [the block seat] and Udaipur [the district capital]. They work in the factory in Gogunda, and as laborers in Udaipur. They can even also go to Surat and Bombay [large cities in other states]. They make tea or work in hotels. Every household has someone working outside.37

Across India, rates of rural–urban migration are rising.38 The United Nations calculates that over one-third of recent urban population growth can be attributed to in-migration from rural areas.39 Most urban migrants are men, many of whom engage in circular movement: seeking employment in cities while regularly returning to their home village.40 One recent national estimate puts the number of circular migrants flowing between cities and villages at 100 million.41 The circular nature of this movement is important, since it allows individuals to remain at least partially embedded in their villages.

35 36 37 38 40 41

Author interview, Bargaon block, Udaipur district, December 14, 2010. Kapur et al. 2010, p. 48. Author interview, Gogunda block, Udaipur district, January 30, 2011. 39 Deshingkar & Farrington 2009; Mahesha 2013; UNESCO 2012. UNESCO 2012. Most permanent migrants, in contrast, are women who move after marriage. Deshingkar & Akter 2009.

Boundary Porousness

127

table 5.3 Movement Beyond the Village Citizen Survey*

Rapid Survey**

0.22 –

0.28 0.18

– – – – –

0.60

– – –

0.78 0.73 0.69

Migration (household) Employment beyond Village (non-migration) Travel beyond Village (non-work) Frequency of Travel beyond Village Monthly Weekly Daily Reasons for Travel beyond Village Access Markets/Shopping Visit Family outside Village Financial Affairs/Banking

0.24 0.19 0.17

* Source: Citizen survey 2010–2011 (n = 2,210). ** Source: Rapid survey 2011 (n = 240).

In Rajasthan, slightly less than one-quarter (22 percent) of survey respondents reported that they or a member of their household live outside of the village for more than thirty days in the year. The number of people who leave the village for work, however, is much higher if we consider those who travel on regular basis to nearby towns and cities. In Udaipur district, for example, substantial numbers of men travel by bus daily or weekly to the district seat or other towns where they are employed on construction sites, in factories, or in other industry. This kind of mobility, which interviewees referred to as “up-down,” is not captured in the main survey question about migration, which imposed a definition that requires sustained periods of residence outside of the village. “Up-down” is a qualitatively different undertaking than the longer-distance and longer-term patterns of migration typically associated with processes of modernization and urbanization, in which citizens are thought to break free from “parochial” rural ties to become embedded in more “cosmopolitan” urban communities.42 Circular and more localized patterns of movement sustain a connection to the village and, at the same time, broaden networks of exposure beyond such local boundaries. An additional rapid survey carried out in Udaipur and Kota districts one year after the large-n citizen survey casts some light on these dynamics (Table 5.3).43 In this second, small-n data set, 28 percent reported that they or a household member travel and stay beyond the village for thirty days or 42 43

See, for example, Lerner (1958), as discussed in Chapter 4. See Appendix I for a discussion of the small-n survey methodology. Research assistants carried out 232 rapid surveys accompanied by structured interviews in six purposively selected case study villages, drawn from the original village sample in Udaipur and Kota districts. Households in each village were randomly selected and were not part of the original citizen survey sample.

128

Encountering the State: Citizens’ Social and Spatial Exposure

more each year. Importantly, though, an additional 18 percent stated that they work outside of the village, commuting back and forth on a daily or weekly basis (“up-down”), including one-third of all men interviewed. Even if residents do not work outside of the village, the vast majority still have reason to leave the village fairly regularly. Sixty percent reported traveling beyond the village once a month or more: of these, 17 percent did so on a daily basis. The most common reasons for this travel were visiting markets or shopping; visiting family; and financial affairs, such as banking. The frequency of this extra-village movement is in large part a product of noted structural conditions – namely, declining land-to-labor ratios and reduced economic opportunities in agriculture – that compel residents to seek alternative sources of livelihood. It also reflects relative improvements to Rajasthan’s roads and transportation services. Road density has increased, almost doubling between 2005 and 2012.44 Much of this expansion has been funded under the Pradhan Mantri Gram Sadak Yojana (PMGSY) – a central government program designed to increase rural connectivity through roads. More than twenty thousand rural habitations have been newly connected by roads in Rajasthan since the inception of this program in 2000.45 Over twothirds (67 percent) of citizen survey respondents live in villages with paved approach roads. An elderly ST man described the development of a new road linking his village to Udaipur city – a distance of about 16 kilometers: Twenty-five years ago, there was no road here. People went on foot. Pregnant women or sick people were carried on stretchers by foot. They would walk for a day to reach the hospital. Then slowly, slowly the roads were made. First villagers gave their own labor to make a kaccha [unpaved] road. Seva Mandir [a NGO] initiated the work. Later, party people became involved and they made it a pucca [paved] road.46

More than half of respondents also lived in villages with a bus stop. A young ST man in Udaipur described the importance of this connectivity: Ten, fifteen years ago, the roads were very bad. There was no bus, no jeeps. It was very expensive to get to Udaipur … Now you can get up in the morning, drink tea, and reach Udaipur before the sun is up. Everyone is going. Each house here has at least one person who goes, who works on buildings [in construction] there. The city is within reach, and so it is our city. In my father’s time, it was a faraway place.47

44

45

46 47

Open-ended answers to the structured interviews were subsequently coded by two other research assistants, each working independently to ensure inter-coder reliability. World Bank 2016. Between 2005 and 2012, road density rose from 438 to 726 kilometers of road for every 1,000 square kilometers of land area. Online Management, Monitoring, Accounting System, Pradhan Mantri Gram Sadak Yojana (PMGSY). Nationally, between 1991 and 2001, the percentage of villages connected by paved roads rose by almost a third (from 42 percent to 64 percent). Author interview, Bargaon block, Udaipur district, January 10, 2011. Author interview, Bargaon block, Udaipur district, January 11, 2011.

Boundary Porousness

129

The effects of this movement beyond the village are both economic and social. Greater rural–urban mobility tightens the rural labor supply. This, coupled with the flow of financial remittances, has the effect of “enhancing the bargaining power of Dalit [SC] households within the village economy and weakening traditional clientelistic political structures.”48 As the boundaries between village and urban centers become more porous, new social norms and practices are also introduced.49 An elderly man in Udaipur described these dynamics: More and more, the younger generations are going outside in search of work. It is changing the face of the village, bringing in new ways. Now everyone who leaves the village comes back and thinks he is Shah Rukh Khan.50

Another man in Kota reflected on the high numbers of people who “go outside” for work, noting that they are “changing the village mentality”: Village people are generally less educated and have limited information. But people who go outside have state-level information. Everybody has their own point of view. Their thinking is wider, they have foresight. They are changing the village mentality, coming back and saying “this should be this way or that way.” And so people in the village are getting more active, more aware.51

In sum, the lines demarcating “rural” and “urban” are becoming less clear over time as village networks penetrate further toward urban centers, and as more mobile individuals venture farther afield, propelled by economic changes in the village.

Advances in Rural Education Increasing access to education among the lower castes – particularly for younger generations – has also made it possible to pursue new economic and social opportunities.52 An initial reading of the data reveals a bleak set of conditions. Rural Rajasthan consistently lags behind urban areas in educational attainment. A thin majority (54 percent) of citizen survey respondents reported

48 49

50

51 52

Kapur et al. 2010, p. 47, noted in a study of neighboring Uttar Pradesh. As Kapur and Witsoe (2011, p. 4) note, social remittances also take the form of new ideas that circulate between urban and rural settings. Rural–urban mobility thus produces not only material but also “cognitive and symbolic resources.” Author interview, Gogunda, Udaipur, February 25, 2010. Shah Rukh Kahn is a famous Bollywood actor. Author interview, Sangod block, Udaipur district, February 20, 2011. The greatest statewide advances in education are among the young: between 2005 and 2012, for example, the proportion of those who completed secondary or higher levels of schooling rose from 21 to 35 percent among eighteen- to thirty-year-olds – the largest increase of any age group (World Bank 2016).

