Deceptive Majority: Dalits, Hinduism, and Underground Religion 9781108494571, 2020049864, 2020049865, 9781108843829, 9781108826662, 9781108920193

469 122 7MB

English Pages [353] Year 2021

Report DMCA / Copyright


Polecaj historie

Deceptive Majority: Dalits, Hinduism, and Underground Religion
 9781108494571, 2020049864, 2020049865, 9781108843829, 9781108826662, 9781108920193

Citation preview

Deceptive Majority How did it come to be common sense that the vast swath of the population of South Asia once known as ‘untouchable’ are and always have been Hindu? Grounded in detailed archival and ethnographic research, Deceptive Majority unearths evidence that well before the emergence of twentieth century movements for Dalit liberation, the subset of ‘untouchable’ castes engaged in sanitation labor in colonial India conceived of themselves as constituting a religious community (qaum) separate from both Hindus and Muslims—a community with its own prophet, shrines, rites, legends, and liturgical songs. This book tracks the career of this tradition alongside the effort to encompass it within a newly imagined Hindu body politic—a majoritarian project advanced in complex, distinct, yet convergent ways by colonial administrators, Hindu nationalists, the Congress Party, and Mohandas Gandhi. A sensitive account of contemporary religious life in the north Indian city of Lucknow illuminates both the embrace and the contestation of Hinduization within a Dalit community. A weaving together of the history and ethnography of religion, Deceptive Majority reveals the cunning both of the architects of Hindu majoritarianism and of those who quietly undermine it. Joel Lee is Assistant Professor of Anthropology at Williams College, Massachusetts.

SOUTH ASIA IN THE SOCIAL SCIENCES South Asia has become a laboratory for devising new institutions and practices of modern social life. Forms of capitalist enterprise, providing welfare and social services, the public role of religion, the management of ethnic conflict, popular culture and mass democracy in the countries of the region have shown a marked divergence from known patterns in other parts of the world. South Asia is now being studied for its relevance to the general theoretical understanding of modernity itself. South Asia in the Social Sciences will feature books that offer innovative research on contemporary South Asia. It will focus on the place of the region in the various global disciplines of the social sciences and highlight research that uses unconventional sources of information and novel research methods. While recognising that most current research is focused on the larger countries, the series will attempt to showcase research on the smaller countries of the region. General Editor Partha Chatterjee Columbia University Editorial Board Pranab Bardhan University of California at Berkeley Stuart Corbridge Durham University Satish Deshpande University of Delhi Christophe Jaffrelot Centre d’etudes et de recherches internationales, Paris Nivedita Menon Jawaharlal Nehru University Other books in the series: Government as Practice: Democratic Left in a Transforming India Dwaipayan Bhattacharyya Courting the People: Public Interest Litigation in Post-Emergency India Anuj Bhuwania Development after Statism: Industrial Firms and the Political Economy of South Asia Adnan Naseemullah

Politics of the Poor: Negotiating Democracy in Contemporary India Indrajit Roy South Asian Governmentalities: Michel Foucault and the Question of Postcolonial Orderings Stephen Legg and Deana Heath (eds.) Nationalism, Development and Ethnic Conflict in Sri Lanka Rajesh Venugopal Adivasis and the State: Subalternity and Citizenship in India’s Bhil Heartland Alf Gunvald Nilsen Maoist People’s War and the Revolution of Everyday Life in Nepal Ina Zharkevich New Perspectives on Pakistan’s Political Economy: State, Class and Social Change Matthew McCartney and S. Akbar Zaidi (eds.) Crafty Oligarchs, Savvy Voters: Democracy under Inequality in Rural Pakistan Shandana Khan Mohmand Dynamics of Caste and Law: Dalits, Oppression and Constitutional Democracy in India Dag-Erik Berg Simultaneous Identities: Language, Education and the Nepali Nation Uma Pradhan

Deceptive Majority Dalits, Hinduism, and Underground Religion

Joel Lee

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. Information on this title: © Joel Lee 2021 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 2021 Printed in India A catalogue record for this publication is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Lee, Joel G., author. Title: Deceptive majority: Dalits, Hinduism, and underground religion / Joel Lee. Description: Cambridge; New York, NY : Cambridge University Press, 2021. | Series: South Asia in the social sciences | Includes bibliographical references and index. Identifiers: LCCN 2020049864 (print) | LCCN 2020049865 (ebook) | ISBN 9781108843829 (hardback) | ISBN 9781108826662 (paperback) | ISBN 9781108920193 (ebook) Subjects: LCSH: Hinduism and politics--India--Lucknow. | Dalits--India--Lucknow--Religion. | Dalits--Political activity--India--Lucknow. | Caste--Religious aspects--Hinduism. | Caste--Political aspects--India--Lucknow. | Social integration--Religious aspects--Hinduism. | Political sociology--India--Lucknow. | Lucknow (India)--Politics and government. | Lucknow (India)--Religion. Classification: LCC BL1215.P65 L44 2021 (print) | LCC BL1215.P65 (ebook) | DDC 305.5/68809542–dc23 LC record available at LC ebook record available at ISBN 978-1-108-84382-9 Hardback ISBN 978-1-108-82666-2 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.

In memory of Sakun

Shakuntala Devi (1953–2013) Portrait by Sanya Darapuri

[A]rtifice was necessary, he had found, for stemming the cold and inhumane blast of the world’s contempt. Thomas Hardy, Jude the Obscure


List of Figures Acknowledgments

xiii xv

I. UNTOUCHABILITY AND ALTERITY, NOW AND THEN 1. Introduction: Signs, the Census, and the Sanitation Labor Castes 2. The Ummat of Lal Beg: Dalit Religion before Enumerative Politics

3 31

II. MAKING “UNTOUCHABLES” HINDU, OR THE GREAT INTERPELLATION 3. Missionary Majoritarianism: The Arya Samaj and the Struggle with Disgust 4. Trustee Majoritarianism: Gandhi and the Harijan Sevak Sangh 5. Hinduization and Its Discontents: Valmiki Comes to Lucknow

77 121 163

III. SEMIOTICS OF THE OPPRESSED 6. Victory to Valmiki: Declamatory Religion and the Wages of Inclusion 7. Lal Beg Underground: Taqiyya, Ethical Secrecy, and the Pleasure of Dissimulation Epilogue References Index

213 252 291 307 327


1.1 Awadh, with traditional area of settlement of the 583 and neighboring qabīle (clans)


2.1 Lal Beg shrine (western style)


2.2 Lal Beg shrine (eastern style)


3.1 “Football of the Untouchables”


4.1 “Left-Handed Compliment”


4.2 “Sweet Mercy”


5.1 Deg


5.2 Bhagaunā


7.1 Sohan Lal (who goes by another name in this book)



A thousand acts of generosity and forbearing have made this book possible. Let me start in Lucknow by thanking Ganesh Kumar Adivamshi, Sneh Lata, and their son Shubh Ambedkar, whose friendship, hospitality, humor, and belief in the value of this project are the bedrock on which this book rests. My gratitude goes likewise to Ganesh’s siblings Nisha, Mukesh, Amit, Vinay, and their spouses. Their mother Shakuntala Devi, known by friends as Sakun, taught us all so much by word and deed, and left us all bereft when she died so early. I think she would have liked this book, and I dedicate it to her. Another friend lost too soon was Sohan Lal “Gupta,” who shared my fascination with history and without whose guidance in the mysteries of Lal Beg this book would not have been possible. He and Shanti and their children welcomed me and my family as their own, as did Pushpalata and Jagdish Atal, Mehphul and Shyama and their sisters, Sant Kumar and Ali Hussein. I thank Sanjeed, Lallan, Rajan, Dilkash Warsi, Kunwar Ranjit Singh, Anup Kalyani, Shyam Lal Vaid, Shyam Lal Pujari, Shyam Lal son of Bhaggan, Change Lal, Khannaji, Sant Kishore, Jagdish Prasad, Devi Prasad, Vinod Valmiki, Mosna, Chunni Lal, and all of their families. I am grateful to Nandini Vaid, Rimpi Gauri, Mayank Khanna, Abhishek Diwan and the whole 2011–12 Spoken English cohort at the Fast Computer Training Institute for being such inspiring students, inviting me into their lives, and keeping in touch. The book begins with observations from the 2011 census in a Dalit neighborhood of Lucknow; I am grateful to Nina Sharma, Superintendent of Census Operations for Uttar Pradesh, for kindly granting me permission to participate in this eye-opening exercise. Chapter 5 relies in part on the writings, photographs, and personal correspondence of the first generation of Balmiki political leaders in Awadh; for sharing their family collections I warmly thank Sarojini Balmiki, Prithviraj Chauhan, Raj Kumar Singh, Ashok Kumar, Chandra Kumari, and P. Chandra—children and grandchildren of Kanhaiya Lal Balmiki (Member of Legislative Assembly) and Narain Din Balmiki (Member of Parliament). My gratitude goes as well to Harsh Singh, grand-nephew of Govind Prasad of the reformist triumvirate, who kindly read my entire dissertation

Acknowledgments xvi

and gave me pages of detailed feedback. In Lucknow I also want to thank Sara Khan, Ahtesham Khan, and the late, much-missed Ram Advani. Conversations with Masood Alam Falahi and C.M. Naim at the latter’s ancestral home in Bara Banki decisively shaped my thinking during fieldwork and beyond, as did many discussions over biryani with Nishat and Nadeem Hasnain in Lucknow. I thank Pushpa Balmiki, her husband Ramesh and their children, who, along with Dev Kumar in Kanpur, included me in their emancipatory projects and encouraged me to pursue the history of Lal Beg and Valmiki. In Delhi, Amritsar, Panipat, Benares and elsewhere the fieldwork for this book took me, I gained tremendously from conversations with and support from Paul Divakar and Annie Namala, Bezwada Wilson, Vimal Thorat, Sukhdeo Thorat, S. Anand, Chandra Puhal, Murtaza Alam, Ali Hussein Ustad, Darshan Ratan Ravan, Jayshree Mangubhai, S. D. J. M. Prasad, Aloysius Irudayam, Shashi Chauhan, Amuda Prakash, Ashok Pandey, J. P. Singh, Priyadarshini Vijaisri, Prem Singh, Praveen Kumar, and Imtiaz Ahmad. This study builds upon foundations laid by scholar-activist Bhagwan Das (1927–2010). An associate of Dr. Bhimrao Ambedkar and participant in and theorist of sanitation labor caste struggles for justice and equality, Bhagwan Das was working in his final years on a monograph on the history of his community. This project was left unfinished; I hope, though, that my book, guided as it is by intuitions found in some of Bhagwan Das’s Hindi and Urdu essays and by an interview I had with him in his last months, might in some small way carry forward his legacy. I thank his daughter, historian Shura Darapuri, and her father-in-law, S. R. Darapuri, and their whole family, for their encouragement of this project. Stepping further back, it was Jebaroja Singh and Prince Singh who first set me on the path of critical caste research; I am always grateful for the life they model and the push they gave. Around the same time Eleanor Zelliot introduced me to Dalit literature and to the life and work of Ambedkar. Royal Rhodes, Wendy Singer, and Miriam Dean-Otting sowed seeds of inspiration. In linguistic training, I benefitted from studying with outstanding pedagogues, especially Aftab Ahmad, Fauzia Farooqui, and Frances Pritchett in Urdu and Andy Rotman, Vidhu Shekhar Chaturvedi, and Premlata Pinki Vaishnav in Hindi. The research on which this book is based was supported by the Social Science Research Council and Fulbright-Hays. Shorter subsequent research stints were enabled by the Max Planck Institute for Human Development and Williams College, and final touches were completed while starting a new project supported by Fulbright-Nehru, the American Institute of Indian Studies, and the National Endowment for the Humanities. I thank the staff at each of these institutions.



I also thank Karola Rockmann at Max Planck and the librarians, archivists, and other staff at the Uttar Pradesh State Archives, Lucknow; Nehru Memorial Library, Delhi; Burke Library at Union Theological Seminary, New York; and the Islamic Studies Library at McGill University, Montreal (especially Anaïs Salamon). It has been a privilege as well to be affiliated with the Anthropology Department of Lucknow University and, in Delhi, the Centre for the Study of Developing Societies and the Indian Institute of Dalit Studies while conducting research for this book. In developing this work I have benefitted immensely from discussions with friends and fellow travelers in graduate school and beyond, especially Anand Taneja, Carla Bellamy, Nicolas Jaoul, Gajendran Ayyathurai, Tabinda Khan, Divya Cherian, Tyler Williams, Joel Bordeaux, Patton Burchett, Jon Keune, Harmeet Kaur, Jayson Beaster-Jones, Ajeet Singh Matharu, Pasha M. Khan, Prashant Keshavmurthy, Laura Brueck, Beatrice Jauregui, and Katyayani Dalmia. At Chicago, Muzaffar Alam, Dipesh Chakrabarty, McKim Marriott, and again C. M. Naim generously shared their time and ideas on the project; Naim Sahib also drew my attention to one of the Urdu texts that would prove crucial to the historical narrative. Owen Lynch modeled intellectual generosity and ethical commitment; he also persuaded me to be an anthropologist. Manpreet Kaur expertly guided me through passages of idiosyncratically transliterated Punjabi sociolect from the colonial archive, as did Gurinder Singh Mann on another occasion. Charu Gupta kindly shared the “untouchable football” image (chapter 3) from her archival discoveries. I warmly thank all of my mentors at Columbia, especially Anupama Rao, Sudipta Kaviraj, Rachel McDermott, Brian Larkin, Katherine Pratt Ewing, Nicholas Dirks, and Frances Pritchett. Conversations with Gopal Guru, Shahid Amin, and Christophe Jaffrelot during their sojourns at Columbia were formative. This book has been immeasurably enriched by the wit and rigor of my dissertation committee: Elizabeth Povinelli, Partha Chatterjee, Brinkley Messick, John Stratton Hawley, and my PhD advisor E. Valentine Daniel. In the metamorphosis of this project into its present form, I have had the good fortune to share my findings with discerning audiences in too many places to name. For their probing and insightful questions I am particularly grateful to: Eric Lott, Irvin Hunt and Karl Jacoby at the Center for the Study of Ethnicity and Race at Columbia University; Kanthi Swaroop Gunti and the Ambedkarite Students Collective at Indian Institute of Technology Bombay; and Demetrius Eudell, Ramnarayan Rawat, Linda Hess, Meena Khandelwal, Philip Lutgendorf, Margrit Pernau, Ute Frevert, Razak Khan, Lisa Mitchell, Kedar Kulkarni, Nandini Gooptu, Martin Fuchs, Sanal Mohan, Julien Levesque, Saswathi Natta, Urvashi Butalia,

Acknowledgments xviii

Suraj Yengde, Surinder Jodhka, and Suryakant Waghmore at conferences from Madison to Milan. I am grateful to all of my colleagues at Williams College for their sustained interest in and encouragement of this project; my writing has been shaped especially in dialogue with Peter Just, Olga Shevchenko, David Edwards, Saadia Yacoob, Aparna Kapadia, Jeffrey Israel, and with my comrades in the ethnography of religion writing group: Meredith Coleman-Tobias, James Manigault-Bryant, and Zaid Adhami. I thank my students, especially Thasin Alam, Grace Fan, Gabriel Silva Collins, Danielle Faulkner, and Eva Asplund, for being a wellspring of hope and reminding me what the scholarly endeavor is all about. Among the peculiar joys of academic life is the informal gift economy of deep, scrupulous, critical readings of one another’s work. I have benefitted more than I can convey from such engagements with this book. In particular I thank Thomas Blom Hansen for his comments on two chapters and Shailaja Paik, Les Beldo, Rupa Viswanath, Jack Hawley, and Nathaniel Roberts for their incisive, mirthful, and profoundly knowledgeable feedback on the entire manuscript. At Cambridge University Press I thank Qudsiya Ahmad, Anwesha Roy, Aniruddha De, series editor Partha Chatterjee, as well as the anonymous reviewers. One of the greatest pleasures in bringing this book to completion has been collaborating with two young artists in Lucknow, Sanya Darapuri and Nikhil Mayur, whose exquisite pencil work appears periodically in the pages ahead. I thank Sanya and Nikhil for the dedication and care they put into these portraits and sketches. I am grateful to Leo and Boaz for the amused skepticism with which they regard everything I do, this book included. My greatest debt is to Joanna, whose reminders always to tell a story are responsible for any sense of plot or pace in the book ahead. Without her steadfast support, companionship, patience, and good humor, none of this would have been conceivable.

I Untouchability and Alterity, Now and Then

1 Introduction Signs, the Census, and the Sanitation Labor Castes

The census of India is a vast undertaking. Once a decade, every person residing in India—roughly one-sixth of humanity—is to be counted, named, and known. In 2011, I found myself in the midst of this monumental endeavor. The scene was Lucknow, famed for its kabobs and culture of politesse yet also the capital of the north Indian state of Uttar Pradesh (or “UP,” as it is called), known for its rancorous caste and communal politics. I had not anticipated being present for the decennial census—its fifteenth iteration since the inaugural British attempt in 1871–72—but I arrived in Lucknow, by chance, on the second day of its implementation. Though observing such a state exercise had not figured in my research design, the potential value it held for an ethnographic study was undeniable, and within a few days I began accompanying census workers on their rounds. My companions were surveying a Dalit neighborhood along a railway track when I began to sense that foundational premises about caste and religious belonging were misplaced. The words with which the enumerators filled their forms told one story, but the silences and circumlocutions of the enumerated seemed to hint at something else. I wanted to understand Dalit religion. I sought, that is, to learn from those who suffer the structural violence of untouchability how their experience of stigma shapes their sense of religious belonging. My interests lay particularly with the caste cluster that supplies virtually all of South Asia’s sanitation workers. Today the sanitation labor castes are widely regarded as simply and self-evidently Hindu. In swaths of north India, indeed, they have a reputation for displays of Hindu zealotry and support for Hindu majoritarianism. Yet little more than a century ago none of this was the case. The sanitation labor castes were known then for defying, in more ways and to a greater extent than other groups, categorization under the religious taxonomy of the colonial state. Far from appearing as straightforwardly Hindu, they featured in the reports of the decennial censuses as a secretive, “chameleon-like” community

Deceptive Majority 4

whose company Hindus abhorred, a community whose syncretic religious observances generated “a great deal of confusion,” making them “the chief disturbing element” in the mapping of India’s religions (Rose 1902: 113). Here, then, was a riddle: how had a community whose social abjection and religious proclivities made it the paradigmatic confounder of order in colonial times come to be regarded in the postcolonial period as commonsense constituents of an unquestioned majority? How had despised outsiders to the house of Hinduism come to be seen as bricks in its very foundation? However this had transpired, the contours of the change seemed to suggest a more fundamental historical relation between the politics of untouchability and the rise of religious majoritarianism—phenomena ordinarily treated as separate or only glancingly related—than is generally admitted. Perhaps observing the census, where caste and religion appear arm in arm as categories through which the state offers its citizens a kind of recognition, might offer some clues. I was therefore grateful when the census director of UP generously granted me permission to accompany enumerators on their rounds. The census, in one major line of argument, bears responsibility for the reification or calcification of caste and religion as categories of social difference in colonial modernity (Appadurai 1993; Cohn 1987; Dirks 2001; Gottschalk 2013; Kaviraj 1992). Bringing ethnography to bear on this largely historical contention might build upon its insights or reveal its limits. Whereas most accounts of the census consider only the remote guise of the state, as a distant power that determines the categorical schema according to which recognition and other political goods will be distributed, firsthand observation would reveal the state in its proximate guise, as a neighborhood schoolteacher or city employee called in for census duty, bringing local relations of power into play in the generation of official knowledge. It was an opportunity to witness how people talk about caste and religion in those brief, tense conversations between enumerator and enumerated that cumulatively produce such seemingly transparent demographic facts as India’s 79.8 percent Hindu majority. Thus I found myself on a grey February morning going door to door with a pair of census workers, participating in a once-in-ten-years irruption of state officialdom into the weekday routines of a working-class, largely Dalit neighborhood or bastī squeezed between the bungalows and bougainvillea of a posh housing colony and the rubbish-strewn tracks of one of Lucknow’s



secondary rail lines. Shankar,1 a municipal clerk, was the enumerator officially responsible for the bastī, but on account of his failing eyesight he had brought along his son Narayan, a mass communications student, who carried the clipboard and forms and conducted most of the interviews. In a lane of small brick apartments, a middle-aged woman fielded Narayan’s questions from her doorway, giving her family’s surname as Gautam. When she disappeared inside to find out her mother-in-law’s date of birth, Narayan turned to his father. “What does Gautam come under?” “Chamar!” Shankar replied in a loud, somewhat theatrical whisper. “SC!” Narayan wrote “SC” in the appropriate box, identifying the woman and her family as Scheduled Caste, the governmental designation for Dalit or “untouchable” communities. When the woman returned, Narayan skipped columns seven and eight; that is, he asked her about neither caste nor religion, but proceeded to literacy status, disabilities and so on before completing the interview and moving to the next home. Though puzzled, I said nothing at the time. Later in the day, in the privacy of the home of a friend and caste fellow of Shankar’s, the enumerators filled in the blank columns, marking everyone in the Gautam family thus: Caste: SC (Chamar) Religion: Hindu

As the father and son explained to me, when Shankar knew (jānte) a person’s caste, there was no need to ask the caste question, and when the caste fell within the Scheduled Castes, there was no need to ask about religion. This method contravened rules in the government’s instruction manual for census workers, rules that underscore that the enumerator is “bound to record faithfully whatever religion is returned by the respondent for herself/himself and for other members in the household” and that warn specifically against assuming a correlation between caste names and religion (Chandramouli 2011: 44–45). 1

Here and throughout the book names have been changed to protect the confidentiality of my interlocutors. Exceptions are public figures (members of parliament and the UP legislative assembly like Kanhaiyya Lal Balmiki, Narain Din, and Achhe Lal Balmiki in chapter 5, Lucknow’s mayor in the afterword), and two individuals, now deceased, who insisted in their interviews with me that their real names be retained: Govind Prasad and Lalta Prasad (chapter 5).

Deceptive Majority 6

In practice, though, Narayan and Shankar’s policy of inferring the caste and religion of Dalit interviewees was the norm—not only for this pair, but for teams of census enumerators I accompanied on their rounds elsewhere in Lucknow as well as in Benares and Mirzapur. The fact that great numbers of Chamars in UP have converted to Buddhism and that the surname Gautam—a name of the Buddha—is preferred by many Dalits precisely on account of its Buddhist resonances was not a consideration for the enumerators. Each Gautam they encountered was recorded as Hindu, without the question having been asked. So it went at the next home, and the next, and several more after that: each family bore a recognizably Dalit surname, rendering the caste and religion questions, from Shankar and Narayan’s point of view, superfluous. A burst of cold rain sent us running for shelter under the blue tarpaulin awning of a chai stall. When it cleared, we made our way to another cluster of brick apartments, where we found a group of women and men watching children play in the puddles while geese noisily snapped up water nearby. As we approached, one of the elderly women in the group, observing us, called out, ‘Panditji has come [Panditjī ā gaye]!’ Not certain I had heard her correctly—and unaware of Shankar and Narayan’s caste—I discreetly asked Shankar what the woman had said. Continuing to walk toward the group, he replied loudly, “She said, ‘Pandit ji has come!’ Because we are brahmins.” “Brahmins,” his son confirmed. “Brahmins by caste,” Shankar added, this time in English. We were now standing before the elderly woman. Shankar’s words seemed to hang suspended in the air during the long, uneasy silence that ensued, until finally one of the men in the group brought over some red plastic chairs, gestured for us to sit, and began to answer Narayan’s questions. “Surname?” “Balmiki.” Hearing this, Narayan marked dashes under the columns for caste and religion—he would fill them in later as “SC” and “Hindu”—and proceeded to other questions. After finishing with this man’s family, Narayan turned to the next-door neighbor, Rajesh, who had just emerged from a bath and answered questions standing in a towel and tee-shirt. After his family’s form was complete—again with everyone marked “Hindu” though the question had not been posed—another neighbor stepped forward to be interviewed, while Rajesh lingered to observe.



Narayan asked the neighbor, “And what work do you do in the municipality?” The man did not reply. “Sanitation worker [Safāī karamchārī],” said Shankar, speaking for the man and gesturing at his son to fill in the space accordingly. “Wait,” said Rajesh, still standing in his towel and watching the enumerators. “You all never asked me what work I do.” “I put you down as ‘worker’ [karamchārī],” said Narayan. Rajesh explained that he worked as a network technician for a telecommunications company. “It’s not as though all of us are sanitation workers,” he continued. “We also have big positions. We have officers.” “Only in a few houses,” Narayan retorted. “But this is discrimination [Yeh to bhed-bhāv hai]. I’m not a sanitation worker.” “I wrote ‘worker.’ ‘Worker’ is alright.” “‘Worker’ is totally misleading. Even big officers are ‘workers.’ Also,” here Rajesh pointed at the column where Narayan had written surnames, “that should be Valmiki, not Balmiki.” “Yes, yes, I’ll fix it,” Narayan replied with unconcealed irritation. But he changed nothing—neither the spelling of the caste title nor the designation of type of labor. Behind this row of brick apartments ran a dirt lane along which stood a line of jhoṁpṛīs: improvised dwellings of brick, mud, thatch, tin, and plastic. Beyond the jhoṁpṛīs lay the railway tracks. In a home on this lane we were met at the door by a woman in a salwār-qamīz who looked the three of us over and asked, “What’s this about? What’s this for?” Ignoring her, Narayan said, “Head of household?” The woman eyed him coolly and disappeared inside. A silver-bearded man emerged wearing a pink tee-shirt and a lungi perforated here and there by cigarette burns. From his threshold he fielded the enumerators’ questions. He worked as a sweeper in a private hospital; his children took up whatever work they could find, in sanitation or anything else. “Caste?” After a substantial pause, he said, “Balmiki.” Narayan came to the religion column, and this time he chose to ask. “You’re Hindu, aren’t you [Āp Hindu haiṅ, na]?”

Deceptive Majority 8

A long silence ensued. The hospital sweeper idly observed children playing in the lane while Narayan looked to his father and Shankar began to fidget. Finally the man said, “Yes, Hindu.” Shankar, visibly perturbed by the man’s hesitation, pursued the matter. “You’re not, for instance, Lal Begi, are you? Because, you know, there are Lal Begis who are Muslim.” “You mean the Dilliwals,” the man replied. He then delivered a roundabout discourse on the essential interchangeability of the terms Lal Begi, Balmiki, Dilliwal, Panch Sau Tirasi (the number 583), and other names by which his caste is known locally. He neither affirmed nor repudiated the allegation of Muslim-ness. Shankar reiterated his contention that some Lal Begis are Muslim, and again probed whether the man was Lal Begi. His interlocutor said nothing but watched Shankar and Narayan impassively. Eventually, Narayan wrote “Hindu” in the religion column of the form and wrapped up the interview. A few doors down we came to a one-room brick structure before which plastic tarps had been stretched to shelter an open cooking area. Stooped beneath this was a woman in a green sweater, stirring a pot of boiling lentils. She stood up, greeted us, and asked, “What will we get out of this?” “This is the census,” said Narayan. “You people are the future of India!” Shankar added. When Narayan came to the caste question she answered, “Balmiki.” Narayan proceeded to column seven, religion, and again decided to ask. “Your religion is Hindu [Dharm Hindu hī hai]?” “No.” She spoke quietly but distinctly. I was startled by her response but tried not to indicate it. Narayan and Shankar gave no apparent reaction. Nobody spoke. The pot of lentils steamed and bubbled. After an interval, Narayan repeated the question with slightly different wording, “You’re Hindu [Hindu haiṅ]?” “Yes.” Shankar turned to me as though to explain the necessity of the question, “Some people do convert [Kuchh log dharmparivartan karte haiṅ].” What was going on here? The woman offered no explanation for her volteface, delivered in the same steady tone as her initial reply. Equally flummoxing was Narayan’s bald disregard of her initial response, as though such words could not be countenanced. If his father sought to assure me—or himself?— of the normativity of Dalit Hindu-ness by pointing to the rare event of formal



conversion to another religion (dharmparivartan) as its only exception, this effort seemed undercut by his own repeated insinuation that the family at the previous house might be crypto-Muslim Lal Begis. And what to make of the man whose reticence and elliptical speech elicited this charge? Caste titles and religious labels mingled and converged in his periphrastic response to Shankar’s queries, suggesting a mode of belonging at variance with prevailing regimes of distinction, indecipherable in the language of the state. Why was the enumerator so vexed by this man’s studied ambiguity? If his silences were to speak, what would they say?

The Story Line in Brief The book that lies ahead attempts to answer this question. Without giving the plot away entirely, let me sketch its trajectory, indicating in brief some of its key historical and ethnographic arguments. This is a study of the disparate yet deeply entwined histories of religion among the sanitation labor castes and Hindu majoritarianism. One cannot be told without the other: no account of Dalit religion in modernity can afford to ignore the past century of interventions in that domain by Hindu nationalists and the state, as those interventions have produced the very terms in which discussion is now legible. Hindu majoritarianism, for its part, has been driven by the fear of Dalit religious autonomy—a fear partly in response to collective practices of the sanitation labor castes in the colonial period—from its very inception. If the interreligious antagonism known in India as communalism has long been animated by the politics of caste (Basu 1996; Hansen 1999; Menon 2010), some of the most foundational sociological assumptions about caste have been manufactured, largely undetected, by communalism. This book is an effort to make sense of that February morning with the enumerators in Lucknow: why the woman stirring lentils first told Narayan that she was not Hindu, and then, when asked again, that she was. Or why the man in the pink tee-shirt replied so obliquely to the question of religion, or, equally, why his long pauses incited the enumerator to say, “You’re not, for instance, Lal Begi are you?” Attentiveness to contradiction and circumlocution, as well as to non-verbal signs like silence and gesture, may guide us toward insights altogether at odds with the “final word” of authorized discourse. It is one of my arguments that a semiotic approach to the study of caste and religious belonging—an approach attentive to signifying practices, the composition and interpretation of signs by which identitarian affiliations are

Deceptive Majority 10

sustained—makes possible the apprehension of social phenomena that have remained opaque to other analytical traditions. These phenomena challenge established paradigms in the study of religion in South Asia and trouble some of the ethical presuppositions that modernity urges on us regarding secrecy, subterfuge, and self-identification. Contemporary politics in South Asia is predicated on the figure of the primordially Hindu untouchable—a figure that conceptually confines Dalits within the framework of Hinduism, securing for Hindus a demographic majority in the present and a claim to religious and cultural hegemony in the past. In this book I argue that the idea of the transhistorically Hindu untouchable emerged scarcely a century ago, and that it ran athwart the collective selfperception of the sanitation labor castes. Drawing on a range of sources from the 1870s to the 1920s, I contend that the sanitation labor castes of north India during that period widely understood themselves as neither Hindu nor Muslim but as members of a qaum or ummat—a cohesive, autonomous socioreligious community—centered on Lal Beg, an antinomian prophet (paighambar) who moved in a largely Islamicate narrative world. Hindus and Muslims, moreover, acknowledged the religious alterity of the Lal Begis, as they were called. Thus Hindu census enumerators in the colonial period often refused to record the sanitation labor castes as their co-religionists. The colonial administrative decision to classify untouchables as Hindus by default contradicted prevailing sociological common sense. In chapter 2, I analyze evidence from the liturgical songs and other oral traditions of the Lal Begis that speak to Dalit perceptions of self and other in that period. All of this began to change as techniques of colonial governance stimulated a politics of numbers in which castes and religious groups, increasingly assuming the politicized character of “enumerated communities” (Kaviraj 1992), vied to constitute majorities in local, provincial, and pan-Indian representative bodies in the early decades of the twentieth century. These conditions gave rise among some Hindus to the “fear of small numbers” that Arjun Appadurai (2006: 52) names as a signal feature in the emergence of majoritarianism globally. It was in the context of a Hindu fear of small numbers—of being a “dying race” demographically and politically threatened by growing Muslim and Christian numbers—that the Arya Samaj, a Hindu reformist organization, systematically took up efforts at achhutodhhār or “untouchable uplift,” and to persuade Hindus and Dalits to reimagine one another as co-religionists. I will show that the idea that the sanitation labor castes and other Dalits are and always have

11 Introduction

been Hindu—an idea that struck some contemporary observers as offensive and others as absurd—was mooted for the first time in the 1910s and 1920s by the Arya Samaj as a strategy of what we may call majoritarian inclusion, an effort to secure a majority against a potential rival by incorporating a heretofore despised outgroup. In chapter 3, I describe this effort and the skepticism with which it was often met through a reading of key Arya Samaj materials, unearthing in the process the degree to which Arya Samajists wrestled with their own ghṛṇā—a north Indian emotion-concept similar to disgust—as they began working with Dalits, and the ways in which Arya Samaj authors encouraged fellow Hindus to suspend the ghṛṇā they felt toward Dalits and to redirect it, instead, at Muslims. It is in these Arya Samaj texts, as well, that the sanitation labor castes were first provided a Hindu pedigree in the form of a genealogical connection to Rishi Valmiki, author of the Sanskrit epic the Ramayana. It was not until the 1930s, though, that the newly conceived figure of the primordially Hindu untouchable came to appear credible to a larger public. Though the colonial state and the Arya Samaj had laid the groundwork, the political maneuvers and representational interventions that were decisive in giving majoritarian inclusion the mass traction it ultimately achieved were those of Gandhi, the Harijan Sevak Sangh (“Servants of Untouchables Society”), the Indian National Congress, and literary figures inspired by Gandhi such as Rabindranath Tagore and Mulk Raj Anand. Their contributions to the discursive and political confinement of Dalits within Hinduism are the subject of chapter 4. “I know infinitely more than you do what Harijans are,” Gandhi (1934d) said to his “untouchable” critics in 1934, referring to their caste fellows with his preferred nomenclature of Harijan or “people of Hari”—Hari being a Vaishnava Hindu name for god—“[I know] where they live, what their number is and to what condition they have been reduced.” The mahatma’s welding together of an enumerative, panoptic, governmental imagination with a decidedly brahminical social ontology set his approach apart; his monological manner of speaking for largely overrode the Arya Samaj’s dialogical effort to speak to and to persuade. Thus the missionary majoritarianism of the 1910s and 1920s yielded to the trustee majoritarianism of the 1930s and 1940s, culminating in the Constitution (Scheduled Castes) Order of 1950, which declared that “no person who professes a religion different from the Hindu religion shall be deemed to be a member of a Scheduled Caste,” and in doing so, elevated the Gandhian representational strategy to the law of the land, securing postcolonial India’s Hindu majority by fiat.

Deceptive Majority 12

How did Hinduization work out on the ground? In chapter 5 we return to Lucknow for a more fine-grained and local study of majoritarian inclusion and its resistance from the late 1940s onward, based on interviews with many of the individuals directly involved and archival materials their families have preserved. Seizing on the institutional space opened for Dalits by the Arya Samaj and the Congress, a subset of the sanitation labor castes took up—and in the taking up, altered—the majoritarian project. Once the Congress’s strategy achieved the status of law, apparent signs of Hinduization, like the wholesale refashioning of names, swiftly followed. The ancient Sanskrit poet Rishi Valmiki, who had no following among the sanitation labor castes of Lucknow before 1947, was introduced to the community as their ancestor. Valmiki became the sign of a new regime of recognition: a government holiday in honor of Valmiki was instituted, streets and parks were renamed after the rishi, and Congress and Harijan Sevak Sangh leaders like Ghanshyamdas Birla began funding Valmiki statues and temples intended for the sanitation labor castes. More contested within the community was the abandonment or repudiation of Lal Beg that the advocates of Hinduization championed, and corresponding transformations in ritual, in the food, drink, and equipment of nuptial and death ceremonies, and in everyday relations with Muslim neighbors. The degree to which leaders of the newly named Balmiki community succeeded in bringing about their reforms correlated with their capacity to deliver concrete goods of housing, access to education, stable employment, and the curtailment of untouchability practices. Leaders tackled the latter by means of one of the strategies of majoritarian inclusion bequeathed them by the Arya Samaj: pursuing legal action against practitioners of untouchability—so long as the offenders were Muslim, not Hindu. Yet this is not only a tale of people coming to inhabit the categorical niches allotted them by the postcolonial state. Part of what I am tracking is a process of this sort—what Ian Hacking (1985) calls “dynamic nominalism” or simply “making up people.” But there is more than this to the history and present of the religious life of the sanitation labor castes, and we will need to turn from historical to ethnographic methods to arrive at other key findings of this book. Having witnessed the decline of Lal Beg and the ascendance of Valmiki over the twentieth century, we turn in chapters 6 and 7 to the practices by which the old caste prophet is remembered and the new caste god is celebrated in Lucknow today. Here we analyze ways in which normative ideas of appropriate modes of signifying and relating to the sacred—semiotic ideologies—structure religious

13 Introduction

practice and self-representation, and ways in which signifying practices render certain social realities hypervisible (and hyperaudible) while making others, equally real, invisible or concealed in plain sight. Through a description and analysis of processions (jhāṅkiyāṅ) and speech-making functions (kāryakram) on Valmiki Jayanti—the annual celebration of Valmiki’s advent in the world— we examine how a declamatory mode of identitarian self-disclosure has come to characterize Dalit religious practice, and perhaps religious ways of being in South Asia more generally, and we obtain a feel for the texture and the limits of the inclusion the sanitation labor castes have been offered as Hindus. Of all the “weapons of the weak” identified in James Scott’s (1987) influential formulation—the “foot dragging, dissimulation, desertion, false compliance, feigned ignorance, slander, arson, sabotage,” and other everyday tactics by which disprivileged groups “avoid direct, symbolic confrontation with authority” even while securing a measure of relief from structural violence—surely the most difficult to study is dissimulation. While all of these “infrapolitical” (Scott 2012) techniques keep a low profile in the historical record, the latter not only does not announce itself as a form of politics, it actively conceals its own tracks. Methodological and epistemological questions fly thick here: when the very definition of successful dissimulation is invisibility, and unsuccessful efforts are necessarily disavowable and disavowed, then on what basis can an enquirer make anything more than a speculative claim about the practice? Chapter 7 contends with these and related questions while giving an account of gestures, ways of signifying the sacred, and what happens now on the day that colonial accounts described as the annual feast of Lal Beg. This chapter contains developments that are better not summarized in advance. My interlocutors taught me that valuable knowledge is not disclosed quickly, in the first week or even the first year of a relationship, but only after a certain thickness of context and commitment to understanding is established. I have structured the book accordingly. For now, suffice it to say that the continuing vitality of traditions of tactical concealment may lead us to reconsider what Hinduization may have meant all along.

The Sanitation Labor Castes To speak of caste is to conjure a babel of discourses all at once. The English word routinely translates two distinct concepts present in Sanskrit and the vernacular languages—varṇa, the four ranked classes or “colors” of brahman, kshatriya, vaishya, and shudra that together constitute the social organism

Deceptive Majority 14

in conceptualizations of society from the Purusha Sukta of the Rig Veda onward, and jāti, meaning species (cognate, in fact, with the Latin “genus”), the multitudinous endogamous hereditary groups that regulate reproduction and set the terms of interaction in actual social life. At the same time caste also conveys the early modern European ideology of blood purity encoded in its own Iberian etymology, as reflected in the application of casta (related to “chaste”) to social groups in New Spain as well as in Portuguese Goa. British imperialism decisively influenced the history of European representations of Indian society, fashioning caste into a trope, a sign of India’s difference from the West, a ready justification of the “rule of colonial difference” (Chatterjee 1993: 18) that has cast a long shadow over popular and scholarly discussions of the social form (Appadurai 1988; Dirks 2001). Yet the critical historicization of the European trope of caste does not preclude analysis of the social form— caste qua jāti—that, in places like Lucknow, continues to exert profound and far-reaching influence over collective life, from sex and diet to political representation and waste management. Though related in complicated ways, the two are distinct intellectual projects, oriented, as it were, toward different “castes.” South Asian history over the longue durée renders a view of caste as a “highly involuted, politicized form of ethnic ranking shaped by the constant exercise of socio-economic power” (S. Guha 2013: 2). An “adaptive structure” (Lynch 1969: 3), it continues to ensure inherited advantage for some and disadvantage for others through interpersonal and institutional networks that have made accommodations with, rather than fallen victim to, such consequential political, economic and technological changes of the last century as the universal franchise, the democratization of education, the integration into a cash economy of agrarian systems of labor exchange, and the mechanization of a host of traditionally caste-based crafts and forms of labor (Natrajan 2012; Subramanian 2019). Organizing social perception and inscribing meaning onto bodies so perceived, caste, in a more intimate register, is both a “state of mind” (Dumont 1980: 34) and “a form of embodiment” (Rao 2003: 5). Like its equally insidious cousin race, caste works its way simultaneously into large-scale institutional systems and the interstices of our bodies and minds. Among the most apposite characterizations of caste for our purposes is that of Bhimrao Ambedkar, India’s first law minister and a towering figure in anticaste theory and praxis, when he describes it as an “ascending scale of reverence



and a descending scale of contempt” (Ambedkar 1990: 26). This framing draws attention to affect, suggesting ways that caste shapes the inculcation of emotion norms and the cultivation of distinctive emotional repertoires according to social location (Guru 2009b; Lee 2020; Lynch 1990). The affective structure of caste will play a significant role in our story of majoritarian inclusion. Ambedkar’s formulation is felicitous as well for its foregrounding of hierarchy, that hoary element of caste analysis that innumerable popular and scholarly accounts, partly in response to its perceived overemphasis in the work of Louis Dumont, have sought to consign to the dustbin of social theory, yet which refuses to cede ground in the empirical domain of quotidian social relations. While there is a desire in several quarters for the concerns of prestige, purity– pollution, and inherent quality-substance in a ranked system to be seen as located in the past, it would be disingenuous to downplay their force in the caste-segregated bastīs of Lucknow at the center of this study, where hierarchy is inscribed in the organization of urban space and the sensory matter that circulates within it (Lee 2017), in the dispensation of resources, in the ubiquity of the vertical metaphor in references to caste status in everyday speech, and in the distribution of waste and death labor. The nether terminus of Ambedkar’s “ascending scale of reverence and descending scale of contempt”—its lowermost extremity, subject to demonstrations of intense affect peculiar to such a location on such a gradation—is inhabited, according to broad consensus, by the sanitation labor castes. Novelist Madan Dikshit (1996: 40) cites a Hindi proverb: there is “no insult worse than whore, no caste lower than Mehtar.” This adunation of sexual and caste hierarchies indexes the constitutive significance for the preservation of caste order of control over women’s sexuality; caste is nothing if not a patriarchal endeavor (Chakravarti 2019). Further, the adage points to a parallelism between subordinated threats to the order of endogamy: the whore whose womb cannot be harnessed to the project of caste reproduction (Ramberg 2014: 146–77), and the Mehtar—a collective term for the sanitation labor castes— long known for accepting into its ranks persons and groups excommunicated from “higher” social locations, frequently for having transgressed prohibitions against sex or marriage outside caste (Shyamlal 1997). As a destination for downward mobility, as receivers of and minglers with refractory elements, the Mehtar community, like the proverbial whore, embodies a principle of mixing that is antithetical to the principle of discernment and separation essential to caste endogamy and prized in brahminical thought more broadly.