130

Encountering the State: Citizens’ Social and Spatial Exposure table 5.4 Level of Education by Caste

Average Years of Schooling No School Some Primary School Some Secondary School Some Higher Education

All Sample

ST

SC

OBC

GC

4.31 0.46 0.17 0.31 0.07

2.89 0.58 0.16 0.20 0.06

3.82 0.53 0.17 0.22 0.07

4.28 0.45 0.17 0.32 0.05

6.05 0.30 0.17 0.43 0.09

Source: Citizen survey 2010 (n = 2,210).

that they have attended school at all – completing, on average, 4.3 years (roughly equivalent to a lower primary or “fourth standard” education). This leaves a striking 46 percent who stated that they have not received any education at all (Table 5.4). There are, moreover, important caste differentials in educational attainment. Majorities among the ST and SC (58 and 53 percent) have not attended school at all, compared to just 30 percent among the GC. And yet, these caste gaps are narrowing – visible in the interaction between caste and age over time. Table 5.5 presents mean levels of “functional literacy” by age cohort, defined as having completed at least five years of school.53 Part A reproduces data from Krishna’s 1997–1998 study in Rajasthan,54 which provide a comparative frame of reference from which to view my own citizen survey data gathered in 2010–2011 (Part B). Together, the data show steady increases in literacy across both caste and age groups over time. A quick comparison of those aged 65 and older to those aged 18–25, for example, reveals dramatic generational gains. Both sets of data, moreover, highlight gains in education among youth from the lower castes and tribes. Younger generations among the lower castes and tribes are rapidly catching up – even as overall levels of educational attainment still reflect caste disparities. Rural Rajasthan is thus home to growing numbers of educated youth from the lower castes and tribes. As Krishna has documented, many of these young people (primarily young men) put their newfound literacy and skills to work in and around the village, serving as intermediaries between politicians, state agencies, and citizens.55 Others seek public or private employment opportunities – venturing into sectors previously off limits. A young SC man interviewed in Kota, for example, explained that he works as a schoolteacher – an occupation that would have been unthinkable one or two generations ago. He reflected

53 54

55

Here I follow Krishna’s (2003, p. 1177) definition of functional literacy. Krishna 2002, p. 40. The data are drawn from a 1997–1998 sample of 2,232 residents across 60 villages in Rajasthan and nine villages in neighboring Madhya Pradesh – described in Chapter 4. Krishna 2002. These lower-caste brokers, or naya netas, are described in detail in Chapter 4.

Boundary Porousness

131

table 5.5 Functional Literacy by Caste and Age Cohort 5 or More Years of School Completed

65+

55–65

45–55

35–45

25–35

18–25

A. Proportion with Functional Literacy by Age Group in 1997–1998* ST SC “Backward” Caste “Middle” Caste “Upper” Caste All Sample

0.00 0.00 0.10 0.09 0.33 0.15

0.07 0.117 0.28 0.14 0.50 0.23

0.12 0.18 0.29 0.22 0.52 0.28

0.15 0.24 0.34 0.38 0.58 0.33

0.24 0.40 0.56 0.43 0.70 0.48

0.53 0.72 0.72 0.50 0.81 0.70

B. Proportion with Functional Literacy by Age Group in 2010–2011** ST SC OBC GC All Sample

0.16 0.15 0.20 0.36 0.23

0.20 0.22 0.22 0.58 0.30

0.31 0.29 0.39 0.59 0.41

0.32 0.34 0.48 0.64 0.46

0.34 0.44 0.60 0.79 0.55

0.57 0.86 0.80 0.92 0.79

* Source: Krishna 2002, p. 40, drawing on 1997–1998 survey data from rural Rajasthan (n = 1,898) and a neighboring district of Madhya Pradesh (n = 334). “Backward” castes are equivalent to OBC in the 2010–2011 sample, while “middle” and “upper” castes are equivalent to the “GC” (or unreserved) category. ** Source: Citizen survey 2010–2011 (n = 2,210).

on how earning a degree not only enabled him to gain employment, but also altered his relationship to upper-caste families in the village: I spent my childhood right here in this village. I grew up next to the Mali [OBC] and the Rajput [GC]. But most would not give the time of day to an Erwhal [SC] boy. But then I left to study. I came back with a degree and earned their respect. Now they send their own children to me. I meet and greet all the parents, pass the time of day with them.56

Barriers of caste, while still an important feature of social identity, thus can no longer adequately predict either a person’s occupation or his or her economic status.

Political Competition Caste boundaries are also shifting in the political arena. The “second democratic upsurge” of the 1980s and 1990s was a caste-driven one, as members of the lower castes began to vote in unprecedented numbers, mobilized by political parties running on caste platforms.57 Caste, during this transition, became an

56

Author interview, Sangod block, Kota district, February 22, 2011.

57

Yadav 2000.

132

Encountering the State: Citizens’ Social and Spatial Exposure

important organizing structure through which parties attracted voters, who sought out their “co-ethnics” on the assumption that they would be treated most favorably by politicians from their own caste communities.58 These patterns, though, have begun to change, and with them the political salience of caste is also shifting. In an increasingly competitive electoral environment, no political party can win on the basis of the support of a singular caste community. Parties that once ran by targeting particular castes thus increasingly turn to mixed-caste coalitions to garner electoral support.59 This is true not only in national and state elections, but also at the local level. As Chauchard observes in his study of Rajasthan’s Gram Panchayats, most candidates for local office must look beyond their co-ethnics for support. This in part reflects the fractionalization of villages along caste lines, making it rare that a single caste is numerically dominant. It also reflects the effects of caste reservations that, by mandating that members of certain caste categories (e.g., SC, ST, or OBC) run for office, ensure that there are often multiple candidates from the same caste backgrounds contesting at once.60 As a result, local elected officials are increasingly likely to allocate benefits along partisan rather than caste lines, targeting multiethnic constituencies.61 In this environment, political brokers that are able to work across caste lines take on new significance. As Krishna noted in Rajasthan, “… caste no longer has primary importance [for brokers]. Non-caste-based political entrepreneurs are more successful than others in delivering economic benefits and providing avenues for greater political participation, … and villagers associate with these entrepreneurs regardless of caste or religion.”62 As an organizing principle of politics, caste is thus both less central and less constraining than it once was. This shift is reflected in my citizen survey data, in which just 18 percent cited co-ethnic identification as a factor informing their voting decision in the most recent panchayat election. Most (72 percent) instead reported voting for a candidate that had a reputation for “good work” in the village, “regardless of caste.” In interviews, villagers also expressed a willingness to look beyond both caste and partisan lines when choosing whom to support in local politics. A Scheduled Caste man in Udaipur (a self-identified BJP supporter) explained: Nowadays people look about them more, they are more aware of the choices they are making. Before, the big Congress men would come and tell us, “give us your votes” and people would do so just like that … Now people here support both parties. And people

58 60

61

59 Chandra 2004. Jaffrelot 1998; Thachil 2014. Chauchard 2017. He notes (p. 85), “Since it is not rare to observe multiple candidates from the same caste group in these elections [given reservations], candidates can in fact rarely assume that they will safely bag the votes of their co-ethnics.” 62 Dunning & Nilekani 2013. Krishna 2003, p. 1190.

Boundary Porousness

133

are also looking outside the community … They have to perform. It is not about party or samaaj [caste community], but about who can get the work done. The old sarpanch was BJP, and he was chosen unopposed not because of his party but because he was a known and active person.63

As mixed-caste political coalitions increase, so, too, do spaces for cross-caste interaction. Candidates seeking office, for example, will bring together members of different caste communities or woo them through celebrations open to all castes. In a village in Kota, for example, I observed the family of a recently elected sarpanch preparing food for a village-wide feast. A Scheduled Caste man, who was accompanying me on a walk through the village, also noted the food preparation, stating: Because of elections, we eat together. In my father’s time, we were never called [to gatherings] but now they call us because without us they do not have the [votes for the] panchayat.64

The local political arena, in sum, is increasingly the site of cross-caste interaction. The result is a politically driven erosion (although clearly not elimination) of some of the social boundaries of caste.

Boundary Crossings by Design: Institutional Drivers of Exposure A final set of changes prompting greater boundary porousness is institutional in nature. There are two broad sets of interventions that, by design, seek to alter norms of social engagement. First, as noted in Chapter 3, political reservations in the Gram Panchayat have created new, formal spaces for the participation of traditionally marginalized groups. Second, nongovernmental and civil society organizations often seek to foster interaction across caste and gender as well as village lines through a combination of targeting, training, and advocacy. Following the Seventy-Third Amendment that constitutionally mandated Gram Panchayat elections, a large number of new actors have entered panchayat politics. Much of this is due to the requirement that panchayat seats be reserved for women (at 50 percent) and for members of the Scheduled Castes and Tribes (in proportion with their share of the population). The material and distributional effects of these reservations are the subject of ongoing debate.65 Some argue that they increase marginalized groups’ control over resources. Chattopadhyay and Duflo, for example, find that the reservation of seats in the GP influences the local allocation of public goods in a way that benefits

63 64 65

Author interview, Gogunda block, Udaipur, February 10, 2010. Author interview, Mavli block, Udaipur, April 21, 2009. For a comprehensive discussion of the distributional and political consequences of reservations, see Jensenius (2017).

134

Encountering the State: Citizens’ Social and Spatial Exposure

traditionally disadvantaged residents. “Local leaders,” they conclude, “seem to have effective control over decisions, even when they are women or SCs.”66 More recently, though, others have questioned the distributional effects of local political reservations. In a multi-state study, for example, Dunning and Nilekani find that the impact of having SC or ST members in local office is attenuated by competitive party politics; allegiances within a multi-caste party organization make it less likely that elected panchayat members will target benefits to their own caste groups and more likely that distribution will occur along partisan lines.67 While the debate over the material impact of political reservations continues, others have argued that that these policies have transformative social effects, simply by bringing traditionally marginalized groups into formal spaces in the public sphere. Critical observers, though, point to the ease with which political office is captured by the local elite regardless of official policy.68 My own field experiences confirm these concerns, as I observed higher-caste groups attempting to influence lower-caste panchayat members.69 This kind of manipulation is even more pronounced where gender reservations are in place; it is a widely acknowledged practice for men to encourage their wives or other female relatives to run for local office.70 When visiting villages with a female sarpanch, I was consistently asked to meet the “sarpanch pati ” (president’s husband), who was treated as a de facto office-bearer. In the words of one such pati, when asked where his wife was: “She’s at home … What does she know of this work? How can a woman sit behind her veil and conduct [panchayat] business?”71 And yet, these very real limitations notwithstanding, the fact remains that many women and members of the lower castes and tribes are being newly inserted into public spaces. Even if largely symbolic to begin with, these

66

67

68 69

70

71

Chattopadhyay & Duflo (2004, p. 979), emphasis added. Also see Duflo and Topalova (2004), and Besley et al. (2004). Pande (2003) finds similar patterns at the state level, where seats in the legislative assembly are reserved for the lower castes. Chin and Prakash (2011) also find an effect of reservations on poverty reduction among members of the Scheduled Tribes (although a corresponding effect for Scheduled Castes is not apparent). Dunning & Nilekani 2013. Also see Chin and Prakash (2011); Chauchard (2014; 2017). Jensenius (2017) argues that reservations have the greatest effect on elite members of the Scheduled Castes, and so should be further disaggregated along socioeconomic lines. Kudva 2003; Palaniswamy & Krishnan 2008; Baviskar & Mathew 2009; Singh 2009. This was sometimes very overt. In one panchayat meeting in Udaipur district, for example, an upper-caste resident (with no official position in the panchayat) sat next to a Scheduled Tribe (and minimally literate) sarpanch, whispering in his ear through the meeting as villagers approached him with concerns or queries. In another case in Kota, two upper-caste groups (one Rajput and one Jat) each “ran” their own Schedule Caste candidate in a panchayat election where the seat of sarpanch had been reserved for a SC woman. Tellingly, local election campaign posters often feature an inset picture of a woman’s husband alongside the photo of a female candidate. Author interview, with translation, Gogunda block, Udaipur district, February 22, 2010.

Boundary Porousness

135

experiences can alter citizen–state relations over time. In panchayat meetings, for example, lower-caste and tribal members, including women, now sit alongside the upper castes on the same jajham (a carpet used to designate a space of public authority). In the words of one Scheduled Caste ward panch: Now we are here in the panchayat, sitting next to them [the upper castes]. We sit on the jajham, they serve us tea. They know that if they need something they will have to come to us, they will have to greet us and say namaskar.72

The long-term social effects of caste reservations are difficult to measure. However, as Chauchard has shown in the Rajasthani context, caste reservations are indeed correlated with a decline in hostile and discriminatory social behavior.73 This, he argues, is less a function of profound changes in personal attitudes about caste, but rather is due to changing perceptions about the acceptability of interaction across caste lines, as well as new understanding of the legal rights of lower-caste and tribal groups. In other words, “when access to political power is redistributed towards members of groups that used to be strictly excluded from it, day-to-day social interactions based on traditional structures of domination also change.”74 The relationships forged may not be deep or sustained, but villagers do indeed interact across caste, class, and gender lines in the affairs of the panchayat. A Scheduled Caste man in Kota who previously served as sarpanch recalled that before his election he had never socialized with members of the upper castes. After his election, he was courted by members of the upper castes. He reflected: When I became sarpanch everyone started calling me, inviting me, as if I turned high caste just by being a sarpanch. But, after all, I am a Meghwal [SC]. They will … call me “Sarpanch Sahib.” They will take food from my hands.75

It is in this sense that Rao and Sanyal argue that village-wide panchayat meetings (Gram Sabhas) are important “discursive” spaces in which the poor and disadvantaged develop civic and political skills. These skills, they suggest, may be put to use in seeking both material and nonmaterial benefits, including increased social standing within the village.76 Gender norms, though in many ways stickier than those associated with caste, are also shifting for women in local elected office. A tribal woman in Udaipur, for example, described her experiences after being elected to the panchayat: Before I never ventured out. My life was only here [in the house], and to the fields and back again. After my election, I went every which way, to Gogunda [the block seat], to

72

73 75

Author interview, Bargaon, Udaipur, December 17, 2010. Namaskar is a formal greeting and sign of respect. 74 Chauchard 2017. Chauchard 2010, p. 35. 76 Author interview, Sangod block, Kota district, February 24, 2011. Rao & Sanyal 2010.

136

Encountering the State: Citizens’ Social and Spatial Exposure

Udaipur. It was my business to go. They even organized buses to take us to Jaipur [the state capital] so we could learn the business of the panchayat.77

A staff member of a NGO in Udaipur commented on similar dynamics, noting that gender transformations unfold over time: Even if at first a woman doesn’t speak, slowly she learns and starts to hold her head a little higher. See this one [indicating to a former sarpanch who now works with the NGO], she used to be quiet as a mouse but now she knows the business of the whole village.78

Panchayat members are often the targets of governmental and nongovernmental training programs that further serve to broaden their fields of exposure. Recall, for example, “Chandibai,” the woman introduced at the beginning of Chapter 1 who, despite her gender, tribal status, and illiteracy, was a wellknown activist in her village. Chandibai was also a former panchayat member, having previously filled a ward panch seat reserved for a tribal woman. Much of her experience and confidence stems from this time in local office. Chandibai, though, did not acquire her political savvy through membership in the Gram Panchayat alone. She was also part of an NGO-led training program that targeted panchayat women in the hopes of instilling in them the skills to carry out their office. Such efforts by civil society organizations create additional spaces in which women and members of the SC and ST gain exposure beyond their local communities. NGOs, as noted in Chapter 4, remain relatively thin on the ground in rural Rajasthan: just 6 percent of those surveyed reported the presence of an active NGO in or around their locality. Yet, these organizations – where present – do appear to play a critical role in broadening citizens’ networks of exposure. I encountered a number of civil society organizations in the course of my fieldwork, all of which were engaged in forms of local “institution building” aimed at promoting the participation of traditionally disadvantaged groups. Different organizations applied different models, ranging from efforts to strengthen the capacity of the Gram Panchayat; to supporting local alternatives to the elected panchayat; to movement-based work that seeks to hold elected officials to account by organizing large-scale protests. While the approach varied, all of these interventions sought to bring marginalized groups into the public arena. In so doing, they promoted new forms of engagement across local boundaries of caste and gender. To the extent that their work spanned more than one locality, they also promoted interaction across village lines. 77

78

Author observation of NGO training session of elected panchayat women, with translation by NGO staff, Gogunda block, Udaipur district, April 14, 2009. Author interview, Mavli, Udaipur, April 21, 2009. Chauchard (2017, p. 92) similarly observes, “‘weak sarpanchs” … may not be as weak four years into their term. Even the ‘weakest’ … visibly gained in confidence: although he originally did not speak very much in meetings, he now clearly leads them.”