Deceptive Majority 16

Mehtar, a Persian word that in the Mughal period denoted prince and now designates sanitation worker, is but one of the welter of names by which the protagonists of this book are known. We have seen already that Lal Begi and Valmiki/Balmiki designate this group as well, and that these terms seem haunted by a past that troubles both the namers and the named. Caste nomenclature is a vexed domain, a field of contestation in which every name bears the affective charge of a history of usage as a term of awe and deference, scorn and revulsion, or gradations of esteem in between. Reflecting on the struggles of Dalits to wrest control of the labels by which they are known on this uneven field, Ambedkar observed: The name “Untouchable” is a bad name. It repels, forbids, and stinks. The social attitude of the Hindu towards the Untouchable is determined by the very name… [T]he Bhangis call themselves Balmikis…. [T]hey give themselves other names which may be likened to the process of undergoing protective discolouration….The name matters and matters a great deal. For, the name can make a revolution in the status of the Untouchables. But the name must be the name of a community outside Hinduism and beyond its power of spoliation and degradation. (Ambedkar 1992: 419–20)

Ambedkar’s portrayal of a grappling over names invested with ideological content and charged with affective force—the name “repels, forbids,” but can also “make a revolution”—assumes that words alter reality rather than merely describing it, and that naming constitutes a tactical maneuver in an agonistic social field more than an exercise in disinterested classification. Ambedkar’s implicit theory of signs, that is, converges with those of mathematicianphilosopher Charles Sanders Peirce (1955: 269–89), for whom signs are assessed not in terms of correspondence to pre-given meanings but in terms of their consequences in the world, and linguistic philosopher Valentin Vološinov (1973: 23), for whom the “sign is the arena of class struggle.” The analysis of the politics of naming, knowledge, concealment and recognition that runs throughout this book is well served by such a framework. It helps account for the striking instability of nomenclature over time at the “lower” reaches of the caste order. When different classes (castes, in our case) belong to the same sign community (that is, speak the same language), the divergent ideological assessments they make of a given phenomenon come to intersect in the sign that represents it. Shot through with these various evaluative “accents,” the sign becomes what Vološinov calls “multiaccentual”: the casteist contempt with



which deployments of the term “Bhangi” are conventionally associated jostles alongside the paternalist sympathy of the Gandhian, the empirical precision of the social scientist, the hate speech concerns of the human rights lawyer and the confrontational reappropriation of the Dalit radical, each vying with the rest for the successful interpretive claim, for efficacy in a given communicative moment and in the habits of interpretation that are built of such moments. The multiaccentuality of the sign is not mere polysemy, where everyone is equally free to select an interpretation or have one’s interpretation validated. Rather, it is ineluctably structured “by the hierarchical organization of communication” wherein socially dominant groups enjoy greater representational agency, especially in the domain of the written word, where they strive “to impart a supraclass, eternal character” to the sign, to make the sign “uniaccentual” (Kockelman 2007; Vološinov 1973: 21, 23). Efforts of this sort help account for the devaluation, over generations, of several of the collective labels by which the sanitation labor castes have been known: Mehtar (“prince”), Jamadar (“head of any body of men,” usually an army or police officer) and Halalkhor (“one who earns an honest living”). These relatively dignified terms adopted in the Mughal and colonial period have all suffered semantic pejoration (Hill 2008: 134–37) over recent centuries of association with sanitation workers (Lee 2018: 10–13). The fashioning of even staid governmental terms like Scheduled Caste into terms of disparagement like “Schaddu” on university campuses (Guru 2009b: 18) bears witness to the ongoing “power of spoliation and degradation” that advantaged castes exercise over caste names in Ambedkar’s diagnosis. Nonetheless, the struggle continues, and if the word is “the most sensitive index of social changes” (Vološinov 1973: 19), then the careers of caste titles—like Lal Begi and Balmiki, whose shifting fortunes this book tracks—may be among our most valuable guides to the past and present of contemporary India. Acknowledging that there are no neutral terms—no caste titles innocent of ideology or untinctured by hierarchical affect—let me clarify those that I provisionally employ. “Sanitation labor castes” is a translation of safāī kāmgār jātiyāṅ (safāī: cleaning, cleanliness, sanitation; kāmgār: worker, laborer; jātiyāṅ: castes), a collective self-designation in Hindi–Urdu that many though by no means all of my interlocutors use, as do a number of non-governmental organizations (NGOs) whose members belong to these castes and writers and journalists of a sympathetic bent. By this term I mean to denote those Dalit castes that perform the great majority of sanitation work in contemporary

Deceptive Majority 18

South Asia. In north India this means the Chuhra (and its Sikh branch the Mazhabi), Dhanuk, Hela, Dom (also Domar), Bansphor (also Basor), and Halalkhor, regional castes affiliated by occupation, overlapping traditions, and myths of shared origin but kept clearly distinct by endogamy (and, in the Halalkhor case, openly Muslim identification). “We are like the bristles of a broom,” my friend Daulat Ram once told me. Like the emblematic tool of their trade, he explained, the sanitation labor castes are tightly bundled at one end, dispersed at the other. In the ethnographic literature this caste cluster has been described as the “Dom group” (Briggs 1953; R. Guha 2010) or simply as “the sweeper castes” or “the scavenger castes” (see, for example, Gait 1902; Ibbetson 1970; Searle-Chatterjee 1981; Temple 1884). In the north Indian vernacular the more popular collective designations are Mehtar and Jamadar, described above, and the much-contested Bhangi. Likely derived from the Sanskrit root bhañj (to break, shatter, split, defeat), this word’s etymology has been speculatively plotted into tales of defeat (the “broken ones”), occupational specialization (basket and broom makers, “splitters” of bamboo—precisely the meaning of Bansphor), or aboriginal revolt (“breakers” of invading Aryan armies) (Das 2007; Kumar 2004). See Table 1.1. My main reservation in using the term “sanitation labor castes” is that it risks reifying the contingent link between a people and an occupation. Table 1.1 North Indian Sanitation Labor Castes Individual castes (jātiyāṇ)

Collective designations (in order of historical appearance)

Titles referring to a caste prophet or tutelary saint (primarily associated with the Chuhra yet sometimes applied collectively)

Bansphor/Basor Chuhra Mazhabi (Sikh Chuhra) Dhanuk Dom/Domar Halalkhor* Hela

Halalkhor* Bhangi Mehtar Jamadar

Lal Begi Valmiki/Balmiki

Note: *Halalkhor from the late sixteenth century—when the Mughal emperor Akbar popularized it—to the early twentieth century collectively designated the sanitation labor castes as a whole; by the late twentieth century the term’s reference narrowed to denote primarily a regional Dalit Muslim caste preponderant in eastern UP and Bihar (see Lee 2017).



Ramnarayan Rawat (2011) cautions against identifying subaltern caste groups primarily by reference to stigmatized “traditional occupations” for empirical and ethical reasons: insofar as only a minority of the group may actually engage in such forms of labor, the identification can mislead; more troublingly, it can reproduce brahminical stereotypes. It is indeed the case that some of the sanitation labor castes entered this domain of work relatively recently; Dhanuks, for instance, appear to have taken up sweeping and scavenging between the 1870s and 1940s (Ibbetson 1970: 266–96, Searle-Chatterjee 1981: 26–30). The most populous of the sanitation labor castes, though—Chuhras and Doms—have been associated with these forms of work for far longer, as attested in their own oral traditions and in Al Biruni’s eleventh century chronicle Tārikh al-Hind, which describes Doms removing filth from villages (Bīrūnī 1983: 85). The expansion of demand for sanitation labor accompanying the unprecedented urbanization of the twentieth century, alongside the withering of these castes’ other “traditional occupations” due to mass production (of sieves, baskets, and chemical fertilizers), has led to urban waste work becoming by far the largest sector in which these castes are employed. In over a hundred interviews in the Balmiki bastīs of Lucknow I came to know of only two families that had no living member employed in sanitation; all other families had at least one and often several or even all adult members working with waste. This is not to gainsay the occupational diversity that obtains among an upwardly mobile minority of the community that lives outside the bastīs. Even they, however, in their dealings with people outside the community, are forced to contend with the stigma of their caste’s association with waste work. In the same spirit that my interlocutors employ it, then, I use the term “sanitation labor castes” not to naturalize the historically produced association between waste work and a cluster of endogamous communities but rather in order to acknowledge the degree to which this domain of labor impinges on the community life of this caste cluster and on the lives and life chances of its members. As the Hindi proverb cited earlier suggests, the ascription of abjectness to the sanitation labor castes obtains not only vis-à-vis the privileged castes but also relative to other Dalit caste formations. In north India, notably, the sanitation labor castes are seen as “lower” than the Chamar-Jatav cluster. Notes of resentment, admiration and envy of this more populous and politically organized caste cluster are not infrequently sounded in the meetings of sanitation labor caste associations. As with the numerically weaker rivals of politically

Deceptive Majority 20

assertive Dalit caste clusters in other regions of India (for example, Mangs, Madigas and Arunthatiyars vis-à-vis Mahars, Malas and Paraiyars, respectively), this resentment has often been exploited by political parties seeking to blunt the force of the critique of Hinduism and caste dominance that the regionally “dominant” Dalit castes, following Ambedkar, have tended to mount. This state of affairs—both the resentment and its exploitation—helps account for the fact that caste-critical social and political movements inaugurated by Chamars in north India have had difficulty retaining significant participation by the sanitation labor castes (Juergensmeyer 2009: 62–63, Rawat 2011: 155, 159; Shyamlal 2016). While there are significant exceptions, the sanitation labor castes on the whole have remained Ambedkar se vimukh or “turned away from Ambedkar,” to use Darshan Ratan Ravan’s (2010) apt phrase. How large is this community? Scheduled Castes as a whole constitute 16.6 percent of India’s population, but census data on individual castes has not been published since the late colonial period. In the 1901 and 1911 censuses the sanitation labor castes of north India amounted to some four million persons and made up 6.3 percent of the overall population of Punjab and 1.9 percent of that of UP (Risley 1902; Gait 1913b).2 If these proportions have remained stable, the sanitation labor castes would now number something in the vicinity of 1.7 million in Punjab and 3.8 million in UP. The fieldwork that informs this book includes interviews with members of most of the north Indian sanitation labor castes—Chuhra, Dhanuk, Dom, Halalkhor and Hela, though not Bansphor or Mazhabi—conducted in Delhi, Bihar (Sasaram and Patna), Haryana (Panipat), and UP (Azamgarh, Bara Banki, Benares, Bhadohi, Faizabad, Gorakhpur, Lucknow, Mau, Meerut, Mirzapur, Pratapgarh, Sitapur, and Tanda). Ultimately, though, this study is based in Lucknow, and grounded foremost in the community life of the regional clan of the Chuhra caste that lives there and supplies half of the city’s sanitation workers (the other half are Dhanuk). Of the sanitation labor castes by far the most populous and transregional are the Chuhra. They figure prominently 2

The sum of the all-India population totals for Bansphor, Bhangi, Chuhra, Dhanuk, Dom, Halalkhor, Lal Begi, Mazhabi, and Mehtar—where they are tallied separately and not as subsets of one another—was 4,142,224 in 1901 and 3,928,504 in 1911. In the United Provinces—the colonial UP, predecessor, with modifications, of postcolonial Uttar Pradesh—their percentage of the population was 1.86 in 1901 and 1.88 in 1911; in Punjab 6.33 in 1901 and 6.29 in 1911. These figures exclude Chuhra converts to Christianity, likely in the hundreds of thousands by 1911 (see chapter 2).



in both the rural and urban population of a region extending from central Pakistan to central UP, concentrated especially in Punjab. Awadh, the region surrounding Lucknow that is the setting of this study, is the easternmost swath of this Chuhra “heartland.” Sections of the caste migrated to cities and cantonments across the subcontinent following the expansion of railway and military networks in the colonial period and the demand for labor they produced. Settled in cities from Karachi to Calcutta to Pune and Hyderabad in the Deccan, Hindi- and Punjabi-speaking Chuhra communities often nurture links of marriage and communication stretching back to the heartland (Campbell 1885; Streefland 1979). In Chuhra oral traditions transcribed in the nineteenth century, the community remembers itself as being none other than the Chandal (chānḍāla), the brahminical discursive tradition’s paradigmatic “untouchable” and archetype of alterity, a figure loathed and feared in two millennia of Sanskrit texts (Aktor 2010; Jha 1986). It was overwhelmingly among the Chuhra caste in the colonial period that the “cult of Lal Beg” thrived and the title Lal Begi was adopted, though there is evidence that Dhanuks and to some extent other sanitation labor castes participated as well. Likewise in the twentieth century it was (and remains) primarily among the Chuhra that the veneration of Valmiki and the adoption of that title obtains, though adjacent castes in the cluster are often assumed to be Valmiki/Balmiki by outsiders and make strategic use of the identification themselves. In Awadh, for reasons absent from the archive and lost to collective memory, the Chuhra caste is constituted by clans (qabīle) known not by names but by numbers: the Hazara (Thousand) of Mahmudabad, the Bara Ghar (Twelve Houses) of Sitapur, the Baisi (Twenty-Two) and the Nau Sau Nawasi (Nine Hundred Eighty-Nine). Most of the people we will meet in this book belong to the Panch Sau Tirasi (Five Hundred Eighty-Three, hereafter 583), easily the largest of the qabīle. The traditional territory of the 583 stretches between the Gomti and the Ghaghra rivers from Sitapur in the northwest to Faizabad in the southeast: the better part of three administrative districts (see Figure 1.1). At the geographical center of this territory stands the Sufi shrine of Dewa Sharif, where the 583 traditionally convene their panchayat or caste council. The clan is organized into exogamous patrilineages, each of which must be prepared to supply men to perform prescribed functions at panchayat meetings: the chaudhury adjudicates, the nāyab deputizes, the pyāda serves summons, the khūnī pyāda enforces summons, and so on. The administrative, even

Deceptive Majority 22

Figure 1.1 Awadh, with traditional area of settlement of the 583 and neighboring qabīle (clans) Source: Map by author.

quasi-military paradigm for the organization of these kinship units sets them apart from gotras (exogamous patrilineages common among north Indian Hindus) and corresponds closely with principles of sanitation labor caste organization noted in the colonial period (Burn 1902: 233; Greeven 1894: 9–14). The 583 trace the origins of this institution to Jumma Mehtarani, a shrewd and sagacious (dimāghdār, chatur) ancestress whose strategic dealings with her royal employers won the caste concrete benefits in land, patronage, and protection. Some say she came from Iran or accompanied Babur’s Mughal army; others place her centuries later, as an accomplice of Begum Hazrat Mahal, the queen of Awadh known for sponsoring native insurgents against the British in the great rebellion of 1857. Unlike the Chuhra caste elsewhere, the 583 do not raise or sell swine, but do rear and trade horses and mules, notably at the annual horse-trading festival at Dewa Sharif. The 583 also distinguish themselves by observing a strict prohibition on the consumption of pig flesh, whereas in many parts of north India pork is central to Chuhra ritual and social life. Finally, the 583 enjoy a reputation for affectations of the lifestyle of Awadh’s erstwhile Muslim elite: a fondness for biryani and other rich and meaty foods, a predilection for hosting



lavish feasts, and a tradition of women bequeathing their daughters beautifully wrought brass pāndāns and consuming much pān, a mild stimulant of tobacco, areca nut and slaked lime wrapped in betel leaf. The incongruity of all this with the abject picture of the lives of Dalit sanitation workers that circulates in the wider society is not lost on the 583, who often make their allegedly nawabi tastes an object of banter. I was introduced to the 583 by Tara Balmiki, a Lucknow-based activist for whose Dalit feminist organization I worked as a volunteer (preparing grant proposals, doing Hindi–English translation and research assistance) in 2003 and later as a collaborator in a four-state advocacy study on caste and gender violence. Her introduction opened doors and conversations. Living in Lucknow in 2011–12 with my family, and then alone in shorter stints spread over the subsequent eight years, I was received in the community with warmth and reciprocal curiosity. That I sought to research and write a book on the sanitation labor castes and their religious history was broadly accepted, by some even encouraged; I was graciously invited to weddings, engagements, panchayat meetings, labor union functions, all-night jāgarans and qawwālī sessions, goat sacrifices to local goddesses, pilgrimages to Sufi shrines, journeys to the hospice of a guru in distant Punjab. My new acquaintances enlisted me as a teacher of conversational English in an afterschool program for youth in one of the bastīs. As relations deepened, friends brought me as well to burials, feasts in honor of the dead, and rituals of a more clandestine kind. My efforts to learn by collecting genealogies and family histories; accompanying sweepers, drain cleaners, carcass collectors and sanitary supervisors on their daily rounds; cross-checking findings from the UP state archives and Amir-ud Daula public library with the narratives and personal collections of community elders; and conversing endlessly on matters of religion and community life were met, on the whole, with forbearance. Yet a sense of the weight of the collective experience being shared—and the expectation that the book I would write would convey the gravity of this history—wove through the tolerant good humor.

Harijan Politics For good reasons, the historical study of Dalit politics and religion has often centered on the Ambedkarite movement and its homologues in other regions, movements defined by bold critiques of caste and Hinduism, an emancipatory discourse and confrontational style, and religious conversion or the assertion

Deceptive Majority 24

of autochthonous religious identities such as Adi-Dravida and Ad Dharmi (see, among others, Aloysius 1998; Jaffrelot 2005; Juergensmeyer 2009; Lynch  1969; Paik 2014; Rao 2009; Rawat 2011; Rege 2006; Zelliot 2010). These were usually led by the regionally most populous Dalit caste cluster, marginally “higher” than the sanitation labor castes in local reckonings: Mahars in Maharashtra, Namashudras in Bengal, Chamars/Jatavs in north India, and so on. Comparatively less is known of the religious politics of somewhat smaller caste clusters, like the sanitation labor castes, that the Congress sought to cultivate as a counterweight to Ambedkar and his allies (although see Dube 1998: ch. 6; Jangam 2017: ch. 5; Jaoul 2011; Prashad 2000: chs. 5–6). In significant ways, our story charts this less studied pattern—a trajectory of politics not of emancipation but of upliftment, not of Ambedkarite confrontation but of Gandhian conciliation, not Dalit politics but Harijan politics. When non-Ambedkarite Dalit engagements with Hinduism are taken into account, the framing question is often this: why do particular Dalit and other disprivileged caste formations, under particular conditions, embrace or appear to embrace a politics that valorizes brahminical ideology, when the ontological inferiority of Dalits is an axiom of that very ideology (Bandyopadhyay 1997; Basu 1996; Gooptu 2001; Hansen 1999; Jangam 2017; Menon 2010; Narayan 2009; Prashad 2000; Rawat 2011)? Vijay Prashad (2000: ix–xi) throws this question into sharp relief by citing allegations of Balmiki participation in the anti-Sikh pogrom in Delhi in 1984 and in subsequent smaller-scale skirmishes with Muslims in Delhi and parts of UP, a concern reignited after the 2002 pogrom in Gujarat, an instance of orchestrated violence against Muslims in which Dalits were reported to participate. There are reasons to exercise caution here: the extent of Balmiki involvement in these incidents has been questioned on evidentiary grounds (Kolenda 2003: 432–53), and it is important not to collapse distinctions between acquiescence to Hindu majoritarian discourse by Balmikis (itself a matter of degree), and active engagement in anti-minority violence. But there is no denying that some Balmikis in some regions have earned a reputation for a full-throated, even militant embrace of Hindu majoritarianism. I began fieldwork expecting to encounter this “counter-intuitive alliance between activists dalits … and the forces of Hindutva” (Prashad 2000: x). What I found, however, has led me to invert the question, important as it is. When one considers the geographical breadth and historical depth of efforts by



hardline Hindu nationalists to enlist the sanitation labor castes in mobilizations against Muslims, what is perhaps more astonishing than their success in a handful of cases is their failure everywhere else. Unlike in the examples Prashad cites, in Lucknow and its hinterland—despite the region being the epicenter of Hindu nationalist mobilizations around the Babri Masjid dispute, and despite militant Hindus’ wooing of Dalits in particular—the sanitation labor castes have repeatedly declined the invitation to majoritarian violence. What is often missed in studies of communal conflict, as Laura Ring (2006) and Bhrigupati Singh (2015) point out, is that everyday amity between groups is as much a consequence of sustained social labor as is violence, and warrants equal analytical attention. Let us ask, then, how it is that the sanitation labor castes in Awadh have not become footsoldiers of Hindutva despite a century of attempts to make them so. What tactics and traditions have facilitated the reproduction of Dalit neighborliness with Muslims when political forces have encouraged its opposite? To understand how the sanitation labor castes in Awadh (and indeed most of India) have resisted elements of the majoritarian project, even while embracing or appearing to embrace others, is at least as important as to explain why some of their caste fellows in other regions have assumed a militant posture.

The Politics of Pahchān It had been a long day of census work. Delayed by episodes of rain rare for a Lucknow winter, Shankar and his son Narayan, with me in tow, had surveyed some twenty-six homes over the course of the morning and afternoon when we arrived at one of the bastī’s provision shops. Its owner Joshi, a friend and caste-fellow of Shankar, greeted us effusively and insisted on taking us home for cup of chai. A brick apartment upstairs from two Balmiki families, Joshi’s was the only brahmin household—and one of very few non-Dalit families—in the bastī. Joshi introduced us to his wife, asked her to prepare tea, and sat us on couches around a table. It was here, out of earshot of neighbors and passersby, that Shankar and Narayan explained to me their method of filling columns seven and eight of the census form by inferring caste from surname, and religion from caste—most of the time, at least. Drawing our host into the conversation, Narayan elaborated that since he knows that Joshis are brahmins and Hindus, he need not ask the shopkeeper’s caste or religion. But what of surnames that have been adopted by more than one caste? The question brought smiles to the faces of Shankar, Narayan, and the Joshis. They named several “Backward

Deceptive Majority 26

Castes” and Scheduled Castes whose members sometimes adopt brahmin and other high-sounding surnames. Such affectations, Shankar assured me, do not escape his notice or lead him to misidentify “backwards” as brahmins. “I catch them [Maiṅ pakaṛ letā huṅ],” said Shankar. Narayan added, “Papa knows [Pāpā jānte haiṅ].” “I have lived here many years,” his father continued, evidently savoring the discussion. “I catch them.” What is it, exactly, that Shankar knows? This book takes as one of its central concerns the politics of knowledge, particularly sociological knowledge instrumental to the operations of the postcolonial state and foundational to scholarly representation. What Shankar knows becomes, through its inscription on the census form and entry into databases of the Ministry for Home Affairs, demographic facts: facts on the basis of which state policy is created, debated, and implemented; facts that determine the allocation of government resources; facts published and republished in textbooks, sociological studies, academic monographs, newspapers, websites, government statements, NGO reports, and a host of other nodes of official discourse; facts that hold aloft the canopy of popular and scholarly common sense that shelters and makes possible the public sphere. Yet the knowledge claims advanced with such certitude by the enumerators at their friend’s home had worn a more anxious aspect earlier in the day. As we have seen, the religious coordinates of the residents of the bastī— their invariable “identities” as Hindus—were obtained over a chorus of silences, ellipses, reversals, questions not asked and questions answered in the ambiguous affirmative. Behind the assuredness of “Papa knows” was the doubt in Shankar’s voice, the perturbation in his manner as he questioned whether the silver-bearded man might be Lal Begi. The enumerators clearly had the last word in determining demographic facts. But in the conversations in the gullies, was Shankar upholding sociological order by “catching” would-be caste imposters, or, in a web of multilayered signs composed so as to simultaneously affirm and undermine the state’s regime of recognition, was it he who was caught? To do justice to our story, we need fit concepts. “Identity politics” is the frame thrust upon collective Dalit action in all too many popular and scholarly accounts. Identity politics implies inferior politics; deployments of the term usually judge the action so described either insufficiently universal (because grounded in the experience of a particular social group rather than an abstract

27 Introduction

collective like “the poor”) or insufficiently radical (because allegedly oriented toward symbolic rather than substantive demands), or both (for example, Gudavarthi 2019; D. Gupta 2005). “Identity” itself, moreover, cannot escape its conceptual mooring in “absolute or essential sameness” (OED Online 2020b) and the homogenizing and dehistoricizing tendencies that follow from it (Brubaker and Cooper 2000). Theoretically more promising is the politics of recognition (Hegel 2018; Povinelli 2002), but its application in the caste context has tended to reduce Dalit politics to an effect of governmentality and to concede too readily the success of the totalizing ambitions of the modern state (for example, Chakrabarty 2002: 84–95; Dirks 2001: 255–96; cf. Certeau 2011: xv, 48). Better analytical purchase may be had from concepts closer to our context. Pahchān is such a concept. This unassuming, everyday word from the north Indian vernacular does what the anglophone social science concepts it partially resembles fail to do; it condenses in a single term the intersubjective action of the identity–recognition dialectic and foregrounds the semiotic terrain on which that action transpires. How so? In official contexts pahchān does what “identity” or “identification” does for the modern state: a police officer may ask for your pahchān patra, identity card, and newspapers publish photographs of unidentified corpses with appeals for their pahchān. But the identification thus secured permits no conceptual traffic with ideas of interior selfhood or enduring self-sameness, as “identity” does; rather, pahchān derives its primary sense from the transitive verb pahchānnā: to recognize, discern, distinguish. This conjures a scene of dynamic intersubjectivity in which it is not the self but the other—the discerner, the conferrer of recognition— who holds the relative advantage in determining a given subject’s status. Well and good; pahchān thus far distills in a word the insights of many a treatise on subject formation, lexically highlighting the asymmetrical transitivity of the encounters through which “identities” are produced, supplying a pragmatic Hindi-Urdu substitute for and improvement upon the English “identity.” Our vernacular concept, however, has two further advantages. First, pahchān has a long history of entwinement with caste. In literature and in everyday speech the two often make joint appearances—jāti as the object of the verb pahchānnā—and have done so since at least the early modern period. In the corpus of the Punjabi poet Bullhe Shah (c. 1680–1758), for instance, we find this oft-repeated couplet:

Deceptive Majority 28

Chal Bullhe chal ūthe chalie jitthe sāre annhe Na koī sāḍī zāt pahchāne na koī sānūṅ manne3 Let’s go, Bullhe, let’s up and go where everyone is blind There no one will recognize [identify, discern] our caste, no one will measure our status (N. Ahmad 2008: 18, translation adapted from Puri and Shangari 1986: 457)

Yearning for liberation from the oppressive social conditions of zāt (that is, jāti, caste), the Sufi poet reveals the extent to which, in South Asia, social recognition is caste recognition—the degree to which, in this milieu, to identify a person is to discern his or her caste. Centuries of development in this social context have shaped pahchān into a more sensitive instrument for taking the measure of caste than concepts calibrated to European and Christian historical experience. Second, pahchān puts the identity–recognition dialectic where, arguably, it belongs: on semiotic conceptual ground. Consider a nineteenth century dictionary entry for the word: “knowledge, acquaintance, ascertainment, recognition, experience, discrimination, discernment; distinguishing mark, characteristic; indication, token, sign” (Platts 2004 [1884]: 284). Indeed, pahchān derives from the Sanskrit root chihn, meaning “sign” (MonierWilliams 2004: 399). An example of pahchān in context may help clarify both points. Omprakash Valmiki (1950–2013), whose autobiography, fiction, and writings on aesthetic theory have decisively shaped the field of Dalit literature, writes in his short story “Andhaṛ” (Sandstorm) of a young Dalit man who works in his father’s piggery. The protagonist rises before dawn each morning and performs the labor of killing pigs, burning off their hair, and cleaning and butchering their carcasses—work that leaves an olfactory trace on the body of the worker. The story explains: The moment he got free, he would bathe with a very thorough scrub down. Despite this the smell of pig flesh did not leave his body. This smell became his pahchān. In school nobody wanted to sit next to him. (Valmiki 2010: 88)4 3 4

I am grateful to Sudipta Kaviraj for directing me to this verse. Here and throughout the book, unless otherwise noted, all translations from Hindi and Urdu are my own.

29 Introduction

A smell—manifestly external in its origin and indifferent to the protagonist’s view of himself—becomes his pahchān. The relationship of odors and other sensuous signs to the caste order will make itself felt periodically in this book; it is a topic I take up more systematically elsewhere (Lee 2017). For our present purposes, what this passage makes clear is that pahchān denotes something very different from the deeply held self-descriptions or enduring intrinsic properties of individuals or subaltern collectives, or from anything like authenticity. The protagonist does not discover pahchān within himself; nor does he alone decide what his pahchān will be. On the contrary, it is his classmates who determine the protagonist’s pahchān, his soapy efforts at suppression notwithstanding. Collectively they know and identify him by an olfactory sign—a faint whiff of pig flesh—which, in brahminical social ontology, can only mean that he belongs to an “untouchable” community, as only those communities rear and butcher swine. Pahchān, then, is a “distinguishing mark,” a sign by which a person or group is recognized by others. Signs mediate between the objects (ideas, statuses, people) they represent and the persons who interpret them. When signs are fashioned by humans (as distinguished from natural signs: the sway of a tree branch indicating the direction and force of wind, smoke indexing fire), the arbitrariness of social convention swings into play, affording the sign-composer room for maneuver, making possible artistry in pictorial, rhetorical, and other modes of representation, introducing the pleasures and perils of ambiguity and multiple meaning. Verbal signs, with their famously arbitrary connection to the objects to which they refer, offer especially wide latitude to their users, but even icons (which represent their objects by similarity) and indexes (which represent their objects by contiguity) enable those who deploy them potentially to misguide even as they guide, or to direct different interpreters down divergent paths of interpretation simultaneously. The boy in the piggery of Omprakash Valmiki’s story, after enduring caste humiliation throughout his childhood, develops expertise in the crafting of such signs. Having excelled academically he obtains a prestigious job in a distant city, changes his name, alters his speech habits, conceals his origins and “passes” as privileged caste. Wresting control of the signs by which he is identified—or better, developing a capacity for the composition of those signs greater than the capacity of discernment of the people among whom he lives and works—he refashions, at great cost, his pahchān. As a concept, then, pahchān directs attention to the semiotic terrain on which identitarian struggle takes place and the malleability of the signs on which recognition depends.

Deceptive Majority 30

Thus when the utterance Valmiki hamārī pahchān hai—one I have often heard in the course of my fieldwork—is translated “Valmiki is our identity,” we are on the cusp of a fundamental misapprehension. This is a statement primarily about strategies of signification: Valmiki is the sign by which we are collectively perceived. It is not a statement about ontology or even, necessarily, belonging, though ontology and belonging are also semiotically produced. Valmiki in this utterance has as little to do with inward or essential selfhood as does the odor of pig flesh that constitutes the pahchān of the protagonist of “Sandstorm” during his school days. Yet scholarship in religious studies has long labored under a misperception of this sort, reading the sanitation labor castes’ traffic in the sign of Valmiki as transparent evidence of a deep, enduring, collective attachment to the Ramayana and to popular Hinduism (for example, Richman 1991: 3). To follow the conceptual path urged on us by pahchān, rather than shoehorning these phenomena into all-too-familiar frameworks of identity and recognition, is to avoid such pitfalls and, attuned to the complex verbal and nonverbal signifying practices of our guides in this book, to be led to unexpected places in the anthropology of caste and religion. In his landmark study of caste slavery and Dalit religion in Travancore, Sanal Mohan (2015: 265) argues for the value “of returning to the sources of the community and redeploying the past in such a manner that historical experiences, however terrible they are, become a resource for imagining a social praxis of liberation.” The story that fills the pages ahead is in no obvious way liberatory—indeed much of it is quite the opposite. Yet, as with Mohan’s account, the experience of the sanitation labor castes—the 583 in Awadh and their caste fellows across north India—stands to make strange some of our familiar truisms of religion and society in South Asia, exposing the contingency of a majoritarian project that thrives on appearing inevitable, and opening possibilities for other social imaginations.

2 The Ummat of Lal Beg Dalit Religion before Enumerative Politics

Who Are Hindus? I am sitting with Gollu, his mother Kanti, and their neighbor Shankar in Kanti’s two-room home in a bastī along the railway tracks in the town of Mirzapur in southeast UP. Gollu, his legs shriveled by polio, assists his mother in splitting the dried fronds of a coconut palm with a small knife; the split fronds will later be bundled and fastened together with a metal ring to become the bristles of a laggā jhāṛū, a long-handled broom, which Kanti will use in her work as a municipal sweeper. I am entertaining Gollu’s seven-month old baby, who has just peed on my lap, while Kanti, working away with the knife, tells me how she officiates at goat sacrifices to their family’s goddess, a goddess whose name I have never heard before. After some discussion of goddesses, I ask whether there are Sufi pirs that her family visits as well. Their neighbor Shankar, a fourth-class employee in the municipal water department, says, “We were Hindu from the very beginning [Ham shuru se hī Hindu the].” He goes on to explain that when the Delhi Sultan Alauddin Khalji forced rebels to convert to Islam, those who continued to resist were made into sanitation workers. He refers me to a magazine in which he recently read about this. A few minutes later, since this is my first day in Mirzapur, Shankar takes me on a tour of the bastī. As we turn a corner, he gestures at a house on the left and says, “That house belongs to a Hindu. All the others here belong to our caste [Yeh ghar hindu birādarī kā hai, baqī sab hamārī birādarī ke hain].” Have I misheard? No, it seems not: in another lane Shankar again points to some houses that belong to “Hindu people” as distinguished from “our homes” on the other side of the street. I do not press the issue, but make a note of it in my notebook. It is Diwali in Lucknow. My friends Sunaina and Jagdish and I are watching our young sons huddled together in a gully setting off all manner of firecrackers while the boom and crack of larger explosions resound in every

Deceptive Majority 32

direction. As the children play, Sunaina introduces me to her eighty-five-yearold maternal grandmother, retired from decades of domestic sanitation work in the city, who has come for a visit. She spent her childhood in a village in Bara Banki, which I ask her to describe. “In our village,” she says, “there were just four families from our community [chār ghar hamārī birādarī ke the]. Then there were also Hindus, and some Muslims as well [phir hindu bhī the, musalmān bhī the].” Here it is again: our community, on the one hand, and Hindus, on the other. Like Shankar’s, Sunaina’s grandmother’s utterance is one of casual contradistinction—the categorical contrast it suggests is not laid forth or explicated in doctrinal fashion but simply implied in the process of sketching social geography. Moments like these keep happening. Sometimes, as with Sunaina’s grandmother, the off-handed opposition of “we” and “Hindus” stands alone, whereas other times, as with Shankar, it sits in tension with explicit avowals of the community’s Hindu-ness. Does it matter that the latter tend to come when Muslims are mentioned, whereas the former seem to be anchored in everyday geography and the embodied knowledge of place? I start to keep track of such utterances, marking them in the margins of my fieldnotes. Then I notice it in the literature. In her extraordinary study of the Partition of India and Pakistan, Urvashi Butalia describes an interview she had with a Dalit woman—a “sweeper” of a Punjab village whom she calls Maya—whose recollections of 1947 trouble Butalia in unexpected ways. In the interview Maya mirthfully describes how, when Muslims and Hindus alternately fled her childhood village as rumors swirled as to whether it would be assigned to India or Pakistan, she and other girls of her caste looted the homes of whichever community departed, stealing food, utensils, and textiles and later selling them for profit. It is not so much the pilferage that Butalia finds startling about Maya’s account, nor even the laughter with which Maya related it. Rather, what upsets the historian’s expectations are Maya’s social categories. “[L]ike all Hindus,” Butalia (1998: 298–300) writes, “somewhere deep down inside me I had assumed that Harijans (Dalits), Gandhi’s supposed ‘children of God’, relegated to the fringes of society, were part of the Hindu community, part of ‘us’. Yet, … Maya was quite clear that they did not see themselves as Hindus or Christians (or indeed anyone else).” Reflecting on her ethnographic study of gender and labor among sanitation workers in Benares, Mary Searle-Chatterjee explains that she began fieldwork


The Ummat of Lal Beg

under the impression that the subjects of her study were Hindus. She writes that her informants “were clearly neither Muslim nor Christian, so I was puzzled when I heard them talking about ‘Hindus’ as if they were people other than themselves, for example, ‘the people who live across the road’” (SearleChatterjee 2008: 188–89). Searle-Chatterjee then posed to her informants the question, “Who are Hindus?” “Their reply astounded me. ‘Hindus’, they said, ‘are Brahmans and Thakurs.’” What is disclosed in moments like these is the discordance between a regnant categorical schema—the socioreligious order of things that SearleChatterjee, Butalia, and I expected to encounter—and the social ontology of those it presumes to categorize. These accounts cast in a critical light the popular and scholarly assumption that “sweepers” and other Dalits, insofar as they are not avowed converts to Buddhism, Islam, or Christianity, somehow belong to the Hindu community and should therefore see themselves as Hindus. They expose this widespread idea as grounded not in Dalit self-understandings or self-descriptions, but as animated by other logics and implicated in other projects—Butalia’s invocation of Gandhi is felicitous, as we will see. These passages also reveal the distance of inherited scholarly common sense, on the whole, from actual Dalit lives and discourse. The ontology of alterity given utterance by Butalia’s and Searle-Chatterjee’s interlocutors is distinct from that featured in the emancipatory ideologies formulated by Bhimrao Ambedkar, Swami Achhutanand, Mangoo Ram, and other twentieth century anti-caste theorists. While these ideologies did assert various kinds of difference between Dalits (as Depressed Classes, Adi Hindus, Ad Dharmis) and Hindus, they also generally worked within the categorical logic of the late colonial state according to which Dalits could only be understood as Hindus of a sort. For example, in advancing the claim to Adi Hindu—original or autochthonous Hindu—status, Swami Achhutanand asserted for Dalits a pahchān that implied difference from ordinary Hindus even as it acquiesced, in the second part of the title (Adi Hindu), to the terms available within the governmental framework. The contrastive sense of “Hindu” we are discussing here, however, is the usage of Dalits who do not espouse such ideologies and who remain largely unlinked to organized anticaste movements—“nonpoliticised” sanitation workers, as Searle-Chatterjee (2008: 189) describes her interlocutors. When Shankar spoke of Hindus as those who lived across the street, and when Maya spoke of looting the homes of Hindus during Partition, they did so not as adherents of an Ambedkarite

Deceptive Majority 34

or Ad Dharmi ideology, self-consciously asserting difference grounded in twentieth century arguments. Theirs, rather, is an older sense of alterity, one with a genealogy distinct from that of the emancipatory project of Ambedkar, Achhutanand, and like-minded figures in north India. This is not at all to say that this older sense of alterity is incompatible with such emancipatory projects, or that the thought of Ambedkar and others does not have roots in parallel traditions grounded in their respective communities (Mahar, Chamar–Jatav, and so on). It is simply to note that the sensibility we are tracking here among the sanitation labor castes is distinct. Not everyone is surprised to encounter the social ontology articulated by Butalia’s and Searle-Chatterjee’s interlocutors. Kancha Ilaiah (2005) explains that no one in the non-elite, laboring-caste milieu of his upbringing understood themselves as Hindu, as this term denoted the dominant castes in contradistinction to the subordinated castes, which were known only by specific caste names. Gyanendra Pandey (1993: 246) observes that in much of south India “and indeed in many parts of North India, the term Hindu is used to this day specifically to differentiate upper-caste Hindus (the ‘Hindus’) from ‘untouchables’ or Harijans as they are now more often called.” In a similar vein, but narrowing the focus from Dalits in general to the sanitation labor castes in particular, Ravinder Kaur (2008: 281–82) reports a joke about Chuhras—the most populous of the sanitation labor castes—during the Partition of India and Pakistan. The joke is “a popular one, often narrated by upper-caste Punjabi migrants in India and Pakistan.” It goes like this: Two Chuhras were busy sweeping the roads of Lahore during the Hindu– Muslim violence. While the Hindus were trying to flee away from the violence, Muslims were pouring into the city from India. One sweeper asked another if he knew why people were running here and there. The other answered that the “Hindus are running to India while Muslims are looking for Pakistan. But we don’t need to escape to another place and nobody is going to touch us”. And they continued sweeping the empty streets. (Kaur 2008: 281–82)

The sense of religious alterity from both Hindus and Muslims expressed by Butalia’s and Searle-Chatterjee’s interviewees seems to be intuitive to the privileged caste migrants who circulate the joke about the Lahore sweepers. In Kaur’s telling it would seem that Punjabi Hindus and Muslims are in agreement with the Chuhras on the matter of their mutual religious difference. Indeed the joke depends on it—only with an awareness of the axis of alterity


The Ummat of Lal Beg

that separates the Chuhra from everyone else does its juxtaposition with the vastly better known Hindu–Muslim axis that fueled Partition have any savor. Clearly, however, what is intuitive for certain classes of people—primarily those writing and speaking from within the social milieus in which the ontology of alterity is readily apparent—is altogether counterintuitive, even startling, to most everyone else. The surprise registered by Butalia and Searle-Chatterjee— and I certainly felt it, too, the first several times I heard the unselfconscious use of “Hindu” to mean “privileged caste other” during fieldwork—is neither unreasonable nor naïve. What was said by Maya (and Shankar, and Sunaina’s grandmother, and the sweepers of Benares speaking with Searle-Chatterjee) flies in the face of the prevailing popular and academic common sense, a common sense that rests on a vast edifice of media and literary representation, governmental data, and scholarship.