Boundary Porousness

137

Take, for example, Vidhya Bhavan, a civil society organization headquartered in Udaipur city that runs an “Institute of Local Self Government.” This organization conducts an outreach and training program for newly elected Gram Panchayat members that specifically targets women and members of the Scheduled Castes and Tribes. Field staff attend and monitor panchayat meetings and reach out to elected members, providing them with training materials. Elected panchayat members are also invited to attend trainings, seminars, and “exposure” trips where they meet with public officials and with members of other panchayats. In this way, Vidhya Bhavan seeks to further draw women and members of the lower castes and tribes into the public sphere. In a second example, another Udaipur-based NGO, Seva Mandir, works to develop village institutions formed around a “self-governance” model influenced by the Gandhian philosophy of swaraj (“self-rule”). In one village, for example, Seva Mandir helped to create and support a “Citizens’ Forum,” which operates as a parallel platform to the elected Gram Panchayat. The Citizen’s Forum draws together men and women from different neighborhoods and castes, holding its own elections and community meetings to decide on programs and the allocation of NGO funds. It also seeks to build support across socioeconomic lines, for example, by working on issues (such as waste management) that will mobilize a broad range of class groups since “everyone can get behind picking up the garbage.”79 In another village, Seva Mandir has worked to create new spaces for women in informal local governance by advocating for their presence in caste association (jati panchayat) meetings that are traditionally male-only. This is slow, incremental work that has met with considerable opposition. However, over time, however, Seva Mandir staff have managed to persuade some local caste leaders to allow women from the community (most of whom have received training from the NGO) to attend caste meetings dedicated to the discussion of issues that specifically impact women, such as child marriage or marital disputes. It is unlikely that women’s presence in these meetings will, in the short term, make any difference in the decision-making of the caste councils. However, the act of simply attending a meeting from which women were formerly excluded marks a powerful departure from traditional practice. In these ways and others, the NGO seeks to challenge caste and gender norms at the village level and to promote movement across such boundaries. The Mazdoor Kisan Shakti Sangathan (MKSS, first introduced in Chapter 3), provides a third and very different example of a civil society intervention that has fostered both social and spatial exposure. MKSS, as noted, campaigns on issues such as the right to work, right to food, and transparency in local governance. In so doing, it attempts to draw together people, often from

79

Author interview with Seva Mandir staff member, Bargaon block, Udaipur district, April 9, 2010.

138

Encountering the State: Citizens’ Social and Spatial Exposure

disparate caste communities, under a common class-based identity of “workers and farmers.” In contrast to Seva Mandir and Vidhya Bhavan’s efforts to build village-level institutions, MKSS seeks mass mobilization via street theater, public rallies, marches, sit-ins and other forms of media-savvy protest. MKSS organizers travel from village to village by jeep, using a bull horn and puppet shows to spread information about their campaigns, and transport people to rallies and other protest events. In this way, MKSS encourages encounters and exchanges across boundaries of both caste and village.

increasing social and spatial exposure The structural, political, and institutional changes described in the preceding section are, taken altogether, producing new levels of mobility and connectivity across established social and spatial cleavages. Within the village and beyond, people are traversing traditional occupational, caste, neighborhood, and even gender lines. These changes are perhaps most pronounced when it comes to caste. As Sheth has noted: “households within a single caste have not only been greatly differentiated in terms of their occupations, educational and income levels, and lifestyles, but these differences have led them to align outside the caste, with different socioeconomic networks and groupings in society – categories which cannot be identified in terms of the caste system.”80 In one of the few empirical studies to document changing social and economic practices over time, Kapur et al. (2010) find dramatic shifts in caste relations in the state of Uttar Pradesh. There, the last twenty years have brought significant changes in the frequency and nature of interpersonal interaction across caste lines. The upper castes, for example, are today more likely to accept hospitality in a lower-caste, “Dalit” (SC) home, while the lower castes are more likely to attend social functions (such as weddings and feasts) alongside members of the upper castes. Kapur et al. are quick to point out that these changes do not represent newfound caste equality. They do, however, mark important social shifts in a relatively short span of time. Chauchard, as noted previously, documents similar trends in neighboring Rajasthan where discriminatory practices against the lower castes are decreasing.81 Villagers’ own accounts from Rajasthan also suggest that boundaries between castes are less rigid than they once were. It is important to stress that these changes hardly signify emerging solidarity across castes; caste discrimination – and even violence – persists. And yet residents in a number of villages cited the declining practice of “untouchability” (chuachoot) as the lower castes become more integrated into the local economy. In the words of one SC man: In my father’s time, everyone lived more apart, and everyone had their place … You wouldn’t dream of arriving at your wedding on a horse. You would be knocked down. 80

Sheth (1999, p. 2504), emphasis added.

81

Chauchard 2017.

Increasing Social and Spatial Exposure

139

Now, they have to show us respect. They know that tomorrow they will be dependent on us since the development and progress of the village depends on us.82

An elderly tribal woman offered another poignant illustration: These silver ankle bangles that I am wearing now, I could not wear them before since only the [higher-caste] Rajputs and Gaayri could wear them. They used to say, how can you wear the same things as us? But now we wear them. People have earned a little bit of money, and now we have more respect. Long ago, we did not even have the right to make lapsi [a sweet]. Once people made it and the Rajputs came and poured dirt in it. They said: “how can you act as though you are equal to us?” But now, everything is changed. Any caste can sit on a jajham now.83

Another man declared: Today people are less scared. There is no untouchability anymore. Everyone will speak his mind. You even see this at the chai [tea] stall. An SC man will drink his tea and speak his mind there.84

This loosening of caste-based strictures, along with a diversification of occupations across caste lines, contributes to the emergence of new social and economic spheres of interaction. In the words of an Udaipur resident: “The higher-up people (uchro gharwalla) have started to mix with the lower-down people (nichro gharwalla).”85 Changing caste relations correspond with, and are intensified by, shifts in spatial boundaries – notably, the greater porousness of neighborhood and village lines. The citizen survey data (to which we return in detail in the next chapter) offer a series of indicators by which to assess the extent to which individuals interact across social and spatial lines. Sixty-eight percent reported that they socialize with people from different neighborhoods, while 81 percent reported working in a mixed-caste setting. Another 22 percent, as noted, reported regular periods of migration beyond the village. Thus, in contrast to an image of a region locked in place by feudal governance and rigid caste structures, the picture that emerges is one of movement and change.86 These changes, though, are far from even; some individuals remain more constrained by social and spatial boundaries than others.87 How, then, does variation in boundary rigidity and in citizens’ exposure affect whether and how they engage the state? I argued in Chapter 2 that more porous boundaries 82

83

84 85 86

87

Author interview, Bargaon block, Udaipur district, December 17, 2010. The groom’s arrival on horseback is a high-status event, traditionally reserved for upper-caste households. Author interview, Gogunda block, Udaipur district, January 30, 2011. The Rajputs and Gaayri are both dominant castes in this woman’s village. Author interview, Bargaon block, Udaipur district, December 14, 2010. Author interview, Gogunda block, Udaipur district, February 6, 2011. Kapur and Witsoe (2011, p. 2) similarly note the dynamism of rural change, contrasting it to the more common “dismal picture of North and East India – poor governance, a feudal structure, stagnant agriculture and incomes and an immobile and static rural society.” Village, panchayat, and individual correlates of exposure are examined in detail in Chapter 6.