Common Sense, Indispensibility, and the Modernist Slot There is not space here to recount the films, plays, novels, short stories, biographies and other artistic representations that figure “the untouchable” as essentially and timelessly Hindu, though we will briefly consider two landmark literary examples from the 1930s—Mulk Raj Anand’s novel Untouchable and the poems Rabindranath Tagore composed for Gandhi’s magazine Harijan— in chapter 4. Nor is this the place to catalog the newspaper and magazine articles, encyclopedia and Wikipedia entries, school textbooks, government policy briefs, NGO reports, United Nations statements, judicial rulings, and reports of parliamentary commissions that treat the transtemporal Hindu-ness of the “untouchable” as a self-evident fact. Such texts rely ultimately on the categorical schema of the postcolonial Indian state, wherein “untouchables” are Scheduled Castes and Scheduled Castes are Hindu by definition (the belated allowance for Dalit Sikhs, in 1956, and Dalit Buddhists, in 1990, to be recognized as Scheduled Castes has done almost nothing to shift the foundations of the regnant categorical common sense). A word must be said, though, about the role of scholarship in upholding the edifice. The logic underlying the prevailing common sense is taxonomic and simple: there is a thing called Hindu society; Hindu society has (or had) a caste system which classifies some people as “untouchable”; “untouchables” thus belong to Hindu society; “untouchables” must, therefore, be Hindu. The reasoning is tractable for the primary school lessons through which India is taught to schoolchildren across the world, but it is by no means absent in

Deceptive Majority 36

more scholarly literature. The chapter on “Hindu Social Structure” in David Kinsley’s Hinduism: A Cultural Perspective, a popular undergraduate textbook, contains two diagrams (figures 8.1 and 8.2): one is a pyramid horizontally divided into five sections representing the four varnas with “untouchables” below them, and the other is a nesting of five concentric circles within a single circle representing the same five groups in order of distance from a center (Kinsley 1993: 154, 156). The pyramid and the circle are bounded, coherent entities representing Hindu society. “Untouchables,” while at the bottom of the pyramid and on the outer ring of the circle, are clearly and unambiguously enclosed within the borders of these entities—their internality with respect to a Hindu whole is readily visually apprehended. Many authors represent this imagined state of affairs not through diagram but through the deployment of terms like “low-caste Hindu,” “folk Hindu,” and “marginal Hindu” as synonymous with “untouchable.” In An Introduction to Hinduism, for instance, Gavin Flood (1996: 219) treats “untouchable” and “low-caste Hindu” as interchangeable terms. Other writers are more explicit still. In her introduction to Mahasweta Devi’s “Draupadi,” Gayatri Spivak (1981: 389) informs readers that the Santal and other tribes in India are “not to be confused with the so-called untouchables, who, unlike the tribals, are Hindu.” Debates over the modernity of Hinduism have both exposed and obfuscated the problem. In response to arguments that Hinduism is a synthetic and exogenous category born of nineteenth century colonial encounter (among others, Hawley 1991; Pennington 2005; Stietencron 2007; Thapar 1989), scholars have sought and found deployments of “Hindu” as a term of selfdescription, often contrasted with “Turk,” in precolonial sources (for example, Lorenzen 2006; Talbot 1995). But as Vasudha Dalmia (2007: 4–5) points out with respect to bhakti texts and Cynthia Talbot (1995) with respect to temple inscriptions, the sense of collective subjectivity or “we-hood” denoted by “Hindu” in this precolonial evidence is decidedly that of a restrictive caste elite often explicitly identified as brahmin. Unfortunately this finding—that the few appearances of “Hindu” as a socioreligious designation between the fourteenth and eighteenth centuries applied to no more than an elite fraction of what that category is now broadly assumed to designate—has been largely ignored in subsequent scholarship, overshadowed by the implications for the history of communalism of the “Turk” in these sources. Most often the idea that “untouchables … are Hindu” is not argued, but is simply asserted, assumed, or implied. In those rare instances when

37 The Ummat of Lal Beg

scholars elaborate on the claim, its justification seems to rest on a notion of indispensability—on the indispensability, that is, of untouchables for the functioning of Hindu ritual and political economy. It is a well-known claim of Louis Dumont (1980: 55) that the impurity of the untouchable is a condition sine qua non for the purity of the brahmin; the latter is obtained, both conceptually and in terms of everyday labor practices, on the back of the former. A similar idea underlies Charu Gupta’s remarks on the partial internality of Dalits to Hinduism in colonial north India: “They were in a sense both outside and within the pale of Hinduism: outside in that they were denied all the privileges of caste-Hindus; within in the sense that their labor was essential for the maintenance of social structures” (C. Gupta 2016: 44, emphasis added). That the exploitation of Dalit labor was (and is) indispensable for the preservation of caste hierarchy is certain, and Gupta and Dumont are right to emphasize it. But does it follow from this that Dalits must be “within the pale of Hinduism,” and thus, in some sense, Hindu? Jews in medieval Europe provided forms of labor that, while forbidden to and held in contempt by Christians, were nonetheless essential for the reproduction of the dominant Christian social order. But their unequal and coerced integration into a shared ritual and political economy has not become the basis for claims that Jews were, in some sense, Christian. As Nathaniel Roberts (2016: 127) points out, “it is unclear why forced ritual service to a cult implies membership in it.” This is a matter, to be sure, of what questions have been posed, in what terms, and to whom. As we will see, in the colonial dispensation British administrators often shared with local elites—including such thinkers as Mohandas Gandhi— the opinion that Dalits were incapable of comprehending religious precepts and determining their own religious views. This prejudice informed colonial knowledge practices. Administrators in the first several decennial censuses admitted that often, during the actual collection of data, Dalits were simply not asked about their religion; instead, enumerators reproduced their own assumptions or reported what local, usually dominant caste headmen told them. Even when the colonial administration did begin to take seriously the matter of the religious status of the “untouchable” masses, the question of how Dalits conceived of their own situation continued not to be asked. Rather, “objective” criteria of observable ritual practice were proposed—whether particular castes were admitted into temples, whether brahmins were willing to officiate at their life cycle rituals, whether they ate beef—that would determine

Deceptive Majority 38

whether they “should be regarded as genuine Hindus or not” (Gait 1913a: 116–17). It is a testament to the long shadow of colonial epistemology that the religious status of Dalit groups continues to be argued, in many circles in India, by reference to ritual practice as stable identitarian index (for example, “they bury their dead therefore they cannot be Hindu”), and that historical and sociological accounts of Hinduism that do not even consider how Dalits represent their own subjective belonging continue to go unchallenged. Articulations of autonomous or non-Hindu religious subjectivity by Dalit and other anti-caste thinkers, meanwhile, have been largely treated as modern, discontinuous with the past, and exogenous in terms of intellectual genealogy. Jotirao Phule, whose repudiation of the brahminical pantheon and celebration of its antagonists has been among the most influential templates for the last fourteen decades of critiques of Hinduism, is treated by biographers as indebted primarily to missionary propaganda and Thomas Paine (O’Hanlon 1985: 50–86). Ambedkar is summed up as an “unalloyed modernist” (Chatterjee 2006: 77). The latter’s assertions of alterity from Hindus, moreover, are often portrayed as an effort “to thrust the ‘untouchable’ irrevocably outside Hinduism”—a formulation that again imputes an interiority of the “untouchable” to Hinduism prior to Ambedkar (Leslie 2003: 54). There are, of course, very good reasons that Phule and Ambedkar are counted among the ardent modernists of nineteenth and twentieth century Indian thought. The point to be noted here is that the almost exclusive emphasis on the European influences on their thought and on its discontinuity from the local—the analysis of their “separatism” in terms of missionary intervention, British strategies of divide and rule, engagement with European Enlightenment thinkers, and the internalization of colonial epistemology—enables an evasion of the question of endogenous modes of Dalit self-description. Before Phule and Ambedkar— or coeval with and after them, but outside their sphere of influence—how did Dalits imagine and speak about their place in a socioreligious order? To what kinds of entities did they describe belonging and attachment, and against what others did they define collective selves? What were the contours of their social ontologies? Questions such as these drop out of view when assertions of religious difference by anti-caste thinkers from Phule onward are plotted primarily into narratives of ruptive modernity. This can happen, notably, both in sympathetic accounts that tend to portray Phule, Ambedkar, and others as representing a broad Dalit consensus—in which case the claim of religious alterity is collective and signals a collective modernist orientation, a shared

39 The Ummat of Lal Beg

eagerness to sever links with the past—as well as in accounts that exceptionalize such thinkers as unrepresentative of their unlettered caste fellows, in which case the latter are assumed to remain “faithful” to Hinduism. In either case, the question of what came before is left in the dark, and the common sense of the timelessly Hindu untouchable is shielded from scrutiny. Cracks are beginning to show in the edifice that upholds the regnant popular and scholarly common sense; historians in recent years have begun to point out that the colonial categorization of Dalits as Hindus was the contingent outcome of decades of political contestation and that it played a key role in bringing into existence the Hindu majority that is so often erroneously imagined as having existed in the past (Adcock 2014; Dalmia 2010; Roberts 2016; Tejani 2008; Viswanath 2014; Zavos 2000), and anthropologists of religion have cast doubt on the attribution of a Hindu subjectivity to Dalits in particular contexts (Juergensmeyer 2009; Prashad 2000; Searle-Chatterjee 2008). Ethnographic and autobiographical accounts of religion by scholars and writers from disprivileged castes have alerted many to the problems of the prevailing common sense (for example B. Das 2007; Ilaiah 2005; Valmiki 2003), though far too often the historical implications of these interventions are avoided by shunting them into the modernist slot. Yet in many domains of scholarship—among them religious studies, where Ambedkarite critique has provoked a kind of need to ethically validate Hinduism by pointing to examples of Dalit participation in it—the wall of common sense holds fast that “untouchables” in some essential and timeless manner belong to Hinduism. This state of affairs, in no small part a consequence of the colonial-and-Gandhian legacy of representing Dalits without having asked for or listened to their selfdescriptions, has been detrimental to our understanding of South Asian history and society.

Beyond the Bifurcation Paradigm Tracing the genealogy of the prevailing common sense is one of the aims of this book. But what of its obverse? What kind of social ontology undergirds Shankar’s usage of “Hindu” as he guided me through the bastī, and what is its history? If there is a tradition of religious autonomy among the sanitation labor castes—and by autonomy I mean to denote neither the sovereign freedom of the Kantian subject nor complete unrelatedness to religious others, but rather, more modestly, self-description as distinct, separate, and largely self-governing in religious affairs—then how has that tradition conceived of the religious

Deceptive Majority 40

landscape of South Asia, and in what terms does it posit difference from Hindus, Muslims, and others? I propose to address these questions in the pages that follow by drawing together and critically examining accounts of the religious imagination of the sanitation labor castes from the late colonial period. This chapter will consider descriptions of beliefs and practices that others have thematized as “conversion,” “syncretism,” and “the Cult of Lal Beg.” My sources are missionary writings; colonial gazetteers and census reports; Urdu literature, journalism, and literary history; bhakti texts in Braj Bhasha along with their commentary in Hindi; writings of Rudyard Kipling; and the “notes and queries” genre of ethnographic and folkloric tidbits. Hovering above all this are the practices and oral traditions that my interlocutors shared with me in Lucknow and elsewhere. In some ways what we will find in the colonial archive seems utterly unlike the stories one hears today, for reasons that we will explicate; in other ways the old themes and melodies continue to reverberate. Much of this material coheres around a religious formation, largely coextensive with the most populous of the sanitation labor castes (the Chuhra) but also including Dhanuks and possibly others (Ibbetson 1970: 296), that British observers called the Cult of Lal Beg. I call it the Lal Begi tradition. What I mean to demonstrate in this chapter is that the myths, rites, and liturgical songs of this tradition give utterance to a collective subjectivity of religious belonging distinct from and on certain points opposed to that of Hindus, Muslims, and Sikhs. I argue, moreover, that the social imagination enunciated in the oral texts of the Lal Begi tradition, while entirely at odds with the sociological common sense with which we now live, did not contradict the prevailing categorical schema of the period. That is, before the rise of early twentieth century enumerative politics, the religious autonomy that the sanitation labor castes gave voice to in their liturgy was generally accepted by Hindus and Muslims. Before the stakes of constituting a numerical majority were dramatically heightened in the initial decades of the twentieth century, ordinary Hindus and Muslims had as little interest in imagining the sanitation labor castes as co-religionists as the latter had the former. To trace the contours of this earlier social ontology is to begin to restore to the historiographic record the self-representations of the sanitation labor castes. It is, by the same token, to assist in a collective project of easing South Asian studies out of an impasse of the imagination, a conceptual cul-de-sac that has been acknowledged but not escaped. This is what Peter Gottschalk

41 The Ummat of Lal Beg

(2013: 18, 183) calls “the bifurcation paradigm,” a scholarly inheritance from the imperial epistemic order, a habit of viewing Hinduism and Islam as “mutually exclusive categories of social belonging that bifurcated nearly every societal and cultural dimension of India”—categories, I would add, that tend to take up all the air in the room, as though they alone exhausted the spectrum of religious possibility in South Asia. Our subjection to this “tyranny of labels,” as Romila Thapar (1996) calls it, distorts interpretations of historical materials by the backward projection of twentieth century common sense. We reflexively gloss the Yavana and Turuska of medieval texts as “Muslim,” for instance, a procedure “methodologically invalid and historically inaccurate” that invites modern conceptions to obscure the often quite different meanings of the earlier terms (Thapar 1996: 7). What needs to be added to the critique is that it is equally erroneous to label subordinated caste groups of the precolonial and colonial periods “Hindu” in the absence of evidence that they saw themselves as such. Identifications like this universalize what was in fact a late colonial development, and bury the self- and other-perceptions of such groups under a categorical schema foreign to their experience. The problem is fairly widely admitted; but in the absence of an alternative conceptual model of the socioreligious landscape of South Asia, the bifurcation paradigm continues to undergird scholarly accounts. To attend to the specifics of the Lal Begi tradition is to draw attention to precisely such an alternative paradigm in the historical-ethnographic record. The social ontology of the Lal Begis evokes not a bifurcated but a multipolar religious landscape—a landscape, moreover, hewn into vertical as well as horizontal divisions. I use the term “divisions” advisedly: this will not be a story that counterposes the hard lines of the communalism of the present with earlier subjectivities characterized primarily by porousness, ambiguity, and “fuzziness” (Kaviraj 1992). However apposite such characterizations may be for many communities through much of South Asian history, they do not answer well the kind of evidence we will encounter, riven as it is by the decidedly non-porous boundary that separates touchable and untouchable. Attention to how the sanitation labor castes in the colonial period perceived their others, and how their others perceived them, may provide a constructive model with which to dislodge the binary thinking that has too long dominated the historical imagination, and to do so without wishing away or collapsing distinctions that obtained in the past. This should help South Asian studies to advance conceptually “beyond Hindu and Muslim” (Gottschalk 2000).

Deceptive Majority 42

“Conversion”: A Brief Sketch of Sanitation Labor Caste Assertion in Sikhism, Islam, and Christianity, 1859–1932 “The chief disturbing element in the return of religions is the Chuhra.” So declared Horace Arthur Rose, Superintendent of Census Operations for Punjab and North-West Frontier Provinces, in his report on the census of 1901 (Rose 1902: 113). There were several reasons that “the Chuhra”— again, the most populous and transregional of the sanitation labor castes of north India—proved troubling to the British effort to categorize its subject population according to religion. They seemed to change religions often; their beliefs and practices appeared to be a promiscuous mingling of the religions of their masters while, at the same time, they were reputed to have a prophet and priesthood of their own; and, finally, census enumerators were loath to count them as co-religionists. Over the next half century these disturbances of state knowledge would be quelled: either resolved into a neat narrative of conversion from Hinduism to Christianity by a subset of the community, or theorized as “syncretism” to which “low caste Hindus” seem perennially to be drawn, or, in the case of the enigmatic prophet and the averse enumerator, actively forgotten. But let us dwell for a while in the period before these clean-up operations, when the disturbance was still afoot. Let us take the causes of official anxiety one at a time, to see what, when read critically but not dismissively, the colonial archive may tell us about the religious history of the sanitation labor castes in the late nineteenth and early twentieth century. First, then, let us turn to sources on conversion. Though our primary interest is the Lal Begi tradition, a consideration of the ways that many among the sanitation labor castes took formal measures toward adopting Sikhism, Islam, and Christianity during the colonial period is essential, for reasons that will later become apparent. Importantly, some of the accounts that we will examine here—especially of ritual accessions to Sikhism and Islam—have never been treated in the secondary literature, and need to be brought to light for the sake of laying broader foundations for Dalit religious historiography. The Punjab Report of the Census of 1911, authored by Pandit Harikrishan Kaul, announced: The Hindus have lost 158,806 Chuhras, due partly to real conversion to Christianity or Islam and partly to misclassification. There can be no doubt about a large number of Chuhras having been converted to Christianity … and there have also been numerous conversions to Musallis (a Chuhra converted to

43 The Ummat of Lal Beg

Islam is usually called Musalli); but the abnormal rise of 252,158, i.e. about 439 per cent. in the number of Musallis, would indicate [that many enumerators failed to adhere to the policy of returning Chuhras as Hindu by default]. (Kaul 1911: 100)

In the census reports of the last two decades of the nineteenth and first two decades of the twentieth century (that is, those of 1881, 1891, 1901, and 1911), passages like this are typical. Administrators struggled to account for the vast numbers of the sanitation labor castes, especially the Chuhra/Bhangi, that appeared simply to switch religions from one decade to the next. For instance, the official population of Sikh Chuhras shrank by 75 percent between 1891 and 1901 in Punjab, only to more than double in the subsequent decade, while the Muslim Bhangi population quadrupled in UP between 1891 and 1901, and then more than quadrupled in Punjab between 1901 and 1911 (Burn 1902: 258; Kaul 1911: 100; Rose 1902: 115). These improbable swings were generated to no small extent by shifts in census policy over time, such as the modification of the official definition of a Sikh in 1891 and again in 1911 (Gait 1913a: 119; Jones 1981; Rose 1902: 124), as well as by disjunctions between census policy and what enumerators were willing to record, a point to which we will return. Quite aside from confusion originating in the census project itself, however, the volatility was also attributed to the widely observed fact that the sanitation labor castes in this period were adopting on a mass scale Islam and Christianity, and to a lesser extent Sikhism. Describing these developments as “mass conversions” or “conversion movements” is, in a way, to cloak them in a deceptive familiarity—as though conversion were a stable, universal, and transparent category, which it is not (Mohan 2015; Roberts 2016; Viswanathan 1998). At the same time, if we specify our meaning and set aside for the moment the problematic of “motive,” it may be fair provisionally to accept the contemporary characterization of these late colonial events as “conversion movements”: movements because they were indeed popular and cohesive in time and space, conversion because they entailed a decisive shift in stated and perceived status vis-à-vis a religious group, generally solemnized by formal, if simple, ritual acts. If the retention, more or less surreptitious, of some of the traditional practices of the preceding religion characterized these movements, it was hardly an exception to the global history of phenomena widely understood to be conversion. It must be emphasized, though, that the prevailing assumption that these conversions were from Hinduism should be

Deceptive Majority 44

treated not as settled fact, but rather as a question for which we might seek answers in the archive. Let us begin with Sikhism. Constantly debated in the decades under consideration was the question of who was and who was not a Sikh, and whether a Sikh was simultaneously a Hindu (Oberoi 1997). To observe that the boundaries of the Sikh community were in flux, though, is not to say that distinctions were not made. Members of the Chuhra caste who adopted Sikh traditions in the colonial period were known as Mazhabis, Mazhabi Sikhs, or Rangretas (Ibbetson 1970: 290–96). Oral traditions suggest that conversions to Sikhism by this caste began no later than the period of the tenth guru (Guru Gobind Singh, d. 1708), who is said to have publicly honored one such Chuhra Sikh (or in some versions, two brothers) for retrieving the decapitated head of the ninth guru from Aurangzeb’s guards in Delhi and delivering it to Gobind Singh in Anandpur (Greeven 1894: 35–36; A. Sharma 1928: 20; Weitbrecht 1886: 736). In our period, Richard Greeven (1894: 52) describes conversions as commonplace. The ceremony, by which a sweeper at Benares becomes a Sikh, is simplicity itself. The technical expression, in use among the scavengers, is, “to assume the motto of Nanak Shah” (Nanak Shah ka mantra lena).

He goes on to describe a rite in which a candidate is prayed over by a sponsor, feasts his caste fellows, and is initiated with a secret mantra into the Sikh community (Greeven 1894: 52–54). Herbert Strickler, writing from Punjab, remarks that those Chuhras who are “formally made followers of the Sikh religion [do so] by ‘pahul’,1 or baptism” (Strickler 1926; on pahul, see Mann 2004: 23–28). The degree to which the converts, as a group, observed Sikh orthopraxy—even as the latter was constantly evolving—apparently varied. Many “wear the ‘kes’ or long hair done up in a top-knot, together with other insignia of the Sikh religion, such as the iron bracelet, comb, etc., prescribed by the Sikh Gurus,” writes one commentator (Strickler 1926), while others insist that “Mazhabi Sikhs, or converts from the Chuhra or sweeper caste … do not wear the kes and have no scruples about smoking” (Gait 1913a: 119; Greeven 1894: 54). It is worth noting the marked presence of Sikh preceptors among 1

Pahul is a Sikh initiation rite dating to the times of Guru Nanak, the tradition’s founder. Baptism is a suggestive parallel, but by no means a strict translation, of pahul.

45 The Ummat of Lal Beg

the sanitation labor castes in accounts from places far from Punjab, such as Poona and Benares, indicating that the engagement with the Sikh tradition was not restricted to regions in which Sikhs constituted a local dominant caste (W. E. M. Campbell 1908: 437; Greeven 1894: 38). What of conversion to Islam? In the period under consideration this is mostly evidenced in Punjab, particularly western Punjab, whereas in the east (eastern UP, Bihar, and Bengal) significant swaths of the sanitation labor castes—then known as Sekras or Sheikh Halalkhors, now known as Halalkhors—appear formally to have embraced Islam in an earlier period (Gait 1902a: 436; Lee 2018: 10–13). Chuhras who embraced Islam in Punjab in the colonial period became known by the titles Musalli, Kutana, or Dindar; census reports claimed that their percentage of the overall Chuhra population grew from 20 in 1891 to 22 in 1901 and 31 in 1911 (Ibbetson 1970: 8; Rose 1902: 113; Strickler 1926; Weitbrecht 1886: 673–74). As for the procedures by which this increase took place, H. U. Weitbrecht (Weitbrecht 1886: 763–64), an Anglican missionary, reports: “A Chuhra observes the Ramazan fast and the time of prayer in the hopes of reception as a Muslim. If successful, he is admitted on condition of renouncing scavenging and the eating of carrion.” Chaina Mall, a correspondent in the Panjab Notes and Queries, elaborates on “the ceremony of initiation” for “converts to Muhammadanism belonging to the Mehtar caste”: He is made to repeat the Muhammadan creed (Kalima) five times after bathing and dressing in new clothes. He … then says Toba (Repentance) in a clear firm voice and vows never to return to his old faith three times before a Maulavi and other witnesses. After this the Maulavi drinks from a vessel … of which the convert drinks also, and is then … pronounced a Musalman. If he be at all literate he calls himself a Nau Musallim. (Mall 1886)

Though the evidence is relatively scant, it appears that no major Muslim organization or institution actively sought converts from the sanitation labor castes until the height of the shuddhi controversy in the 1920s, and even then only in pockets of UP and Hyderabad (see chapter 3). Before the 1920s the Muslim group most thoroughly embroiled in the polemics wars with Christian missionaries and the Arya Samaj in Punjab, and which we might therefore expect to have entered the competition for sweeper caste converts, was the Ahmadiyya. While there is evidence that the Ahmadiyya supported some Depressed Class causes and even published a pamphlet advising Dalits not to

Deceptive Majority 46

be duped by the Arya Samaj (Juergensmeyer 2009: 28, 40), these interventions do not appear to have included conversion efforts. Moreover, Mirza Ghulam Ahmad, founder of the Ahmadiyya, vigorously denounced the suggestion that sweepers were among his followers in terms that suggested antipathy to the caste as a whole. More extensively documented in this period was the large-scale conversion—the “mass movement” as it came to be called in missionary circles—of sanitation labor castes to Christianity in Punjab and western UP. In western UP the movement began in 1859 with Mazhabi Sikhs adopting Christianity, followed in the 1880s and 1890s by their Lal Begi caste fellows. There the Methodist missions played a primary role (Alter 1986: 166–212; Burn 1902: 68; Webster 1992: 38). In Punjab, 1873 is remembered as the inaugural moment and the Presbyterian missions predominated, though all denominations were ultimately affected. In both regions the sanitation labor castes rapidly outstripped all other groups to become the overwhelming majority of the Christian population. By 1900, in western UP, “the terms ‘Christian’ and ‘Sweeper’ became in many places synonymous” (Messmore 1903: 265). By 1931 in Punjab roughly one quarter of all Chuhras had converted to Christianity (Webster 1992: 38), especially in the west, such that in post-Partition Pakistan as well, “sweeper” and “Christian” would become, in popular usage, interchangeable (O’Brien 2012; Streefland 1979). Chuhra mass conversions figure significantly in scholarly accounts of Indian Christianity (for example, Bauman and Young 2014; Cox 2008; Webster 1992, 2007, 2012). One important aspect of these movements that the secondary literature rightly emphasizes is that by all accounts they were inaugurated by the sanitation labor castes themselves; it was they who first approached the missionaries, rather than the other way around. Indeed the first wave of individuals and families seeking baptism took the missionaries by surprise, and led, as their numbers grew, to a change in policy from proselytizing the privileged castes to managing the spiritual and material needs of the dispossessed (Alter 1986: 178; Gordon 1886: 421–23, Webster 1992: 38; Weitbrecht 1886: 763–66; Wiser 2000: ch. 4). But there is another, related aspect of the movements that historiography has not fully acknowledged, an aspect relating to the pahchān of the converts. There was a common perception at the time, both within and without church circles, that a mission admitting “the very lowest castes … has to be satisfied with a lower standard of appreciation of the tenets of Christianity” (Alter 1986: 199, 205; Burn 1902: 68; Messmore


The Ummat of Lal Beg

1903: 264–65). To combat this estimation, “Mass Movement Christians,” as the subaltern converts were euphemistically called, were generally made to undergo special tests of knowledge and sincerity before baptism, or after baptism but before admittance to communion (Griswold n.d.; Weitbrecht 1886: 765). In several major denominations, the admission of Mass Movement Christians to the Eucharist was simply indefinitely deferred, thus creating a relatively permanent subclass of Christians unfit for the Lord’s Supper. The marking of Chuhra converts as less than complete members of the Christian community was not only a ritual matter but a nominative one. Not unlike sanitation labor caste converts to Islam (who, as we have seen, were called Musalli, Kutana, or Dindar) and Sikhism (who were then called Mazhabi or Rangreta), the new Christians, too, were hailed by other members of the religious community they had joined with a name that indexed their belonging to a distinct and subordinate category (“Mass Movement Christians,” “village Christians”). Among the relatively small community of native Christians, some (in Amritsar, for instance) opposed Chuhras joining the church, others “found it very difficult at first” but later reconciled themselves to it, while still others embraced the change in the Christian community’s composition (Gordon 1886: 429; Weitbrecht 1886: 766). As we have seen, “village Christians” generally remained unauthorized to partake in communion. Though a growing majority of the Indian church, they bore the double marks—ritual and nominative—of marginality vis-à-vis the wider Christian community. One contributing factor to this marginalization, at least from the perspective of missionaries, was that Chuhra Christians were, more than privileged caste converts, reluctant to abandon their ancestral religious practices (Messmore 1903: 264–65). Destruction of the shrine of Lal Beg became, in many cases, the ultimate demonstration of fidelity to the new faith. “When all the Chuhras of a place become Christians, then with some fear and trembling the shrine is destroyed” (Griswold 1934: 231). Yet many converts resisted taking this dramatic step, and even for those who did, the matter rarely ended there. “A group of sweepers may on baptism destroy the than [shrine] … but that does not prove that all idolatrous practices will cease” (Griswold n.d.); reports continued to emerge “that images and shrines of the Lalguru are still resorted to in secret” (Alter 1986: 202; Burn 1902: 99). That the worship of Lal Beg proved the most difficult practice for new Christians to give up, and that the ongoing locus of tension with missionaries was the clandestine continuation of the rites of the Lal Guru, should help illuminate the spuriousness of the

Deceptive Majority 48

Hindu majoritarian claim that would, decades later, become an article of common sense. It was not Hinduism from which the sanitation labor castes were converting to Christianity, but the Lal Begi tradition. The final point to be made regarding the conversion movements of the colonial period requires us to rethink the concept of conversion itself. The sources make clear that for many converts, assertions of social and economic agency accompanied the transformation of religious status. These included, notably, efforts to abandon permanently the occupation of manual scavenging—the removal of human excrement (“night soil”) from privies— as well as attempts to obtain land and access to education. “His occupations change somewhat with his religion,” noted Ibbetson (1970: 293) of “the Chuhra” in Punjab in 1881. He “refuses to touch night soil and becomes a Musalli or a Kutana” (Baines 1893: 83: Ibbetson 1970: 293). Likewise Mazhabis “after conversion continue to perform only the less offensive parts of their traditional duties” (Baines 1893: 83). The Gazetteer of the Gurdaspur District (1892: 66) notes that when local Chuhras converted to Christianity, “the converts were unwilling to continue to perform their customary village service [that is, scavenging], and the villagers refused to give the customary due.”2 The link between conversion and acts of occupational rebellion was sufficiently widely understood that landlords often took measures to prevent intercourse between their dependents and missionaries in the area, or to punish those who converted. Mazhabi converts to Christianity were thrown off their tiny plots of land in Moradabad and Bijnor Districts in UP, for instance (Alter 1986: 183). In a village near Batala in Punjab, locally dominant Muslim Rajputs intimidated the Chuhra population, which had been seeking Christian instruction from an Anglican mission, and thwarted an effort to open a school for them (Weitbrecht 1886: 765–66). Landlords elsewhere in Punjab had Dalit converts beaten nearly to death, and abducted their family members (Gordon 1886: chs. 14, 18).


This strike, clearly an act of sanitation labor caste initiative, was brought to a halt by the missionaries, many of whom were ambivalent or antipathetic to disruptions in the social and economic order. The gazetteer account continues: “This state of things threatened to produce awkward complications, but, thanks to the energetic action of the Missionaries of all sects, who at once pointed out the unreasonableness of their position to the converts, the danger was averted and no further complaints have occurred” (Gazetteer of the Gurdaspur District 1892: 66).


The Ummat of Lal Beg

The apprehensions of the landlords were well founded: in approaching Christian institutions, the sanitation labor castes were indeed attempting to improve their economic position, decrease their dependence on local dominant castes, and gain a foothold in the alternative patronage system of missionary networks. The missionary archive is peppered with examples of Dalit converts or even mere inquirers persuading missionaries to grant them rent-free or reduced-rent land, or to settle new jungle tracts (see, for example, Alter 1986: 184; Youngson 1896: 196). Education was another persistent demand of new and would-be converts. Sanitation labor caste delegations frequently requested that missionaries establish village schools for their children or arrange the admission of their children in religious or government schools in the towns (Alter 1986: 178; Gordon 1886: 430; Weitbrecht 1886: 765; Wiser 2000: 39–58). The missions often obliged these demands, but faced fierce opposition from dominant castes. As Rupa Viswanath (2014) observes, efforts such as these constituted an assault on the political economy of untouchability. When Chuhras formally embraced Islam, Sikhism, and Christianity in the late colonial period, they simultaneously accessed education, obtained land for cultivation, and, above all, put distance between themselves and the labor of removing human ordure—all of which directly threatened the dominant caste monopoly over Dalit labor. This put missionaries in an awkward position. The framework of conversion predisposed the missionaries to question and debate their charges’ motive for conversion, and to find them either “worldly” or “spiritual” or some combination of the two. In missionary discourse, then, the acts of Dalit assertion that accompanied conversion could be interpreted as signs of material motives in those seeking baptism—a point that landlords were quick to exploit. C. S. Adcock (2014), however, observes that the missionaries’ framing of the problem does not appear to reflect the perspective of the converts. A close reading of the sources suggests that sanitation labor caste converts saw labor insurrection and educational assertion as fused with, rather than in tension with or a possible betrayal of, baptism (or pahul, or reciting the Kalima). Adcock (2014: ch. 2) proposes that, in place of “conversion,” we think of engagements of the sort we are describing as “ritualpolitical assertion.” This analytically felicitous move serves to highlight that subaltern acts of staking social and economic claims were inseparable from the ritual act of entering the fold of the new religion. Bearing this in mind will help make sense of the sanitation labor castes’ later engagements with

Deceptive Majority 50

Hinduism, initially mediated by the Arya Samaj. Indeed, the twentieth century Hinduization of the sanitation labor castes with which this book is centrally concerned may be seen—in some ways, at least—as foreshadowed by the ritual-political assertion of the mass movements to Christianity in the colonial period.

“Syncretism”: The Sanitation Labor Castes as “Half Muslim, Half Hindu” After conversion, the second trope under which sanitation labor caste religion has been subsumed in the literature is syncretism. Consider the following excerpt from the Gazetteer of Poona in 1885: They are known as Halalkhors or all-eaters, Bhangis … and Mhetars [sic] or princes. They are also called Lal Begis or the followers of Lal Beg, their religious head or guru…. In religion they are half Musalmans and half Hindus, going to mosques and repeating prayers and at the same time having as family deities Khoriyal of Gujarat, Khandoba of Jejuri, Khajapir, Baba Makdumba, and the goddesses Kalsar and Ghochati.3 They pay equal respect to Musalman saints and to Hindu gods and offer them fowls whose throat has been cut by a Musalman…. They keep both Hindu and Musalman fasts and festivals…. Their religious teachers or gurus are either men of their own caste or belong to the school of Nanakpanthi [that is, Sikh] beggars. (J. M. Campbell 1885: 437)

Or this, a description of Lal Begis from the report of the 1901 Census in Bengal: The truth is that they are on the borderland between the two religions, and they worship both Muhammadan pirs, such as Pir Jahar and the Panch Pir, and also Jagadamba and other godlings of the low-caste Hindus around them. Their priests are men of their own community. They eat the leavings of all, whether Christians, Hindus or Muhammadans, and also pork, except


A combination of Muslim saints, and local, non-brahminical heroes and goddesses. Khoriyal: divinized woman of the Chāran caste, associated with crocodiles, worshipped in parts of Gujarat and Maharashtra. Khandoba: martial folk deity, popular among peasant and Dalit castes as well as Muslims, associated with termite mounds and Bhairav, worshipped in much of Maharashtra. Khajapir: Khwaja Khizr, figure of Islamic mythology associated with water and immortality. Baba Makdumba: Baba Makhdūm, fifteenth century Sufi buried in Mahim, Maharashtra.

51 The Ummat of Lal Beg

in Eastern Bengal, where their leaning towards Muhammadanism is most marked. (Gait 1902a: 436)

Or this, a summation of “the religion of the scavenger castes” from Richard Temple’s 1884 collection of folklore, The Legends of the Panjab: This religion may be best styled hagiolatry pure and simple, as it consists merely of a confused veneration for anything and everything its followers, or rather their teachers, may have found to be considered sacred by their neighbours, whatever be its origin. Thus we find in the Panjab that in the religion of the scavenger castes the tenets of the Hindus, the Musalmans and the Sikhs are thrown together in the most hopeless confusion. (Temple 1884: 529)

In these and other colonial descriptions, the ensemble of religious precepts and practices attributed to the sanitation labor castes is regarded as an admixture of Hinduism and Islam (and sometimes Sikhism). That the mingling of elements of disparate religions is a sign of “hopeless confusion” is the implication of many—though not all4—such accounts, whose authors, steeped as they are in the classism native to colonial epistemology, seem unable to conceive of intellectual coherence among unlettered sweepers. But the deeper problem with these accounts is their assumption, reproduced in decades of subsequent scholarship that glosses these same phenomena as “syncretism,” that the ensembles they describe are any more or less promiscuous admixtures of beliefs and practices than the “religions” from which they purportedly derive. The concept of syncretism has been justly faulted for imputing unity and even a kind of purity to the Hindu and Islamic traditions from which “syncretic” groups or movements allegedly borrow or inherit elements, when, in fact, all religious traditions are synthetic mixtures, and the distinction between religion and syncretic cult, like that between language and dialect, reflects histories of asymmetrical power relations more than inherent properties (Bourdieu 1991, 1977; H. Geertz 1975; Stewart and Ernst 2003; Weinreich 1945). To these concerns I would add that a teleology of failure is embedded in the term “syncretism.” In English, usages of the word emerged in the context of a seventeenth century attempt, during the European wars of religion, to reconcile theologically the Protestant sects with each other and with the Catholic Church. 4

Civil servants Richard Greeven and Altaf Husen are notable exceptions, as is the Scottish missionary John F. W. Youngson.

Deceptive Majority 52

It thus carries with it the whiff of a doomed project. This is problematic insofar as it erases the contingency of history, painting phenomena such as the Lal Begi tradition as failures-in-the-making, and treating the bifurcation paradigm— the partitioning of everything along a Hindu–Muslim divide—as the necessary telos of South Asian history. Is there another way to think of the “confused veneration for anything and everything” that the “half Hindu” and “half Musalman” sanitation labor castes were seen as observing in the late colonial period? Beyond exposing the foundations of colonial epistemology in the prejudices of particular groups in a racial and class order, what can sources like these tell us? I suggest that the variety of religious life noted among the sanitation labor castes at the turn of the twentieth century may profitably be understood not as a token of the type “syncretism,” but rather as a distinctive ensemble of precepts and practices that reflect the ethical and religious needs of this uniquely socioeconomically positioned group. Read critically, the colonial ethnographic record (including, here, missionary accounts) is actually quite helpful in rendering visible the particular contours of this ensemble. It has two modalities: the first, which varied in content from region to region, we might call mimetic, while the second, which held consistent across regions, we might call elective. What distinguished the first modality was the imitation, by sanitation labor castes, of the religious practices of the locally dominant caste or community. “The Chuhras are chameleon-like in their copying of the externals of other faiths,” wrote the American Presbyterian Hervey DeWitt Griswold (1934: 227) of the people among whom he had spent thirty years in Punjab and western UP. “Those serving Moslem landlords gradually adopt Moslem customs; those serving Sikh landlords, Sikh customs” (Griswold 1934: 227). According to another observer: In Hindu villages their worship approximates that seen at Hindu shrines. In villages owned and inhabited by Muhammedans, the Chuhras dress, and cut their hair, like Muhammedans…. In Sikh communities even those not formally made followers of the Sikh religion by “pahul”, or baptism, often wear the “kes” or long hair done up in a top-knot, together with other insignia of the Sikh religion, such as the iron bracelet, comb, etc., prescribed by the Sikh Gurus. Those Chuhras who live among a mixed population often present to the onlooker a bewildering variety of customs. They may claim to follow all religions, as did the Athenians of Paul’s day, and often they take part in the feasts and other observances of several religious sects. (Strickler 1926)

53 The Ummat of Lal Beg

If, in descriptions like these, talk of “copying the externals” and “claim[ing] to follow all religions” has the ring more of allegation than of observation, we may provincialize these Christian authors’ prejudice against imitation by recalling, with Gabriel Tarde (1903), that many if not most phenomena in the history of religions, Christianity as much as others, can fairly be described as mimetic. The literature on “syncretism” in South Asian religions provides a range of paradigms, some quite compelling, for making sense of one group’s direct borrowing of another’s practices regarding the sacred (Bellamy 2011; Eaton 1982; Flueckiger 2006; Khan 2004; Lawrence 1984; B. Singh 2015; Taneja 2018). What I am describing as the mimetic modality of the religious practice of the sanitation labor castes, though, cannot be adequately accounted for in these explanatory frames, and directs attention, instead, to the dependency inherent in the political economy of caste. My interlocutors in Lucknow knew of their caste’s historical reputation as “chameleons” and spoke on the theme with some enthusiasm. Gauhar Lal, an elderly man born in Sitapur district north of Lucknow, explained to me that in his village, where the landlords were brahmin, his father and grandfather all maintained a choti (a hair braid traditionally worn by brahmins), had hinduānā (Hindusounding) names, and cremated their dead. When he was married to a woman (of the 583 clan) from the adjacent district of Bara Banki, where Muslim landlords predominated, Gauhar Lal was startled to find that his in-laws had musalmānā (Muslim-sounding) names, buried their dead, and some even wore skullcaps. Gauhar Lal and other older members of the community frequently invoked an aphorism to explain this significant aspect of their ancestors’ religious practice: jaisā rājā vaisī prajā—as the king, so the people. Mimesis, in this folk theory, is normative to conditions of dependency, and subordinate caste laborers can be expected to imitate the religious style of the landlords—petty “kings”—who control land and labor. This recasts “syncretism” as a kind of politic dissembling. Importantly, though, my interlocutors’ invocation of jaisā rājā vaisī prajā passes no moral judgment on the ancestors; it does not imply the charge of hypocrisy latent in “dissembling,” “dissimulation,” and related concepts, freighted as they are with centuries of deployment as accusation in Christian history. Rather than pointing to the moral responsibility of the imitator, the aphorism directs attention to the structure of power relations that gives rise to imitation. In this respect Gauhar Lal’s interpretation of his community’s history draws close to taqiyya, the Shi’i doctrine of tactical

Deceptive Majority 54

dissimulation—a point to which we will return. Note also that the terms hinduānā and musalmānā—adjectives denoting “Hindu-like” or “Muslimesque”—address the sound or appearance, but not the ontological status, of the objects they modify. By using hinduānā and musalmānā rather than Hindu and Muslim, Gauhar Lal can speak eloquently of his parents’ and inlaws’ pahchān—the signs by which they are recognized or misrecognized— while saying nothing about being, belonging, or belief. The second modality of the ensemble of religious practice discernible in the colonial record, which I am calling elective, stands outside the logic of jaisā rājā vaisī prajā. Upon comparing descriptions in accounts from Poona, Benares, Delhi, Lucknow, Saharanpur, Muzzafarnagar, Jodhpur, Bengal, and numerous distinct locations in (undivided) Punjab, and checking these against local configurations of dominance in these places, one begins to see that conformity to the practices of regional landlords accounts for only part of the picture. Beyond that, the descriptions resolve into a pattern with these elements: (a) veneration of chthonic goddesses, Muslim pirs, and divinized heroes (notably Zahir Pir and Khandoba, figures resistant to categorization as Hindu or Muslim); (b) animal sacrifice, usually with Muslims performing the role of sacrificer; and (c) affiliation with a loose network of peripatetic mendicants—bhagats or gurus—of the Chuhra community, who were either formally Sikh (udāsī, nānakpanthī) or versed in Sikh traditions, who linked the sanitation labor castes spread across north India and the Deccan, and who provided religious instruction and ritual leadership. We noted earlier the marked presence of Sikh preceptors in places like Poona and Benares with no landed Sikh presence to speak of. Likewise devotion to non-brahminical earth goddesses appears consistently even in places where the sanitation labor castes overwhelmingly worked for Muslim landlords, while devotion to pirs and Muslim (or musalmānā) naming practices prevailed even where Hindu and Sikh landlords were the primary employers. This pattern of practice is not “the confused veneration of anything and everything”: significant features of popular north Indian religion—Krishna, Ram, and the whole constellation of Vaishnava objects of worship, for example—are noticeably absent in these accounts. To the extent that some local objects of worship were identified also with the brahminical pantheon (as avatars, local manifestations, and so on), it was the “fierce” deities associated with deviance and tantra—Kali and Bhairav—who provided the link (Griswold 1934: 212; Strickler 1926: ch. 2).

55 The Ummat of Lal Beg

The oral histories I collected in Awadh correspond to this picture drawn from colonial sources. My eldest informants spoke of their parents and grandparents, who had been alive at the turn of the twentieth century, fasting for Ramadan, mourning during Muharram, and lighting oil lamps on Diwali. They respected the religious leadership of bhagats of their own caste who spoke in an idiom inflected by Sikh tradition and traveled, at times, to Punjab and western UP. Those generations were devotees of Zahir Pir, Ghazi Miyan, the earth-goddesses Mari Mata and Purvi Mata (Dead Mother and Eastern Mother), the local Sufi pir Waris Ali Shah, and other Muslim pirs of Bara Banki and Lucknow. This constellation of religious practice held alongside, or underneath, the adoption of the traditions of the landlords for whom they labored. Thus, even in Gauhar Lal’s home in rural Sitapur, where his father and grandfather wore the choti in imitation of their brahmin jajmans, Zahir Pir and Lal Beg were honored. And at the home of his in-laws in Bara Banki, employed by Muslim landlords, the goddesses Mari Mata and Purvi Mata were by no means abandoned in the prevailing Islamic environment. Now let us pause and observe that everything in the preceding discussion— the entire exercise of rehearsing and interrogating the enduring trope of the “half Hindu, half Muslim,” and sifting through descriptions of precepts and practices—avoids the question of who the sanitation labor castes understood themselves to be. This is again the trap of colonial-era taxonomy, and the “syncretism” framework that has extended its life in scholarship. When the only evidence admitted is beliefs and practices, the question of subjective belonging is occluded. The nature of the evidence shifts, though, when we turn from accounts of conversion and syncretism in the colonial record to descriptions of “the Cult of Lal Beg.”