140

Encountering the State: Citizens’ Social and Spatial Exposure

contribute to an increasing likelihood and broadening repertoire of claimmaking: the greater the exposure beyond the immediate community and locality, I theorized, the greater one’s encounters with and linkages to the state, and the broader the flows of information, ideas, and beliefs that inform citizen action. The next chapter turns to the citizen survey data to empirically investigate these relationships. Before doing so, however, the remainder of this chapter examines villagers’ qualitative accounts to illustrate how social and spatial exposure might influence claim-making activity. These accounts are drawn from a diversity of villages and residents. The experiences of these men and women, from high and low caste and class backgrounds alike, highlight the power of porous boundaries in influencing citizens’ aspirations and capabilities.

exposure and claim-making aspirations Different approaches to the state are associated with fundamentally different sets of expectations about what the state can and should do: that is, with different notions of entitlement and of political and personal efficacy. To illustrate these concepts (first introduced in Chapter 2) in the terms in which they are articulated by citizens, I begin with a comparison of two men in Udaipur district, both members of the same Scheduled Tribe. The first man resides in an isolated hamlet of his village. He works locally as an agricultural laborer and so rarely ventures farther afield. As a member of my research team began asking his opinion on a range of public programs and services, he responded bleakly: Seeing is believing. Do you see those things here? As far as I am concerned they don’t exist. Why would I waste my breath on all that?88

The second man, who also lives in a small hamlet, painted a very different picture. This man is a laborer in a rock quarry located about ten kilometers away and so travels regularly beyond the village. He begins each day by walking from his hamlet to the central (main or “mukhya”) village to catch the bus. A heated exchange with a General Caste man, observed at a tea stall adjacent to the bus stop, highlights the impact of the ST man’s exposure to people and settings beyond both his hamlet and village: gc man: This village has everything. Roads, drinking water, schools. There is so much development! st man: Those facilities are just for rich people. Even if you have to go to the doctor you have to wait! gc man: Just compare this village to other places and you will understand how much you have here.

88

Research team member interview, Gogunda block, Udaipur district, February 9, 2011.

Exposure and Claim-Making Aspirations

141

st man: I do compare! That’s how I know how many facilities there are, and how little we [in the tribal hamlet] have. You might live your whole life in some far off basti [hamlet] and never know what all schemes you mukhya [central village] people are getting here. But every day I am coming here and seeing your riches.89

As this exchange reveals, social and spatial exposure not only generates information about public services but can, in fact, create new expectations. Citizens’ interests vis-à-vis the state change in light of levels of (and exposure to) service delivery. Much of this rests on observation of services and resources beyond one’s locale. In the preceding exchange, the tribal man articulated a sense of entitlement that stemmed directly from having noted the treatment of others in his village. Importantly, though, a sense of entitlement does not develop through an abstract notion of “rights” or even of “citizenship,” but rather takes hold in response to the perception of unequal or unfair treatment. It is, in other words, an expression of a comparative grievance that develops as knowledge of public resources expands faster than one’s own access to goods and services. An Udaipur woman, for example, described how demands for water had spread throughout her village: Today we have five hand pumps in the village. It happened slowly, one by one they were made. People would see the hand pumps and started to demand their own, and so would say it to the sarpanch. First one was made out of the MLA funds. Then people started saying it in the quorum [panchayat] meetings. Now each mohalla [neighborhood] has one.90

These dynamics are not new, but have become more pronounced since the 1990s as social budgets and programs have expanded, resulting – as demonstrated in Chapter 3 – in a broad but uneven terrain for social welfare provision. Krishna captured these dynamics in a 1999 interview with a party leader in a village in Rajasthan. Now people want a bus to come to their village; they are no longer willing to walk five kilometers. They want electricity to run their pumps … They want schools … If they have a primary school, they want a middle school … [then] a high school … They have seen these things [available] in other villages, so why not in their village?91

Grievances alone, though, will do little to spur citizen action absent any expectation of a governmental response. Claim-making thus also rests upon a notion of efficacy that – like sense of entitlement – also develops in reference to the experiences of others. Observing that other people have successfully secured access to public resources keeps alive the hope that the state might deliver.

89

90 91

Conversation observed during author village visit, Bargaon block, Udaipur district, December 15, 2010. Author interview, Gogunda block, Udaipur district, January 29, 2011. Krishna 2003, p. 1182 (emphasis added), quoting a Congress party activist and local labor leader.

142

Encountering the State: Citizens’ Social and Spatial Exposure

However, while notions of entitlement develop with regard to how different people are treated, a sense of efficacy most often develops in reference to the experiences of similarly placed people. A poor, tribal woman, for example, is much more likely to respond to examples of successful claim-making activity among other tribal women than, say, among upper-caste men. A person’s points of reference, in this sense, cannot be too far removed from the day-today reality of his or her life. Take, for example, this Meghwal (SC) woman in Kota, who refers to levels of service delivery in other villages when explaining her own claim-making intentions. Importantly, though, she references the experiences of other low-income people from her same caste community: We see how in other villages people are getting things. Poor people, Meghwals, like us. That is how we have come to know about all the schemes that are available … But only one man in this neighborhood got a house [under a government program targeting the SC] … Next time my son will go to the meeting and demand that our name is on the list. If he is getting these things, why not us?92

As this woman illustrates, citizens’ expectations develop in regard to ideas and beliefs that emerge in particular “cognitive neighborhoods,” understood as the people and places to which they compare themselves.93 In order to inspire a sense of efficacy, then, exposure requires a degree of local and personal resonance.

exposure and claim-making capabilities In addition to aspirations, a person also requires knowledge of how to proceed: information about the goods and services available, procedural knowledge of how to seek them, and tacit knowledge of what to do when formal channels fail to produce the desired results. There are many sources of information, not all of them social in nature. Media or public outreach campaigns require little if any social exposure on the part of individual citizens. Government agencies and NGOs alike carry out outreach campaigns through a variety of media, including radio, newspaper articles, and even loudspeaker announcements. It is also common to see information and slogans about public programs painted on village walls. And yet it is clear that information about the state and its services is rarely enough, in and of itself. Consider, for example, a man in an Udaipur village who lives in a house adjacent to a large government billboard covered in slogans about the Right to Information (RTI) Act – legislation passed in 2005 under which citizens can request information from government agencies.

92 93

Student interview, Sangod block, Kota district, February 24, 2011. Ray 2003, as discussed in Chapter 2.

Exposure and Claim-Making Capabilities

143

He spent a long time telling me about his effort to settle a land dispute and complained repeatedly that he had not received a response from the Land Registry. When I asked if he had thought of filing a petition under the RTI Act, he gave me a blank look. At first, he said he didn’t know what RTI was. When I pointed out the government billboard, he shrugged and said, “They are forever painting slogans, but I don’t know what these things are. I don’t know anyone who has ever done such things.”94 As is clear from this example, the provision of information does not guarantee its consumption. Indeed, numerous studies in rural India have demonstrated a gap between citizens’ knowledge of public services and their willingness or ability to act to secure them. In a study of Rajasthan, for example, Banerjee, Deaton, and Duflo (2004), find that providing information about health services did not generate a corresponding increase in demand for those services. They attribute this to the fact that the information campaign was not tied to wider efforts to inform citizen action, but instead was delivered in an atomized and abstract manner. Niehaus and Sukhtankar (2013) similarly find that simply knowing about MGNREGS wage increases by no means guaranteed that workers received – or even demand – correct wages. The gap between citizen knowledge and action reflects, in large part, the insufficiency of procedural information in complex and uncertain service delivery environments. In settings marked by the discretionary and often arbitrary behavior of officials, citizens have little reason to place their faith in public outreach or media campaigns, the likes of which a NGO staff member in Udaipur referred to as “yellow paint promises” (referring to the yellow-colored paint often used on panchayat buildings where announcements are posted). Instead, people are much more likely to believe, and so act upon, what they see and hear around them. Socially embedded sources of information are, in this way, more powerful than impersonal sources. Highly localized ties of kin, caste, neighborhood, or village can limit the flow of information and ideas; within these tightly knit circles, certain narratives about the state circulate and are reinforced. These constrained dynamics were common in interviews among women, whom, as we have seen, are the most restricted socially and spatially. A Rajput (GC) woman, a self-described housewife living in a remote Ajmer village, summed up these dynamics with a sentiment echoed by many other women. I was in the process of asking her about a range of public services, but she stopped me short: I don’t know about any of these things. How would I know? My life is here [she gestures around her to the house]. I do housework and nothing else. You come from where? From Ajmer? Maybe they have all those things there. I don’t know anyone who has all that.95

94 95

Author interview, Bargaon district, Udaipur block, April 16, 2009. Author interview, Peesangan block, Ajmer district, April 2, 2010. Interestingly, this woman kept saying that I came from Ajmer (the district’s capital city), even though I had introduced myself as

144

Encountering the State: Citizens’ Social and Spatial Exposure

In other cases, citizens draw on ideas and narratives that circulate beyond their local communities when considering how to engage the state. It is a difficult task to trace stories of social learning and the spread of practices. Claim-making repertoires are developed over time, through the slow accretion of experiences and encounters that inform a person’s understanding of the state. Given these dynamics, most people are hard put to reflect on what they learned, when, and how. It is one thing to ask a person what he or she has done (and even this is plagued with problems of recall and reliability). It is quite another to ask them why they did that thing. There are, nonetheless, instances in which I gained glimpses into these largely internal processes. For example, a member of a youth group in a village in Udaipur described how they began to take up issues related to village sanitation with the panchayat: interviewer: How did you start working on these [sanitation] issues as a youth group? youth leader: Most of us go to study in Udaipur [city], and we heard of such things there, and so thought we should make our own group here to improve the work of the panchayat … If the GP won’t get our work done, we will make an andolan [movement]. We will lock down the panchayat: lock the door and say, “when you do the work we will open it.” interviewer: Have you actually done this? youth leader: No, but we have heard of it. A group of women here in the village threatened to do this. There had been almost five days without water. The women went together to the panchayat and said, “fix this problem or we will lock you out.” And the problem got resolved! We learned from them. If even the ladies of this village can do this … [he trails off, implying “if women can do this, so can we”].96

I later spoke to one of the women involved in the event referenced by this youth leader. I asked her where the idea for this kind of protest originated. She explained that a group of women who met regularly at a government day care center (anganwadi) began discussing the water issues. One of the women, “Gita,” used to be a Gram Panchayat member; at her urging, they organized themselves to complain about the lack of water. They drew, she explained, on Gita’s past experience in the panchayat and the knowledge she gained in that position: We demanded water from the officials … At first you might feel shy. People were thinking, what’s the use? Why should we ladies attend these meetings? But Gita ji has experience in the panchayat. She has traveled and attended trainings, and she knows how the administration works. She told us, “We have elected those people. We have votes, just as the men!”97

96 97

coming from America. Each time I would say “America,” she would nod her head and say “Ajmer.” This perhaps suggests that her worldview did not extend far enough to include notions of America. To her, “Ajmer” (about 20 kilometers away) represented the outside world. Author interview, Bargaon block, Udaipur district, December 20, 2010. Author interview, Bargaon block, Udaipur district, December 16, 2010.

Exposure and Claim-Making Capabilities

145

In this series of examples, ideas about claim-making spread within a group of women through exposure to narratives that, themselves, were informed by experiences beyond the village. Gita, in this case, formed a “bridge” that linked the group of women to broader political narratives that informed their practice within the village. The experiences of these women, in turn, shaped the frames of reference employed by the youth group, which were reinforced by their own, external ties to Udaipur city. Multiple sources of information are often at play. A recent school graduate in Udaipur, for example, described the process through which he came to organize a protest over teacher absenteeism in his school. He organized a group of students and others from the village who blocked the local highway with rocks, stopping traffic. The District Collector came to the protest site, and promised to take action to discipline the absent teacher. He recalled: I read in the newspaper about some blockades. It was in Jaipur, I think. I read how the people there had blocked the roads and how they were throwing stones because the wells were dry and no leader would hear them. I thought, here we have wells. But our school is broken. No one hears us, but let us try … My cousin-brother has also worked in Ahmedabad and Mumbai, and he tells us how people there do such things, blocking up the traffic when there is some issue to resolve. So I thought, let us try this here.98

This young man acted, first, on knowledge of protest practices gained through reading the newspaper. Second, though, this impersonal source of information was reinforced by personal knowledge of someone who had witnessed protest activity beyond the village. Together, these experiences shaped his understanding of the strategies of action he might undertake when making claims on the state. Claim-making also requires outward-reaching linkages to the state: that is, either contact with officials or with intermediaries who, in turn, facilitate access to officials. The extent and nature of these contacts reflects, in part, the local opportunity structure. The sociologist Ronald Burt describes this in terms of the structural holes – or gaps – in a social network that inhibit access to particular people or resources, and the “gatekeepers” who bridge those gaps.99 These gatekeepers, however, are not uniformly distributed: access to them varies from place to place and from person to person. The more expansive a person’s networks, the more likely he or she is to encounter potential linkages to the state. Greater social and spatial exposure, in other words, builds an individual’s opportunity structure. To illustrate, consider this resident of Jodhpur district, who described how his contact with political parties changed when he began working as a manual

98

Author interview, Gogunda, Udaipur, February 22, 2010.

99

Burt 1992; 2000.

146

Encountering the State: Citizens’ Social and Spatial Exposure

laborer in the city, traveling there on a daily basis from his nearby village in Mandor block – a peri-urban area that is adjacent to the district capital: The city is where you meet all the political people. I’m Congress. My family is Congress. This whole village is Congress. But stay put in the village and it does not matter who you are. No one is coming to hear you. They will eat up your votes, that’s it. Travel to the city and you can meet the MLA face to face.100

He went on to explain how, when working on construction sites, he heard other laborers recount their experiences visiting the offices of the local Member of Legislative Assembly (MLA). It is relatively common in urban settings for the MLA or his or her staff to meet with members of the public, often during designated constituency service hours.101 This kind of direct contact with politicians, however, is rare in village settings. In moving beyond his village, the man in question gained new contacts (on construction sites) and, through them, new information about potential claim-making practices (through parties). His presence in the city, moreover, put him in greater physical proximity to politicians, increasing their accessibility. The man visited the local MLA office to enquire about a pension for his elderly mother, the application for which had been stuck for long periods of time. He met with a “peon” in the MLA’s office, who wrote down his complaint. Two months later, he reported, his mother’s pension was approved. It is not possible to know whether this man’s advocacy had any direct effect on the pension application. But in his own assessment, “the money came only after I went straight to the MLA. It was a matter of going straight to the top.” Greater exposure, in this case, lead to greater claim-making opportunities. Other sites of exposure (and of claim-making) are more local. One of the most common sites, particularly for lower-caste women, is the National Rural Employment Guarantee worksite. As a tribal woman in Udaipur described: Before in the dry season we would sit at home. Now there is 100 days of work. Everybody comes there when we are working. There is the mate, the technician, the supervisor. The higher-ups from Gogunda also come and check on the progress. Sometimes politicians also come to show their faces.102

MGNREGS, in this sense, provides a platform for local citizen–state engagement. Officials often visit worksites in the course of public outreach campaigns. For example, when attempting to inform local residents about an upcoming administrative camp, a block official in Udaipur reported sending his staff to visit as many worksites as possible in order to spread the word among workers.103 Politicians also visit MGNREGS sites in the course of their campaign work or when trying to recruit people for political rallies. Citizens 100 102 103

Author interview, Mandor block, Jodhpur district, April 8, 2010. Author interview, Gogunda block, Udaipur district, February 2010. Author interview, Gogunda block, Udaipur district, January 2011.

101

Bussell 2015, 2018.