“The Most Obscure of All Indian Objects of Veneration”: Lal Beg and the Lal Begi Ummat “It is well known,” wrote Richard Temple in 1884, “that the scavengers, or at any rate a large proportion of them in Northern India, are Lalbegis or followers of Lal Beg, and that they have a religion of their own, neither Hindu nor Musalman, but with a priesthood and a ritual peculiar to itself ” (1884: 529). What does it mean to have a religion of one’s own? Temple, as we have seen, was an early advocate of the “syncretism” interpretation of sanitation labor caste religion, wherein the community’s tradition is a hodgepodge of “the tenets of the Hindus, the Musalmans and the Sikhs … thrown together.”

Deceptive Majority 56

Yet somehow, alongside his insistence on the derivativeness of the tradition, there is something in what Temple observes that gives rise, in his characterization as in others, to phrasings like “a religion of their own.” What provokes such descriptions? What is the quality of own-ness, and of distinction from Hindu and Muslim traditions, that finds its way into even those colonial accounts that simultaneously insist that the sanitation labor castes are “half Hindu, half Musalman”? Or is the problem more fundamental still? Is the question of what “a religion of one’s own” means doomed from the start, insofar as it traffics in the troubled category of “religion”? For some time, it has been argued that the concept of religion, with its European Christian genealogy, distorts and transfigures many of the global phenomena to which it has been applied by the social and humanistic sciences (Asad 1993; Daniel 2000; Josephson 2012; Masuzawa 2005). We noted earlier how native converts to Christianity in South Asia, as indeed in much of the world, did not carve their phenomenological experience into domains entirely commensurate with those their missionary patrons assumed to be universal. The question of what counted as “religious” and what as “political” or “social” was, at the very least, “haunted by problems of translation” between the conceptual categories of the colonial state and its subjects (Adcock 2014: 50). Moreover, the promulgation of an official British policy of religious neutrality in the wake of the rebellion of 1857 made the definition of the religious a subject of hot controversy and an object of political maneuver for the rest of the colonial period. Attempts to construe the religious domain as encompassing caste, and counter-efforts to define caste as a social rather than religious institution, proved particularly consequential for Dalit life under colonialism (Dirks 2001; Viswanath 2014; Viswanathan 1998). Rather than presuming its stability and transparency, then, it behooves us to regard religion as a category troubled—as all categories are—by specific histories of entanglement in human struggle. Alert to these concerns, I will nonetheless, in what follows, venture some assertions about the Lal Begi tradition as a religious formation, and one rightly understood as “of their own”: as autonomous in the sense outlined earlier. What I hold to be “religious” about this formation will reveal itself as we review the Lal Begi oral traditions from the colonial archive. Suffice it to say here that while the evidence does point, in part, to matters of belief—that keystone of Protestant conceptions of religion which, in a telltale sign of the Protestant moorings of the social and humanistic sciences, continues to dominate

57 The Ummat of Lal Beg

accounts of “religion” worldwide—it points with perhaps greater emphasis to ritual practices and to the sense of subjective belonging that animates Islamic conceptions of collectivity: ummat and qaum. Before turning to the songs and stories that circulated among the Lal Begis in the colonial period, it may be helpful to know something more about the axial figure of their tradition. Who was Lal Beg? Waris Shah’s eighteenth century Punjabi romance Heer Ranjha identifies Lal Beg as the pir of the Chuhras, just as ‘Ali is pir of the Shi’a and Ram the guru of the Hindus (Shah 1966). In the late nineteenth century “the god of the sweepers” seized the imagination of the anglophone reading public when he appeared briefly in Rudyard Kipling’s novel Kim as well as in his short story “The Vengeance of Lal Beg,” which tells us that “Lal Beg is the mehter-god, and his image is the Glorified Broom made of peacocks’ feathers, red cloth, scraps of tinsel, and the cast-off finery of English toilette tables” (Kipling 1887, 1901). But Kipling was following an already lively discussion among Indian and European enthusiasts of the burgeoning field of ethnology and folklore— many of them administrators, civil servants, and missionaries—who took an interest in the oral traditions of the sanitation labor castes. Richard Temple, in his capacity as editor of the widely read Panjab Notes and Queries played no small role in provoking debate when he declared that “Lal Beg, owing to the ignorance of the scavenger classes, is the most obscure of all Indian objects of veneration” (Ibbetson 1883). Sayyid Altaf Husen and Richard Greeven, colleagues in the Bengal civil service who “were both nettled at the jealous mystery with which scavengers have hitherto guarded their secrets,” picked up Temple’s gauntlet “by taking the trouble of seeing the ceremonies for ourselves, and having them explained by the sweeper-priests” (Greeven 1894: i–ii). The result of their collaboration, published after Husen’s untimely death, is among the most valuable sources on the sanitation labor castes in the nineteenth century. Yet however detailed the genealogies and legends they collected, Husen and Greeven were unable to establish with any certainty the secular origin story for Lal Beg that they and their contemporaries sought; it was in a speculative mode that they proposed he might be “the sweeper’s garbled version of Ghazi Miyan,” that is, Sayyid Salar Masud, the eleventh century Ghaznavid prince buried at Bahraich in UP (Greeven 1894: i–ii). Denzil Ibbetson, whose work as deputy superintendent for the 1881 census in Punjab gave him a professional interest in the subject, hypothesized that Lal Beg could be “a misreading by Persian-writing Munshis for Bal

Deceptive Majority 58

Mik … which written in a fast-moving hand would appear to be similar.” While the two do share a degree of resemblance—Bal Mik: Lal Beg: —one wonders how the religious life of a then almost universally illiterate population could be determined by a clerical mistake. In his monumental Urdu encyclopedia, Farhang-e Asafiya, Sayyid Ahmad Dehlawi (1898–1918: 165) drew the conclusion that “obviously [zāhiran]” the historical Lal Beg was an educated Mughal of Ghazni who, either due to a contrarian disposition (zidd) or love of a woman (ishq), became a sweeper “and for the sake of his dignity and honor made himself famous as their prophet.” Temple, for his part, speculated that the name was a corruption of “Lal Bekh (bhikshu), the red (or saffron clothed) monk” (Ibbetson 1883). This position was endorsed later by Ambedkar’s close associate Bhagwan Das (2007), who further suggested that Lal Beg was in all likelihood a Buddhist monk, and the shrines built in his memory were originally miniature stupas. Attempts such as these to provide a disenchanted biography of Lal Beg continue (O’Brien 2012). Among his followers, though, Lal Beg’s story was a decidedly enchanted one, and the question of his origins, when posed at all, had a theological rather than historical answer. In the corpus of sanitation labor caste songs and stories documented in the colonial period, Lal Beg is described as a nabi and paighambar—a prophet in Arabic and Persian—as well as a pir and guru, that is, a religious preceptor in Sufi and Sikh/Hindu idioms. He was created out of light (nūr) by Allah (God in Lal Begi songs is Allah, Rabb, and Khuda) and contained in an egg (baiza) before the creation of angels and mankind. He shared the conditions of the sweepers in earthly life and ensured their passage to paradise in the next. His birth was miraculous: he emerged from a pot, or the cloak of a great Sufi saint, or a hair of Allah’s beard, or a drop of Shiva’s semen, or he was born of a barren Mughal woman blessed with fertility by a saint. He was suckled by a hare. Clad in red and given to visions and intoxicants, Lal Beg swept the steps of heaven with a golden broom, drove camels in Kashmir, led his people to conquer Kabul, ate bread baked by the Prophet Muhammad’s daughter Fatima, won praise from ‘Ali, resurrected a dead and eaten horse, rescued the sweepers from the wrath of a king, and will intercede for his people at the day of judgment (qayāmat). Alongside Lal Beg in the oral traditions are several other ancestral and tutelary figures: Pundri, a maiden who herds swine and cures leprosy; Bhadli Halalkhori (or Sati Chuhri), a woman who protects food and chastises the


The Ummat of Lal Beg

gods; the ancestor Khwaja Jhaumpra who bargains with Allah; and others (Lee forthcoming). The oral traditions associate Lal Beg particularly closely with two other names: Bala Shah (or Shah Bala, Bala Shah Nuri) and Bal Mik (sometimes Bal Nek, Bal Rikh, Valmik). The relationship between the three varies from one oral text to the next, but in most of the oral traditions, Bala Shah is simply another name for Lal Beg—they are one and the same. It is for this reason that the terms Shahi—follower of Bala Shah—and Lal Begi are functional equivalents, and why “Shahism” was sometimes offered as another name for the Lal Begi tradition (Youngson 1896: ch.6). Bal Mik, on the other hand, while occasionally conflated with Bala Shah/Lal Beg, is more often a distinct personage. The majority of stories cast Bal Mik as a magician of Ghazni (and father of the swineherd Pundri) who brings about Lal Beg’s miraculous birth, whether out of his own loneliness and desire for a son, or at the request of a barren woman or a sheikh of Multan. If Bal Mik has a narrative double, it is the prophet Ilias (Elijah), who performs the same fatherly function vis-à-vis Lal Beg. After Lal Beg’s advent in the world, he and Bal Mik weather adversity and triumph as they travel with their followers to cities and provinces suggestive of an early Mughal itinerary: Kabul, Kashmir, Kashi, and Thanesar. A few of the folklorists who collected Lal Begi oral traditions in the late colonial period were quick to infer that Bal Mik must be Rishi Valmiki, author of the Sanskrit Ramayana—an idea that, a generation later, would be seized upon by Arya Samaj activists working among the sanitation labor castes in Lahore, and would subsequently spread across north India. We will follow the development of this idea in the following chapters, but it is important to flag from the beginning that it had scant basis in the oral traditions themselves: they simply do not contain references to the Ramayana or its composition, or to Valmiki being a Bhargava brahmin (as he is consistently depicted within the Ramayana text), or to Sita or Lav and Kush, or indeed to any of the narrative tropes associated with the Rishi Valmiki that emerges from the brahminical epic tradition. Acknowledging the multiplicity of unrelated Valmikis in the literary traditions of South Asia (Goldman 1976; Leslie 2003; Sahdev 1997), let us specify this figure—the rhyming rishi of the Ramayana—Valmiki One. Valmiki One is nowhere in the Lal Begi oral traditions. Centuries later, in post-epic bhakti narratives, the same name is grafted onto the ancient Buddhist trope of the contrite highway brigand, earlier associated with the name Anguli Mala (Leslie 2003: chs. 3–4).

Deceptive Majority 60

This Valmiki, the repentant robber—let us call him Valmiki Two—does make appearances in Lal Begi oral traditions, though in these Bal Mik’s change of heart is induced not by brahminical sages but by Guru Nanak. What colonial commentators found irresistible was largely external to the content of the Lal Begi tradition: the apparent similarity of the names (Bal Mik and Valmiki) and the popular tradition of Rishi Valmiki’s mean origins—even though these origins, when specified, were with the Bhil tribe of Rajputana, and not with the sanitation labor castes. Meanwhile the British observers who asserted the identity of the Ghazni magician with the rhyming rishi were either unaware of or ignored indigenous traditions that recognized two or multiple Valmikis, and that actively distinguished Valmiki One from the Bal Mik associated with the sanitation labor castes (Lee 2019). Indeed the dissonance between the oral traditions these colonial commentators recorded and the conclusions they subsequently drew is marked in their writing. Greeven (1894: 67), for example, after transcribing and translating a wealth of Lal Begi songs and stories, discards the exegesis of his informants in order to conclude, “Absurd as it may appear, Balmik is none less than Valmiki, the author of the Ramayana.” Having introduced “the most obscure of all Indian objects of veneration,” his comrades, and the narrative worlds in which they moved, let us turn to

Figure 2.1 Lal Beg shrine (western style) Source: Drawing by Nikhil Mayur.

61 The Ummat of Lal Beg

particular passages in the oral traditions of the sanitation labor castes that illuminate their collective sense of self and other. We might begin with what Lal Begi songs say about one of this tradition’s remarkable features: its shrines. In the late nineteenth century, the shrines of Lal Beg (often simply called sthān or thān) dotted the landscape of a vast swath of the Indian subcontinent. The shrines were small, square, earthen platforms on which were raised either five domes (more common in the east, for example, Benares), or a set of triangular niches in which candles could be set (more common in the west, for example, Gujranwala), or both (in central north India, for example, Saharanpur, Delhi) (see Figures 2.1 and 2.2). In form they partly resembled the marker of a humble Sufi’s grave (the western type) or shrines to the Panch Pir (the eastern type), and were at times mistaken for these (for example, Azad 1907 [1880]: 394). According to one interpretation, the shrine represented the five-domed mosque in Ghazni where Lal Beg prayed as a child (for a detailed discussion of colonial accounts of the shrines of Lal Beg, see Lee 2014). When these shrines are mentioned in the Lal Begi liturgy, they are contrasted with the religious geography of Hindus and Muslims. For example, in one of the songs that Husen and Greeven transcribed when it was performed at the annual feast of Lal Beg in Benares, ‘Ali declares to Lal Beg:

Figure 2.2 Lal Beg shrine (eastern style) Source: Drawing by Nikhil Mayur.

Deceptive Majority 62

Hindu ka dehra, Mussulman ki masjid; Main sewun teri kachchhi marhei The Hindu has his temple, the Muslim his mosque, but I give to you this altar of mud (Greeven 1894: 43, 48)5

Another song, transcribed by the Scottish missionary John Youngson in Punjab, presents a similar sentiment: Jon hindu ganga nu parsann Jon makka mussalmanan Shahi nam tere nu nun mannan Pind pind than banavan As Hindus revere the Ganga, as Muslims have their Mecca, so the Shahis adore your name and build your shrines in every village (Youngson 1906:343)6

These verses put forward a series of indexical relations between places and communities. Two are familiar and instantly legible to the dominant historiography: the temple (dehra) and the river Ganges (ganga) index Hindus, while the mosque (masjid) and Mecca index Muslims. But to this the Lal Begi songs add a third: the humble earthen shrine (than, kachchhi marhei) indexes the Lal Begis/Shahis. Alongside places, religious practices, too, serve to contrast Lal Begis from Hindus and Muslims in oral traditions. In a text from a village near Sialkot we find Khwaja Khizr, the long-lived, green-clad Islamic saint, saying to a 5


For purposes of clarity I have retained the spelling but set aside the system of diacritics used by Greeven and other colonial sources cited in this chapter, based as they are on conventions long out of use. Greeven’s translation of this verse is: “Yet leave thou unto the Hindu his temple, and unto the Moslem his mosque. For thee will I maintain a clay-built tabernacle….” Youngson’s own translation is: “The Ganges Hindus fear, and Muslims make their weary pilgrimage to Mecca far, but thee the Shahis love and build to thee unnumbered shrines o’er all the crowded land.”

63 The Ummat of Lal Beg

progenitor of the sanitation labor castes named Jhaumpra (literally Hut or Shack): Tere shahi dan denge main nun pas bahan Jhalak laggega chandoe da munh pani lan Shahian de bihishti jan da ih pakka nishan Na roza na ashtami na tur makke jan Your Shahis, with me close, will offer alms When water, lifted to the mouth, flashes like silver It is a sign that the Shahis go to paradise Without fasting [for Ramadan] or observing Ashtami [a Hindu fast] or going to Mecca (Youngson 1907: 27)

Or, in another version of the Jhaumpra-Khwaja Khizr dialogue: Shahion nun farmana Na main varton asthmi Na tur makke jana Allah alif saman da Sabhho ih biyan Command the Shahis: Neither do I fast for Asthami Nor make pilgrimage to Mecca Allah who is first like the letter Alif Gives this one statement (Youngson 1906: 351)

In these songs it is the fast of Ashtami (the “Eighth” of the Hindu lunar month) that indexes Hindus, while it is Roza (the Ramadan fast) and the Hajj that index Muslims. Shahis/Lal Begis are set apart from both by their abstinence from these practices. Another basis for the Lal Begi assertion of religious distinction appears in a description of an incident in a village in western Punjab when a Chuhra intervened in a discussion between a Muslim and a Christian missionary about whether there was any profit in proselytizing Chuhras. The missionary was John Youngson, whose meticulous transcriptions of the songs, stories, laments, and genealogies of the sanitation labor castes among whom he lived from 1891 to his death in 1920, are, alongside Husen and Greeven’s collection, among

Deceptive Majority 64

the most valuable sources from the period. Unfortunately, Youngson’s account of the incident we are about to consider, which is taken from his memoirs, is not characterized by the same attentiveness to the vernacular as are his transcriptions of Chuhra oral traditions published in The Indian Antiquary. Problems of translation and paraphrasing mar Youngson’s account, and we have no way of knowing precisely what words the Chuhra in the story uses. Nonetheless, for a Chuhra in rural Punjab in the 1890s to have conveyed even the general thrust of this passage is striking. Here is what Youngson reports the Chuhra to have said: Our religion is the best of all. Did you ever see a clean village where there were no Chuhras? What sort of a place would Heaven be without us? The Chuhra is better than the Hindu, at any rate. I will prove it. If a Hindu’s cow dies during the night the Hindu that tied her up the evening before is excommunicated, and must wash his sins away in the Ganges. He goes on foot, with the cow’s tail round his neck, and he begs as he goes. Arrived at the Ganges he is beaten by Chuhras with a shoe seven times, more or less severely, according to the amount of money he gives, while they cry, “Gao hatya!” (Cow murder!). Now when the beating with Chuhras’ shoes takes away sin, it is demonstrated that the Chuhras’ religion is the better, for the Hindus’ religion could not do it. (Youngson 1896: 226–28)

The custom described here is one of a number of historically attested practices in which the radical impurity ascribed to the sanitation labor castes is mobilized in a punitive or, counterintuitively, purificatory rite. Exploiting the ambiguity of the custom, the Chuhra speaker—as Youngson represents him at least— uses it to argue for the superiority of his “religion” to that of the Hindus in the latter’s own terms: sin and purification. Some of the clearest assertions of Lal Begi difference from other religious communities are soteriological in nature. There are a number of stories and songs that either imply or explicitly state that Hindu, Sikh, and Muslim paths to salvation are false or inferior to that made possible through Lal Beg. Within the tradition, Lal Beg is the principal mediator between God and humanity, whose intercession secures entrance to heaven. Karma and rebirth make no appearance here. The question of Muhammad’s place in Lal Begi theology is ambiguous; in some songs he appears in catalogs of the praiseworthy agents of God in an unambiguously positive light. Yet, while the question of the finality of Muhammad’s prophethood is not explicitly raised, the implication

65 The Ummat of Lal Beg

that his message has been superseded by that of Lal Beg is difficult to escape. Thus Husen and Greeven report the following “common anecdote” from late nineteenth century Benares: “[A] Lalbegi, when asked whether Mussulmans could obtain salvation, replied:—‘I never heard of it, but perhaps they might slip in behind Lalbeg’” (Greeven 1894: 8). Elsewhere the soteriological distinction is drawn more sharply, as in several verses in a book Youngson consulted in the village of Kharolian, which tell of the followers of Bala Shah/Lal Beg attaining paradise (bihisht) while Musallis—who, it will be recalled, are Chuhra converts to Islam—are admitted to hell (narake dakhil honge) (Youngson 1907: 76, 107). The Lal Begi path of salvation was likewise described as preferable to the Sikh path. In Punjab around 1920, a Chuhra informant explained to Herbert Strickler (1926: ch. 3): “If any follower of Bala Shah (a Shahi) shall repeat the Muhammedan creed, he is to be counted an unbeliever; if Baba Nanak’s, he is to be cast off. But all that profess the creed of Bala shall go straight to heaven (dargahe).” In terms of soteriology, then, as well as with reference to sacred places (Ganga, Mecca, temples, mosques) and ritual practices (fasting for Ashtami and Ramadan, brahminical purification rites), Hindus, Muslims, and Sikhs appear, in Lal Begi oral traditions, as others against whom the collective self is defined. Alongside these articulations of alterity, the sources also give utterance to a positive sense of group selfhood. We have already seen how the terms “Shahi” and “Lal Begi,” used in contrast with Hindus and others, denote a collective subjectivity. There is also, in the oral traditions, a striking pattern of usage of the terms ummat (ummah) and qaum. These terms, often rendered “nation,” “tribe,” or “religious community” in translation, are concepts of collectivity grounded in and indexically representative of the historical experience of the Islamic community—the original and paradigmatic ummah—and that suggest strong bonds of cohesion and solidarity. Lal Beg and several progenitor figures in the Lal Begi pantheon are described as inaugurating the ummat or qaum of Shahis/Lal Begis. Mere shāh jī kī ummat baṛhāyā (My great Lord’s community prospered/expanded)—so recites the Lal Begi bhagat in a liturgical song transcribed by Husen and Greeven in Benares (to which the assembled Lal Begis reply with the refrain Bolo, momino, vohī ek: Say, o believers, it is the one) (Greeven 1894: 42). A similar verse in a kursinama, or genealogy, that Lal Begis of Ambala and Karnal districts shared with Richard Temple reads sarvat ummat pāī, which Temple (1884: 531), imparting a Christian flavor to the Islamicate original, renders: “The Saviour hath obtained a following.”

Deceptive Majority 66

An origin myth that appears with startling regularity in the earliest documented oral traditions of Dalit communities across the subcontinent has it that untouchability began when one of four brothers was made, often through an act of deception, to remove a cow’s carcass from the presence of the remaining three (Deliege 1993; Prakash 1991). This story appears in the Lal Begi corpus as well. In these versions, the outcasted brother (or his son and daughter-in-law) confronts the brothers with the injustice of their action before setting off to live a separate and cow-eating life. At this point in the story, in a telling that Youngson transcribed, the narrator says: Te qaum hoi ih wakhri, “Thus a separate qaum arose.” This community or tribe (qaum), separate or different (wakhri) (B. M. Singh 1982 [1895]: 1190) from the offspring of the duplicitous group that exploited their brother’s loyalty and labor, would become the Lal Begis. But the story does not end with exile and abjection. In the oral traditions of the sanitation labor castes, the excommunicated brother—who is none other than Khwaja Jhaumpra, the Lal Begi caste progenitor—takes his case before Allah himself. Acknowledging Jhaumpra’s petition Allah affirms that there is nothing immoral (haram) about eating the flesh of the cow. The dialogue that then ensues between Jhaumpra and Allah draws together several of the threads we have been following: deployments of “Hindu” and “Muslim” as contrastive from the collective self, usages of ummat and qaum to characterize the Lal Begi community, and a concern with ritual practice and divergent paths to salvation. Jhaumpra remonstrates: Mainun Hindu na nere aun denge Mussalman na parhnge janaza Meri kaun shifa’at bharega Tu sun khuda raja Main ummat rakhna chahunan Hindus forbid my approach Muslims refuse to bless my dead Who will stand witness with me? You listen, lord god I want to establish an ummat

The sting of Jhaumpra’s exclusion from the society of his brothers in mythical time is here merged with the anguish borne of maltreatment by Hindus and

67 The Ummat of Lal Beg

Muslims in lived Lal Begi experience, and juxtaposed with a yearning for community. Allah replies: Roz qiyamat waqt de Tainun milegi vadidi ... Ram te Rahim kian Chhap chhap jana Sava neze te din avega Haoe dozakh pana Par bihisht banake Samhne vikhana Ummat teri bhajjke Bihishti var jana On the Day of Judgment You will obtain greatness … The partisans of Ram and Rahim Will be scattered When the sun sinks to a length-and-a-quarter Hell will be their path But to you I will show paradise Which I have created beyond Your ummat will swiftly enter Into this paradise (Youngson 1906: 350–51)

Precious little in the historiography of Indian religions prepares us for findings like these. From one direction our expectations are molded by the ubiquitous common sense that “untouchables … are Hindu,” that “low-caste Hindu” and “untouchable” are synonyms. From another, we are led to anticipate that all anti-caste thought must follow the boldly modernist path charted by Phule and Ambedkar. From a third, we are told that caste subalterns of the colonial period did have their own, endogenous, oppositional consciousness, but one which is “implicit, barely stated” in their oral traditions, and whose “marks, faint as they are, of an immanent process of criticism and learning,” must be rescued from inchoateness by careful scholarly decoding (Chatterjee 1993: 197). But what the Lal Begi tradition enunciates in passages like these is neither implicit

Deceptive Majority 68

nor inchoate, nor modernist, nor reflective of a sense of self that can in any fair way be called Hindu. The most sensitive scholarship prepares us to find “ambiguity” in the religious history of the subordinated castes, and often for good reason. But the voices that speak from the archive of sanitation labor caste thought can also, at times, be all too clear. Such is the dialogue of Jhaumpra and Allah, and the latter’s pronouncement that on the Day of Judgment the long-suffering Lal Begis will attain paradise while their earthly tormentors, the followers of Ram (Hindus) and Rahim (Muslims), will find themselves in hell.

Reactions to “the Hindu Chuhra” One part of the answer to the question with which we began this chapter is now beginning to take shape. When Shankar, showing me around his neighborhood of Mirzapur, off-handedly said, “That house belongs to a Hindu. All the others here belong to our caste,” he voiced a discourse with a history. I was not imagining things when I started scribbling “Contrastive usages of ‘Hindu’” into the margins of my fieldnotes; such usages, in fact, were standard in the stories and liturgical songs that circulated among my interlocutors’ not-so-distant ancestors. The view of the world put forward by Maya, Urvashi Butalia’s interviewee in Punjab who pilfered Hindu and Muslim homes as a child during Partition, was neither idiosyncratic nor new, but in meaningful ways continuous with the social ontology developed and given expression in the Lal Begi tradition, as documented in the decades when Maya’s grandparents would have been alive. Let us turn now to the other side of the question. If the sanitation labor castes in the late nineteenth and early twentieth centuries spoke of Hindus and Muslims as others, how did the latter view the former? Butalia (1998: 298–300) notes that “like all Hindus … I had assumed that Harijans (Dalits) … were part of the Hindu community, part of ‘us’.” Was this assumption operative at the turn of the twentieth century? Among Hindus, was it common sense, in the 1880s or 1910s, that sweepers and other “untouchables” were “part of ‘us’”? A systematic treatment of this question would be of immense value for the study of South Asia, but it is beyond what can be attempted here. Rather, let us consider just a few glimpses that the colonial archive affords us of what some classes of Hindus had to say about the “untouchables” that the colonial state had begun to categorize as their co-religionists. Since Dalits were rarely discussed as a pan-Indian collective before the twentieth century, we will need to consider sources that speak to Hindu

69 The Ummat of Lal Beg

perspectives on particular “untouchable” castes, rather than Dalits as a whole. Of the “Pariah” community in southern India, the editor of the influential daily The Hindu wrote in June 1891 that “[t]he Hindus do not recognize them as part of their community and nothing can be more humiliating and intolerable than the treatment that the Pariahs … receive from the Hindus of higher castes” (Viswanath 2014: 1). While the formulation here leaves the door open to the colonial categorical scheme—if those who mistreat the Pariahs are “Hindus of higher castes,” rather than simply Hindus, the reader may still be allowed to imagine Pariahs as also Hindus in some sense—the editor is unambiguous that, in terms of a subjective sense of belonging and difference, Hindus do not see Pariahs as their own. Regarding the Chamar community in central India, census officials H. V. Russell and Rai Bahadur Hira Lal (1916: 315) observed in 1916 that in Chattisgarh to call a man a Hindu conveys primarily that he is not a Chamār, or Chamara according to the contemptuous abbreviation in common use. A bitter and permanent antagonism exists between the two classes, and this the Chamār cultivators carry into their relations with their Hindu landlords by refusing to pay rent.

Again we encounter, in reportedly popular usage, “Hindu” as a term contrastive to, rather than encompassing of, Dalit communities. One group of Hindus whose view of “untouchables” made a clear mark on the colonial archive is those who were government servants involved in the decennial censuses. Many were provoked when the colonial state obliged them, in what Vasudha Dalmia (2007: 14–15) rightly considers one of the most consequential administrative decisions to be taken in South Asia in the modern period, to categorize untouchables as Hindus by default, that is, unless they declared themselves Muslim, Christian, or Animist (Kaul 1911: 97; Prashad 2000: 67–68). This move put the enumerator and the Dalit into the same religious category, officially representing them as co-religionists. As the author of the report of the Census of 1881 noted (in characteristically prejudiced terms): [M]any of the more bigoted high caste Hindoos employed as census enumerators or supervisors objected to record such low persons as of the Hindoo religion. This was illustrated by numerous instances brought to my notice of such persons having been recorded as of the Dher, Mang or

Deceptive Majority 70

Chandal religion by mere repetition of their caste in the column for religion…. [P]robably they were not even asked [what their religion was]. (Plowden 1883: 17)

Ten years later, in the 1891 census, enumerators demonstrated a “marked disinclination to apply the term Hindu to members of the lower and impure castes, such as sweepers and village menials” (Baines 1893: 158). Again in 1901, the superintendent of census operations in Punjab noted that “Hindu enumerators have the greatest objection to returning [the Chuhra] as a Hindu by religion, and so he is often either entered as a Chuhra or as a Mussalman” (Rose 1902: 113). In 1911 as well it was remarked, for north India generally, that Hindu census workers were loath to identify members of the sanitation labor castes as their co-religionists (Gait 1913a: 115). Census officials were aware of the irony of what they had been tasked to do. Charged with implementing the imperial government’s decision to categorize “untouchables” as Hindu by default in this principal technology of colonial knowledge, several administrators remarked on the policy’s chafing disjunction from the social imagination of those represented in the resultant “data.” Census commissioner E. A. Gait, for instance, wrote in 1910, “It is obviously absurd to enter without comment as Hindus persons who do not worship the Hindu gods and are not admitted to Hindu temples, and who are not regarded by others and do not themselves profess to be Hindus” (quoted in Dalmia 2007: 14). He noted further that the census data represented as Hindus even those “who actually objected to being so classed” during enumeration, such as Satnami Chamars (Gait 1913a: 114). In describing the “scavenger castes” for his report on the 1881 census in Punjab, Denzil Ibbetson introduced “the Hindu Chuhra,” a category sufficiently strange to necessitate an immediate qualification: “[T]hat is to say the Chuhra who follows the original religion of the caste and has been classed by us as Hindu” (Ibbetson 1970: 294, emphasis added). Did Hindus in Punjab—aside from census enumerators—accept the new categorization? Ibbetson (1970: 268) remarked that in spite of the Chuhras’ mimetic adoption of the outward observances of local dominant castes—despite, in particular, their taking up burial when being employed by Muslims and cremation when serving Hindus—they were nonetheless “not recognised by their masters as either Hindu or Musalman.” Similarly, Husen and Greeven in Benares found that while some of their “sweeper” respondents claimed to be Muslim, Hindu, or Sikh, “[t]hese pretentions are, however, equally rejected by Hindus, who

71 The Ummat of Lal Beg

exclude them from temples, and by Mussulmans, who exclude them from mosques” (Greeven 1894: 7–8, 38). Thus, whether we take the sanitation labor castes in particular or Dalit communities more broadly, a preliminary review of evidence suggests that significant classes of Hindus at the turn of the twentieth century did not imagine such people to be their co-religionists, as fellow Hindus. What, though, of Muslims? Given the depth of Lal Begi engagement with the Islamic narrative and theological imagination, and the fact that many if not most Lal Begis bore musalmānā names in the period under question (Shyamlal 1984: 30), one might probe further the Muslim side of observations like those of Ibbetson and Greeven. Might not the degree of Lal Begi investment in their tradition have moved some Muslims to acknowledge between them a brotherhood of sorts? Some colonial administrators assumed as much: in 1901, the census commissioner for the North-Western Provinces and Oudh described the Lal Begis as the third largest sect among Muslims, after Sunnis and the Shi’a (Burn 1902: 96). Again, no systematic enquiry is possible here, but two examples hint at a range of ideas current at the time. In his Ab-e Hayat (Water of Life), published in 1880, the Urdu literary historian Muhammad Husain Azad relates an anecdote from the life of the Delhi poet Shah Nasir (d. 1829–30) intended to illustrate the liberal quality of the poet’s piety. In his peregrinations, whenever Shah Nasir came across a niche (tāq, mokhā) before which a garland of flowers was arranged, he would stop and reverently recite the fātiḥa, to which the orthodox among his apprentices objected. Azad explains that once, [i]t even came to this that an apprentice knew [the place] and said, “Teacher! I know that opposite this is the home of a Halalkhor and this is the Lal Beg niche (tāq) that he built!” At that point Shah Nasir chuckled and said, “Well, I have recited the word of God. Its blessing cannot just vanish into the air. Where there is a place for it, it will settle. My good deed has not been in vain.” (Azad 1907 [1880]: 394)

Here the Lal Beg shrine, an alarming sign of idolatry (shirk) to the orthodox apprentice, provides a site for Azad to have his broad-minded poet exhibit his liberality. Yet his remarks clearly register religious difference—Shah Nasir concedes that a Lal Beg shrine is not an appropriate place to utter the fātiḥa, and by implication that the Halalkhor/Lal Begi who maintains the shrine is not a fellow Muslim.

Deceptive Majority 72

A less generous averment of the exteriority of the sanitation labor castes from the brotherhood of Islam comes from Mirza Ghulam Ahmad, founder of the Ahmadiyya sect. When the Punjab Census Report of 1901 suggested that Ahmad had a particular following among the sweepers, he vehemently denied association with that “criminal community [jurāim pesha qaum]” and “most degenerate class of people,” representing them not merely in terms of class contempt but also as a category apart from Muslims, even “low Muslims of inferior morals [adna darje ke musalmān aur razīl safāt]” (Ahmed 1904, 1984).

Antecedents of Alterity The social orderings spoken of by Hindus, Muslims and Lal Begis in north India in the late nineteenth and early twentieth centuries were no less deeply felt, and no less traversed by axes of difference, than those that animate the everyday politics of caste and religion today. But they were certainly structured differently. What glimpses we have obtained of Muslim and Hindu perspectives on the religious standing of the sanitation labor castes, and of Lal Begi perspectives on their own ummat and its relation to the religious communities of others, suggest that the categorical schema that enjoys the status of popular and scholarly common sense today, and that a vast edifice of twentieth century representations portrays as timeless, would have struck as outrageous many of the people whose religious life in the colonial period it retrospectively presumes to classify. The Partition “joke” that Ravinder Kaur relates—the one about the cool indifference of Lahore sweepers to Partition violence since they were in no danger of being seen as Hindu or Muslim—turns out to have a better grasp of the socioreligious landscape across a wide swath of South Asia in the colonial period (and, in more fugitive ways, beyond) than the disquisitions on social structure in our Hinduism textbooks. We have only begun to sense the contours and qualities of the social ontology articulated in Lal Begi oral traditions of the colonial period. How deep into history does it reach? How does it relate, moreover, to the understandings of self and other that obtained among subordinated caste-clusters other than the sanitation labor castes—how did they imagine and speak of their relations with Hindus and Muslims? Groups like the Satnamis of central India continue to be written about within the framework of “marginal Hinduism,” even when we know that many of the former actively objected to being classified as Hindu in the census (Dube 1998). If we return to the archive of subaltern religion


The Ummat of Lal Beg

disabused of the common sense of the present, attentive to the presence of concepts like qaum and ummat in the speech of the subaltern, and alive to questions of collective self-description, what might be discovered? Let us conclude this chapter by noting recent findings that suggest that the colonialera Hindu perspective we sketched above—that the sanitation labor castes, and Dalits in general, were decidedly not “part of ‘us’”—had firm footing in the categorical schema of polities that preceded and continued alongside the colonial state. In her extraordinarily rich study of the adjudication of caste and gender by the precolonial Rajput state of Marwar, Divya Cherian describes a late eighteenth century royal edict that helps throw into relief just how sharp a break from local precedent was the British practice of clubbing Hindus and Dalits into the same category. The royal order, issued by the Marwar crown in 1785, mandated a particular recitation practice for all Hindus. In doing so, the edict categorically divided the kingdom’s subjects into two kinds: Hinduvan (Hindus), for whom the practice was to be mandatory, and Achhep (untouchables), for whom it was forbidden (Cherian 2015: 90–92). The order went further and enumerated which castes were Achhep: nomadic pastoral groups, leatherworking castes (ḍheḍh, chamar), Muslims (turak), and the sanitation labor castes (halalkhor). In the eyes of the Rajput regime, not only were the sanitation labor castes not Hindu, they were the antipode of the Hindu; the actions that the state required of its Hinduvan subjects were precisely those that it forbade its Achhep subjects ( Cherian 2015: 90–92). Notably, moreover, Muslims in this schema were not the primary “other” against whom the Hindu self was constructed, as the twentieth century bifurcation paradigm has primed us to expect. The turak, rather, was a subset of the Achhep; Muslims were secondary “others” assimilated to the brahminical tradition’s antecedent category of alterity, the Chandal, the paradigmatic untouchable. Was Marwar alone among precolonial polities in conceiving—and enforcing a conception of—the socioreligious landscape in this way? The appearance of Muslims as a kind of Achhep has precedent in earlier developments in brahminical normative literature; Medatithi, commenting on the Manusmriti, explains that mlechhas were to be regarded as Chandals (Aktor 2010: 44–53). In the Maratha empire, in which the dharmaśāstra tradition was explicitly drawn upon for legislative purposes, the Peshwa issued orders enforcing the exteriority of “Atishudra” (that is, “untouchable”) subjects to Hindu space and ritual (Fukazawa 1991: ch. 4). Similar thinking is in evidence in Nepal’s

Deceptive Majority 74

legal code of 1853 and in the legislation of caste slavery in nineteenth century Travancore (Gaborieau 1972: 85–86; Mohan 2015). More attention to this question is needed, but there are hints that the categorization of Hindus and “untouchables” as antipodes extended across several kinds of regimes, not only those like the Peshwas that cited dharmaśāstra in their rulings but also those like the Rathors of Rajputana that did not. It may be that what Cherian (2015: 91) says of the 1785 edict held beyond Marwar—that it was not an isolated measure, but rather part of a broad effort to enforce a vision of society properly ordered, an order in which “Hindu-ness … was defined in opposition to the ‘untouchable.’”

II Making “Untouchables” Hindu or The Great Interpellation

3 Missionary Majoritarianism The Arya Samaj and the Struggle with Disgust

c. 1928 Arya Samaj tract addressing the sanitation labor castes: “In the census you were returned as Hindu. You people are Hindu, your religion is Hindu, the Veda is your holy book, Balmiki ji is your guru. So those of you who circumcise, who marry and bury by Muslim rites, are perpetrating a great sin.” (S. Sharma n.d.: 23) 1933 Gandhi, discussing temple-entry legislation with Ambedkar: “But I must say that you ought not to say that you are not a Hindu. In accepting the Poona Pact you accept the position that you are Hindus.” Ambedkar: “I have accepted only the political aspect of it.” Gandhi: “You cannot escape the situation that you are Hindus in spite of your statement to the contrary.” (Ranga 2000: 131) 1950 Government of Independent India: “[N]o person who professes a religion different from the Hindu religion shall be deemed to be a member of a Scheduled Caste.” The Constitution (Scheduled Castes) Order, 1950

From Alterity to Identity In the previous chapter we saw that the nineteenth century British innovation of categorizing the “untouchables” as Hindus contradicted prevailing perceptions of those so categorized. Hindu census enumerators objected to recording untouchables as co-religionists, Lal Begis spoke of themselves as a religious community (qaum) separate from and oppressed by the Hindus, and colonial administrators charged with implementing the new policy described it as “absurd.” How is it, then, that the central premise of the new policy attained,

Deceptive Majority 78

by the latter half of the twentieth century, hegemonic status? When neither untouchables nor Hindus perceived each other as co-religionists in 1900, how did it come to pass that virtually everyone presumes them to be members of the same religious community a hundred years later? As we have seen, scholars over the decades continue to be startled when they discover that Dalits whom they meet in remote corners of north India—Dalits who are neither converts to Islam or Christianity nor committed to a repudiation of Hinduism following Ambedkar’s example—speak unselfconsciously of Hindus as others. But when did this become a matter of surprise? How did a notion that apparently remains foreign to some of the very people about whom it speaks become a foundational assumption for everyone else? The preceding epigraphs mark stages in the historic transformation in question, the transformation that the next three chapters will chart: the Arya Samaj’s campaigns to Hinduize Dalits in the 1920s, the “untouchable uplift” efforts of Gandhi’s Harijan Sevak Sangh in the 1930s, and the exclusion of Dalits who “profess a religion different from the Hindu religion” from statutory positive discrimination by the Congress Party upon Indian independence. They are thus mileposts in the progress of what I will argue was a project of majoritarian inclusion: a dominant group’s forging or maintaining a majority status vis-à-vis a perceived rival by the induction of a heretofore despised outgroup. The epigraphs also suggest distinct modes of majoritarian inclusion. The first, an Arya Samaj tract targeting the sanitation labor castes, asserts with passionate redundancy that its addressees are Hindu, only to belie its own certitude by the admission that some among them circumcise and practice Islamic rites of marriage and death. This, with true missionary conviction, it labels as sin. In the second epigraph, when Gandhi makes the same assertion—“you are Hindus”—he, too, meets resistance; but rather than seeking to convince Ambedkar he simply declares him overruled, insisting on Ambedkar’s Hindu-ness “in spite of your statement to the contrary” (emphasis added). The third formulation retains the indifference to criticism of the second, and endows it with the authorless authority, fixity, and force of official state discourse. We glimpse, then, a spectrum of modes through which majoritarian inclusion was brought about, from the dialogical exhortation of the proselytizer to the monological fiat of the legislator. These three utterances offer clues to the answer of our question, the question of the ascendance to sociological common sense of the idea that “untouchables” were Hindus, an idea that was scarcely thinkable a generation

79 Missionary Majoritarianism

earlier. This chapter and the following two chart the career of this idea at the hands of the individuals and institutions that most successfully purveyed it in the crucial decades of the 1920s, 1930s, and—in Lucknow—in the years following independence. Through a careful reading of key texts written by the Arya Samaj and Congress leader Swami Shraddhanand, by his bête noire Khwaja Hasan Nizami, by grassroots Arya Samaj prachāraks working among the Chuhras (in this chapter), and by Gandhi, his supporters in the journal Harijan, and his lieutenants in the Harijan Sevak Sangh (in chapter 4), I hope to illuminate the conceptual, representational, and political labors by which the despised antipode of the Hindu was resignified as the Hindu’s own, and the affective labors by which this resignification was sought to be worked into embodied dispositions. Having tracked this project at the national level from its early days on the radical fringe to its insertion into the political mainstream and to its elevation to the law of the land, we will then turn to Lucknow (in chapter 5) for a detailed study of how Hindu majoritarian inclusion played out there in the late 1940s and beyond—attending both to the convulsive transformation of relations between communities generated by the project, as well as to the limits of social restructuring and the recalcitrance of older practices and ways of life. This is a story in which streams of Indian nationalism often imagined to be both separate and opposed—the Hindu nationalism of the Arya Samaj and Hindu Mahasabha, and the Gandhian nationalism at the heart of the Indian National Congress—are seen not only to flow in parallel channels, but at points to commingle and converge. To draw attention to this convergence is not to ignore or underplay a host of crucial distinctions, most centrally those between Gandhi’s advocacy of a politics that embraces the religiously plural condition of the nation, and Hindu nationalists’ pursuit of a politics that would purge the nation of religious difference. The consequences and legacies of such cleavages in political vision are as significant as they are well known. What is far less well recognized is that, on the caste question, Hindu nationalists and Gandhian proponents of religious tolerance—figures on opposite poles of debates over communalism—joined ranks. On the necessity of securing the Depressed Classes as Hindus in the political framework of the emerging nation, and, to that end, of undertaking “untouchable uplift” programs that inculcated Hinduism in its beneficiaries and narratively constructed them as (lost or fallen) Hindus, the Gandhians and the Hindu nationalists, for all their substantive differences in other domains, found themselves in strong accord.

Deceptive Majority 80

I mean to map, in this and the following chapters, the grounds of this accord. Under what conditions did the language of alterity—of the “untouchable” as the antipode of the Hindu—yield to the language of identity? How did Swami Shraddhanand, Gandhi, and rank and file Congress and Arya Samaj activists frame their inclusionary overtures toward Dalits? How did they justify their action to Hindu skeptics? What provoked and sustained the “untouchable uplift” campaigns of the 1920s and 1930s, and what ideas and practices were shared among Gandhian and Hindu nationalist institutions engaged in it, despite at times sharp public disagreement between their leaders? While emphasizing convergences, I will also draw a significant distinction between the Arya Samajists’ mode of majoritarian inclusion and that pursued by Gandhi and some of his Congress followers. As hinted at in the opening epigraphs, the latter adopted a monological style of speaking for and speaking over, of enacting the newly imagined community through performative utterances—nominations, in this case—that depend for their efficacy not on the assent of those named but on the categories and operations of the modern state. In terms of affect, the Gandhian mode of majoritarian inclusion promoted social relations modeled on the narrative world of bhakti religiosity, where the supplicant meekness of the Harijan is answered by the benevolent compassion of the Hindu. The Arya Samajists, while also drawing on bhakti tropes in an effort to interpellate Dalits as Hindus, did so in a manner more exhortatory, dialogical, and responsive to the moral and material concerns of the objects of their “uplift.” For Shraddhanand and his fellow Aryas, moreover, it was not enough to give utterance to the imagined community or to appeal to the kindliness of Hindus or the logic of the state to bring it to fruition. Rather, they acknowledged and sought to uproot in themselves an embodied, visceral impediment to the project of inclusion that Gandhi’s monological approach simply dodged. For the Arya Samajists, majoritarian inclusion necessitated a mighty struggle with their own disgust. Or rather, a struggle with their ghṛṇā, an affect in many ways comparable to the anglophone emotion-concepts of disgust, revulsion and abhorrence, yet also distinct in its linguistic usages and cultural presuppositions. Ghṛṇā, as I have argued elsewhere, regulates boundaries on two distinct planes of experience, the material-corporeal and the social-moral (Lee forthcoming). It springs into action when the body’s contents are exuded or sloughed off, when hair and nails shed or when the inside reveals itself to the outside as pus, blood, saliva, mucous, vomit and excrement—when the body envelope

81 Missionary Majoritarianism

leaks and presents the world with what, adapting the well-known insight of Mary Douglas (2006), we might call corporeal matter out of place. At the same time that it patrols the perimeter of the self, ghṛṇā also guards the frontier of acceptable social relations, warning against contact with socially or morally abject categories of person. In north India to ask who feels ghṛṇā for whom is often to be guided along the boundary protecting touchable society from its notionally polluted exterior: vegetarian from meat-eater, clean from unclean, savarn from Dalit. The homology between ghṛṇā in the two domains draws sustenance from a labor economy and spatial organization that relentlessly juxtapose waste matter and the social group expected to both handle and live alongside it. Practices of what I call environmental casteism—from dumping toxins in Dalit neighborhoods to providing municipal housing for sanitation workers between rubbish depots and public toilets—forge in the lived experience of inhabitants of caste society indexical chains bonding together material and social elicitors of revulsion (Lee 2017). Ghṛṇā on the social-moral plane is thus inculcated at least as much through the implicit pedagogy of space and the everyday experience of an unflaggingly caste-based division of stigmatized labor as it is by the explicit passing forward of casteist ideology. It is inculcated, moreover, not merely as reaction to external stimuli—as disgust is often imagined to operate in anglophone theories of emotion—but also as an active practice aimed at a given object. In English you are disgusted by something, whereas in Hindi-Urdu ghṛṇā either can be reactive or, joined with the verb karnā (to do), can function as a transitive conjunct verb: you do disgust, you practice disgust, intersubjectively conveying your revulsion at the ostensibly disgusting thing, person, or group. This formulation usually appears in the present indefinite, suggesting not a singular episode but a constant practice, a habit. It is this usage of ghṛṇā that appears in what is among the most ubiquitous remarks in everyday speech in Dalit neighborhoods in north India on the continued prevalence of untouchability: ve ham se ghṛṇā karte haiṅ—they practice disgust on us, they revile us. The collective and conventional character and outwardly-directed force of ghṛṇā are indicated by this utterance, as is, more broadly, the anchoring of everyday caste relations in the viscera. Oral traditions among the sanitation labor castes in the colonial period feature a number of stories that foreground and critique the advantaged castes’ cultivation of ghṛṇā toward Dalits—and indeed toward other forms of life (Lee forthcoming). These oral traditions anticipate later anti-caste theory that posits untouchability as “a kind of

Deceptive Majority 82

repulsive feeling, a sort of nausea, that sits deep at the bottom of Brahminical mind” (Guru 2009a: 50). As Arya Samaj activists in the twentieth century took up the effort to reconceive of Dalits as their Hindu “brothers,” they too were compelled to contend with the ghṛṇā they had been socialized to feel toward this category of person, and the difficulty in unlearning a habit of revulsion lodged deep in the gut.

Shraddhanand and Shuddhi: Early Years Until recently, virtually all studies of the Arya Samaj’s early twentieth century shuddhi or “purification” campaigns among Dalit populations accepted, more or less uncritically, the Samaj’s own characterization of the process as one of “reclaiming” “low-caste” or “outcaste” “Hindus” (or “the lowest segments of Hindu society”), reproducing, if in somewhat hazy terms, the idea that Dalits were primordially Hindu, that they had enjoyed a previous condition of Hinduness to which they could sensibly be said to have been re-claimed (see, for example, Graham 1943; Jones 1976; Jordens 1977, 1981; Rai 1915). Scholars in recent years have increasingly brought non-elite sources to bear on this period, being led thereby to train a more skeptical eye on the Samaj’s framing of Dalit religious history (Adcock 2014; Rawat 2011; Sikand 2003). In fact, as the previous chapter of this book demonstrates, the language of “reclamation” rides roughshod over the sanitation labor castes’ understanding of themselves as reflected in Lal Begi liturgy and oral traditions; moreover, “reclamation” obscures the radical novelty of what the Arya Samaj was attempting in these decades. A sketch of the history of shuddhi and its tireless exponent Swami Shraddhanand will throw into relief just how ruptive this effort was, and will illuminate the ways in which it was a project of majoritarian inclusion. This brief history will be followed by a more detailed analysis of Arya Samaj tracts composed and deployed in the majoritarian project, as well as one counter-tract by Khwaja Hasan Nizami, Shraddhanand’s rival and proponent of Islamic tabligh (propagation) among Dalits. These are previously untranslated Hindi and Urdu texts that detail the specific challenges posed by the sanitation labor castes and by the problem of ghṛṇā. Founded in 1875, the Arya Samaj was an organization committed to the transformation of Indian society on the model of an imagined Vedic Aryan past. As Vasudha Dalmia (2010: 1–10) argues, each of the various labels scholars have applied to the Samaj—“fundamentalist,” “neo-Hindu,” “revivalist,” “militant,” “Hindu nationalist,” and, especially, “Hindu reformist”


Missionary Majoritarianism

(Adcock 2014; Graham 1943; Hacker 1978; Hansen 1999; Jones 1976; Kothari 2009; Llewellyn 1993)—makes a valid point, yet none of them captures the complexity and internal heterogeneity of the organization; moreover, several of these labels apply equally well to those Hindu groups that called themselves sanātan dharmī or sanātanist—advocates of the “eternal religion” of Hinduism—that mobilized in direct opposition to the Arya Samaj and its engagement with Dalits at the beginning of the twentieth century. Though founded in Bombay by a peripatetic brahmin sannyāsī (renunciate) of Kathiawar, it was in Punjab among the urban, educated members of the Hindu commercial castes that the Arya Samaj took root and grew into an organization with significant membership and influence. The Lahore Arya Samaj, which would become the epicenter of Samaj activity in the subcontinent in the following decades, was formed in 1877. Its early members were the English-educated sons of the urban Hindu elite engaged in business, law, and government service, drawn largely from Arora, Khatri, Aggarwal, and related mercantile castes, as well as brahmins (Jones 1976: 50–66). When Swami Dayanand Saraswati, the Samaj’s founder, died in 1883, these young Punjabi Hindus assumed leadership of the organization, developing its network through print publications, educational institutions, public confrontations with Christian missionaries and Muslim ‘ulema (scholars), and a variety of efforts at Hindu reform. One of Swami Dayanand’s successors was Lala Munshi Ram, who would later, upon formally taking sannyās in 1917, assume the title Swami Shraddhanand. Born in 1856 into a Punjabi Khatri family, Munshi Ram joined the Arya Samaj in the early 1880s in Jullundur, where he practiced law. He swiftly rose to positions of leadership in the Samaj, steering the more radical of its two major factions (the “Mahatma Party”) through a series of clashes and reconciliations with its rival (the “College Party”), associated with Lala Lajpat Rai, in the 1890s and 1900s. In these conflicts Munshi Ram’s faction favored strict vegetarianism over meat-eating, promoted the education of girls, and sought to increase the Samaj’s institutional support of prachār— the preaching or propagating of the Arya message, often using methods popularized by Protestant missionaries. Munshi Ram undertook extensive dharmyātrās, or preaching tours, and represented the Arya Samaj in public debates with the traditionalist sanātan dharmīs who opposed the Arya Samaj’s attempts to reform Hinduism. In the several journals he founded and edited (Saddharmprachārak, the Arya Gazette, the Liberator) and in numerous

Deceptive Majority 84

pamphlets and short books, Munshi Ram wrote prolifically, defending the Arya Samaj’s position on various public controversies, promoting Vedic education, and criticizing Muslim and Christian efforts at evangelizing. By the 1900s the canvas of Munshi Ram’s activities extended well beyond Punjab; the Gurukul, or Vedic college, that he founded at Kangri (near Haridwar) in the United Provinces, attracted admiring visitors that included Gandhi (as early as 1915), C. F. Andrews, and British prime minister Ramsay Macdonald. Munshi Ram helped create the Arya Samaj’s national-level body, the Sarvadeshik Arya Pratinidhi Sabha, and served as its president from 1910 to 1917. By the mid-1920s he was also elected vice president of the Hindu Mahasabha, led by his friend Pandit Madan Mohan Malaviya (Adcock 2014; Jones 1976; Jordens 1981). Munshi Ram was also a Congressman. Active since 1888, when he helped form the Jullundur committee of the Indian National Congress and was briefly its secretary, his most intensive period of involvement followed Gandhi’s ascent to nationalist leadership and his own transformation into Swami Shraddhanand (Jordens 1981: 25, 39). Between 1919 and 1922 the ochre-robed renunciant became a prominent figure in the Congress, playing leading roles in the Delhi Rowlatt Satyagraha and Amritsar Congress session of 1919, before being appointed head of the committee on Depressed Classes relief in 1922, an episode to which we will return. It was under the direction of Munshi Ram/Swami Shraddhanand that shuddhi, the “purification” ceremony by which the Arya Samaj inducted nonHindus into the Hindu fold, grew from a relatively rare event involving small numbers of people to a mass campaign and magnet for national controversy between 1900 and 1926. Swami Dayanand had improvised shuddhi ceremonies twice in the late 1870s for youths who, though born into privileged caste Hindu families, had been excommunicated for breaking faith; in both cases, the ceremony was “a means to rejoin them with their families after they had lost caste by converting to Christianity or Islam” (Adcock 2014: 117). In the last quarter of the nineteenth century the Arya Samaj in Punjab, at times in collaboration with a Sikh organization, attempted to standardize ritual procedures for shuddhi and thereby “purified” small numbers of Christians and Muslims, again mostly men and women who had been born Sikh or Hindu but had converted and later sought to reinstate themselves in their community of birth (Adcock 2014; Graham 1942; Jones 1968; Jordens 1977, 1981). Talk of religious conversion was in the air in this period in Punjab. The more aggressive

85 Missionary Majoritarianism

of the Protestant missions challenged Muslim and Hindu spokesmen to public debates on the merits of their respective faiths; conversions to Christianity of individuals from prominent Muslim or Hindu families were celebrated, condemned, and debated in pamphlets and newspapers, at railway stations and bazaars. As the Arya Samaj, the Singh Sabhas, and Muslim associations such as the early Ahmadiyya (followers of Mirza Ghulam Ahmad) responded to the Christian missionaries, Punjab’s public spaces and vernacular print spheres were increasingly shaped into arenas of polemic contestation—and competition for followers—between religious groups (Imad-ud-Din 1885; Jones 1968; Powell 1993, 2003). In this context of religious rivalry, the ambit of shuddhi was expanded to include individuals who had been born into Muslim families but sought to join the Aryas. Pandit Lekh Ram, a contemporary and friend of Munshi Ram and one of the Arya Samaj’s early controversialists, depicted shuddhi as a centuries-old procedure for inducting members of other religions into the Aryan faith—that is, as a means of religious conversion. “The Shastras are clear that Muslims, Christians, Jews, Jains, and Buddhists by birth … were made pure [shuddh] just by the Gayatri mantra or the [agnihotra] fire sacrifice, and were then able to enter into the Arya dharm” (cited in Adcock 2014: 118). In the environment of multipolar religious competition in late nineteenth century Punjab, shuddhi emerged as Hinduism’s answer—though it remained highly controversial among Hindus outside the Arya Samaj—to Christian baptism and the Islamic recitation of the shahāda. Religious rivalry also framed the Arya Samaj’s earliest imagining of the possibility of admitting untouchables into the fold. Munshi Ram and a few others first mooted the idea in the 1890s, though no formal action was taken. Early discussions in Arya Samaj publications made no secret of the radical novelty of the idea or of its intimate connection with the Samaj’s competition with Christian missionaries. As the editor of the English-language Arya Patrika put it in 1896, [t]he admission of the Churas [sic] to the Samaj and their consequent elevation may appear the height of absurdity to ignorant minds … but the Vedic religion is for all. But it is confessed that no one has yet thought of turning the direction of the Samaj toward these matters. The very idea of converting such low and downtrodden people seems abominable. We live amongst people who would not tolerate such conduct on our part for a moment, and our instinct too treats it with utter repugnance. Yet the Christians do it. (Quoted in Graham 1942: 490)

Deceptive Majority 86

The practical remoteness of the idea is underscored in another entry in the same journal a year later: “For our Upadeshaks [preachers, instructors] to compete with Christians, they must give up Chhut Chhat [practices of untouchability] and caste restrictions” (quoted in Graham 1942: 473). Three points are key here: first, in Arya Samaj writings in the late 1890s, the idea that untouchables are in some way co-religionists of the Hindus is nowhere in sight. On the contrary, the idea that they could be reconceived of as part of a collective Hindu self remains “the height of absurdity” to the unenlightened and “abominable” to the progressive. Second, it is precisely the fact that “the Christians do it” that impels the Samaj to entertain the possibility of consorting with Chuhras and other untouchable castes. Third, when the idea of drawing close to untouchables first appears in Arya Samaj sources, talk of “utter repugnance” appears with it. Disgust enters the archive of privileged caste thought arm in arm with the Dalit, signaling the rootedness of caste “mentality” in the viscera and portending the depth and intensity of the ordeal ahead. Unsurprisingly, then, the first shuddhi ceremony in which untouchables joined the Aryan fold in 1900 was a result not of the initiative of Arya Samaj activists—who had not yet been able to bring themselves to pursue “such low and downtrodden people,” despite their recognition of the necessity of doing so to compete with missionaries. Rather, it was the initiative of locally organized Punjabi untouchable groups that brought about their first “elevation” to Hindu status by means of shuddhi. By 1900, tens of thousands of the sanitation labor castes in UP and Punjab, along with some Chamars and other Dalit castes, had converted to Christianity. These “mass movements,” as we found in chapter 2, were initiated by the sanitation labor castes, who had approached the often ambivalent missionaries, heretofore focused on proselytizing the elite, with concrete requests for education, legal assistance, land, employment, and baptism. When members of an untouchable community first sought admission in the Arya Samaj, they did so in a similar manner. In 1896, a delegation of Rahtias, a regional caste of Dalit Sikhs, approached Munshi Ram in Jullunder with a request for “help in getting justice” with respect to their access to wells and their being prohibited from sitting on the same carpet as Hindus and Sikhs in public forums (Adcock 2014: 120). The Rahtias had petitioned the Sikh Sabhas for several years to obtain such basic rights vis-à-vis their fellow Sikhs and had grown frustrated with the lack of progress. In Munshi Ram and the Jullundur Arya Samaj they saw—as the sanitation labor castes in neighboring parts of Punjab saw

87 Missionary Majoritarianism

in Protestant missionaries—a network of educated, highly placed men whose support could facilitate access to government and resources, who were willing to defy Hindu and Muslim landlords by extending such support, and who offered a ritual means of effecting a desired change in status. The request split the already fractious Samaj. Munshi Ram, living up to his reputation as a radical, replied to the Rahtias that “if they could see their way to embrace the Vedic religion, the Arya Samajists would willingly suffer every hardship for their sake” (quoted in Adcock 2014: 120). But many of his fellow Aryas remained unwilling to take the extreme step, and Munshi Ram was blocked twice from accommodating the Rahtias’ request. Ultimately the Rahtias turned from Jullundur to the Lahore Arya Samaj, where, in June 1900, with Munshi Ram’s support, nineteen Rahtias undertook shuddhi. In the ceremony the hair and beards of the “purified” Rahtias were shorn and they distributed sweetmeats to and ate alongside their Arya Samaj patrons from privileged caste Hindu families. Rancor ensued. A number of the Arya Samajists who welcomed the Rahtias were excommunicated by their families, shunned by service castes, and boycotted by local Hindus (Graham 1943: 492). Sikhs, too, were outraged, and the earlier Sikh–Arya cooperation in shuddhi ended. “There was even a false rumour in the Tribune that the Sikhs had massacred the Jullundur Aryas and that Munshiram was in hospital in a critical condition” (Jordens 1981: 52). When more Rahtias in the following months were admitted to the Arya fold in Lyallpur, Jullunder, and Ropar, Aryas in those places were boycotted and prevented from drawing water from Hindu and Sikh wells; some, upon disavowing all connection with the Arya Samaj, were subsequently readmitted to their castes (Adcock 2014: 124; Graham 1943: 492). From its inaugural moment, then, shuddhi as a means of rendering untouchables Hindu convulsed the Hindu body politic and raised the specter of communal violence. As a ritual of bringing in and eating with the Dalit, shuddhi provoked rituals of outcasting and severance of commensal relations. Mirroring the somatic drama of disgust, acts of incorporation triggered acts of expulsion. One might ask, though, if becoming Arya was seen as the same as becoming Hindu. It is true that in the 1890s and early 1900s some Arya Samajists, in a bid to align the Samaj with the “universal” side of the European distinction between “universal” and “national” religions, promoted use of the Arya title at the expense of and in rhetorical contradistinction to “Hindu,” deployed as an

Deceptive Majority 88

index of a more parochial, unreformed and “national” religion (Adcock 2014: 90–112), particularly in the context of polemical sparring with sanātanists. But efforts to contrast Arya and Hindu as identitarian categories were neither consistent nor long-lived, whereas most Arya Samajists appear to have remained content to be Hindu and Arya, and accepted the latter as a subset of the former: not all Hindus were Arya Samajists, but all Arya Samajists were Hindus. Lala Lajpat Rai, for instance, firmly stated: “The Arya Samaj, being a Vedic church, [is] as such a Hindu organization.…” (Rai 1915: 220). In the eyes of non-Arya Samajists, meanwhile, it seems there was never a question of the Arya Samaj being something other than a Hindu—albeit reformist Hindu— organization, or of shuddhi not involving becoming Hindu. Thus the very first shuddhi ceremonies prompted a Hindu correspondent in a Lahore newspaper to write of the Arya Samaj: “Of late they have begun to admit low caste persons—chamaars, sweepers etc—in their religion, which is nothing short of defiling Hinduism and destroying its purity” (quoted in Adcock 2014: 123–24). Munshi Ram, for his part, wrote that “the Rahtias, who consisted of some thousands, were all absorbed into the Hindu society”—making quite clear that shuddhi meant becoming Hindu, even while underscoring the thenregnant common sense of such groups’ exteriority to Hinduism prior to the rite (Shraddhanand 1946: 88). In her landmark study of Arya Samaj involvement in the caste question, C. S. Adcock (2014) argues persuasively that shuddhi in this period must be understood at least in part as a mode of “ritual-political assertion” through which untouchable groups like the Rahtias collaborated with Munshi Ram’s radical faction of the Arya Samaj to secure long-desired rights and respite from the long-hated rituals of humiliation enforced by the agrarian caste order. That such concerns significantly underwrote the initial wave of Rahtia interest in the Samaj, and that this exposed the Samaj to the same accusation of “material enticement” that the Samaj itself had leveled at Christian missionaries working among the sanitation labor castes nearby, was by no means lost on Munshi Ram and his comrades. As the Arya Gazette put it in 1900, “Some people are giving out that the Rahtias have no belief in the Vedic Siddhants [principles]. They want to join the Arya Samaj simply with a view to give themselves a lift in society” (quoted in Graham 1943: 492). The Aryas answered critics, interestingly, in language similar to that of the missionaries. Insisting that the aspiring converts were “sincere believers,” the Aryas acknowledged that a concern for material and social improvement also played a role in their seeking shuddhi, and, sobered by their encounter with the conditions of life of their


Missionary Majoritarianism

new fellows, they suggested that this desire for improved status was a legitimate desire (Graham 1943: 492). As more Rahtias sought Arya Samaj support, more shuddhi ceremonies followed. Within three years, delegations from two other regional untouchable castes, the Ods and the Meghs, approached the Arya Samaj for membership and were admitted. Within a few years, untouchable participants in shuddhi events grew from dozens to hundreds at a time. As the scale and pace of shuddhi ceremonies strained Arya Samaj resources, the Aryas formed a separate All-India Shuddhi Sabha. According to this organization, in the first decade of the twentieth century some three to four thousand Rahtias had undergone shuddhi in Punjab, as had two to three thousand Ods, and thirty thousand Meghs (Kaul 1911: 149–50). Unlike the Rahtias, the Ods and Meghs had not adopted the Sikh tradition; nor were they Muslim or Christian. Nor, at this point, had their neighbors and Arya Samaj patrons begun to think of these castes as in some way already Hindu; rather, their admission to the Arya Samaj was seen as a moment of Hindu becoming. Most Arya Samajists, including Munshi Ram, had yet to clearly or consistently formulate the idea that untouchables were primordially Hindu. In an interview in 1908, for instance, Lala Lajpat Rai distinguished two procedures for joining the Arya Samaj: “With Hindus merely signing the declaration of faith is sufficient. In the case of non-Hindus a shuddhi ceremony of a simple kind is obligatory” (quoted in Graham 1942: 465). Since the thousands of Ods and Meghs who had joined the Arya Samaj by that time had all done so by means of shuddhi, the implication that they were previously “non-Hindus” was inescapable. Remembering these early days of shuddhi, Munshi Ram—now Swami Shraddhanand—in his 1919 pamphlet Jāti ke Dīnoṅ ko Mat Tyāgo, arthāt 7 Kroṛ Dīnon ki Rakshā (Do Not Abandon the Suffering Castes, or the Defense of the Wretched Seventy Million) writes: At the very first stage, when untouchable uplift [achhūtoddhār] work began in Punjab and when the Rahtias were purified in Lahore and granted admission into the Arya Samaj, at that time there was tremendous opposition from the Hindu people. The shuddhi of Meghs in Sialkot and of Ods in Multan and Muzaffargarh took place in the teeth of fierce resistance. (Shraddhanand 1987b: 93)

Given what we know from the previous chapter, such a state of affairs is not unexpected. When the prevailing common sense was that Hindu and

Deceptive Majority 90

untouchable are opposed categories, Hindus could be expected to be riled when radical Aryas supported untouchable assertions of the right to access Hindu wells and wear the brahminical sacred thread. This phase of the shuddhi movement garnered little support, and much ill-will, from the Hindu community. What Shraddhanand says next, however, is key: “But at this time [that is, the time of writing of the pamphlet, 1919], a political wave has transformed the entire situation” (Shraddhanand 1987b: 93, emphasis added). That is, the situation (avasthā) that obtained in 1900 underwent dramatic transformations over the next two decades—transformations provoked by political (rājnaitik) events—and it was these transformations that began to reverse Hindu opposition to the incorporation of untouchables. The political developments that provoked this sea change have been characterized by Sudipta Kaviraj (1992) as the emergence—for the first time in the history of South Asia—of “enumerated communities.” Shraddhanand’s account of the “political wave” well illustrates Kaviraj’s point. Shraddhanand emphasizes three developments that were decisive in making Hindus more receptive to his radical proposals between 1900 and 1919. One was the 1905 partition of Bengal along religious lines. This imperial act, unprecedented in British India, made vivid to a mass population the stakes of being classified as a “minority” or “majority” community—a classification based on the enumerative techniques of the colonial administration. A second major development was the introduction and expansion of institutions of selfgovernance with popularly elected members, along with provisions for communal representation, or separate electorates. Among the key effects of the Indian Councils Act of 1909 and the Montagu–Chelmsford Reforms of 1918–19 was the diffusion of the idea that the governance of the emerging nation would be a matter of molding or mobilizing numerical majorities. The third development, closely related to the previous two, was the birth of the idea, first propagated by Colonel U. N. Mukherji, that Hindus were a “dying race.” Borrowing the language of population science from the colonial state and making unsound statistical projections based on census data, Mukherji advanced the thesis that without serious social reforms, the Muslim population of India would continue to multiply while the Hindu population would dwindle and become extinct within 420 years. This number-based discourse of extinction became such a centerpiece of Shraddhanand’s thinking that he entitled his English-language manifesto Hindu Sangathan: Saviour of

91 Missionary Majoritarianism

the Dying Race (Datta 1999; Shraddhanand 1926).1 The popularization of the “dying race” thesis exemplifies what Arjun Appadurai considers one of the key processes in the emergence of majoritarianism: the transformation of “a benign social identity” of a large cultural group into the “predatory identity” of an enumerated community that perceives itself as threatened: Predatory identities … are based on claims about, and on behalf of, a threatened majority. In fact, in many instances, they are claims about cultural majorities that seek to be exclusively or exhaustively linked with the identity of the nation…. The discourse of these mobilized majorities often has within it the idea that it could be itself turned into a minority unless another minority disappears, and for this reason, predatory groups often use pseudodemographic arguments about rising birthrates among their targeted minority enemies. Thus, predatory identities arise in those circumstances in which majorities and minorities can plausibly be seen as being in danger of trading places. (Appadurai 2006: 52)

Colonial north India, and especially Punjab and UP in which Shraddhanand and the Arya Samaj flourished, provided precisely such circumstances. It is important to pull back the curtain of teleology that subsequent history has thrown over this historical moment and to acknowledge its contingency. So  much of what the post-1947 experience casts as inevitable—the end of imperial rule, the Partition of India and Pakistan, the existence and viability of a Hindu majority in one and a Muslim majority in the other—was, at this time, far from certain. Depending on how untouchables were categorized, undivided Punjab was as readily imaginable as a Muslim- or Sikh-majority province as it was one with a Hindu majority; pre-Partition UP was similar. The Hindu majority that Mukherji and Shraddhanand, their eyes on the census, construed as fragile, was indeed plausibly seen as “in danger of trading places” with the Muslim minority, though not for the reasons Mukherji cited. Rather, as Shraddhanand diagnosed the situation, it was the religion of the Depressed Classes that would be decisive in determining majorities and minorities in the nation(s) to come. The rise of enumerative politics, then, and the concomitant transformation of the socioreligious landscape into competing blocs of enumerated 1

As Shraddhanand would later put it in his Hindu Sangathan: Saviour of the Dying Race (1926: 14), “within the next 420 years the Indo-Aryan race would be wiped off the face of the earth unless steps were taken to save it.” His text makes clear that by “Indo-Aryan race” he means precisely “the Hindus.”

Deceptive Majority 92

communities, began to give Shraddhanand’s inclusionary message traction. These were the political conditions that began to make possible the wholesale reshaping of popular conceptions of the boundaries of the Hindu community.

Shraddhanand and Shuddhi: The Second Phase By 1907, the initial wave of untouchable-led shuddhi in Punjab subsided, but within the Arya Samaj enthusiasm for shuddhi was on the rise. The radicals within the Samaj who had welcomed and encouraged the Rahtias, Ods, and Meghs enjoyed a kind of vindication. On one front, they prevailed against sanātan dharmīs in a series of legal battles over the excommunications, which they framed as defamation and inciting religious hatred (Adcock 2014: 125). On the other, by demonstrating their capacity to attract converts from among the oppressed castes, they put Christian missionaries on the defensive. They also enabled a tremendous numerical increase in the membership of the Arya Samaj. Buoyed by these unanticipated successes and the Samaj funds that were consequently made available for sustaining them, Shraddhanand and his colleagues adopted a different strategy and began active proselytizing (prachār) among untouchable communities (Graham 1943: 495–96; Jordens 1977: 152). This marked a significant departure; whereas the Rahtia, Od and Megh wave of shuddhi had been the result of untouchable initiative, it was the Samaj that usually took the first step in the mass shuddhi events of the next three decades. The Aryas expanded the ambit of their operations socially and regionally: in the 1910s Arya Samaj activists took up missions among the Dumna caste in Punjab, among Basiths in Kashmir, and among Doms in the hill tracts of UP. Overcoming the initial skepticism of their new audience—some groups suspected them of “mercenary motives”—the Arya prachāraks ultimately won converts on a scale even larger than that of the previous decade (Graham 1943: 496–504). Their success continued to depend on their willingness to secure their new brethren access to wells and other substantive signs of elevated status; shuddhi, that is, continued to function as a mode of ritual-political assertion. And insofar as it did so, it continued to provoke Hindu resistance. In the 1910s and well into the 1920s, Arya Samajists and their initiates faced organized resistance by local Hindu communities (Adcock 2014: 115–42; Rawat 2011: 143) and occasional episodes of violence, such as the 1923 murder of an Arya Samaj prachārak in Jammu by a Rajput mob outraged by his work among untouchables (Jordens 1981: 131). At the same time, the penetration of the political logic of enumerative identity into the consciousness of an increasingly wide swath of the population


Missionary Majoritarianism

meant that the political value of untouchable castes—as enumerated communities whose numbers could enhance or deflate the “strength” of Hindus or Muslims as population blocs—began to attract growing attention. Because shuddhi inducted untouchables into the Hindu fold, it came increasingly to be seen by both its proponents and its critics as a mechanism for manufacturing a Hindu majority. Hindu resentment of the rights to well access, education, and so on, that shuddhi secured for new Aryas now began to be weighed against the value obtained by shuddhi as a catalyst of Hindu growth, a cure for the “dying race.” Shraddhanand’s 1919 pamphlet makes clear—not least in the centrality of number in its very title, 7 Kroṛ Dīnon kī Rakshā (Defense of the Wretched Seventy Million)—that the enumerative revolution was well underway by the late 1910s, and that the specifically numerical significance of untouchables was a central plank of the Arya Samajists’ arguments in favor of shuddhi by that time. Controversial from the beginning, shuddhi was by this time attracting increasing criticism from Muslims who perceived it—accurately, I am arguing—as a key strategy in a Hindu majoritarian project. But it was in the early 1920s, with the Moplah Rebellion and the Malkana controversy, that shuddhi was catapulted into the center of nationalist debate as one of the primary contributing factors to communal discord and violence between Hindus and Muslims, a status it would continue to hold into the 1930s (Adcock 2014; Freitag 1989: ch. 7; Gooptu 2001: ch. 6; Gould 2005: ch. 3). Following the Moplah Rebellion of 1921, in which Moplah (that is, Mappila) Muslims on the Malabar Coast attacked not only colonial institutions with British personnel but also killed Hindu landlords and forcibly converted some local Hindus to Islam, the Arya Samaj sent prachāraks to re-convert the new Muslims to Hinduism. Shraddhanand championed the move and excoriated Congress Muslims for hesitating to condemn the Moplahs (Graham 1943: 508–11; Jordens 1977: 152–53; Shraddhanand 1946: 150–51). Then in 1922–23 open competition broke out between the Arya Samaj and Muslim organizations when a Hindu-leaning subset of the Malkanas, a Muslim Rajput caste in western UP whose traditions combined Hindu and Muslim practices, sought shuddhi and unification with Hindu Rajputs. A new Bharatiya Hindu Shuddhi Sabha was formed with Shraddhanand as president; the swami and his team camped in the region and conducted massive shuddhi ceremonies while Muslim organizations in Bombay and elsewhere protested (Jordens 1981: 131–34; Sikand 2003). The ensuing strain on Hindu–Muslim relations prompted some of Shraddhanand’s admirers in the Congress to urge their colleague to desist, at least until a less sensitive time (Jordens 1981: 133).

Deceptive Majority 94

The Malkana controversy also put Shraddhanand in direct conflict with Khwaja Hasan Nizami, arguably the most influential of the Arya Samaj’s Muslim critics, and the only Muslim writer of the period to argue for the propagation of Islam specifically among the sanitation labor castes. Nizami, a hereditary member of the Chishti Sufi lineage that served as caretakers of Nizamuddin Auliya’s shrine in Delhi, developed a following as a prolific Urdu journalist, diarist, publisher of religious magazines, and Sufi reformer. In the 1920s he advocated tabligh, or propagation, as an Islamic response to the Arya Samaj’s shuddhi campaigns, and in his 1923 book Da’i-i Islam (The Missionary of Islam) outlined a practical program for conducting tabligh among the masses, with particular attention to the Malkanas and various non-elite Muslim castes and communities (Hermansen 2008; Sikand 2003). In the same year Nizami published an essay arguing that Sikh doctrine had more in common with Islam than with Hinduism and that Sikhs and Muslims were thus essentially co-religionists (Sikand 2004), as well as Halalkhor, a collection of ethnographic and folkloristic materials on the sanitation labor castes interspersed with speculations as to their Islamic, rather than Hindu, provenance. In these and other texts—and we will return to Halalkhor later in this chapter—Nizami put forward refutations of Hindu majoritarian claims and proposed means of revitalizing Islam that displayed close knowledge of the Arya Samaj’s methods. More writer than activist, Nizami rarely undertook the equivalent of Shraddhanand’s preaching tours or shuddhi camps, but he did organize a publicity tour for a Malkana Rajput prince who, rejecting the Arya Samaj’s overtures, had publicly affirmed his Muslim pahchān and encouraged other Malkanas to do the same (Hermansen 2008: 173–75). Shraddhanand saw in Nizami’s writing an existential threat not only to his shuddhi campaign but to the future of Hinduism. Within days of encountering Da’i-i Islam, Shraddhanand published a pamphlet entitled Hinduo Savadhān! (Hindus Beware!) that selectively quoted from Nizami’s book to suggest a widespread yet clandestine Muslim plot to generate, through conversion, a Muslim majority (it quickly went into a second edition, retitled Khatare kā Ghanṭā, arthāt Muhāmmadī Ṣaḍyaṅtra kā Rahasyabhed [The Warning Bell, or, Exposure of the Muhammadan Conspiracy]) (Jordens 1981: 140–42). Hinduo Savadhān set the tone for the bulk of Shraddhanand’s writing in the final years of his life; from anti-Islamic pamphlets like Blind Faith and Secret Holy War to his best-known work, Hindu Sangathan: Saviour of the Dying

95 Missionary Majoritarianism

Race, what animated virtually all of the Swami’s output from 1923 to his death in 1926 was the fear of small Hindu numbers. “[I]f all untouchable castes become Muslims,” Shraddhanand warned in Hinduo Savadhān!, “then the Muslim party will become equal to that of the Hindus” (quoted in Jordens 1981: 161). This became a frequent refrain in Shraddhanand’s writing and speeches, a refrain that circulated widely through Arya Samaj networks. In another pamphlet, Vartamān Mukhya Samasyā (Today’s Foremost Problem), the swami elaborated that the numerical danger of Muslim growth and Hindu diminution posed by the unsettled religious status of the Depressed Classes was fused to the specter of untouchable selfreliance in the emergent nation: “If all untouchables became Muslims then these will become equal to the Hindus, and at the time of independence they will not depend on the Hindus, but will be able to stand on their own legs” (quoted in Jordens 1981: 144). Importantly, Shraddhanand’s explicit invocation of the “time of independence” here indicates that his concern with numbers in the 1920s attached to a clear vision of the nation qua state. No longer a provincial contest over converts in Punjab, as some Aryas understood the first wave of Rahtia shuddhi in 1900, the stakes of shuddhi were now nothing less than the possibility of a Hindu majority in an independent nation-state. The obvious adversaries to the project of securing this majority, as Shraddhanand saw it, were Christian missionaries and Muslim proponents of tabligh like Nizami. “The suppressed classes of South India can no longer bear Hindu tyranny,” wrote the swami on a shuddhi tour in 1925. “The Christian missionary with his millions is at their door. Mahommedan Mullahs are appealing for lakhs and rushing to the scene. The Arya Samaj mission alone can save the situation by taking them into the bosom of the Vedic church” (quoted in Jordens 1981: 161). Yet for Shraddhanand perhaps the most vexatious obstacle to securing the untouchables for Hinduism was the Hindu community itself. Time after time he upbraided his Hindu colleagues in the Congress for failing to condemn or curtail the practices of untouchability that daily reinscribed and publicized untouchable alterity from the Hindu (Shraddhanand 1946: 121, 133–38, 147–48, 179–88). It was above all the unwillingness of Hindus to permit untouchables basic rights, the swami insisted, that thwarted the realization of a robust Hindu majority. Accordingly, in his protreptics directed at nationalist Hindus, Shraddhanand conjoined pleas for the extension of rights to untouchables with reminders of their numerical vastness as a bloc and the danger of their numbers being captured

Deceptive Majority 96

by another constituency: Muslims, Christians, or the colonial government. For example, as chairman of the Reception Committee of the 1919 Session of the Indian National Congress in Amritsar, Shraddhanand (1987a: 165) exhorted his Congress colleagues to “reflect on how your six and a half crore [65 million] brothers—those pieces of your heart that you have sliced off and flung away— how those six and a half crore sons of Mother India can become the anchor of the ship of a foreign government.” Or again, in Jāti ke Dīnoṅ ko Mat Tyāgo, Shraddhanand wrote of untouchable converts to Christianity: Instead of seeing themselves as a part of the Indian nation [Bhāratīya rāṣṭra kā ek bhāg] they dream of entering its halls of governance…. If the seven crore [70 million] untouchable women and men of India [Bhāratvarṣ], harried from maltreatment by the twice-born, become Christian, then the Home-Rule, Hindu leaders [swarājyavādī sanātandharmī līḍar] will be able to do nothing but regret it…. So long as we do not understand those shudra brothers to be a part of our selves, and so long as they are not treated fairly, the Aryan race will remain politically in mortal peril. (Shraddhanand 1987b: 96)

Again in this passage Shraddhanand signals a concern with the Indian nation as a political and territorial entity—a rāṣṭra, as distinct from a qaum, or nation in the sense of a people or community—that is within sight of independence (Home Rule). The potential of seventy million untouchables to become Christian, however—and we have seen earlier that the fear is equally of their becoming Muslim—presents the prospect of a Christian check to the hopedfor Hindu majority of this imagined rāṣṭra, a mortal blow to both “HomeRule, Hindu leaders” and the “Aryan race” itself. The solution, Shraddhanand suggests, is to end the maltreatment of the untouchable, and to “understand those shudra brothers to be a part of our selves.” Other nationalist Hindus began to heed and amplify Shraddhanand’s message of the political danger of perpetuating untouchability, though with less appreciation of Depressed Class agency. In a popular Hindu reformist journal of this period, a political cartoon entitled “Achhūtoṅ kā Football” (Football of the Untouchables) appeared, in which a Hindu (a pot-bellied brahmin wearing a sacred thread) is kicking a black-colored football on an open field while a Muslim maulvi and a Christian priest position themselves to catch it (C. Gupta 2016: 118) (see Figure 3.1). The cartoon dramatizes the Hindu majoritarian imagination of untouchables as the decisive variable in a zero-sum game between competing religious blocs, and implicitly chastises the brahmin for unwittingly giving the Muslim and the Christian an advantage. It also

97 Missionary Majoritarianism

Figure 3.1 “Football of the Untouchables” Source: Chand, January 1929. Image courtesy of Charu Gupta.

underscores the continuing regnancy, even in the late 1920s, of the common sense of the untouchables’ exteriority to the Hindu body politic—it would make little sense to represent untouchables as a black football being kicked about by a Hindu if they were conceived of as an organic part of the Hindu self. In warning his nationalist colleagues of widespread Depressed Class antipathy toward the Congress and (qualified) preference for British government, Shraddhanand offered an astute and fundamentally correct assessment of the subaltern political scene, an assessment grounded in his considerable experience engaging with local-level untouchable leadership in rural and urban Punjab, UP, and Delhi. By the 1920s, most major regional Dalit castes in the subcontinent had formed political organizations that offered pointed critiques of the Congress as a party of the privileged castes, and that mobilized opposition to Congress-led nationalist actions at various levels. In Punjab it was the Chamar-dominated Ad Dharm that played this role (Juergensmeyer 2009), in UP it was Swami Achhutanand’s Adi Hindu Mahasabha (Rawat 2011), in Bengal the Namasudra movement (Bandyopadhyay 1997), and in Chattisgarh the Satnamis before the Satnami Mahasabha (Dube 1998). In western and southern India, sophisticated anticaste critiques of Congress-led nationalism had a depth of two generations by the 1920s, going back to Mahar activists in the Satyashodhak Samaj (Rao 2009; Paik 2014) and Tamil Buddhists associated with Ayothee Thass (Aloysius 1998). Of the major regional and pan-regional Dalit castes, arguably

Deceptive Majority 98

the most significant outliers in this political landscape were the sanitation labor castes. Unlike the Mahars, Chamars, Pariars, Namasudras, and others, the sanitation labor castes had not launched their own regional caste-based political organizations by the 1920s, and their earliest locality-based political groups, the Valmiki Sabhas in urban Punjab, were directly sponsored by the Arya Samaj, and were thus pro-Congress. Without the forceful representation of distinctive group claims by a Mangoo Ram, an Ayothee Thass, or an Ambedkar, the sanitation labor castes were seen, relative to other populous regional Dalit castes, as politically quiescent and tractable to Hindu leadership, a fact not incidental to their prominence in the representational practices of Gandhi and his admirers, as we will find in the next chapter. The fact that organizations based in major regional Dalit caste-clusters— with exceptions like the sanitation labor castes—were in the 1920s portraying the Congress as a privileged caste interest group antithetical to Dalit flourishing posed, for the nationalist movement, a pressing problem of legitimacy. Well ahead of most nationalist leaders, Shraddhanand recognized this and repeatedly raised the issue, chiding Gandhi, for example, for not prioritizing untouchability removal and thus deserving the heckling he received from politically well-informed untouchables on his tour of Madras Presidency in 1920.2 In arguing that nothing short of a radical undoing of the regime of disabilities imposed on untouchables would attract the latter to the Congress and to the Hindu fold, Shraddhanand summarized two decades of the Arya Samaj’s experience with shuddhi while simultaneously urging a program of action that the Congress would not take up in earnest until 1932. Even as he criticized shuddhi as a “puerile measure” aimed at “increasing [Hindu] numbers” while evading the fundamental problem of caste, Ambedkar (1992: 425) characterized Shraddhanand—not Gandhi—as “the most revolutionary and ardent reformer of the Hindu society.” Arguably more than any other Hindu reformer or nationalist in the 1920s, Swami Shraddhanand attempted to dismantle the system of practices on which untouchability rested, and to secure for untouchables tangible rights. And almost no one articulated with such force the motivation for undertaking such 2

“Even Mahatma Gandhi had not realized its importance [the importance of the removal of untouchability] and was taken up with his resolution of non-violent Noncooperation…. While touring in the Madras Presidency he was heckled by the socalled Panchama Untouchables of the South with questions about their position on the attainment of the national right of self-determination” (Shraddhanand 1946: 121).