Exposure and Claim-Making Capabilities

147

working on MGNREGS sites thus come into contact with new political actors. The same woman in Udaipur summed this up powerfully, saying: “On the [MGNREGS] job, we are seen. At home, we are ignored.” This does not, of course, in any way ensure that officials at MGNREGS sites are actually responsive to citizens’ demands. But the worksites are, at the very least, spaces where citizens encounter officials who are otherwise markedly absent from the village.104 In other instances, channels of access to the state are less direct. Another woman in Udaipur explained how she came to learn about potential intermediaries in her village, which informed her decision about where to turn when facing a drinking water shortage. She explained: I move about and meet people. I talk to people on the way, and in the fields, and in this way come to learn about who is good, who will help you. Someone will say, “So-and-so, he is jagruk [aware], he can get things done.” And so when the water pump was broken, I knew where to turn.105

Greater exposure, in this case, increased the claim-making opportunity structure by providing information about and channels of access to brokers. Taken altogether, these accounts illustrate how encounters beyond the immediate community and locality can serve to build knowledge, generate expectations, and expand channels of access to the state. Importantly, there is no single driver of exposure; individuals are set in motion or constrained by a wide-ranging and often contingent set of geographic, economic, social, and institutional factors. Once drawn beyond their local communities, however, they gain resources (information, ideas, contacts) that build aspirations toward the state and capabilities for state-targeted action.

104

105

MGNREGS sites can also be sites of NGO advocacy. An Udaipur-based NGO, for example, would visit worksites to advertise new programs. A number of other civil society organizations also regularly visit MGNREGS sites to document working conditions and to provide information on workers’ rights. Author interview, Bargaon, Udaipur, December 16, 2010.

6 Claiming the State Exposure as a Catalyst for Citizen Action

The theory of active citizenship developed thus far predicts that individuals who come into contact with people and settings beyond their immediate community and locality will be more likely to engage in state-targeted claimmaking, and will do so through a broader array of practices, than those who operate in more circumscribed spheres. The preceding chapter offered qualitative accounts, drawn from village-level interviews, of the ways in which exposure beyond one’s own caste, neighborhood, and village might influence citizens’ approaches to the state. This chapter examines the theory’s plausibility among a broader sample of rural Rajasthanis, utilizing data from the citizen survey along with village case studies. I begin by presenting an empirical strategy through which to test the theory, before turning to the survey data to develop a series of metrics of social and spatial exposure in order to examine their relationship to claim-making. I go on to present the results of multivariate analysis, demonstrating that greater exposure is broadly associated with both more and more diverse claim-making practice. The remainder of the chapter turns to the question of causality, asking: what’s driving what, and how? I start by distinguishing underlying from more proximate causal processes, recognizing that in the long run there is a powerful feedback loop wherein the state shapes citizen claim-making activity and claimmaking, in turn, affects the actions of the state. In the short term, though, citizens’ networks of exposure appear to have a demonstrable independent effect on whether and how particular individuals engage the state. To address concerns over endogeneity in this more proximate sense, I introduce a “proxy” variable – the village land-to-labor ratio – that, by capturing an underlying structural driver of mobility, allows us to indirectly assess exposure’s effects on claim-making. I then examine whether there are omitted (or “hidden”) variables that might confound the analysis by influencing both citizens’ exposure to and engagement of the state, as well as the possibility of a differential effect of 148

Testing the Relationship

149

exposure on differently situated groups of citizens. Together, the analysis highlights the persistence and consistency of exposure’s catalytic effects on groups set apart by class, caste, gender, and other features. Last, I turn to qualitative data drawn from comparative village case studies to further illustrate the relationship between social and spatial exposure and claim-making, highlighting complexities not captured in the survey data.

testing the relationship In the analysis that follows, I treat measures of social and spatial exposure as explanatory variables, examining the extent to which each one is predictive of a range of claim-making outcomes related to incidence, practice, and breadth of repertoire.1 To isolate the potential effects of exposure, I introduce a range of controls at the level of the individual, the household, the village, and the Gram Panchayat, along with block fixed effects. Outcome Variables: Claim-Making Incidence, Practice, and Repertoire Claim-making is defined, in the terms laid out in Chapter 1, as state-targeted citizen action in pursuit of social welfare. I begin, following the measurement strategies laid out in Chapter 4, by examining the incidence of claim-making: that is, whether or not a person reports having directly or indirectly contacted an official for assistance related to the provision of welfare goods or services.2 Claim-making, as we have seen, takes place through an array of channels and strategies. I therefore also examine the practices employed. These include direct, face-to-face contacting of public officials as well as contact mediated through a range of non-state actors and institutions. Each claim-making practice is measured in binary terms, coded one where the respondent reported having personally contacted the actor or institution in question, and zero otherwise. I utilize the index of the claim-making repertoire to measure the breadth – or number – of practices that an individual undertakes. 1

2

Full specifications of the models are provided in Appendix II. The basic empirical strategy is: Y = α + β(exposure) + γ(individual) + λ(household) + θ(locality) + Ω(block fixed effects) + ε, where Y is the claim-making outcome of interest (incidence, practice, or repertoire); where β estimates the predictive effect of a series of indicators of social and spatial exposure; where γ, λ, and θ represent vectors of controls at the level of the individual survey respondent, their household, and their locality (village and Gram Panchayat); all with the inclusion of block fixed effects. Standard errors are clustered at the village level. Each measure of exposure is separately assessed due to collinearity among the measures. An additional model includes all measures of exposure together to test for their joint significance in predicting claim-making. In the analysis that follows, I examine claim-making in regard to any kind of welfare service, collective or selective. As a robustness check, I run additional analysis in which collective and selective goods are treated separately. In these models, the main effects of social and spatial exposure on claim-making persist regardless of the type of good.

150

Claiming the State: Exposure as a Catalyst for Citizen Action

table 6.1 Indicators of Social and Spatial Exposure Mean

Std. Dev.

Min

Participation in Mixed-Caste Cultural Group Social Engagement beyond Neighborhood Engagement in Mixed-Caste Workplace Migration beyond Village (household) Index of Exposure

0.14 0.68 0.81 0.22 1.85

0.34 0.47 0.39 0.41 0.91

0 0 0 0 0

Village Land-to-Labor Ratio (proxy for exposure)

1.52

1.21

0.08

Max 1 1 1 1 4 10.12

Source: Citizen survey 2010–2011 (n = 2,210). Notes: Index of exposure is a composite of the binary measures that precede it. Village land-to-labor ratio is hectares of cultivable land relative to adult working population, drawn from the 2001 Census of India (Primary Abstract and Village Directory).

Explanatory Variables: Measuring Social and Spatial Exposure Exposure involves the experience of difference: that is, encounters with people and places beyond one’s own community and locality. In the Rajasthani context, as we saw in Chapter 5, this exposure is gained through movement across caste, class, and gender, as well as neighborhood and village lines. There are, as noted, many drivers of this exposure that vary across localities and among individuals.3 It follows that there is no single, encompassing metric of exposure. I do not, moreover, make any attempt to comprehensively map an individual’s sources of exposure; to do so would impose a false sense of fixedness on a complex set of engagements and relations that are subjective and shifting in nature, and so, like most social networks, which cannot be “mapped as a whole entity.”4 Instead, I employ four separate indicators that reflect different dimensions of exposure, summarized in Table 6.1. These indicators do not comprise an exhaustive list. Instead, they simply capture different forms of interpersonal engagement that span social and spatial boundaries. I begin by examining social engagement that extends across caste and neighborhood lines. The literature on social capital has long noted the importance of “bridging” forms of association that cut across social cleavages and so generate new resources for civic as well as political participation.5 Within this 3

4 5

Panchayat, village, and individual-level correlates of social and spatial exposure are examined later in this chapter. MacLean 2010, p. 20. As discussed in Chapter 2, social and spatial exposure does not require trust or reciprocity, nor does it assume horizontality in social relations. As such, the resources generated through exposure are closer to alternative conceptions of social capital that emphasize the unequal distribution of benefits by virtue of participation in unequal social networks (Bourdieu 1986; Portes 1998; Levien 2015).