99 Missionary Majoritarianism

efforts: securing a Hindu majority in an independent India. Only the cessation of untouchability could forestall the threat of seventy million untouchables anchoring, as Muslims, a Muslim majority or allying themselves, as Christians, with the colonial government. Just months before his death the swami put forward perhaps his most succinct plea to his nationalist Hindu colleagues to take up the campaign of majoritarian inclusion; in April 1926 he wrote that “[t]he uplift of the untouchables and their assimilation in the Hindu polity is the very plinth on which alone the edifice of free India can be constructed” (Jordens 1981: 163, emphasis added).

Birth of a Myth It was in the context of aggressive efforts at majoritarian inclusion in the 1920s that Arya Samaj preachers put into circulation one of their more astonishing— and astonishingly enduring—claims: the thesis that Muslims had created untouchability. Swami Shraddhanand gave utterance to this idea as early as 1891, when he wrote that the caste system “is a direct outcome of the advent of Muslim rule in India” (quoted in Prashad 1996: 556). Though contradicted by the entire corpus of normative Sanskrit literature prior to the eleventh century—a corpus with which educated Arya Samajists like Shraddhanand were well acquainted— this falsehood well suited the Samaj’s narrative of an ancient Aryan golden age subsequently corrupted by foreign rule. It also made possible a cynical appeal to untouchable traditions of militant resistance to oppression (Prashad 2000: 78–79, 107–11). As the shuddhi movement grew, Arya Samajists embellished the trope: the untouchables, they said, were the descendants of Hindus who opposed Muslim conquest, resisted conversion, and were forced by their new masters to undertake degrading tasks. By the 1920s, the trope was a staple feature of Hindu majoritarian overtures to the Depressed Classes. Arya Samajists in Pilibhit in 1924, for example, attended a “sweeper” wedding and “made speeches informing the sweepers that they were the descendants of Valmiki muni who had been forced to become a sweeper by the conquest of Hindustan by the Muhammadans” (C. Gupta 2016: 59). A prachārak in Delhi told sweepers there: “There were no sweepers before. They came into existence when the Muslims entered this land. As Muslim women observed seclusion, the Muslims needed sweepers. Therefore, they made others into sweepers” (Prashad 2000: 78). The claim in its most general form—that untouchability was a product of Muslim rule—could be adapted by Samaj activists to suit the circumstances of particular untouchable groups. For the sanitation labor castes the claim took

Deceptive Majority 100

the particular form of a spurious history of sanitation labor. More sophisticated versions tied labor to technology and suggested that Muslims, by introducing the enclosed dry latrine to the subcontinent, created a demand for manual scavenging where none had previously existed, and thus generated a new class of degraded laborers. This, too, was false: sweeping and cleaning privies are specified as duties of Chandals and other despised social groups in dharmaśāstra texts that predate Muslim presence in South Asia by centuries (Aktor 2010: 45; Chakravarti 2006: 83; Jha 1975: 22; Nārada 1989a: 108, 1989b: 150), and archaeological evidence in South Asia of enclosed latrines that would have required routine manual removal of human ordure goes back to Indus Valley Civilization (Kenoyer 1998: 60).3 Nonetheless, as we will see, the idea that the sanitation labor castes were created under an Islamic dispensation circulated widely through Arya Samaj networks, attained mainstream respectability in Gandhi’s untouchable uplift organization, the Harijan Sevak Sangh, and became a truism for the first generation of Balmikis to be patronized by the Congress. Even Ambedkar accepted the myth until his young supporter Bhagwan Das, citing evidence of dry latrine usage in ancient Buddhist monasteries, disabused him (B. Das 2009: 44–48). Though fictitious, the trope of the Muslim invention of untouchability proved effective as one of the means by which Arya Samaj activists worked to dismember solidarity between Muslims and the Depressed Classes, and to attract the latter toward militant Hindu leadership instead. The trope also exonerated Hindus of the bulk of the responsibility for untouchables’ condition, clearing the path for an appeal to Hindus to rethink their entire affective orientation to this heretofore despised community.

Patitoddhār: Rerouting Disgust, Cultivating Fellow-Feeling An examination of Patitoddhār (Upliftment of the Fallen), an influential Arya Samaj manifesto on untouchable uplift, will help illustrate the claims we have been making in this chapter. Though neither author nor publication date is attributed, the 192-page Hindi text appears to have been written in 1918 by one Shri Ram Sharma, most likely in UP. Patitoddhār, along with a similar text entitled Patit Prabhākar (The Fallen Light-Giver) penned by Devadutt 3

While the cities of the Indus Valley Civilization are renowned for their drainage networks, Kenoyer (1998: 60) explains that the latrines of only some homes were connected to drains, whereas many domestic latrines were sump pots sunk into the earth that would have necessitated regular manual cleaning in much the same manner as dry latrines in subsequent centuries.


Missionary Majoritarianism

Sharma in 1925 in UP, and Amichand Sharma’s Śrī Bālmīki Prakāś (Light of Shri Balmiki) to which we will turn shortly, circulated widely in north India through Arya Samaj networks. The ideas that these texts put forward found their way into many a subsequent account of Dalit life, from those of journalists, novelists, social workers, and Dalit Congressmen writing in Hindi to the head of the Harijan Sevak Sangh writing in English. Patitoddhār begins with a riddle. If a key feature of India’s ancient civilizational genius is the perfectly ordered four-fold society of Vedic religion— and Max Müller is quoted at length in support of this premise—then whence came the cruel aberration of untouchability? What a surprising and grievous situation that in this holy land where … sages reside in their incomparably splendorous and peaceful ashrams, where the system of four stages of life prevails, where such divine avatars as Ram, Krishna, Buddha, Shankar and Dayanand have dwelt—that tens of millions of the offspring of this very land of Bhārat should be considered fallen [patit], untouchable [aspṛśya]? And this to such an extent that among them, tens of millions should become hostile toward their own people? O Time! Who comprehends your Play [līlā]? … In this Bhārat, who is “fallen” and why? This is one question whose answer this book provides. (S. Sharma n.d.: 6–7)

The author reveals to his readership that those typically considered fallen or untouchable—the terms are used synonymously throughout the text—are in fact the forgotten descendants of the kshatriyas, the prestigious second tier of Vedic society. Today those whom we call by the names lowest-born [antyaj], fallen [patit], untouchable [aspṛśya] and so on, who in Punjab are called Megh, Rahtiya, and Chamar and who in Gujarat, the South, Madras and elsewhere are called Dhed, Mahar, and Pariya, and to touch whom is considered against religion [dharm] in some places, and some of whose members weave cloth and plow the land and tan leather—on the topic of these people if sufficient research were to be done then they would be proven not to be low. Their [gotra/lineage names]—Pawar, Solanki, Jadav, Chauhan, Makwana, Chauwda, etc.—give evidence that they were Kshatriyas of the ancient period. The particular manner in which their wedding and other rituals are conducted are manifest proof of their being ancient Hindus. If today the Hindu race/nation [jāti] would think about this important subject, it would be established that in the ancient period these [communities] were pure Kshatriyas and thus a leading member of the Hindu jāti. (S. Sharma n.d.: 25–26)

Deceptive Majority 102

If this is the case, though—if today’s untouchables are in fact yesterday’s Hindu ruling class—then how did this profound loss of status occur? The author of Patitoddhār offers only hints in the early part of his treatise. He quotes, for instance, a speech that Arya Samaj leader and Congress stalwart Lala Lajpat Rai made in 1914 in Allahabad, in which Rai declares that “the idea of untouchability was either not present or very weak before the invasions of the Muslims” (S. Sharma n.d.: 74). Later, however, the author spells it out. After asserting that rural and urban Indians in the millennia before Muslim rule always defecated in the outdoors, he continues: When there were no toilets in our country [deś], then where did this business of picking up excrement come from? The necessity arose due to those brothers enamored of purdah. [As the proverb goes,] “The buffalo is his who carries the club [jis kī laṭhī us kī bhais].” Taking up this unjust, merciless rule, they seized whomever they pleased, and coerced them into cleaning their toilets. The  unjustly fallen twice-born people became a separate caste [jāti] as the fellow-feeling and support of their co-religionists [sahdharmī] was gradually, over great lengths of time, extinguished. Over hundreds of generations they wholly forgot their ancient glory. But happily, due to some good fortune in history, there remains in these Bhangis that old Kshatriya hauteur…. If somehow, despite enduring this oppression, they had had the opportunity to fully nurture their lost twice-born dharm or shudra dharm, then the condition of those who remove excrement would not be looked upon with such disgust [ghṛṇā]. (S. Sharma n.d.: 166–67)

Here we begin to glimpse the affective project of the text, made possible by the trope of the Muslim invention of untouchability. The Hindu reader, relieved of historical responsibility for the Bhangi’s condition, is invited to suspend his disgust at this figure and to entertain, instead, the possibility of feeling sahānubhūti, empathy or fellow-feeling, toward this long-lost brother. The term sahdharmī is crucial here. The rights and responsibilities specific to a given moral community in the social order are that community’s dharma; the brahminical tradition famously distinguishes Brahmin dharma, Kshatriya dharma, Vaishya dharma, and Shudra dharma (as well as Stri dharma and Purusha dharma for women and men, respectively). Being a sahdharmī, literally of the same dharma, means enjoying the same rights and shouldering the same responsibilities as the rest of a given moral community does. The Arya Samaj was deeply invested in consolidating the various dharmas under the banner of a single Hindu or Arya dharma—Patitoddhār itself asserts that “Vedic

103 Missionary Majoritarianism

dharma, Arya dharma, Brahmin dharma and Hindu dharma: these words are equivalent” (S. Sharma n.d.: 4). This project was facilitated by the colonial state’s policy of defining “Hindu” capaciously and by the rendering of dharma as religion (and vice versa) in colonial knowledge practices. Thus by the time of Patitoddhār the term sahdharmī suggested a fellow member of the (newly consolidated but imagined as primordial) moral community of Hindus, with powerful connotations of shared rights and responsibilities, at the same time that it served as a workable Hindi translation of “co-religionist” with purchase in the enumerative politics of the colonial state and anglophone nationalism. The novelty of the suggestion of the untouchables’ sahdharmī-ness with the Hindu makes itself felt frequently in the text. When describing what is entailed in untouchable uplift, for example, the author approvingly quotes a 1916 article from the Arya Samaj journal Abhyuday that “they should be initiated into the Hindu Religion [unhe Hindu dharm kī dīkśā dī jāe], we should give even more respect to them than the Muslims and Christians are giving because they are ‘ours,’ the foundation of the body of the Hindu race/nation [jāti]” (S. Sharma n.d.: 70–71). Dīkśā can mean instruction or initiation; in either case, that the article insists that untouchables undertake dīkśā in Hindu religion bespeaks the still regnant common sense of their exteriority to the Hindu fold. Likewise the article’s use of scare quotes around “ours” [apne], a rarity in Hindi in this period, signals that the author is alive to how strange the word must sound to his readers in this context. Strange, yes, but necessary. Like Shraddhanand, the author of Patitoddhār loses few opportunities to drive home the urgency of untouchable uplift, and to remind the Hindu reader of the political peril he faces should majoritarian inclusion fail. Sometimes he does this in his own voice; elsewhere he relies on others, as in this passage, which again quotes Lala Lajpat Rai: The question of the uplift of the untouchable castes is the question of the life or death of the entire Hindu race/nation [jāti]…. If these untouchable castes, due to your injustice and apathy, leave us, then the communal [qaumī] body will become very weak and the basis of our claim to greater numbers, too, will leave us. If you lift them up, educate them, give them a “social status” [English in original], then you will not be doing some great charity, but rather you will be protecting yourself and preventing your own destruction. (S. Sharma n.d.: 83, emphasis added)

Lest the threat of political minoritization not suffice, Rai also adds the other rationale often offered in appeals to Hindus to soften their treatment of

Deceptive Majority 104

untouchables in this period. “It is in your own interest that you elevate them. If they depart, then this labor that they currently do, you will have to do yourself” (S. Sharma n.d.: 81). Thus far, Patitoddhār illustrates what I am arguing is the Arya Samaj’s formula for majoritarian inclusion: the assertion of the untouchable’s previous Hindu-ness, the trope of the Muslim invention of untouchability, and the introduction—albeit wavering—of a discourse of brotherhood and sahdharmī-ness, all backed by the fear of small Hindu numbers. But then the text takes a sensuous and affective turn. Acknowledging the embodied depth of his readers’ objections to fraternizing with untouchables, the author engages with these objections on their own terrain. Chief among these objections, he says, is the sense among his fellow Hindus that “these people [untouchables] live filthily.” The author addresses this concern at length. As dirtily and filthily as these people live [jitne gande maile kućaile yeh rahte hain], there are people of other castes that live in an equally filthy manner, and they are not considered untouchable. Ask truly, and what is the main cause of the spread of so much disease these days? It is due [not to untouchables but] to Halwais [sweet-makers], Rasois [cooks], Dhobis [launderers] and Hajams [barbers]. When the eye observes the Halwai’s soiled clothes and his hands, moist with the saliva from the chillum he’s been smoking, shaping the sweets, who isn’t exceedingly disgusted at the idea of eating [khāne se kise atyant ghṛṇā nahīn hotī]? (S. Sharma n.d.: 14)

The passage then describes how, at night, men urinate along the edges of the streets in the bazaar. At dawn of the next day, at the very time that the Halwai is arranging his sweets for display, [t]he Bhangi, applying his broom, scatters the dried, urine-saturated dust not only over the Halwai’s sweets but all over the neighborhood. Does anyone ever consider this filth? (S. Sharma n.d.: 14)

With the mention of the Bhangi, the author here seems momentarily to forget that his object is to arouse the reader’s disgust at the actions of touchable castes. This quickly passes as he turns to vivid portrayals of the sweat of the Rasoi dripping into the wheat of the roti he bakes and the squalid heaps of unwashed laundry at the Dhobi’s home. “The good barber,” he continues, “for his part, is the very living definition of filth.” Having catalogued the bodily exuviae of the “disease-afflicted” customers of the barber to be found on his shop floor, the author finally sums up the passage thus: “Is a bathed


Missionary Majoritarianism

and cleaned, pious Chamar filthier than the above-mentioned forms of filth?” (S. Sharma n.d.: 15). The sarcasm with which this passage is laced should not be confounded with farce; the author of Patitoddhār is in earnest here, and his concern with ghṛṇā is both serious and at the heart of his project. If the talk of fraternity with the Depressed Classes is to have any traction, the text suggests, we Hindus must contend with ghṛṇā. Pointing out the double standards of untouchability by drawing attention to ghṛṇā-eliciting behaviors of some of the artisanal and service castes that hover slightly above the line of touchability is the author’s initial tactic. Soon, though, it is clear that he is only warming up to his topic, and the Halwai and the Hajam are secondary targets. What of those eaters of cow-flesh that we call Yavanas [Muslims]? Shall we accept with relish sugar cubes and molasses cakes made by their hands, and shall we slurp up with pleasure milk from the vessels of Muslim Gaddis [dairy farmers] and such castes, and yet not even touch the cow-worshipping, Hindu Chamar? … Is it a sin to even touch the hand of a Chamar and such castes, but we should cling daily to the hands of Yavanas? It’s as though the chillum licked by Muslims is the household deity of the proud Hindu. (S. Sharma n.d.: 21)

The everyday exchanges—even intimacy—between Hindus and Muslims in north India’s integrated, religiously pluralistic society are here presented as a problem: there is something wrong with accepting food and milk from, or sharing the pipe or holding hands with, Muslims. Such quotidian contact, moreover, is figured as a perverse inversion of the Hindu’s non-contact with the untouchable. Gesturing to the common practice of Hindus consulting and accepting the healing remedies of Muslim religious specialists, the author remarks: The Hindu who becomes polluted [bhraṣṭ] by touching Chamars, does he become pure by consuming the ashes spit on by the Mulla and the sweet cakes distributed by the [Sufi] Pir? (S. Sharma n.d.: 21)

The one happens at the expense of the other. It is as though intimacy and disgust were fixed quantities, subject to the law of conservation of matter, in a closed affective economy. From the material domain of sweat and saliva, the author now moves to the moral register of ghṛṇā. When a Qasai [Muslim butcher], who day and night lets loose rivers of blood by murdering harmless goats and other beneficent animals that were only waiting for human affection, is touchable, then for what sin, even of criticizing

Deceptive Majority 106

some aspect of Hindu dharma, is the cow-worshipping Chamar untouchable?’ (S. Sharma n.d.: 25)

Time and again, the reader of Patitoddhār is presented with sensuously descriptive juxtapositions of the untouchable and the Muslim. If the former is “bathed and cleaned,” the latter are “eaters of cow-flesh” who “spit on” ashes, “lick” chillums, and let “loose rivers of blood.” If the Hindu reader is being tasked with the formidable challenge of confronting the ghṛṇā he has been socialized to feel toward the untouchable, the burden is eased, in a way, by the provision of a new object of disgust. Rather than eradicating this most visceral and corporeally rooted of emotions, he is encouraged to redirect it at the Muslim. At the center of Patitoddhār, then, is a project of emotion work, an effort to restructure the pattern of affective bonds that organize north Indian social life, and to do so from within. The Hindu is asked not only to reimagine his relations with Muslims and the Depressed Classes, but to cultivate the emotional habits that would work this newly imagined community into the flesh, into embodied knowledge. Studies of the deployment of ghṛṇā and karāhat in anti-Muslim discourse in present-day Gujarat and Maharashtra may measure the success of this aspect of the Arya Samaj project over the past century (Ghassem-Fachandi 2012; Mhaskar 2018). At the same time as ghṛṇā is rechanneled toward Muslims, though, a new repertoire of emotion practices must be developed in relation to the untouchable, a repertoire that includes fellow-feeling (sahānubhūti) and even love (prem). Thus, after quoting a 1917 speech by C. F. Andrews on the wretched conditions and moral depravity he observed among indentured Indian laborers working on British plantations in Fiji, the author of Patitoddhār asks his readers to practice feeling toward untouchables what they feel reading Andrews’ account: If one of our relatives were compelled to go to one of those islands [like Fiji] and do all of this [coerced] labor, would we excommunicate them from the caste? Would our hearts not melt upon seeing their plight? Surely our love [prem] would rise wavelike and we would embrace him with our heart—in precisely this way we should swiftly embrace and adopt [apnānā] these who were made to be fallen for a longer period. These are our ancestors’ relations, not strangers. (S. Sharma n.d.: 38)

Patitoddhār concludes with a section entitled “Kartavya” (Duty). Here, after quoting the Bhagavad Gita and Booker T. Washington—whom he glosses as

107 Missionary Majoritarianism

“America’s auspicious, dedicated, ideal Shudra”—on the importance of selfreliance, the author lays out an eleven-point program of action for Hindus to save the Hindu nation by reaching out to the “fallen” untouchable. As with the rest of the treatise, these proposals attend to the affective anchoring of caste practice; for instance, when exhorting Hindus to cease hailing untouchables with degrading terms of address (such as nīć, lowborn, or achūt, untouchable), he writes that “the feeling of disgust that comes with the caste names connected to professions should also be obliterated” (S. Sharma n.d.: 190). And in the spirit of Shraddhanand, the eleven proposals include some frontal attacks on the ritual-political economy of untouchability: the institutions of jūṭhan (leftover or half-eaten food as “payment” for sanitation labor) and begārī (coerced labor) are to be summarily abolished. The final proposal is to instill in untouchables that “ancient manliness” that is to be the sign, in the new political dispensation, of their Hindu heritage. “Give them this bold identity: ‘You are the descendants of Kshatriyas!’” (S. Sharma n.d.: 191).

Śrī Bālmīki Prakāś: Ingenious Genealogies, Draupadi’s Disgust The call to radical action among the untouchables, and to the invention of new genealogies, did not go unheeded. Rank and file Arya Samajists, responding to appeals like that by the author of Patitoddhār, Swami Shraddhanand, and others, increasingly stepped forward to conduct dharm prachār among Depressed Class communities in north India in the 1910s and 1920s; in this period they also for the first time extended their overtures to the sanitation labor castes. We have seen that the sanitation labor castes figured in Arya Samaj debates as the limit case of the radicalism of Shraddhanand and his supporters as early as the 1890s. As it happened, it was smaller, regional castes—Rahtias, Meghs, and Ods—that supplied the radicals with their initial proving ground. Yet for a variety of reasons—their great numbers, trans-regional demographic spread, and reputation as the backbone of the Christian mass movements in north India— the sanitation labor castes posed for the Arya Samaj both a challenge and a prize that could not be ignored. In Lahore and later Delhi, a team of Arya Samajists headed by a UP brahmin named Amichand Sharma and a bhajan composer called Sadhu Yodharam (or Yodhnath) launched a mission to the sanitation labor castes that would prove decisive in the subsequent representation of this community as Hindus (B. Das 2007: 34–38; Prashad 2000: 91–92; A. Sharma 1928: 7–10,

Deceptive Majority 108

R. Sharma 1995: 137–41). Their initial efforts met with skepticism, but when Sharma joined sanitation workers in a sweepers’ strike and was jailed along with the workers, he gained credibility and a more sympathetic audience (Prashad 2000: 91–92). As we have seen, this is entirely in keeping with the pattern of Arya Samaj (and Christian missionary) “success” among the Depressed Classes: it hinged on willingness to support untouchables’ acts of ritual-political assertion, and to weather with them the retributive violence that so often followed. A revealing window into the grassroots campaign among the sanitation labor castes in Lahore and Delhi is provided by a tract that Sharma composed entitled Śrī Bālmīki Prakāś, or the Light of Shri Valmiki. If Patitoddhār seeks to persuade Hindus that untouchables are their co-religionists, Śrī Bālmīki Prakāś tackles the problem from the other side. Didactic and polemic in equal measure, the text illustrates with sometimes startling transparency Arya Samaj strategies for persuading the sanitation labor castes to imagine themselves as Hindu, even as it supplies evidence of the informed skepticism with which this effort was met. Dating the text remains tricky, partly due to the fact that Śrī Bālmīki Prakāś went through at least three editions and circulated in lowcost, often undated prints in both Hindi and Urdu. Scholars have suggested its composition as early as 1912 and as late as 1936; evidence internal to the text points clearly to 1928, though this may reflect passages added by Sharma for the second or third edition. In any case, by the 1930s the tract had found its way into a vast number of sanitation labor caste homes; it became “the staple tract of the Balmiki community” (Prashad 2000: 92), the “bible of this new religion” (O’Brien 2012). Even where the text itself did not travel, its central premise and many of its key ideas circulated and took hold. Śrī Bālmīki Prakāś begins with an autobiographical preface and concludes with a collection of hymns (bhajanmālā) intended for Sharma’s sanitation labor caste audience to sing together. Between these is the centerpiece of the tract, a dramatic dialogue between a twice-born Hindu devotee of Ram (Ram sevak) and a man of the sanitation labor castes, in which the former attempts to persuade the latter that his people are descendants of the disciples of Rishi Valmiki, composer of the Ramayana, and that they are therefore Hindu and have been since ancient times. This idea—Sharma’s artful contribution to the Arya Samaj practice of supplying untouchable castes with Hindu genealogies— would become the most enduring legacy of the tract. Though Lal Begi and Mehtar were by far the more common terms of self-description at this time, Sharma gives his untouchable character the title

109 Missionary Majoritarianism

“Balmikiya”—follower of Balmiki.4 The dialogue between the Ram sevak and Balmikiya echoes and allegorically represents the pedagogical relationship Sharma understood himself to have with the sanitation labor castes of Lahore and Delhi. In the opening scene the Ram sevak is singing a bhajan in praise of Rishi Valmiki (written, throughout the text, as Balmiki). As he was singing the above song, a Balmikiya who was passing by heard Balmiki ji’s name and stood nearby. He folded his hands and began to entreat the devotee, “Maharaj [Great King]! The bhajan you are singing is to which Balmiki ji?” The Ram Sevak replied, “What, brother, have there been many Balmiki jis?” Balmikiya: “I am not learned, but I have heard that in the Bhaktamāl the stories of two Balmikis are told. The people of the Granth Sahib [that is, Sikhs] say there were seven Balmikis. Why, is this untrue?” Ram Sevak: “In the Bhaktamāl the two stories of Balmiki do not prove that there were two Balmikis, because the names of the mother and father of either Balmiki are not mentioned. Actually they are two stories about the same Balmiki. And the matter of the Granth is pure gossip.” (A. Sharma 1928: 11)

Immediately we are confronted with the multiplicity of Valmikis in Indic traditions, and the problem this poses for linking the Rishi Valmiki of brahminical tradition to the Balmik of Lal Begi oral tradition. Hearing the Hindu protagonist sing a bhajan to the former—an unlikely scenario, Valmiki never having been a popular object of devotion in South Asia until after the events with which this chapter is concerned—the untouchable character asks the first in a long series of questions. While deprecating his own knowledge, he rightly points out that the literary traditions of South Asia are populated with a number of Valmikis. These include, among others, a Prakrit grammarian; 4

Small numbers of Punjabi Chuhras were identified as Balmiki or Balmikiya as early as the 1901 census, which names “Lal Begi and Balmiki” as a sect of Sikhs numbering less than 3,000 persons, mostly in Ludhiana and Ferozepore districts (the total number of Muslim, Hindu, and Sikh Chuhras in Punjab in 1901 was officially 1,255,679) (Rose 1902: 183–84). Given the variability of naming practices within the caste this is not surprising—as we noted in the last chapter, the Lal Begi liturgy supplied several names for self-reference, including Lal Begi and Bala Shahi or simply Shahi. Balmiki seems to have been, at this point, another such epithet grounded in the Lal Begi oral tradition (in which Balmik plays a significant role), though far less widespread than Shahi or Lal Begi. For Sharma, of course, it was essential to give his protagonist this relatively uncommon, if easily recognizable, caste title.

Deceptive Majority 110

a snake-eating, supernatural bird-of-prey; and a son of Brahma in a medieval Oriya text; in addition to the better-known Bhargava brahmin composer of the Ramayana (whom I have called Valmiki One); the robber-turned-ascetic of late redactions of the Skanda Purana (Valmiki Two); and Balmik the magician of Ghazni in Lal Begi oral tradition. Moreover, there is not only a lack of evidence that Valmiki One and the Balmik of Lal Begi oral tradition were associated in this period, there is positive evidence that they were understood as distinct and separate figures: Bhaktamāl commentators like Priyadas (1712) and Rupkala (1910) speak of “both Balmikis” (ubhay Bālmīki), the “Great Sage” Balmiki and the “Other” Balmiki (Mahārṣi Śrī Bālmīkiji, Dūsre Śrī Bālmīkiji) (Rupkala 1962 [1910]: 2, 248; see also Lee 2019). It is thus on good authority that Sharma’s untouchable character asks the Ram sevak which Balmiki he is singing about. In response, the latter confidently asserts that Sikh literature is false and the distinct Balmikis of the Bhaktamāl are in fact the same. Sidestepping the cycle of stories featuring Balmik as a magician of Ghazni, the Ram sevak contends that all the Balmikis are one: the brahmin sage who composed the Ramayana. This sweeping move makes possible the conclusion at which the text arrives some twelve pages later: Balmikiya: “Maharaj! From everything you have said this has become obvious that we, too, are Hindus, and our guru is eminently worthy.” Ram Sevak: “What doubt is there in this? You people are Hindu. At the last census a pandit of Kashi delivered this opinion [vyavasthā] that [you] people are Hindu. In the census you were returned as Hindu. You people are Hindu, your religion is Hindu, the Veda is your holy book, Balmiki ji is your guru. So those of you who circumcise, who marry and bury by Muslim rites, are perpetrating a great sin.” (A. Sharma 1928: 23)

Sharma’s hectoring repetition of the averment “You people are Hindu” underscores its indispensability to his project. At the same time, in his insistence, the Ram sevak appears to protest too much, rendering visible the very religious alterity he seeks to exorcise. Here he admits—even as he condemns—the fact that some of the sanitation labor castes circumcise, marry by nikāh, take out Muslim-style funeral processions (janāza), and bury their dead. Elsewhere in the tract the Ram sevak deplores the reality that so many of his untouchable interlocutors have become Christian and Muslim and that they accept food from Muslims. The Balmikiya, for his part, demonstrates a deep familiarity with specific Christian and Sikh texts while he appears to be learning about

111 Missionary Majoritarianism

his putative Hindu-ness for the first time.5 The more the tract insists on the Hindu identity of its intended audience, the more it reveals their distance from actual Hindus. Meanwhile in repeatedly mentioning the census, Sharma flags the majoritarian anxiety that animates his text. A further tension is produced by Sharma’s attempt to preserve brahminical social structure even while extending the offer of communal inclusion to its paradigmatic other. The issue is forced on Sharma by his own ingenious device: if his untouchable interlocutors enjoy a genealogical link with Rishi Valmiki, then how can the latter’s impeccable brahmin pedigree, amply attested in the Sanskrit sources, be maintained? To stand by the veracity of the Sanskrit Ramayana—that is, to maintain that Valmiki was a Bhargava brahmin—and at the same time to assert the Balmikiya’s descent from Valmiki would amount to claiming that India’s sweeping and scavenging castes were, in fact, brahmins. Sharma (1928: 19–20) offers an alternative solution: Balmikiya: “Maharaj! Pray do tell: is Balmiki ji our guru by descent or our guru by discipleship?” Ram Sevak: “Shri Balmiki ji is not your people’s guru by descent, rather he is a guru who gives you teachings.” Balmikiya: “From whom are we people descended?” Ram Sevak: “Regarding your ancestry, it is written in the Ośanaś Smriti that: Brahmanya śūdra sansargāta jātah chānḍālamuchyate. Meaning the offspring of the union of a brahmin woman and a shudra man is a Mehtar; and further in this very scripture it is written that your people’s work is sweeping [jhāṛū denā] and removing excrement [mal uṭhānā]. Therefore you are Hindu.”

The deft insertion of discipleship into the tract’s central genealogical premise enables Sharma to overcome the logical impasse, to have it both ways. If the ancestors of the sanitation labor castes were Valmiki’s disciples, then the genealogical imperative can be satisfied—the guru lineage being a culturally valued mode of genealogy—even as the moral and ontological chasm between the brahmin and the Chandal is reaffirmed. Lest there be 5

For example, “If we are Hindu then why do Hindus not touch us?” “Maharaj … I am extremely grateful to you for your kindness, but then this question rises in my mind: why was all of this not known before today?” (A. Sharma 1928: 14, 20–21, 31)

Deceptive Majority 112

any confusion on this point, the Ram sevak invokes shastric authority for the moral propriety of the social abjection and stigmatized labor attached to the sanitation labor castes: they are rightly designated Chandal (again, the paradigmatic antipode of the brahmin, here seamlessly conflated with the modern category Mehtar), appropriately engaged in sanitation work. Two pages later, when the Balmikiya asks, “If we are Hindu then why do Hindus not touch us?” the Ram sevak replies that untouchability is the consequence of the bad deeds (bure karm) of the ancestors of the untouchables, and that if they reform their behavior the Hindus will cease practicing untouchability with them (A. Sharma 1928: 21). As with Patitoddhār, here, too, the Hindu is absolved of historical responsibility for untouchability—though here it is not Muslims but untouchables themselves who have brought about their abject condition. Understandably curious how this situation began, the Balmikiya in Sharma’s dialogue asks, “When and why did we make Balmiki ji our guru?” One of the Ram sevak’s responses is that “when Balmiki ji made the Pandavs’ sacrifice complete, the hearts of the Mehtar people were … drawn toward Shri Balmiki ji, and your elders made him their guru.” Elsewhere in the dialogue the Ram sevak narrates the story to which he here refers, a story that places Balmiki in the world of the Mahabharata. Found both in Priyadas’s commentary on the Bhaktamāl and in the oral traditions of sweepers in the late nineteenth century, this is the one place where the Balmik of the Lal Begi tradition—as opposed to Valmiki One or Valmiki Two—makes an appearance in the canon of Hindu devotional literature, the one narrative site in which Hindus and the sanitation labor castes glimpse one another dimly. Here is how Sharma, through the Ram sevak, tells it: One time the Pandavs held a sacrifice. In order to test its completeness Shri Krishna hung up a conch that would sound of its own accord when the sacrifice was complete. If it did not sound, then it was understood that the sacrifice was incomplete. All the preparations for the sacrifice had been completed; the brahmins and sages had already taken the gifts presented to them, yet the conch did not sound. King Yudhishthir said to Shri Krishna, “Maharaj! The conch did not sound. The sacrifice is incomplete, what can be the cause?” Krishna ji replied, “King! Some devotee of mine remains without food from your sacrifice.” Yudhishthir said, “Where shall we search for him?” Krishna said, “Search everywhere! He will be found.” Arjun and Bhimsen began searching and in a stable they found Shri Balmiki. They returned and

113 Missionary Majoritarianism

told Raja Yudhishthir and Krishna ji. Shri Krishna ji took all five of the Pandavs to Shri Balmiki ji and said to him, “Maharaj, come [to our] home.” He replied, “Maharaj, I am a Chandal, you people become corrupted upon seeing me. For this reason I did not go to your place.” Krishna ji said, “O great sage! To me even a Chandal devotee is very dear, therefore please come and make complete the sacrifice of the Pandavs.” Shri Balmiki ji came to the house of the king. Krishna gave Draupadi the order to make the food. With great love Draupadi prepared thirty-six kinds of dishes. Having served the plates she placed one before Shri Balmiki ji. All five of the Pandavs stood in attendance upon him. Shri Balmiki ji ate the whole mass of it in a few mouthfuls [sab padārth kuch grās khāye]. Seeing Balmiki mix together lentils, cooked vegetables, halwa, pudding, and the rest, a feeling of revulsion arose in Draupadi’s heart [Draupadī ke dil mai ghṛṇā paidā huī] that this man is truly lowborn [saćmuć hī nīć]. Balmiki ji finished his meal, yet still the conch did not sound. Then the Pandavs said to Krishna ji, “Maharaj! The conch still has not sounded.” [Krishna said,] “Ask Shri Draupadi ji.” Draupadi ji said, “Toward the great sage I acted with disgust [maine rishi jī se ghṛṇā kī hai], that is why the conch did not sound.” [Krishna said,] “Now again if Draupadi ji will make food and serve it to him with a pure heart, then the conch will sound.” Draupadi ji again prepared food and with pure emotions [shuddh bhāv se] served it, at which point the conch sounded and the Pandavs’ sacrifice was complete. Shri Balmiki ji, praise be to your devotion and humility which destroyed the pride of haughty brahmins and made the devotees rejoice. (A. Sharma 1928: 17–18)

There is far more to say about this narrative than is possible here. In terms of the support it offers Sharma’s genealogy for the sanitation labor castes, the story seems a poor choice. While throughout Śrī Bālmīki Prakāś Sharma has insisted on a singular Balmiki whose untarnished brahminhood is not in question, the Balmiki in this story unambiguously identifies himself as a Chandal. This is, after all, precisely the figure that the devotional literature tradition labels the “Other” Balmiki, whose Chandal-ness the Bhaktamāl commentator Priyadas further underscores in his influential (c. 1800) version of the story by having this Balmiki speak of sweeping doorsteps and accepting leftover food (jūṭhani lai ḍārau sadā ko buhārau) (Rupkala 1962 [1910]: 155). In order to avoid logical contradiction, then, Sharma is compelled to argue that Balmiki here is no less brahmin than ever, but calls himself Chandal in a kind of rhetorical conceit intended to underscore the inclusivity of bhakti. Thus the value that the magical conch narrative holds for Sharma’s tract appears to be something other than as further “evidence” for his central claim.

Deceptive Majority 114

When one bears in mind the structure of Sharma’s text as a nested series of narratives that thematically reinforce one another (the story of Draupadi and Balmiki is situated within the dialogue of the Ram sevak and the Balmikiya, which is itself situated within Amichand Sharma’s autobiographical account of his time among the Mehtars of Lahore), another possibility emerges. What makes the story such a compelling fit for Sharma’s project is the allegorical purchase it affords both teller and listener on their immediate situation—as actors in a drama of majoritarian inclusion—and its insight into the hard emotional labor this effort entails. It is difficult to miss the resonance of the story of Balmiki’s inclusion in the Pandav brothers’ sacrifice with the experience of Arya Samajists like Amichand Sharma, the author of Patitoddhār and others, along with the tens of thousands of untouchables inducted into the house of Hinduism by shuddhi in the 1910s and 1920s. The Pandav brothers, who have neither heard of nor have any business with the Chandal who lives out in the stables, find themselves suddenly in need of him to accomplish a political end, a sacrifice. They ultimately achieve their goal by going to him, persuading him to join them, and interdining with him amid much fanfare. This is an almost perfect parable of the logic and practice of Hindu majoritarian inclusion. Like Balmiki, the sanitation labor castes and other Depressed Classes in the 1910s and 1920s were abruptly told that a community who, until a moment before wanted nothing to do with them, now needed them, and moreover that they now belonged to that community. The magical conch announcing the formation of a Hindu-majority India could not be sounded without them. The story also provides an allegory of the emotion work that such a radical refashioning of community boundaries requires. Whatever the Pandav brothers may be imagined to have felt in the story, Sharma’s telling of it (as also with Priyadas’s and Rupkala’s) sets the burden of revolutionary feeling onto the narrative’s only woman, Draupadi. Tasked not only with preparing and serving food to the unusual guest, but with doing so out of love (prem) and pure emotions (shuddh bhāv), Draupadi nonetheless finds herself wrestling with ghṛṇā. In subtle but significant ways the Hindi here underscores ghṛṇā’s embeddedness in the flesh—Draupadi does not simply “feel disgust” as one might in an anglophone imagination but rather ghṛṇā emerges or is birthed in her heart (dil mai paidā huī). Sharma’s prose also emphasizes the activeness of ghṛṇā as an intersubjective practice: “Toward the great sage I acted with disgust,” Draupadi acknowledges, using the transitive compound ghṛṇā karna,


Missionary Majoritarianism

to do disgust, to practice disgust, an action that requires an object. For a twiceborn woman or man to practice disgust with a Chandal is, of course, entirely appropriate in the normative ethics of the brahminical tradition. Disgusting— nirghṛṇāh—is a defining epithet of the Chandal in no less of a textual authority than, ironically, Valmiki’s Ramayana (Jha 1986: 18). Nirghṛṇāh is what the Chandal is; ghṛṇā is what one is supposed to feel, and the way one is supposed to act, toward this category of being. Draupadi’s initial behavior toward the food-mixing Balmiki is thus neither aberrant nor idiosyncratic, but normatively correct and understandable in a conservative Hindu milieu. Like the author of Patitoddhār, who understandingly portrays the revulsion his Hindu readership has heretofore felt toward the untouchable, Sharma offers in Draupadi a figure who embodies prevailing norms yet who models a trajectory of transformation, of converting ghṛṇā into prem. Like the Arya Samaj itself, which in publications in the 1890s admitted “utter repugnance” at the idea of fraternizing with untouchables but which by the 1920s was known as one of their most outspoken allies, Sharma’s Draupadi exhibits a back-and-forth struggle with her own ghṛṇā, an effort to unlearn and disembed those affective habits the social order has most deeply sunk into the body.

A Muslim Mission? Khwaja Hasan Nizami’s Halalkhor Even as the Hinduizing message of Sharma and his Arya Samaj colleagues began to gain traction in sanitation labor caste neighborhoods in Lahore and Delhi, a publisher in another part of Delhi was printing two thousand copies of an Urdu book entitled Halalkhor, the cover of which promised “an elucidation of the entire religious circumstances of the Halalkhor community for the knowledge of Muslims and the propagators of Islam.” The author was Khwaja Hasan Nizami, author of Da’i-i Islam (The Missionary of Islam), the manual of Islamic tabligh (propagation) that Swami Shraddhanand condemned as “the Muhammadan Conspiracy” and that drew Nizami and the swami into their highly publicized rivalry. Published in 1923—the same year as Da’i-i Islam and Shraddhanand’s response Hinduo Savadhān! (Hindus Beware!)—Halalkhor is in some respects an exact antithesis of Sharma’s Śrī Bālmīki Prakāś. Where Sharma insists on the transhistorical Hindu-ness of the sanitation labor castes, Nizami argues that these same castes have long followed the path of Islam. While Sharma wishfully gives his protagonist the title Balmikiya—the most Sanskritic and

Deceptive Majority 116

Hindu-sounding of names adopted in various regions by the sanitation labor castes—Nizami selects the most venerable of the Islamicate titles for the same community, the Arabic-Persian Halalkhor, “one who earns an honest living.” And just as Sharma’s tract opens with an attempt to elevate one key figure in the Lal Begi pantheon (Balmik) above others and to provide him with Hindu credentials, Nizami’s begins by foregrounding the other key figure (Lal Beg) and an effort to certify his Muslim identity. The opening chapter, “Halalkhors and Their Lal Beg,” begins thus: Lal Beg was the Halalkhors’ pir and guide [hādī], in other words the guru of those people whose occupation was cleaning filth [ghalāzat sāf karnā] and rendering servile labors. The name of their religious guide [dīnī rahbar] was the famous Lal Beg. When did he live? Of what pedigree was he? What was his education? These are some of the many important questions that will be considered here. (Nizami 1923: 1)

From here Nizami launches into an academic debate over Lal Beg’s name—was it a misreading of Balmik written in nastaliq, or perhaps a corruption of Lal Bhikshu, red monk?—indicating his familiarity with discussions on this topic in the pages of Panjab Notes and Queries and monographs authored by colonial folklorist-administrators. Contending that the people of north India are not so dull as to mistake the well-known Mughal title Beg for partial homonyms or similarly written lexemes, and that the speculations of Denzil Ibbetson and others thus hold little water, Nizami (1923: 2) opines that “Lal Beg was never the Hindu Balmik; rather he was some Muslim elder [buzurg].” The Muslimness of Lal Beg is then a step toward a broader claim, clearly signaled in the title of the book’s next subsection: “Stories of the Halalkhors That Prove They Are Muslim.” The stories that follow are indeed—as so many of the oral traditions of the Lal Begis documented in the late colonial period were—populated with Islamic figures (Ali, Bibi Fatima, Imam Hussein, the Prophet Ilyas, the Sufi “Pir of Pirs” Abdul Qadir Jilani) and redolent with Islamic themes. In later sections, though, Nizami also includes the story of Draupadi’s disgust, and that of Balmik manifesting himself as a corpse in Kashi, and other narratives that, as his own chapter headings acknowledge, are “hinduānā”—Hindu-esque, or situated in a Hindu milieu. Nizami makes no effort to delegitimize these tales as spurious, or to reclaim them for Islam; rather, he simply recounts them without comment. As the text proceeds, it becomes increasingly apparent that


Missionary Majoritarianism

Nizami’s project is only partly, or half-heartedly, an Islamic mirror-image of Sharma’s Hinduizing protreptic. Whereas Sharma seeks to subordinate Lal Beg as “simply another disciple of Shri Balmiki ji” and exhorts the sanitation labor castes not to identify with the former (“those who call themselves Lal Begiyas are making a mistake; you all should always and forever call yourselves Balmikiyas”), Nizami makes no parallel admonishment, and seems curiously untroubled by the presence of the Hindu-esque alongside the Islamicate in the Lal Begi narrative world. Indeed, as Halalkhor unfolds, the argumentative tenor of the opening passage drops out of view entirely, and the promise to “prove” that the sanitation labor castes are Muslim is quietly forgotten. Instead, the text is overtaken by the accumulative impulse and organizational conventions of late colonial ethnography; what begins as a controversialist’s polemic becomes a folklorist’s miscellany. The reader is led through a compilation of oral traditions (“The Name Halalkhor Bestowed by Imam Hussein,” “Another Strange Tale of Their Being Muslim,” “A Hindu-esque Legend,” “Stories of the Halalkhors of Benares,” and so on), followed by genealogical songs (“The Kursīnāmās of the Mehtars of Punjab”), followed by accounts of the customs and rituals of the community (“Birth Rites,” “Funerary Rites,” “Their Way of Marriage,” and so on). Quite damaging to Nizami’s initial claim (or for that matter Sharma’s) is the section entitled “Rites of Admission to the Lal Begi Religion” (“Mazhab-e Lāl Begī ke Dākhila kī Marāsim”), which details the sacral offerings that a man must make before Lal Beg’s shrine (here called a gurūdwāra) to become a Lal Begi. Yet Nizami includes it, and neither denounces the rite nor attempts to cast it in an Islamic light. The theme of disgust arises in several of the stories Nizami includes. There is a narrative in which the young Imam Hussein suffers a bout of diarrhea and his mother, Bibi Fatima, asks a sweeper woman to clean up after the child. When Bibi Fatima sees that the woman does the cleaning “cheerfully” and “without aversion” (zarā karāhiyat na āyī), she blesses the woman and her people (qaum) with invulnerability to disease (Nizami 1923: 10). If this story reflects a set of oral traditions in which the sacred status of the Prophet and his family enables the usual affective entailments of manual scavenging to be held in abeyance—for piety to overcome revulsion—others in Nizami’s collection center on privileged caste disgust toward the sanitation labor castes. In the version of the horse-with-three-legs legend that Nizami recounts, Lal Beg makes a speech on the sinfulness, in the eyes of God, of disgust practices like the refusal to eat with sweepers. Lal Beg tells his host Jivan,

Deceptive Majority 118

Man is born pure [pāk]. There is no thing that can make him impure…. You people clean the filth of the external world [zāhir kī ghilāzat], and from this your interior [bātin] becomes clean, whereas those people who avoid external dirt, and treat you as dirty, their internal self becomes impure [najis], and nothing but the flames of hell can cleanse that impurity” (Nizami 1923: 13).