Testing the Relationship

151

body of work, scholars frequently distinguish institutionalized voluntary associations from more quotidian kinds of social ties.6 I therefore employ two distinct measures of social engagement, first of a formal and second of a more informal nature. Participation in a Mixed-Caste Cultural Group A first metric assesses participation in cultural groups, defined in terms of membership in music, film, sports, youth, or other such voluntary organizations. Formal organizations of this kind are relatively rare: just 15 percent of the citizen survey respondents reported participating in any such group. However, of those who do, 90 percent stated that they participate alongside people from different castes.7 For these respondents (who account for a total of 14 percent of the survey sample), these cultural groups offer critical sites of exposure to people beyond their own caste communities. Social Engagement beyond the Neighborhood A second set of survey questions sought to shed light on more quotidian – but also more common – forms of social engagement, by asking about the frequency with which respondents “meet or sit with people from different mohallas (neighborhoods).” This question was followed by a series of examples intended to anchor the notion of “meeting” or “sitting” with others, which included visits to people’s homes, playing cards or other games, performing kirtan (collective worship), or gathering for puja (religious ceremonies).8 As noted, most villages are internally divided along both caste and class lines; wealthier, upper-caste homes are located in the center, while lower-caste and poorer homes are located on the outskirts or in hamlets. Within the “main”

6

7

8

Varshney (2002), for example, argues that institutionalized associations have a more powerful and lasting effect in the maintenance of ethnic peace than more informal, day-to-day interethnic contact. MacLean (2010), in contrast, emphasizes the role of “everyday” informal social relations in shaping political participation. In India, formal voluntary associations are largely an urban phenomenon, participation in which is most often the purview of the wealthy and middle classes (Chhibber 1999; Krishna 2002; Harriss 2005). This may reflect the fact that cultural associations are often established or supported by external organizations that seek to attract mixed membership. These include NGOs as well as political organizations such as the Rashtriya Swayamsevak Sangh (RSS) – a voluntary organization that promotes Hindu nationalist ideology while attempting to attract members from different castes, particularly among the lower castes (Thachil 2014). These examples were offered as a way to situate the survey question in the local context. However, individual respondents were left to decide for themselves both what constitutes social engagement as well as what constitutes their neighborhood. This degree of subjectivity makes it impossible to precisely map these forms of cross-cutting engagement, but instead provides insight into the conceptual categories at play and the degree to which a person, according to his or her own assessment, traverses such divides. It provides, in other words, a powerful sense of the degree to which a person engages with people that he or she perceives to be “different.”

152

Claiming the State: Exposure as a Catalyst for Citizen Action

village, neighborhoods also often have a distinctive caste character, and it is common to hear people refer to areas of the village by caste name (for example, the “Brahmin basti,” the “Jain mohalla,” the “Bheelwada,” and so on). Despite these clear demarcations, more than two-thirds (68 percent) of survey respondents reported engaging socially (“meeting” or “sitting”) with people across neighborhood lines.9 Engagement in a Mixed-Caste Workplace Labor markets and workplaces are also key sites of interpersonal interaction. A third measure of exposure therefore examines the extent to which citizens’ occupational networks extend across caste lines. Survey respondents were asked to report whether they worked primarily with people from their own or other caste communities (jatis).10 Rural occupations, as noted in Chapter 5, were historically caste-bound. Caste and occupation, moreover, were traditionally linked to place (particularly under systems of bonded labor). Working alongside people from different caste communities therefore creates a key opportunity for exposure across communal (caste) lines, as well as beyond one’s neighborhood or village. This, in fact, is the most common source of exposure: 81 percent of those surveyed reported working “side by side people from other caste communities.” These large numbers likely reflect the diversification of rural livelihoods and decoupling of caste and occupation, discussed in Chapter 5. Migration beyond the Village In an attempt to more directly capture the spatial dimensions of exposure, respondents were also asked about migration. Slightly less than one-quarter (22 percent) reported that they or a member of their household lived and worked outside of the village for more than thirty days in the year. This varied by region within the state, reflecting differences in local demography, geography, and economy. Where agricultural opportunities are scarce, people are more likely to migrate in search of new opportunities. Thus, in

9

10

Most (64 percent) reported only “occasional” social contact of this kind, while a much smaller number (4 percent) reported “regular” or “frequent” engagement beyond the neighborhood. The remainder of the sample stated that they “almost never” meet or sit with people from other neighborhoods. For ease of interpretation and consistency with the other indicators of exposure, social engagement beyond the neighborhood is reported in Table 6.1 in binary terms. It was left to the respondents to decide what constituted the boundary of their own caste community, defined in the survey as “jati samudaay. ” The question thus has the potential to capture interactions both across broad categorical caste divisions (e.g., SC, ST, OBC, and GC) and across specific caste or even subcaste lines within those broad categories. It does not, however, tell us anything about the relative strength or weakness of particular caste boundaries. This data limitation is discussed later in this chapter.

Testing the Relationship

0

10

Percent 20 30

40

50

153

0

1 2 3 Number of Sources of Exposure

4

figure 6.1 Distribution of the Index of Social and Spatial Exposure

Udaipur (the poorest district sampled, in which productive agricultural land is most limited), 25 percent of those surveyed reported that they or a family member migrated, compared to just to 16 percent in Kota (the most developed agricultural district). The survey, however, offers only a blunt measure of these dynamics at the household rather than individual level. We cannot, moreover, distinguish between long-distance migration and more local, circular forms of “up-down” (described in Chapter 5). I return to these issues, and to the potentially divergent effects of different kinds of extravillage movement, below. Index of Social and Spatial Exposure As should be clear, one source of exposure does not preclude another. Working in a mixed-caste environment, for example, might lead to social engagement across neighborhood lines, just as migration might lead to employment in a mixed-caste setting. I therefore introduce an index of the sources of exposure reported by an individual (each assessed in binary terms). The index ranges from zero to four; the vast majority (92 percent) reported at least one source of exposure (Figure 6.1), with an average of just under two sources (Table 6.1).11 Control Variables To capture the relationship between exposure and claim-making, it is necessary to distinguish exposure’s effects from other factors that could also influence citizen action. I therefore introduce a range of control variables at the

11

In addition to this index, an additional model conducts principal component analysis to assess the combined effect of the four indicators of exposure.

154

Claiming the State: Exposure as a Catalyst for Citizen Action

individual, household, village, and panchayat levels.12 Individual controls hold constant a wide range of features that might influence a person’s relationship to the state and hence their claim-making practice, among them: age, gender, caste, and level of educational attainment. Other individual controls include access to media (controlling for an effect of information that operates independently from exposure); as well as party membership (controlling for the effects of partisanship on claim-making). Household controls include family size; wealth, assessed in terms of landownership and an index of assets (including durable goods, livestock, and the materials with which one’s home is constructed); and sources of livelihood, including both farm and nonfarm economic activities (with a separate control for employment on a MGNREGS site). Village controls reflect demographic features including population size and density, as well as caste composition (the fractionalization or “mix” of castes in a village and, assessed separately, the proportion of SC and ST in a village). Geographic features include the distance to the closest town, whether the location is a hamlet or part of the “main” (central) village, as well as measures of village connectivity (for example, whether there is a bus stop in the village). Other village controls are socioeconomic, including literacy rates, average levels of asset- and landownership, and an intra-village Gini coefficient measuring inequality. Other variables control for village partisanship and for whether the village is the Gram Panchayat’s headquarter. Panchayat-level variables control for the GP’s population size, as well as for caste and gender reservations for the seat of sarpanch. The core models presented below also include block fixed effects, which in hold constant everything at the level of the administrative block and above, including differences in bureaucratic governance, geography, or level of development.

analysis and findings Analysis of the citizen survey data reveals, in broad keeping with the theory’s predictions, that social and spatial exposure is associated, all else equal, with a greater incidence of claim-making and with a broader repertoire of claim-making practice (Table 6.2). Across a range of metrics, encounters beyond the immediate community and locality are correlated both with an increased likelihood of statetargeted citizen action and a wider array of strategies employed.13 12

13

The full range of controls is described in Appendix II, with descriptive statistics in Tables A1.1 and A1.2. The same controls are presented as correlates of claim-making in Tables A2.2 and A2.3. Full results are in Appendix II, Table A2.4, which presents the association between the indicators of exposure and the full range of claim-making practices, both direct and mediated. The effect of each indicator of exposure is separately assessed in its own model. When included in a regression together, the indicators of exposure also strongly predict claim-making incidence and repertoire, with the p-values of tests of joint significance