It is only in the closing pages of the book, in a section entitled “A Movement among the Halalkhors” (“Halālkhoroṅ mei Ek Tehrīk”), that Nizami revisits the question of the sanitation labor castes’ relationship with Islam. He writes: Approximately twenty-five or thirty years ago a person named Mirza Imam al-Din Sahib, who was a resident of Qadian [Punjab] and was cousin of Mirza Ghulam Ahmad [founder of the Ahmadiyya sect], observing his relation’s claim to being messiah and Mahdi, wanted to become the pir and guru of the Halalkhors. (Nizami 1923: 75)

In attributing to Imam al-Din a kind of guru envy, a desire for religious fame on par with his controversial cousin, Nizami portrays his subject in the same unflattering light in which he appears in the writing of British administrators, Christian missionaries, Arya Samajists, and Mirza Ghulam Ahmad himself (although the administrators, missionaries, and Arya Samajists paint both Imam al-Din and Ghulam Ahmad as dubious self-promoters). Nizami explains that Imam al-Din, in his writings and speeches, advised the sanitation labor castes that their Lal Begi religion was “superior to all the religions of the Hindus, Muslims, Sikhs and Christians” and that they had “no need to choose any new religion.” Nonetheless, Nizami insists, even though the eccentric Imam al-Din here and there spoke badly of Islam and found fault with its values like a nonMuslim would, still he advised the Halalkhors to uphold Tauhid [the unity of God, Islam’s cardinal doctrine]. From a number of his words it is clear that he skillfully furthered the propagation of Islam [tabligh-e Islam]…. Mirza Sahib had launched a propagation movement. (Nizami 1923: 76)

In other words, despite his unorthodox methods, Imam al-Din’s preaching to the sanitation labor castes could be considered a form of Islamic tabligh, and thus a precedent for Nizami’s effort to counter Hindu majoritarian shuddhi in the 1920s. Nizami concludes the section—and the book—as follows. I have heard that Halalkhors used to seat Mirza Imam al-Din on a chārpāī [cot] and carry him from village to village and treat him with great reverence. If he were alive today, or if some individual were to be his successor, then this


Missionary Majoritarianism

movement for the improvement of the Mehtars [tahrīk-e islāh-e mehtarān] would grow and profit greatly. But perhaps after the death of Mirza Sahib the movement too has ceased, because these days one hears nothing of it anymore. (Nizami 1923: 78)

This is the damp note on which Halalkhor ends. Nizami draws his book to a close by blandly observing the ephemerality of Imam al-Din’s legacy with the sanitation labor castes, a legacy in any case questionable from the point of view of normative Islam. If, in Nizami’s talk of a successor to Imam al-Din, there is a wishfulness or even an implicit plea to would-be promoters of tabligh, it is still a far cry from the systematic call to action we have seen in Patitoddhār, Śrī Bālmīki Prakāś, and other writings and speeches by Arya Samajists in this period. The difference is symptomatic, ultimately, of the gulf between Nizami’s foray into Depressed Class missionizing and the sustained campaigns of the Arya Samajists. While impressive as a compilation of colonial-era ethnography, Nizami’s treatise reflects nothing of the grassroots attempts by Amichand Sharma and other Arya prachāraks to persuade the sanitation labor castes that they were Hindu. Whereas Sharma joined a sweepers’ strike and was jailed along with those he sought to proselytize, Nizami’s was a comparatively detached engagement; he “only rarely went on walking tours to villages to teach isolated and marginalized Muslim communities” (Hermansen 2008: 159). Halalkhor, moreover, attained nothing like the “staple tract” status that Śrī Bālmīki Prakāś came to enjoy; one does not find copies (or even rumors of copies) of Nizami’s book in sanitation labor caste neighborhoods in urban north India. If, instead, Nizami’s object was more akin to that of the author of Patitoddhār—if, that is, he sought to inspire fellow Muslims to reimagine the sanitation labor castes as co-religionists and to encourage untouchable uplift by Muslims—then, too, the results appear to have been meager. Despite his prolific advocacy of tabligh in print, Nizami “was apparently not able to mobilize many Muslim volunteers to take up his cause” (Hermansen 2008: 160). While his writings may have indirectly inspired tabligh among Dalits elsewhere in the subcontinent—such as the remarkable grassroots movement led Bahadur Yar Jang in the princely state of Hyderabad in the late 1920s (Muhammad 1973: 57–65; Pernau 2000: 251–53)—Nizami does not appear to have generated a volunteer movement in the north Indian heartland of the sanitation labor castes. But if Nizami’s efforts failed to check the progress of the Arya Samaj’s shuddhi campaigns, his writings illuminate the degree to which the religious

Deceptive Majority 120

identity of the sanitation labor castes remained, in public discourse, an unsettled question. It was not unthinkable, in the 1920s, to claim in print that this transregional swath of the Dalit population was, in practice, Muslim. Nor was it outrageous for as prominent a leader in the Indian National Congress as Mohammad Ali to suggest—as he did in his presidential address in Kakinada in 1923—that Hindus, Muslims, and Christians could “divide the suppressed classes and separately convert them” (quoted in Prashad 1996: 554). If remarks like Ali’s or treatises like Nizami’s were vulnerable to the accusation of communalist motives, they were no less so than the contrary claims of Shraddhanand, Amichand Sharma, and other Arya Samajists. Together their writings illustrate just how accurately the cartoon of Hindu, Christian, and Muslim sportsmen competing over “the untouchable football” reflected public debate in this period. Like the cartoon’s ball in the air, the issue of Dalit religious identification was very much in play. Swami Shraddhanand, murdered in 1926, did not live to see the Arya Samaj’s first significant successes among the sanitation labor castes. It was only at the very end of the 1920s that Amichand Sharma’s mission began to bear the kind of numerical fruit the Arya Samaj so desired. In 1930, over six thousand members of the sanitation labor castes underwent shuddhi in Lahore, followed by another 1,530 three years later (S. K. Sharma 1985: 70). In Delhi the early 1930s saw the first cremations in the community (which had previously buried) and the replacement of Lal Beg shrines with Valmiki temples (Prashad 2000: 109–10; R. Sharma 1995: 141). In Jodhpur in the same years members of the sanitation labor castes, in response to Arya Samaj efforts, banned beef-eating and replaced “their Muslim names like Akbur, Jamal, Multan, Data Deen, Aladeen, etc., by names like Jagdish Kumar, Ganesh Ram, Ram Chandra, and likewise” (Shyamlal 1984: 29–30). The Valmiki movement was afoot. Still, these were acts of thousands, whereas hundreds of thousands of the sanitation labor castes by this time professed Christianity and Islam, and even greater numbers continued to call themselves Lal Begi. Ultimately, as we will see, Valmiki would require another patron: the Arya Samaj would need allies to achieve, on a mass scale, its majoritarian ends. The uphill struggle to reinvent the category of the Hindu to include its despised antipode would necessitate more than the Samaj’s missionary efforts at persuasion, vigorous though they were. It would require a canny intervention in and deployment of the categorical schema by which the colonial state recognized its subjects, an exercise in thinking and acting like a state. In this endeavor a decisive role would be played by Mohandas Gandhi.

4 Trustee Majoritarianism Gandhi and the Harijan Sevak Sangh

Bhangi as Synecdoche In a speech at a “Suppressed Classes Conference” in Ahmedabad in 1921, Mohandas Gandhi, after expressing regret at how few “untouchables” were in attendance, related the following vignette from his childhood in western India. A scavenger named Uka, an untouchable, used to attend our house for cleaning latrines. Often I would ask my mother why I was wrong to touch him, why I was forbidden to touch him. If I accidentally touched Uka, I was asked to perform the ablutions, and though I naturally obeyed, it was not without smilingly protesting that untouchability was not sanctioned by religion, that it was impossible that it should be so. I was a very dutiful and obedient child and so far as it was consistent with respect for parents, I often had tussles with them on this matter. I told my mother that she was entirely wrong in considering physical contact with Uka as sinful. (Gandhi 1966a: 570)

The story of Uka supplies a fine example of Gandhi’s carefully calibrated message to traditionalist Hindus: insisting that he was himself a sanātani Hindu and thus one of them, Gandhi denied the validity of religious sanction for untouchability and encouraged his co-religionists—not by denunciation but by patient persuasion—to follow his example in abandoning the practice. His portrayal of the “tussles” over untouchability as a relatively non-rancorous affair within a traditional family can be read as further support for the argument of Nishikant Kolge and others that Gandhi’s approach to reforming caste was a strategic gradualism, calculated not to alienate the orthodox Hindus whose cooperation he felt to be crucial, but rather to invite them to adopt more liberal ways (Brown 1990; Kolge 2017; Lelyveld 2011; Nanda 1985). What, though, of Uka? The latrine cleaner, having provided the occasion for Gandhi’s display of progressivism, vanishes from the narrative. One wonders what Uka may have thought about the debate over sin in the Gandhi household,

Deceptive Majority 122

or in what terms Uka understood his own relation to the religious community of the family whose toilets he cleaned. One might ask how Uka got his name. Given what we now know of the sanitation labor castes in this period, it would be unsurprising if “Uka” was the name bestowed by the senior Gandhi or other Hindu employers of Uka’s parents—an exercise of the “lordly right of giving names” (Nietzsche 1989: 26) that privileged caste householders generally enjoyed with the Dalits on whose labor they depended. But in his own home, was “Uka,” like many of his counterparts elsewhere in north India, known by a name with Islamic referents—Ali, Akbar, Eidu, Ramzanu? Questions such as these cannot be asked within the narrative structure of Gandhi’s account, where the true protagonists are Gandhi and his mother, reformist Hindu and traditionalist Hindu. A brief review of the verbs in Gandhi’s narrative makes clear that Uka’s agency is circumscribed to “attend[ing]” and “cleaning latrines,” whereas touching, abluting, protesting, and tussling are Gandhi’s domain. The question of who touches whom and with what consequences is not Uka’s to decide. The story of Uka is one to which Gandhi returned not infrequently. Uka, as Vijay Prashad (1996: 552) observes, stood for “the generic [B]hangi” in Gandhi’s thought, and the Bhangi, in turn, stood for all of the Depressed Classes. “With his gift for symbolism,” writes Eleanor Zelliot (2010: 154), “Gandhi selected the Bhangi, a scavenger caste of north India, to represent the problem of untouchability.” In other words, it was the figure of the panregional sanitation labor caste par excellence, the Chuhra or “Bhangi”—and not the Mahar, or Chamar, or Pariar, or other major Dalit castes of colonial India—that would serve as synecdoche for all of the Depressed Classes in Gandhi’s sociology. It was the very community with which this book is primarily concerned, the caste that produced and was largely isomorphic with the Lal Begi qaum or religious community, that would stand for the whole of the untouchable masses in Gandhi’s representational practices. The index of the Government of India’s one hundred volume Collected Works of Mahatma Gandhi identifies more references to the word “Bhangi” in Gandhi’s oeuvre than to such other caste (jāti and varna) groups as Pariar, Chamar, Dhed, Mala, Madiga, Namasudra, Arunthathiyar, Vankar, Shudra, Rajput, and (Gandhi’s own) Baniya all combined.1 Indeed the only caste category that appears more 1

These remarks apply to the Second Revised Edition of 2001. Later editions of the Collected Works of Mahatma Gandhi (CWMG) have removed the Bhangi entry from the index.

123 Trustee Majoritarianism

frequently than Bhangi in Gandhi’s collected works is Brahmin. Gandhi spoke about the Bhangi almost as often as he spoke about the Bhagavad Gita, the scripture he considered his “eternal mother” and most profound inspiration.2 The consequences of Gandhi’s intensive involvement in the category of the Bhangi, and his deployment of the Bhangi as synecdoche for all untouchables, have yet to be fully appreciated. I certainly failed to appreciate them in the early days of my research. I was baffled, for instance, when some of the older men among my Balmiki interlocutors in Lucknow told me that the caste title Valmiki/Balmiki was Gandhiji kā den, the “gift of Gandhiji.” But what did Gandhi have to do with it? The particular majoritarian strategy of Hinduizing Dalits by linking them with Rishi Valmiki was, I knew by then, developed and pursued by the Arya Samaj. Gandhi disapproved of the Samaj’s shuddhi campaigns; he even broke with Swami Shraddhanand over the issue. The rare appearances of Valmiki as a caste moniker in Gandhi’s collected works suggested that he approved of the title, but there was no evidence that Gandhi himself sought to promote its use among the sanitation labor castes. On this point, I thought, the elders surely must be mistaken. Eventually, however, the meaning that these men intended, which was both broader and less direct than I initially understood, became clear. It was not the forging of the association between the subaltern caste and the exalted rishi to which they were pointing, nor the Hinduizing campaigns in Lahore and Delhi that first popularized Valmiki in Dalit bastīs. Rather, they were gesturing to the conditions under which the Valmiki title was finally adopted on a mass scale. These conditions, which elevated the idea of the transhistorically Hindu untouchable—the conceptual lynchpin of the Hindu majoritarian project— from the radical fringe to the political mainstream and ultimately the law of the land, were in an unobvious yet crucial sense the legacy of Gandhi’s discursive framing of untouchability and his mode of engagement with Dalits, as institutionalized in the Congress, the Harijan Sevak Sangh, and the legal framework of independent India. They were, as my interlocutors rightly perceived, Gandhiji kā den. Among both detractors and admirers of Gandhi it has become commonplace to acknowledge a certain “paternalistic” or “patronizing” quality in his dealings with the Depressed Classes. While there is a certain truth to these claims, their analytical purchase is limited, and this way of formulating 2

In the CWMG index, “Bhangi” has 163 entries and the “Bhagavad Gita” 166.

Deceptive Majority 124

the problem has hobbled a more robust examination of the effects of Gandhi’s interventions in the social order. When what is meant by “paternalism” is an attitude of condescension tempered with affection, the claim does illuminate an important aspect of Gandhi’s relation with untouchables, but hardly the only one of consequence, and with insufficient attention to the particular cultural meanings of the relation. When what is meant by “paternalism” is a political doctrine in which one group or individual arrogates to itself the capacity to determine the best interests of and act on behalf of another group construed as dependent or subordinate, the shoe again fits, but still obscures more than it reveals. In telescoping the complex problem of untouchability into the singular figure of the scavenger, and in naming and asserting over this figure specific kinds of governmental-enumerative and brahminicalontological knowledge, Gandhi put in motion a representational politics much more complex, culturally particular, and far-reaching in its consequences than the usual remarks about “paternalism” suggest.

The Poona Pact: Hindus “in Spite of Themselves” The Indian National Congress passed its first resolution on untouchability in December of 1917, seventeen years into the Arya Samaj’s shuddhi campaigns and two years before Gandhi’s rise to nationalist leadership. Resolved that this Congress urges upon the people of India the necessity, justice and righteousness of removing all disabilities imposed by custom upon the Depressed Classes, the disabilities being of a most vexatious and oppressive character, subjecting those classes to considerable hardship and inconvenience. (Zaidi 1987: 400)

The landmark resolution signaled that the Congress would no longer sidestep the caste question by delegating it to the Social Conference or other bodies. Little followed in the way of concrete action—Ambedkar (1991: 18) summarized it as “a dead letter”3—but five years later the Congress formed a committee to formulate a practical scheme for improving the conditions of the Depressed Classes. In a decision indicative of the acceptability Hindu majoritarian views enjoyed in the Congress, the person appointed by the Congress to head the new committee was none other than Swami Shraddhanand. Within a month, 3

“Congressmen … forgot the Resolution the very day on which it was passed. The Resolution was a dead letter. Nothing came of it.”


Trustee Majoritarianism

though, Shraddhanand resigned, alleging that Congress leaders denied the committee sufficient funds, part of what he saw as a long pattern of Congress duplicity and prevarication on the caste question (Shraddhanand 1946: 133–38, 147–48, 179–88). Within a year, the 1922 committee was dissolved; like the 1917 Depressed Classes Resolution it was intended to implement, the second formal effort by the Congress to address untouchability had minimal effect. It was not until a decade later, when the British government accommodated the Depressed Classes with the “Communal Award” urged by Ambedkar, that the existential crisis Shraddhanand had forewarned became, for most members of the Congress, a palpable reality. No small amount of ink has been spilled in narrating and interpreting the sequence of events that culminated, in September of 1932, in Gandhi’s “epic fast” against “separate electorates” and the Poona Pact that brought it to an end. The basic contours are well known: in a series of Round Table Conferences in London in 1930–32, political leaders representing a range of communities in India met with representatives of the British government to formulate a framework for constitutional reforms. At the second conference Gandhi and Ambedkar clashed sharply and publicly over whether the Depressed Classes were “entitled to recognition as a separate entity” in the legislatures (Ambedkar 2009: 52). Ambedkar argued extensively in favor of such recognition and built a coalition of minorities persuaded of its necessity; Gandhi, while acquiescing to representation by special constituencies for Muslims and Sikhs, resisted it for the Depressed Classes. The two also argued over who genuinely represented the untouchables. Ambedkar (2005: 661) claimed, “I fully represent the claims of my community,” while Gandhi countered: I claim myself in my own person to represent the vast mass of the Untouchables. Here I speak not merely on behalf of the Congress, but I speak on my own behalf, and I claim that I would get, if there was a referendum of the Untouchables, their vote, and that I would top the poll…. Let this Committee and let the whole world know that today there is a body of Hindu reformers who are pledged to remove this blot of untouchability. We do not want on our register and on our census Untouchables classified as a separate class…. I would far rather that Hinduism died than that Untouchability lived. Therefore, with all my regard for Dr. Ambedkar, and for his desire to see the Untouchables uplifted, with all my regard for his ability, I must say in all humility that here the great wrong under which he has laboured and perhaps the bitter experience that he has undergone have for

Deceptive Majority 126

the moment warped his judgement…. I say that it is not a proper claim which is registered by Dr.  Ambedkar when he seeks to speak for the whole of the Untouchables of India. It will create a division in Hinduism which I cannot possibly look forward to with any satisfaction whatsoever…. I want to say with all the emphasis that I can command that if I was the only person to resist this thing I would resist it with my life. (Gandhi 1971: 297–98)

When the conflict over minority representation in the constitutional reforms proved intractable, the delegates signed an agreement to abide by a decision that the British government would make. This decision—known as the “Communal Award” and announced several months later in August 1932—permitted qualified members of the Depressed Classes to vote both in a general constituency and, “in selected areas where the Depressed Classes are most numerous,” in a special constituency in which only the Depressed Classes could vote. The arrangement was to be temporary, with the special constituencies dissolving in twenty years. The Communal Award was thus a compromise of sorts between Ambedkar’s and Gandhi’s positions: the Depressed Classes would be part of the general electorate (unlike Muslims and Sikhs, who had actually separate electorates), but they would also, in a concession to Ambedkar’s demand for political safeguards, have the capacity to elect their own representatives in the “special constituencies,” at least in some places and for a limited time. True to his vow, Gandhi announced “a perpetual fast unto death” against the provisions of the Communal Award, clarifying that “what I am against is [the Depressed Classes’] statutory separation even in a limited form, from the Hindu fold” because “in the establishment of separate electorate at all for the Depressed Classes I sense the injection of poison that is calculated to destroy Hinduism and do no good whatever to the Depressed Classes” (Gandhi 1972a: 383; 1972c: 31). Privately, the Mahatma elaborated his concern that if the Depressed Classes were granted separate electorates, “‘Untouchable’ hooligans will make common cause with Muslim hooligans and kill casteHindus” (quoted in Roberts 2016: 143). He declared that he would cease the fast only if the government “of its own motion or under pressure of public opinion,” withdrew the provision for special Depressed Class constituencies altogether (Gandhi 1972a: 383–84). Ramsay MacDonald, the British prime minister, made clear that the government would only retract its arrangement if an “agreement of the communities themselves” were brought forward (Ambedkar 2009: 84). The onus to “save Gandhi’s life” thus fell abruptly


Trustee Majoritarianism

onto Ambedkar’s shoulders; as the highly publicized fast began, immense nationalist pressure was brought to bear on Ambedkar to reverse his position and abandon the demand of his community for special constituencies (which, despite being misleading in the context of Communal Award, continued to be called “separate electorates”). Ultimately, in a series of bedside meetings with the enfeebled Mahatma, Ambedkar negotiated a further compromise in which the special constituencies were given up but an increased number of reserved seats for Depressed Class candidates in the general constituency were secured. The terms were written into a nine-point agreement—the Poona Pact— which Ambedkar signed with Madan Mohan Malaviya and other Hindu representatives on the fifth day of Gandhi’s fast. When the British government accepted the Poona Pact as a replacement of the Communal Award the following day, Gandhi broke his fast. The Communal Award and Gandhi’s fast then provoked a surge of nationalist engagement with the untouchability problem unprecedented in scale and intensity. What Shraddhanand’s dire predictions of Dalit political assertion had failed to do a decade earlier, the realization of his predictions in the form of the Communal Award succeeded. A temple entry bill and an untouchability abolition bill were introduced in the legislature. Gandhi undertook another highly publicized fast4 and a nine-month, all-India “Harijan tour” to raise funds for “untouchable uplift.” In not insignificant numbers, schools, wells, and temples were opened to the Depressed Classes, and these openings were publicized in three new weeklies—the Harijan (English), Harijan Sevak (Hindi), and Harijan Bandu (Gujarati), with overlapping but not identical content—launched in 1933, replacing Young India as Gandhi’s mouthpiece. “Untouchable uplift” leapt from the periphery to the center of nationalist discourse. Five days after the signing of the Poona Pact, Gandhi’s close colleagues inaugurated an All-India Anti-Untouchability League, later renamed the Harijan Sevak Sangh—an organization we will examine in some detail later in this chapter. Within months of the threat of Dalits voting for their own candidates, the Congress had done more to address untouchability than it had in the previous three decades.


Between May 8 and 29, 1933, this fast was intended “to make it clear that the movement [against untouchability] is purely religious and to be prosecuted by religious rectitude” (Gandhi quoted in Verma 1971: 59).

Deceptive Majority 128

In the highly charged debates over the causes and consequences of the Poona Pact, what has generally been overlooked is the role it played in conceptually confining the untouchability problem within the house of Hinduism.5 How this is so can be thrown into relief by comparing the pact with what came before. Recall that in the Depressed Class Resolution of 1917, when the Congress urged “the necessity, justice and righteousness of removing all disabilities imposed by custom upon the Depressed Classes,” it urged that necessity “on the people of India” (Zaidi 1987:400, emphasis added). That is, the structural injustice of untouchability was conceived of here as the responsibility, not of Hindus alone, but of “the people of India” as a whole—sensibly so, given the prevalence of untouchability practices in churches, mosques, and gurudwaras as well as in temples, and in the everyday exploitation of Dalit labor by landlords of all faiths. The idea of collective responsibility for untouchability was no stranger to political discourse before 1932. While some prominent Muslim leaders denied that Muslims shared culpability with Hindus for caste, others publicly acknowledged it—Sarojini Naidu’s protégé Bahadur Yar Jang, for example, often criticized fellow Muslims for practicing untouchability, while Muhammad Iqbal deplored that “we have out-Hindued the Hindu himself” in the matter of caste prejudice (Muhammad 1973: 57–65; Pernau 2000: 251– 53; Titus 1930: 171–72). Swami Shraddhanand, for his part, never missed an opportunity to call out Muslims, alongside Hindus, for denying untouchables access to wells. And claims like these of shared blame for untouchability implied a shared obligation to act. Whatever else one takes from Mohammad Ali’s 1923 speech as Congress President—wherein he proposed that Muslims, Christians, and Hindus could “divide the suppressed classes and separately convert them” (quoted in Prashad 1996: 554)—it clearly indicates that Hindus were not considered the sole stakeholders in the untouchability problem. With the Poona Pact, the collective responsibility of “the people of India” drops from view, to be replaced by Hindu responsibility. The pact was presented by Gandhi and its signatories as an agreement between the Depressed Classes and “the Hindu community” (Pyarelal 1932: 153, 313–14). Malaviya, as principal signatory, was acting “on behalf of caste-Hindus” 5

Recent, welcome exceptions are Shabnum Tejani (2008) and C. S. Adcock (2014: 14), who notes that “the premise, now enshrined in the Constitution of India and in much common-sense understanding, that Untouchables are Hindu by religion … was put in place with the agreement between Gandhi and Ambedkar in 1932.”

129 Trustee Majoritarianism

(Verma 1971: 48). Six days later, when the signatories reconvened in Bombay with other supporters to ratify the pact and launch the Anti-Untouchability League, they did so unambiguously as a “public meeting of Hindus” (Verma 1971: 48–52). What had been a problem for all of India’s people to address was now presented as a matter to be adjudicated by Hindus. In part, of course, this framing of things was a response to the prime minister’s insistence that only an “agreement of the communities themselves,” and not the threat of Gandhi’s fast, would impel government to retract the Communal Award. But such an agreement of communities could have been configured any number of ways—Sikhs, Muslims, and Christians could well have been consulted, as indeed they had been by Ambedkar in the Round Table Conferences. Gandhi, however, specified the configuration of parties he wanted to deal with—the parties who would play the only roles available in his scripting of the story—in his “Statement to the Press” on the eve of his fast: “I should … abide by an agreement … that may be arrived at between the responsible leaders of caste Hindus and the ‘depressed’ classes and which has been accepted by mass meetings of all Hindus” (Gandhi 1972c: 64). It was not that Gandhi was unaware of Muslim, Sikh, or Christian practices of untouchability—elsewhere he writes that “by looking down upon the Bhangi we—Hindus, Mussalmans, Christians and all—have deserved the contempt of the whole world” (Gandhi 1936). But in his deliberative framing of the untouchability problem at the time of his fast, talk of collective responsibility is eschewed; Hindus alone emerge as bearing the burden of culpability and, with it, the moral obligation to act. This obligation was the Hindu’s guarantee of a politically agentive role—in fact, the only politically agentive role—in the drama to come. There is no question that the shift from the national-cultural frame of the 1917 Congress resolution to the religious-communitarian frame of the Poona Pact involved a complex series of moves by multiple actors in a political field fundamentally structured by imperial interests. Yet in that complex series of moves, Gandhi played no small part in assiduously steering the debate away from the collective implication of “the people of India” in untouchability and toward a framework of exclusively Hindu responsibility and reform. Gandhi was the first political actor of stature to apply with consistency the still-novel idea that untouchables were already or transhistorically Hindu. The Arya Samajists, as we have seen in their texts and pronouncements, continued to fumble over the idea even as they aggressively promoted it in the 1910s and 1920s. Their very practice of shuddhi, which marked the ritual participant as

Deceptive Majority 130

initially exterior to the Arya/Hindu community in the manner of a Muslim or Christian, remained a stark reminder of the logical contradiction of their position. Other reformers who sought to bring about a merger of Hindus and Depressed Classes continued to talk about the two as distinct and contrastive categories. “Hinduism absorbed the Shudras,” wrote B. G. Tilak, Gandhi’s political mentor, in 1918. “[C]an it not also absorb the untouchables?” (quoted in Zelliot 2010: 154). Untouchables are not only not internal to the Hindu community in this articulation of pre–Poona Pact common sense, but the possibility of their becoming so is far from certain—it is quite literally an open question. Gandhi, on the other hand, deployed the logic of the transhistorically Hindu untouchable with an unprecedented clarity, confidence, and uniformity. The Depressed Classes “are part of an indivisible family” with the Hindus, he insisted when explaining the reasons for his fast (Gandhi 1972c: 63). Treating this “indivisible family” as synonymous with “the Hindu fold” and “Hinduism,” Gandhi was able to present any kind of state recognition of the distinctness of the Depressed Classes as a “separation” of two classes of Hindus: “So far as Hinduism is concerned, separate electorates would simply vivisect and disrupt it” (Gandhi 1972b: 191; see also Roberts 2016: 142). While still at variance with the popular social imaginary, this portrayal of things was consistent with the British administrative categorization of untouchables as Hindus; leveraging colonial sociology against itself, Gandhi was able to depict the Communal Award as a cynical application of the policy of divide and rule. If the Depressed Classes were already members of an indivisible house of Hinduism—as the imperial census, taken seriously, would imply—then the special constituencies could be painted as an imperialist stratagem, rather than the political safeguard against the tyranny of the majority that Ambedkar insisted they were. A subset of Gandhi’s deployments of the idea of the untouchable’s timeless belonging to (an equally timeless) Hinduism warrant particular note. In one of his statements to the media on his fast, he wrote, “There is a subtle something— quite indefinable—in Hinduism which keeps them [the Depressed Classes] in it even in spite of themselves” (Gandhi 1972c: 63). A few months later, in February 1933, when discussing with Ambedkar a proposed piece of legislation regarding temple entry, Gandhi used similar language: Gandhi: “But I must say that you ought not to say that you are not a Hindu. In accepting the Poona Pact you accept the position that you are Hindus.”

131 Trustee Majoritarianism

Ambedkar: “I have accepted only the political aspect of it.” Gandhi: “You cannot escape the situation that you are Hindus in spite of your statement to the contrary.” (Ranga 2000: 131)6

These peculiar formulations—that untouchables are Hindus “in spite of themselves” and Ambedkar is Hindu “in spite of [his] statement to the contrary”—may be understood simply as indexing the tenuousness of Gandhi’s position vis-à-vis the reigning common sense. But is there something more here? What does it mean to belong to a community in spite of oneself? Under what conditions, for Gandhi, can people be said to be something in spite of their statements to the contrary? Before we can answer this, we must briefly consider the founding of the Harijan Sevak Sangh and the distinction Gandhi drew between its work and that of the Arya Samaj. According to its own historians, the Harijan Sevak Sangh was founded precisely in response to Gandhi’s fast-unto-death against special constituencies for the Depressed Classes in the Communal Award (Nehru 1940: 6–7; Verma 1971: ii). Originally named the All-India Anti-Untouchability League, its founding charter was a three-point resolution issued by the “public meeting of Hindus” presided over by Madan Mohan Malaviya in Bombay on September 30, 1932, a mere four days after the British accepted the Poona Pact. The resolution felicitates Gandhi on “the happy termination of his fast,” lays out the mission of the organization as “carrying on propaganda against the observation of untouchability,” and appoints the first president of the new League. The choice of president indicates a great deal about the ideological orientation of Gandhi’s inner circle on questions of caste: it was the industrialist Congressman Ghanshyamdas Birla, a “fast devotee” of Gandhi who was also an exponent of Hindu sangaṭhan and the single largest contributor of funds to the Arya Samaj’s shuddhi efforts (Birla 1953: xv; Jordens 1981: 165). Birla was a favorite of Hindu nationalists; Lala Lajpat Rai laid his hopes on him “for the future leadership of the Hindus in politics,” and when Birla contested from Benares against Shraddhanand’s son Indra in the communally polarized elections of 1925–26, Shraddhanand, opposing his own son, joined Lajpat Rai and Madan Mohan Malaviya in campaigning for the more stridently majoritarian Birla 6

It was after this meeting that Gandhi wrote to Rajagopalacharia, “I had, what must be described in one way as a very unsatisfactory interview with Dr. A. He is irreconcilable” (quoted in Birla 1953: 83).

Deceptive Majority 132

(Birla 1953: 23; M. Hasan 1981: 204–05). Birla would remain president of the Harijan Sevak Sangh for its first twenty-seven years, until 1959. Birla’s Arya Samajist leanings would leave an impress on the Harijan Sevak Sangh; at the same time, Gandhi exercised considerable influence over Birla on the administration of the new organization, and Gandhi insisted that it carve a path of untouchable uplift distinct from that hewn by Shraddhanand. Though long an admirer of the swami, Gandhi distanced himself from Shraddhanand in an article in Young India in 1925, precisely over the question of shuddhi, which he repudiated as an imitation of Christian proselytism that is inherently intolerant and “has done more harm than good” (Gandhi 1925: 180). Gandhi’s characterization of Arya Samajists in general and Shraddhanand in particular as “having a narrow outlook and a pugnacious habit” (Gandhi 1925: 178) gave voice to well-grounded concerns that Gandhi shared with Muslims and others about the Samaj inflaming Hindu–Muslim tensions. At the same time, as Adcock (2014) has demonstrated, Gandhi’s figuring of all proselytization as coercive by nature, along with his portrayal of shuddhi as inauthentically Hindu, enabled him to claim for Hinduism a monopoly on the virtue of tolerance even while obscuring or delegitimizing the structural assaults on untouchability that accompanied shuddhi (and conversion to Christianity and Islam) and gave such conversion much of its appeal to the Depressed Classes. Gandhi’s “more subtle method,” Adcock (2014: 163–64) writes, “would encompass the Untouchables within the Hindu political constituency by definition, without the political risk or social provocation of very substantial changes in the lived relations between castes.” The particulars of Gandhi’s “more subtle method”—the components of his strategy for bringing about the Depressed Classes’ “complete merger in the Caste Hindus and the latter’s in the former” (Gandhi 1972c: 348–49)— can profitably be sought in his correspondence with Birla. Both before and after assuming leadership of the Harijan Sevak Sangh, Birla repeatedly raised the question of the legitimacy of shuddhi in his letters to Gandhi. Birla (1953: 5) acknowledged that the shuddhi campaigns that he bankrolled had frankly majoritarian aims and that coercion might be involved, but found this justified: “I do not see any harm if the Arya Samajists do a bit of proselytizing by resorting to force and thereby add to the number of the Hindus.” This position emerged in part from Birla’s admiration of Arya Samaj and Hindu Mahasabha militancy, which he found to have the salutary effect of keeping Muslims on


Trustee Majoritarianism

the defensive.7 But each time Birla reiterated his support for shuddhi, Gandhi cautioned him against it. When Birla asked Gandhi for permission for the Sangh to provide funds to the Arya Samaj for “reconversions” of Doms in Benares (as Birla was already doing in his personal capacity), Gandhi (1973b: 330) replied decisively in the negative: “However desirable therefore the return of the Doms to Hinduism may be, I am quite clear that its encouragement by the Society is outside the scope laid down at its very foundation.” In one of their more illuminating exchanges on the subject, after his public break with Shraddhanand but before the Poona Pact, Gandhi wrote Birla in 1927: I have been doing some serious and deep thinking on the subject of conversion…. Our movement should be directed against conversion to Christianity or Islam. That requires a fundamental change in our outlook. Once we accept that certain methods employed in conversion are blameworthy, we should not follow them ourselves…. By furthering the shuddhi movement we only add to the filth and at the same time prevent the spontaneous urge for reform among the followers of Hinduism. (Birla 1953: 39)

Thus far we find both convergences and divergences between Gandhi’s and Shraddhanand’s approaches. Like the swami, Gandhi feels that conversion to Christianity and Islam must be blocked; the field must be cleared for “the followers of Hinduism” alone to enact their “spontaneous urge for reform” on untouchables. This is not anomalous: in a persuasive analysis of Gandhi’s writings on missionary work, Nathaniel Roberts (2016: 142–47) has shown that the Mahatma portrayed Muslim and Christian efforts among Dalits as necessarily “harmful”—harmful simultaneously for Hinduism and for India. While Gandhi avoided the conflation of religion and territory in other contexts, it emerges quite clearly in his thinking about untouchables, as it did for Shraddhanand. But whereas Shraddhanand counters Muslim and Christian missionizing with the Arya Samaj’s own proselytizing methods, Gandhi in his letter to Birla insists on maintaining a principled opposition to conversion across the board. If Christian and Muslim proselytization is “blameworthy” and productive of 7

“Now that the Hindu Mahasabha and the Aryasamaj have exhorted people to wield the sword, the Muslims are a little bit afraid of attacking them. I know this intensifies the quarrel now, but I am not sure that this would not be the last of the quarrels” (Birla 1953: 4). See also Birla (1953: 24, 270).

Deceptive Majority 134

“filth” (a striking concept to apply to the context of untouchable conversions to any religion), then Hindu proselytization cannot be the answer. Instead, Gandhi proposes an alternative approach: It is unnecessary to convert those who changed their religion through compulsion or out of ignorance. Such people are, in fact, Hindus—the only thing that remains to be done is for Hinduism to be more catholic in its outlook. (Birla 1953: 39, emphasis added)

Conversion, here, is unnecessary. Insofar as the Depressed Classes’ current religious situation can be construed as a product of their ignorance or vulnerability to compulsion, it can be declared invalid. Irrespective of beliefs, practices, or subjective experiences of belonging—whether, that is, they circumcise, attend Catholic mass, give their children Muslim or Christian or Sikh names, proclaim Tauhid or the Trinity or consider themselves Lal Begis or Ad Dharmis or Satnamis—“such people are, in fact, Hindus.” Once this is accepted, the controversial business of securing their assent to a particular religious identification can be dispensed with altogether. In Gandhi’s considered view, all that is needed is for this act of renaming—this assertion of nominative power over a population constitutionally incapable of speaking for itself—to be framed as a moral good, as Hindu catholicity. To trace the moves that Gandhi makes in these reflections—(a) bypassing the problem of untouchable consent by (b) figuring untouchables as incapable of agency (qua victims of ignorance, compulsion), in order (c) to cast the untouchables, in an assertion of nominative power grounded in Gandhi’s moral authority, as willing Hindus in a reforming Hinduism—is to identify what would become the dominant representational strategy of the Harijan Sevak Sangh and the Congress vis-à-vis the Depressed Classes for the next twenty years.

“What Harijans Are”: Gandhian Ontology For the most part, Gandhi did not spell out the approach outlined above so much as live it by example. His mode of engagement with educated members of the Depressed Classes in the pages of his weekly publication Harijan illustrate this. Repeatedly in the 1930s and 1940s, Gandhi was confronted by Depressed Class individuals and organizations that questioned and criticized his and the Harijan Sevak Sangh’s ideology and methods. Consider the Ad Dharm.


Trustee Majoritarianism

This popular Depressed Class movement in Punjab staunchly opposed the Congress, supported Ambedkar against Gandhi at the Round Table Conferences, and maintained a very clear position on its religious autonomy from Hinduism and Sikhism. In the 1931 census, hundreds of thousands of Ad Dharmis, in spite of intimidation, boycott, rape, and other forms of violence from Sikhs and Hindus bent on stopping them, returned themselves as a separate and distinct religion (Juergensmeyer 2009: ch. 7). While Gandhi made no specific mention of the Ad Dharm in this context, he declared only a few months after the 1931 census—in his speech at the Round Table Conference noted earlier: “We do not want on our register and on our census Untouchables classified as a separate class” (Gandhi 1971: 297–98). In 1934 a Harijan Sevak Sangh worker in Punjab wrote Gandhi asking for advice on how to deal with assertive members of the Ad Dharm. He sought counsel particularly on “how to meet or counteract” the propaganda of Ad Dharmis, “who wean [Harijans] away from Hindu society.” In a reply published in Harijan, Gandhi advised him that “Adi-Dharmis [sic] are themselves Hindus…. They will return to the fold when they see that untouchability has been entirely removed” (Gandhi 1934b). Dismissing in a word the self-definition of the Ad Dharm, Gandhi pronounced this organized, mass demonstration of Depressed Class will to be precisely what it said it was not. Again, we find the discursive production of Hindus in spite of their statements to the contrary. On the southern portion of his nine-month “Harijan tour” of 1933–34, Gandhi (1934a: 8) wrote that he was “flooded with communications” from untouchable correspondents who objected to his insistence that eradicating untouchability could and should be done in the name of Hinduism. In a speech addressing Depressed Class critics at Palluruthi in January 1934, printed in Harijan a week later under the heading “To Harijan Sceptics,” Gandhi wrote: You have advised me not to conduct this campaign in the name of Hindu religion. I am very sorry I cannot endorse your advice. It is wholly wrong to say or even think that this movement is conducted in order to consolidate Hindu religion or consolidate anything.

The language of “consolidation” here refers to sangathan, the concept of majoritarian Hindu consolidation popularized by Swami Shraddhanand. Having disavowed the numerical thinking and communalist attitude associated with sangathan, Gandhi proceeds to make a crucial set of claims about knowledge, representation, and being.

Deceptive Majority 136

I have not come to help those who feel their strength…. I know infinitely more than you do what Harijans are, where they live, what their number is and to what condition they have been reduced. You can speak of this place, possibly of the whole of Cochin, possibly of Cochin and Travancore, still more possibly of Malabar. But I claim to be able to speak of Harijans from the North, to the South, the East and the West of India, and I know their abject position. My  only business is, if it is at all possible, to lift those who are in the mire. (Gandhi 1934d)

The first step here is a distinction drawn between “those who feel their strength” and those who do not. Gandhi portrays his interlocutors here as the former: a subset of the Depressed Classes who, relative to most of their brethren, are privileged by literacy and learning to the extent that they can confidently mount a political critique—they “feel their strength” as political subjects. This comparative advantage, though, appears to render them “infinitely” less capable of understanding the condition of their caste fellows—for Gandhi, the educated untouchable is perforce alienated from kin and community. (This anticipates, or perhaps lays the groundwork for, the “creamy layer” argument that has come to be mobilized against the Dalit intelligentsia from the 1990s onward in debates over reservations.) Gandhi impugns the representational capacity of his Depressed Class interlocutors also on the grounds of their being parochial. Even if they could speak for their unlearned cousins, the authority of the organized untouchables of Palluruthi is limited, at most, to the region of Malabar. Gandhi uses this tactic with Ambedkar as well. “I may inform you that Dr. Ambedkar speaks for that particular part of the country where he comes from,” Gandhi said at an interview in London on the sidelines of the Round Table Conference. “He cannot speak for the rest of India” (Gandhi 1971: 161). Against the purblind provincialism of his Dalit critics Gandhi contrasts his own, more global, gaze. The comprehensive knowledge to which Gandhi has access, importantly, combines geography and enumeration: “I know … where they live” and “what their number is” in “the North, to the South, the East and the West of India.” Gandhi’s knowledge, that is, speaks the language of the imperial census—of populations distributed across territory, measured by numbers—and derives a measure of its authority therefrom. If Gandhi’s “I know” thus operates, in part, in a governmental-enumerative register, its other register is ontological. Gandhi knows “what Harijans are.” The particular content of their ontic status is left relatively unelaborated in his

137 Trustee Majoritarianism

Palluruthi speech, but “in the mire” would seem to characterize it, as would the “abject” and “reduced” quality of their “position” or “condition.” Elsewhere the picture is more fully fleshed out. “The poor Harijans have no mind, no intelligence, no sense of difference between God and no-God,” explained Gandhi in a dialogue with Indian Christians over the question of conversion (Gandhi 1976a: 18, quoted in Roberts 2016: 146). “Would you … preach the gospel to a cow?” asked Gandhi on another occasion, adding that “some of the untouchables are worse than cows in understanding…. They can no more distinguish between the relative merits of Islam and Hinduism and Christianity than can a cow” (Gandhi 1976a: 37, quoted in Roberts 2016: 146). The nature of Harijan-ness as Gandhi understood it is perhaps most fully revealed in his engagements with and thinking about the sanitation labor castes. As with the “Harijan,” so too with the “Bhangi,” we find a proximity to cows and other non-human animals in Gandhi’s thought. In an article centrally concerned with cruelty to animals, prompted by a bullock-cart journey in which the driver used a goad with an infixed nail, Gandhi turns abruptly to the subject of the Bhangi: If it is our desire to win swaraj [self-rule] through self-purification, where shall we fix a limit to the process? Will the limit be reached when we have come to treat our Bhangi brethren as we do our own blood brothers? What of our other brothers and sisters—the animals? What is the difference between the soul in them and in us? They eat and sleep, feel happiness or suffer, just as we do. (Gandhi 1966a: 518)

Here the Bhangi is plainly human and not animal; yet reflection on nonhuman animals invites reflection on the former. The Bhangi represents a limit of sorts—a political limit, since self-rule rests on traversing it, but also a conceptual-categorical limit: the far edge of the human community, before it shades into the animal. Elsewhere, referring to a story from the Mahabharata featuring the eldest of the Pandava brothers, Gandhi writes: Securing swaraj is like ascending to heaven. Yudhisthira refused to enter the gate of heaven without his dog. Do we hope to get into the temple of swaraj ourselves, leaving our Bhangi brethren behind and running at top speed towards it? (Gandhi 1966b: 143–44)

Here again the animal and the Bhangi are presented as homologous, juxtaposed  on account of their structurally parallel alterity from and

Deceptive Majority 138

dependence on the agentive “we.” It must be acknowledged that the otherness of the Bhangi from the presumed subject of Gandhi’s discourse is mitigated somewhat by the assertion that they are “brethren.” Yet for Gandhi, this is a condition they share with non-human animals, who are also—indeed in the same breath—“brothers and sisters.” What we have then is a cosmology in which all sentient beings share a kind of kinship—a view of things thoroughly grounded in Jain, Buddhist, and Upanishadic thought—in tension with an insistence on categorical difference: “we” exist, as do Bhangis over there, and animals still further in that direction. It is well known that Gandhi claimed to be a Harijan by choice. What has been less noted is that these claims were often more specific: frequently, what Gandhi claimed to be was not simply untouchable but more precisely Bhangi. In an argument with a Hindu of unspecified caste at a temple in Delhi, Gandhi (1983: 189, emphasis added) said, “Let me tell you with all due respect that you can’t represent the Bhangis. I am a Bhangi by choice. I have removed night-soil.” “I myself did a Bhangi’s duty continually,” Gandhi (1972d: 47) wrote a friend in 1932. “I used to do it dressed like an ordinary labourer. In the Ashram it is done with the dhoti tucked in.” In a speech to sanitary volunteers in 1938, Gandhi (1976b: 363) said, “I am an expert Bhangi. I have been doing this work for over thirty-five years and I have done it in the proper spirit.” In these and many other instances, Gandhi anchors his claim to speak for the sanitation labor castes in his having performed the labor of disposing of human excrement. Admirers take Gandhi’s scavenging as evidence of his commitment to untouchables or of his determination to deconstruct the castebased regime of labor by example (for example, Kolge 2017). Critics note that his much publicized scavenging valorizes a labor practice widely experienced as dehumanizing (for example, Ramaswamy 2005). Both of these positions miss a vital point. Gandhi’s embrace of night-soil removal as the grounds of his own legitimacy in representing the sanitation labor castes binds the Bhangi in a brahminical universe, one in which castes are constituted by their labor, in which night-soil is a matter of Bhangi being. To propose, as Gandhi does, that his Bhangi-hood rests on his manual scavenging—and that others’ incapacity to represent the sanitation labor castes rests on their not having scavenged—is to advance an ontology in which occupation and caste status are fused. Gandhi (1936) explains, My ideal Bhangi would know the quality of night-soil and urine. He would keep a close watch on these and give a timely warning to the individual

139 Trustee Majoritarianism

concerned. Thus he will give a timely notice of the results of his examination of the excreta…. Such an ideal Bhangi, while deriving his livelihood from his occupation, would approach it only as a sacred duty. In other words he would not dream of amassing wealth out of it. He would consider himself responsible for the proper removal and disposal of all the dirt and night-soil within the area which he serves and regard the maintenance of healthy and sanitary condition within the same as the summum bonum of his existence.

In the manner of the Chandal in the dharmaśāstra tradition, Gandhi’s Bhangi  fulfills his prescribed function of removing night-soil in a sacralized division of labor, and does not amass wealth in the process. Scavenging being constitutive of Bhangi being, those who do it—and more narrowly, those who do it “in the proper spirit”—have the capacity to speak as and for the sanitation labor castes, and more broadly untouchables, whereas those who do not engage in scavenging lack that representative capacity. It bears pointing out that this was by no means the only way in which the sanitation labor castes, their labor, and their relation to the rest of society were imagined in the 1930s and 1940s. Gandhi’s vivid projection of the brahminical imagination onto this category of persons in the mid-twentieth century was not a foregone conclusion. Unions and the Left hailed these same persons as “sweepers” and “sanitation workers,” interpellating them in a universalizing narrative of global organized labor. When confronted with such a narrative, as he was by the Bombay sanitation workers’ strike of 1946, Gandhi took pains to shift the discourse from one of “sweeper’s grievances” to one of “the duties of Bhangis”—to ease the worker back into a brahminical frame (Gandhi 1946c, 1946d). Having denounced the strike as “coercive” and reiterating his view that “A bhangi may not give up his work even for a day,” Gandhi (1946d) writes: I tried to show … the duties of Bhangis as well as of citizens. I have often said that every kind of injustice is meted out to Bhangis. I have no doubt that citizens do not fulfil their obligations to them. Thus it is their duty to see that Harijan dwellings are built properly, the means employed for cleaning are decent, that they have a special working uniform given to them, that they and their children have facilities for education etc.… The Bhangis may not go on strike for lack of these amenities but it is up to all citizens to raise their voice on behalf of them.

In upbraiding “citizens” for their treatment of “Bhangis” and asking them to do better, Gandhi unselfconsciously presents the latter as a category separate from and contrastive with the former. He also insists, in yet another

Deceptive Majority 140

instantiation of the structure of agency we first saw in the Uka story, on the necessity of “uplift” originating from above: it is for citizens to ensure housing and working conditions for Harijans, and not for the latter to demand these things themselves. Another way in which the sanitation labor castes imagined and publicly represented themselves in this period, overlapping with yet distinct from both brahminical and Leftist narratives, is suggested by the enduring popularity of the titles Mehtar and Halalkhor as terms of self-designation in the 1930s and 1940s. These Arabic- and Persian-derived titles suggested occupation first and caste somewhat more loosely—unlike Bhangi, which emphatically denoted both—and conjured an urbane Mughal mode of sociality in which conventions of politesse countervailed, however slightly, the general contempt. The sanitation labor castes in this period appear to have much preferred Mehtar and Halalkhor over Bhangi, perhaps on account of the affirmation of the dignity of labor embedded, albeit faintly, in their Islamicate etymology. On numerous occasions, in person and in correspondence, members of the sanitation labor castes objected to Gandhi’s use of “Bhangi.” In place of the offensive term, several correspondents and the residents of the sweeper colony in Delhi with whom Gandhi briefly resided, asked him to use “Mehtar.” “I tried to make them understand,” he replied in Harijan, “that it mattered little as to which of the current words was used for the same occupation” (Gandhi 1946b: 1). Yet the retention of “Bhangi” seemed to matter a good deal to Gandhi; he continued to use it consistently, going so far as to propose changing the name of his journal from Harijan to Bhangi (Gandhi 1981: 347). The only alternative name for the sanitation labor castes that Gandhi did countenance was, unsurprisingly, Valmiki (Gandhi 1981: 148).8 Gandhi’s exertion of nominative power over the sanitation labor castes thus reinforced his discursive confinement of the Depressed Classes within Hinduism. If Mehtar and Halalkhor (and for that matter Lal Begi) indexed an Islamicate imagination and a potentially Muslim affiliation, and sweeper or sanitation worker signaled a secular-Left orientation, the Sanskrit-derived Bhangi conjured a brahminical and thus Hindu web of meaning. Dismissing the 8

The few references to Valmiki (qua caste title) in Gandhi’s CWMG come, as we might expect, only after the mid-1930s, by which time the Valmiki movement had begun to gain traction. In 1946 Gandhi (1981: 148) informs the audience of Harijan: “Readers must know that Valmiki is another word for bhangi.”

141 Trustee Majoritarianism

sanitation labor castes’ designations for themselves, Gandhi’s representational practices wove them into that web—making them, again, Hindus “in spite of themselves.” The argument thus far may be illustrated in Gandhi’s correspondence with a self-described Mehtar in the pages of Harijan in 1946. Political independence was imminent and the correspondent wrote with concerns about the representation of untouchables, and particularly the sanitation labor castes, in the Constituent Assembly soon to be formed. He wrote: It is believed that Scheduled Castes are also to be represented (adequately?). But is there any proposal from you or from Congress to elect adequate or at least some members from the Mehetar [that is, Mehtar] Community? Who, I am sure will discharge their duty of citizenship and pick up their legitimate share in the future constitution of Free India. You might say you have been and will do everything for us, but I wish to say “let us be with you when everything for us is to be done. Let us be represented democratically.”

Putting his finger on the structure of agency we have been highlighting in Gandhi’s thought—wherein the capacity to act as a political subject resides exclusively with the citizen/Hindu, whose acts of benevolence are to be mutely accepted by the Harijan/Bhangi—the correspondent proposes, instead, that untouchables inhabit the role of citizen and participate democratically in the solving of the caste problem. After reproducing the full text of the letter, Gandhi replied: I have reproduced the foregoing in order to show what havoc dangerous knowledge of English has produced in our society. This is a specimen not of English English nor yet of Indian English. It is bookish English which the writer probably half understands. I suggest to him that if he had written to me in the national language Hindustani or in his provincial language, it would not have evoked an unfavourable response from me. The writer has paid me a left-handed compliment and that perhaps in order to teach me how to express my love for the bhangi, otherwise known as mehetar. The writer is a discontented graduate, setting no example or a bad example to bhangis. He has isolated himself from them, though he professes to represent them.

Deceptive Majority 142

He will certainly become my teacher if he will be a graduate in the art of being a good bhangi. I very much fear that he does no scavenging himself, he does not know what scientific scavenging is. If he became an expert in the art, his services would be wanted by all the cities of India. When bhangis really rise from the slumber of ages, they will successfully sweep the Augean stables everywhere and India will be a pattern of cleanliness and there will be in India no plague and other diseases which are the descendants of filth and dirt. In the place where I am living in Bombay,9 my room and the adjoining lavatory are fairly clean, but I am in the midst of suffocating dirt. I have had no time to examine the tenements in front of me. They are as crowded and as dirty as the ones in the quarters where I was living in New Delhi. Had my graduate fellow bhangi been an expert in the art, I would, without a doubt, have requisitioned his services as my guide and helper. As it is, not only have I no use for him, I have to risk his displeasure by telling him that he should not think of the Constituent Assembly or other assemblies. Let those go to them who are wanted there. Instead of getting rid of the wretched caste mentality, he argues that any Harijan is not good enough for the purpose but preference should be given to the mehetar caste. I suggest to him that it is a harmful method, doing no good to anybody. (Gandhi 1946b)

The themes that we have been tracing in Gandhi’s mode of engagement with Depressed Class interlocutors here are drawn together: A member of the Depressed Classes is told that he cannot represent his community on account of being educated. A man who identifies himself as Mehtar, an occupational category redolent of an Islamicate historical context, is renamed Bhangi, a caste name that drapes him in the garb of Hindu culture (see Figure 4.1). A question about the representational rights of disadvantaged citizens is met by an invitation to sweep the tenements of Bombay. And a demonstration of political knowledge elicits not only a counter-assertion of superior knowledge, as we by now expect, but also in this case the contention that, qua “Bhangis,” the sanitation labor castes have no business writing in English, taking interest in democratic politics, or involving themselves in pursuits other than the 9

For the days of this Bombay visit, Gandhi was residing in a house in ‘the Bhangi quarter.’ He spent part of the previous month (June 1946) residing at ‘Bhangi Niwas,’ his name for a sanitation labor caste colony in New Delhi (Gandhi 2000: vols. 91, pp. 207, 222, 255–56).

143 Trustee Majoritarianism

Figure 4.1 “Left-Handed Compliment” Source: Harijan, July 14, 1946. Image courtesy of the Burke Library at the Union Theological Seminary, Columbia University, New York.

fulfilment of their (Hindu) caste dharma. While Gandhi’s reply to the Mehtar correspondent is somewhat uncharacteristic of his engagements with Depressed Class critics in the biliousness of its tone, it does illustrate patterns in the substance of these engagements. The pages of Harijan thus illustrate clear patterns in Gandhi’s mode of engagement with Dalit critics, patterns that, read alongside his narrativization of the problem of untouchability, amount to a representational strategy. It is difficult to overstate its influence. The strategy was replicated in multiple spheres of activity; its effects continue to ramify today.

Deceptive Majority 144

“Raidas, the Sweeper” Among the domains in which Gandhi’s representational practices exerted extraordinary influence is literature. Rabindranath Tagore—the Bengali novelist, playwright, mystic and Nobel Laureate whose stature was such that his contemporaries often referred to him simply as “The Poet”—was sufficiently moved by Gandhi’s fast-unto-death that, though frail at seventy, he journeyed from Calcutta to Poona by train to be at Gandhi’s side (Lelyveld 2011: 232–33; Pyarelal 1932: 77). In the following months he lent his talents and reputation to the cause of Gandhi’s new journal, composing or translating poems that would be featured on the cover of Harijan, including its inaugural issue. While this is not the place for an extensive analysis of this series of compositions, a few of their features must be mentioned for the work they did in amplifying and giving literary heft to Gandhi’s framing of untouchability. “The Cleanser” was the title of the poem that graced the cover of the Harijan’s inaugural issue of February 11, 1933. This was Tagore’s “free rendering from the Bengali” of Satyendranath Datta’s “The Scavenger,” and as both titles suggest, the poem’s protagonist is a manual scavenger. The labor of this figure is praised as “making the earth and air sweet for our dwelling” and likened to that of a mother cleaning her infant as well as to the god Shiva who “saved the world from a deluge of poison by taking it himself” (Tagore 1933c). In virtually every respect—the valorization of the labor of excrement removal, the foregrounding of the scavenger as a synecdochic sign of all that belongs under the heading “Harijan,” and the comparisons to both Shiva and mothers (comparisons Gandhi himself offered on multiple occasions)—“The Cleanser” reproduced Gandhi’s representational practices. In a cycle of poems centered on figures from the bhakti tradition of Hindu devotionalism, Tagore made the extraordinary choice to graft onto the poetsaint Raidas (also known as Ravidas)—famously a leatherworking Chamar and thus not of the sanitation labor castes—the figure of the Bhangi. Consider “Sweet Mercy,” Tagore’s composition on the cover of issue fifteen of Harijan: Raidas, the sweeper, was tanner by caste whose touch was shunned by the wayfarers and the crowded streets were lonely for him. Master Ramananda was walking to the temple after his morning bath, when Raidas bowed himself down before him from a distance.

145 Trustee Majoritarianism

“Who are you, my friend” asked the great Brahmin and the answer came, “I am mere dust dry and barren, trodden down by the despising days and nights. Thou, my Master, art a cloud on the far away sky. If sweet mercy be showered from thee upon the lowly earth, the dumb dust will cry out in ecstasy of flowers.” Master took him to his breast pouring on him his lavish love which made a storm of songs to burst across the heart of Raidas, the sweeper.

(Tagore 1933b)

The Raidas of bhakti tradition is a complex figure whose poetic oeuvre includes critical deconstructions of caste ideology and visions of urban utopia as well as the kind of self-effacing love songs with which bhakti is often associated; his hagiography paints an equally mixed picture of a character inclined sometimes to meekness and other times to confrontation in his encounters with brahmins. The bhakti tradition thus offered Tagore a range of interpretive possibilities. The Raidas that Tagore ultimately chooses to portray—scrupulous in his deference to the brahmin, observant of the law of keeping distance, solicitous of voluntary acts of mercy, agreeable to Ramananda’s embrace when it comes, and a sweeper (even while still “tanner by caste”)—needs to be understood in light of the context which provoked its composition in early 1933. In a historical moment defined by the clash between Ambedkar and Gandhi over Dalit representation, Tagore supplies a scene of tearful untouchable acquiescence to reformist Hindu leadership in full conformity with Gandhi’s insistence that change be driven by Hindu benevolence and not Depressed Class assertion. This parable of reconciliation is then set in the timeless Hindu milieu of bhakti lore, implying the Hindu identity of the characters involved and further circumscribing the Depressed Classes within narratives of Hinduism (see Figure 4.2). The other poems in Tagore’s bhakti cycle are markedly similar in their adherence to the Gandhian script. In “The Sacred Touch,” Tagore (1933d) again portrays an encounter in which a “grizzly old tanner bow[s] himself down to the dust from a distance” before “Ramananda, the great Brahmin Teacher,” who then unilaterally embraces the submissive untouchable,

Deceptive Majority 146

Figure 4.2 “Sweet Mercy” Source: Harijan, May 20, 1933. Image courtesy of the Burke Library at the Union Theological Seminary, Columbia University, New York.

explaining that the sun’s “light descends on your forehead as well as on mine.” The third in the cycle, “Love’s Gold,” returns to Raidas, and unexpectedly takes up the hagiographic theme of confrontation with orthodox brahmins. But even here Tagore manages to maintain the Gandhian structure of agency: the criticism of the brahmin for his caste pride is put in the mouth of the Rani of Chittor, Raidas’s royal disciple. For his part, “Raidas, the sweeper, sat still, lost in the solitude of his soul,” his passivity in the poem mirroring the political quiescence of the sanitation labor castes to which Tagore had reassigned him


Trustee Majoritarianism

(Tagore 1933a). Relative to Raidas’s actual caste of Chamars, or indeed to Ambedkar’s Mahars, or Namasudras, Paraiyars, and others, the “sweepers” that Gandhi and now Tagore projected as the exemplary untouchables posed no organized threat to the Congress or to Gandhi’s narrative framing of the caste question. In Tagore’s hands, Raidas takes on the contours of Uka, the apolitical sweeper waiting, deferentially, for his Hindu employers to decide whether he is to be touched. Tagore was not alone in giving literary wings to Gandhi’s vision of untouchability. What the august poet did in the world of verse, the young Mulk Raj Anand did in the novel. Untouchable, Anand’s widely acclaimed 1935 novel, was directly inspired by Gandhi’s story of Uka, which Anand read in Young India in 1921 (Christopher 2015: 67). Set in Punjab, Untouchable is the story of a day in the life of Bakha, an uneducated sweeper and scavenger. Gandhi himself appears in the novel; his post–Poona Pact “Harijan Tour” brings him to Bakha’s town where he delivers a speech that Bakha observes from a tree with mixed but largely positive feelings. The speech, importantly, is a suturing together of Gandhi’s actual writings and speeches on untouchability—it includes, for example, the Uka story, faithfully reproduced from the pages of Young India. Anand’s commitment to realism, however, also leads him to present a social world in some ways at odds from that which Gandhi sought to portray. For instance, the pre–Poona Pact sociological common sense makes more than a few appearances in the novel, as the novelist and his characters unselfconsciously speak of untouchables and Hindus as contrastive categories, for instance when Bakha’s father reminds him that “[Ram Charan] was a Hindu, while Bakha was a mere sweeper” (M. R. Anand 1935: 71, see also 28, 38, 39, 68). In crucial ways, though, Anand’s novel does precisely the work of popularizing Gandhi’s framing of the problem of untouchability, and elevating it to the truth of literature. It does this centrally by making its eponymous hero a scavenger—or, to put it differently, by titling his novel about a scavenger Untouchable. Equally important, this boldly synecdochic figure is presented as entirely innocent of politics. Though the Punjab in which Untouchable is set had been roiled by anti-Dalit violence in reaction to the Ad Dharm’s mass mobilization and assertion of autonomy in the most recent census, Bakha knows nothing of it. Indeed, in Anand’s portrayal of early 1930s Punjab, there is neither Ad Dharm, nor Ambedkar, nor even Arya Samaj—for Bakha, as for the reader of Untouchable, there is not a whisper of autonomous

Deceptive Majority 148

Dalit mobilization, and liberatory politics begins with Gandhi. As Gauri Viswanathan (1998: 223) observes, “[t]he novel’s occlusion of Ambedkar is consistent with a certain tradition of writing about untouchability that has roots in the antagonistic rhetoric of the Indian National Congress” in the wake of the Gandhi–Ambedkar conflict. Anand’s framing of things went on to enjoy immense popularity—E. M. Forster introduced the novel and it became probably the single most widely read novelistic account of Dalit life. K. W. Christopher (2015: 74–75) notes that Untouchable set a precedent for future representations of the struggle over untouchability, notably Richard Attenborough’s influential 1982 film Gandhi, which also removes Ambedkar from the narrative. Thus it was partly through literature that the Gandhian narrative acquired color, depth, and poignancy—that it began to seem real.

“Self-Appointed Trustees” If Tagore and Anand offered a kind of literary solution to Depressed Class criticism of Gandhi’s approach to untouchable uplift, the Harijan Sevak Sangh offered an institutional solution. At the first, ad-hoc meeting of the All-India Anti-Untouchability League in October 1932, it was decided that Ambedkar and two other Depressed Class leaders (M. C. Rajah and R. Srinivasan) should be included on the central board. All three resigned from the body within a few months, Ambedkar on account of his proposals to the board being not even acknowledged (Ambedkar 1991: 133–41). The lack of Depressed Class representation in an organization dedicated to Depressed Class uplift was discussed in the annual meetings in 1933 and 1934. Gandhi, in response to concern expressed by some members as to the non-democratic nature of the constitution he proposed, replied: I have come to the conclusion that in an organisation like ours, there is no room for election, democracy or anything of that sort. Ours is a different kind of institution. It is not a people’s organisation in the ordinary sense. We handle money as self-appointed trustees, using it solely for the benefit of the Harijans…. Ours is an organisation formed with a view of doing our duty by those whom we have despised. (Verma 1971: 130)

Constructing the Harijan Sevak Sangh as an organization of savarn (high status) Hindus discharging a moral debt obviated the question of untouchable representation; the burden of leadership would necessarily fall on penitent Hindus. This freed the organizers from having to countenance Depressed


Trustee Majoritarianism

Class criticism regarding the prioritization of temple entry and the framing of untouchable uplift as a religious and particularly Hindu enterprise. In the same stroke it bolstered that very framing: how could untouchability be anything but a Hindu problem when the All-India organization designed to solve it was precisely an organization of caste Hindus? Accordingly, the Harijan Sevak Sangh, constituted by a “public meeting of the Hindus,” adopted an organizational structure in which all positions were appointed—not elected, and not subject to review by the organization’s intended beneficiaries—from the president downward. In his capacity as president of the central board, Birla solicited Gandhi’s approval for each appointment that he made (Birla 1953: 96, 103). In terms of organizational structure, the Harijan Sevak Sangh was, as Birla (1953: 104) put it, a “restrained autocracy.” After the resignation of Ambedkar and his Depressed Class colleagues shortly after the first formal meeting of the central board, there were no Depressed Class members; they were simply not appointed. Even at the regional and local levels, the Harijan Sevak Sangh rarely included Depressed Class individuals in decision-making positions, a policy Gandhi explicitly defended. “The reader is aware that I have dissuaded Harijans from pressing for representation on Harijan Boards [that is, local branches of the Harijan Sevak Sangh], for the very simple and complete reason that these boards are meant to be composed of savarna Hindus who regard untouchability as a sin and who would do reparation to Harijans for past wrongs done to them” (Gandhi 1934c). The Harijan Sevak Sangh thus institutionally embodied Gandhi’s mode of rendering untouchables Hindu: bypassing the difficulties of securing Depressed Class consent by figuring them as objects of, rather than subjects of or partners in, reform, the field was cleared for Gandhi and his league of penitent Hindus to frame what had been a question of rights and representation in the emergent nation as a problem of attitudinal reform within Hinduism. On this problem only Hindus—as self-appointed trustees, neither chosen by nor institutionally accountable to their ostensible beneficiaries—could speak and act. What I have sought to elucidate thus far is the difference between the strategy of majoritarian inclusion pursued by Swami Shraddhanand and the “more subtle method” modeled by Gandhi. Both men described Hinduism as in mortal peril, a peril from which it could be rescued only by the removal of untouchability and the inclusion of the Depressed Classes in the Hindu

Deceptive Majority 150

fold. Both feared the prospect of alliances between Muslims and the Depressed Classes and saw conversion to Islam and Christianity by Dalits as a problem that needed to be fought against. Both, moreover, directed a great deal of their reformist rhetoric toward fellow Hindus—that is, toward non-Dalits. Their modes of engagement with the Depressed Classes, however, differed rather profoundly. Shraddhanand and his Arya Samaj colleagues were engaged, ultimately, in an effort of persuasion—that is, an interlocution that presumes the moral agency of those spoken to. Like Shraddhanand in the Chamar bastīs of Delhi, Amichand Sharma in the Lal Begi bastīs of Lahore attempted to convince his audience that they were Hindu and should identify themselves as such on the census. This required listening to the Depressed Classes, learning how they perceived themselves vis-à-vis Hindus and others, and presenting arguments in favor of a new conception of community—arguments their interlocutors were understood to be able to comprehend and assess. Sharma was compelled to take seriously Dalit skepticism—recall that the untouchable protagonist of his tract asks, “If we are Hindu then why do Hindus not touch us?”—and to offer reasoning, however controvertible, in support of his case. While relations of power between the Arya Samaj prachārak and his Depressed Class interlocutors were unquestionably asymmetric, theirs was nonetheless a dialogue, a dialogue in which the concerns and objections of the latter shaped the ongoing efforts of the former. Gandhi, by contrast, represented as misguided the effort to persuade the Depressed Classes of their belonging to Hinduism. Their incapacity for discernment and moral agency, akin to the cow, was essential to “what Harijans are,” and made them appropriate objects not of dialogue and persuasion, but rather of guidance and reform. Gandhi’s ontology of the Harijan and the Bhangi rendered it unnecessary to countenance Dalit self-designations (Ad Dharmi, Lal Begi, and so on) or the claims to representativeness of individuals like Ambedkar, the Mehtar graduate, or other educated members of the Depressed Classes, whose very education had deprived them of their Harijan-ness. In place of dialogical persuasion, Gandhi offered monological nomination. The speaker in this act of naming needed to be morally and politically agentive, and thus not Harijan in the strict sense of Gandhi’s term. But he needed to have authority, and not merely the coercive power of the colonial state, of which Gandhi was famously critical. This difficulty was resolved in the figure of the Harijan-by-choice, the penitent Hindu who, by voluntarily taking up traditionally untouchable forms of labor, partook in Harijan-ness and thus


Trustee Majoritarianism

acquired the capacity to speak as a Harijan. Of the small band of Harijans-bychoice inspired by Gandhi to take up such labor, it was the Mahatma himself, on the strength of his scavenging—“I am an expert Bhangi”—who would assume the leading role in the act of nomination. In an important sense, then, Gandhi’s remark to Ambedkar in 1933—“You cannot escape the situation that you are Hindus in spite of your statement to the contrary”—in fact addressed not only Ambedkar but all of the Depressed Classes. The contrast between the two approaches, however, should not be overdrawn. In practice, neither was the Arya Samaj averse to sometimes calling the Depressed Classes Hindu without their consent, nor did Gandhians always abstain from efforts to persuade untouchables to identify as Hindu. We have seen that Amichand Sharma resorted, at times, to monological declarations that the sanitation labor castes are Hindu irrespective of their own thoughts on the matter, citing shastras, the census, and a pandit of Kashi as authorities. And as we will see in the next, penultimate section of this chapter, the Harijan Sevak Sangh adopted some practices that were difficult to distinguish from those of the Arya Samaj.

Hinduization as “Uplift” The Harijan Sevak Sangh has received scant treatment in the historiography of the 1930s and 1940s. This lacuna cannot be attributed to the obscurity of the organization’s leadership, which included not only Gandhi and Birla but Rameshwari Nehru, Dr. Bidhan Chandra Roy, Amritlal Thakkar (Thakkar Baba), N. R. Malkani, and other prominent members of the Congress.10 Nor was the organization insignificant in the funds it channeled, the property it owned, the network of schools and hostels that it operated, or in its intimate interrelation with the Congress ministries of 1937–39 and the Social Welfare Ministry of independent India. Most importantly from the point of view of this study, the Harijan Sevak Sangh had no small impact as a model for and agent of the integration of the Depressed Classes into the political and educational structures of the emergent nation. 10

Nehru was editor of Strī Darpan, founder and president of the All India Women’s Conference, and marital cousin of Jawaharlal Nehru; Roy was the second chief minister and “architect” of West Bengal; Thakkar was a renowned social worker and member of the Constituent Assembly until his death in 1950; Malkani was member of Parliament (MP) from 1952 to 1962, chairman of the Government of India’s committee on manual scavenging (1960), and a prolific writer in Sindhi and English.

Deceptive Majority 152

The few descriptions of the organization that do appear in studies of the period take note of two of its major endeavors: providing educational opportunities for the Depressed Classes and spreading propaganda against untouchability among Hindus. Christophe Jaffrelot’s (2005: 70) summary is representative: The League, which was renamed the Harijan Sevak Sangh (Association for Serving Harijans), focused its activities—thanks to Birla’s funds—on helping Untouchables in a paternalistic manner: it aimed to help their social advancement, notably in the field of education, while promoting a change of heart among the upper castes.

Characterizations such as these (see also Bayly 1999: 250; Pandey 2002: 57; Zelliot 2010: 169) are largely correct: the organization did devote much of its energy and resources to education (acquiring, building, and managing schools and residential hostels for Depressed Class students, as well as establishing and distributing scholarships for them), and to anti-untouchability propaganda targeting Hindu audiences (organizing and funding speeches, conferences, and publications like Harijan). These are also the priorities of the Sangh as laid out in its constitution (Verma 1971: 232). One very significant facet of the organization’s operations, however, is missing from these accounts. This is the Sangh’s inculcation of Hindu religious teachings and ritual practices, and its efforts to eradicate “non-Hindu” practices, among the Depressed Classes. The oversight may be attributed partly to sources: the well-known resolution in which the Sangh was chartered, as well as its later Constitution, are silent on the promotion of Hinduism, whereas it appears abundantly in the Sangh’s less well-known annual reports and the dispatches of regional Sangh branches in the pages of Harijan to which we will turn shortly. More deeply, the oversight indexes the stunning success of the very project in which the Sangh was engaged. With the congealing of the post–Poona Pact common sense and the normalization of the idea of the transhistorically Hindu untouchable, the entire religious dimension of the Sangh’s labors conceals itself, as it were, in plain sight—if the beneficiaries of the Sangh’s efforts are assumed to be already Hindu, then the organization appears to be engaged in a straightforward project of internal Hindu reform, exactly as Gandhi characterized it. The matter takes on a different light when we recall that the targets of this Hindu reform often had names like Hussein, Anwar, Mariam, and Nazira, and sang hymns that foretold paradise for the followers of Lal Beg but hell for the followers of Ram. When we take into account that the


Trustee Majoritarianism

sanitation labor castes, among others (for example, Ad Dharmis, Satnamis), understood themselves to constitute religious communities separate from and critical of the Hindus, we are in a position to reevaluate a number of aspects of the received historiography, including the seemingly transparent claims of “untouchable uplift.” Consider this report from a regional Sangh branch in central India in 1934, published in Harijan. MAHAKOSHAL (C.P. HINDI) REPORT FOR JUNE AND JULY 34. Education: The Harijan Ashram at Burhagar (Jubbulpore), where five Harijan boys are being trained in handicrafts as well as in the 3 R’s, is working satisfactorily…. Religious: Ten kathas [religious dramas] were held and two havans [Vedic fire ceremonies] performed in Harijan quarters. On rathyatra day [a Hindu festival], Harijan bhajan [devotional song] parties were also invited and their procession followed the rath [the god’s chariot]. Mahaprasad [food blessed by the god] was also distributed, which both Harijans and caste-Hindus took without distinction…. The Sangh loses no opportunity of taking Pandits and Mahants to Harijan quarters and giving discourses on religious subjects. Mahant Laxmi Narayan Das—a well-known orthodox Mahant of Raipore— paid a visit to the Harijan quarters in Jubbulpore and distributed prizes to them. (B. R. Singh 1934)

Such an account of activities is representative, and illustrates several of the methods by which the Harijan Sevak Sangh disseminated Hindu religious instruction and encouraged Hindu religious practices among the Depressed Classes. These included sponsoring religious discourses by orthodox Hindu leaders, encouraging Harijan observance of Hindu festivals alongside Hindus, and, in what appears to be a sign of Arya Samaj influence, the introduction of the Vedic rite of the havan. The promotion of temple-based Hindu ritual practice is another prominent feature of Sangh reports. This involved both the opening of temples previously closed to the Depressed Classes as well as the construction of new temples exclusively for Harijan use. “One New Ram temple has been built for Harijans in Rajamundry,” announced the editor of Harijan in April 1933, for example (Sastri 1933b). Another report declared, “One Mariamman temple was opened at Kallikudi, in Madura district, for the exclusive use of Harijans, and three temples are being built at Vellore, Kavanur and Arcot, in the North Arcot district” (Gopalaswami 1934). Temples built for “the exclusive use of Harijans,” of course, appeared to encourage

Deceptive Majority 154

temple-based practices even while perpetuating untouchable segregation. But all of the Sangh activity oriented toward temple-entry attracted controversy; a number of Depressed Class observers criticized it as a misplaced priority with an unwarranted religious thrust. In reply Gandhi repeatedly and categorically denied that Sangh funds were used for temple construction, even as the reports of regional Sangh boards at times appeared to contradict this, and the Sangh’s general secretary Amritlal Thakkar, famously scrupulous in his accounting, at times mentioned temple construction as among the planned and budgeted activities of the Sangh. The method of Hindu pedagogy most ubiquitously attested in Harijan Sevak Sangh sources was the teaching and circulation of Hindu devotional songs—bhajans and kirtans. The Sangh in Mysore, notes a typical report, “published 5,000 copies of Harijana Bhajanavali in Kanarese, being selections from Saint Purandaradas. The Harijan school masters were particularly taught these songs” (Ramchandra 1933). Like other aspects of “untouchable uplift,” bhajan sessions were painstakingly quantified, as in this monthly report from Rajputana: “7 bhajan kirtans were held in Harijan mohallas. 8 kathas from religious scriptures were recited to Harijan gatherings” (C. B. Sharma 1934). Teaching and circulating bhajans as a means of imparting religious knowledge was not the initiative of only local Sangh activists; it was promoted at every level of the organization. In her account of The Harijan Movement, Rameshwari Nehru—a member of the Sangh’s Central Board from the beginning and its president (after Birla) from 1959 to 1965—elaborated that “Harikirtans and Katha recitals (the reading of the sacred books)” were key to the “cultural side” of the Sangh’s activity (Nehru 1940: 9, parentheses in original). In a conversation with M. C. Rajah on bhajans as a vehicle for “intensive religious propaganda among Harijans,” Gandhi praised both southern and northern provincial Sangh leaders for their promotion of bhajans (Desai 1936b: 50). Moreover, in the pages of Harijan Gandhi (1933a) exhorted Hindu students to spend their school vacations among Harijans, “reading to them simple stories from the Ramayan and the Mahabharat” and “teaching them simple bhajans,” in addition to cleaning the Harijans’ quarters, bathing their children, giving them lessons in hygiene, and providing them medical aid. Alongside teaching Hinduism, the Sangh also devoted resources to convincing the Depressed Classes to give up “un-Hindu practices.” Propaganda, Gandhi explained in an article directed at Depressed Class critics, meant “holding Harijan conferences and the like for the purpose of telling

155 Trustee Majoritarianism

the Harijans what is being done by the caste-Hindus and what is expected of Harijans in the way of internal reform, such as observance of the laws of sanitation and giving up carrion-eating and other un-Hindu practices” (Gandhi 1973b: 262, emphasis added). Gandhi did not go on to enumerate what other un-Hindu practices he had in mind. This was in contrast with Arya Samajists, who, as we have seen, described and deplored among the sanitation labor castes circumcision, Islamic marriage and death rituals, giving children Muslim and Christian names, the rites of Lal Beg, and so on. To discuss these in the pages of Harijan would be to invite Muslim criticism of the Sangh’s efforts as communally motivated. It would also be to draw attention to the de facto distance of Depressed Classes from Hindu sociality, exposing a fundamental flaw in the premise on which the Sangh’s efforts were hung. Instead, Gandhi and his colleagues restricted their discussion to practices unlikely to attract attention on communal grounds: the consumption of liquor and carrion. In much Sangh literature, abandoning liquor and carrion appears to be a selfevident good, a matter of hygiene and health, a desideratum for the Depressed Classes not contingent on a particular religious or communal position. Depressed Class organizations largely held this view as well; Ambedkar, for example, exhorted his caste fellows to give up drink and carrion not on the grounds that these were un-Hindu practices, but because their effects were deleterious. The efforts of the Harijan Sevak Sangh on this front featured a slippage between this apparently religiously neutral reason and specifically Hindu reasons. Neither Depressed Class nor Hindu reformists of this time disputed the damage caused by heavy alcohol consumption or the desirability of curbing it. But Gandhi (1973a: 245) also recognized the significance of liquor in the religious life of the Depressed Classes: “I know that alcohol is used even in their rituals.” He knew, for example, that among “Harijans of Mysore,” liquor routinely played a role in ritual animal sacrifice before deities (Gandhi 1946a). This tradition of subaltern religion constituted, for Gandhi, an “evil that has been handed down through the ages” that must “be wiped out of existence” (Gandhi 1946a). Similarly, Gandhi (1933b) understood carrion-eating in religious terms: “[T]he eating of carrion is a most filthy habit, regarded as one of the heinous sins in Hindu Scriptures, and it is essential that at this hour of self-purification our Harijan brethren should be helped to get rid of this habit.” Yet the abhorrence of carrion-eating that elites in this period often couched in the language of hygiene, upon scrutiny, amounted to nothing but the religious or communal objection against beef-eating.

Deceptive Majority 156

As Swami Shraddhanand (1946: 134) had astutely pointed out to Gandhi in 1921, the discourse of hygiene was disingenuously applied to this context, since untouchables, in the ordinary course of things, consumed the flesh of cattle immediately upon the latter’s death. In practice the difference between the meat of slaughtered and naturally expired kine was fine indeed. And in fact, Gandhi’s own comments and the reportage of Sangh secretaries confirms that the overt goal of eradicating carrion-eating frequently was conflated with the goal of eradicating beef-eating, and ultimately meat-eating altogether. Thus in the last analysis, beef “being forbidden in Hinduism … carrion and beef eating must be given up” (Gandhi 1972c: 429). The endeavors of the Harijan Sevak Sangh to transform the dietary practices of the Depressed Classes—to have them eat and drink like Hindus—resulted in reports like this from Bengal: ABSTINENCE: – As a result of propaganda work, 450 Harijans of Bolpur, in the district of Birbhum, have given up drinking habits and 1,275 Muchis (Chamar) have taken a vow not to take beef. In the Midnapur town, the Mehtars brought flesh of a dead horse to eat; but at the endeavour of the district committee [of the Sangh] they gave it up. As a result of propaganda work, in almost all the districts the Harijans are giving up eating flesh of dead animals and, in many places, drinking. (Roy 1933)

The Lal Begi oral tradition gives sanction, perhaps even sacrality, to the idea of a communal feast on horse flesh, a central event in a popular legend of Lal Beg. That the Bengal Sangh prevails upon the Mehtars of Midnapur to desist from this tradition is suggestive of the kinds of shifts being brought about by Gandhian “untouchable uplift,” shifts inadequately represented by the secular-sounding trope of carrion and liquor. At times, the Sangh’s efforts to promote Hinduism among the Depressed Classes went beyond the dissemination of religious teaching and dissuasion from “un-Hindu” practice—we might call these “soft Hinduization”—and took on the more openly communal rhetoric and tactics of the Arya Samaj and the Hindu Mahasabha, or what might be called “hard Hinduization.” A summary of the Sangh’s annual report for 1936, documenting activities in the months following Ambedkar’s announcement that he would not die a Hindu, reflects a ramping up of the organization’s efforts in the religious domain: All branches of the Sangh devoted attention to the religious education of Harijans. Reading of the Ramayana, Mahabharata, Geeta and other religious

157 Trustee Majoritarianism

books was undertaken at nearly 1300 centres run by the Sangh. More than one thousand bhajan parties were organised in Kerala on the temple-entry question. In Mysore State, 40 bhajan mandirs were built and 19 were under construction. Vedic rites were introduced in the marriage customs of Harijans so as to make them simple and give them a really religious character. Harijan children were given opportunity to attend caste-Hindu marriages. At Allahabad and elsewhere, the Diksha Sanskar was held. (Verma 1971: 142)

The first several entries indicate a mere acceleration in the Sangh’s usual promotion of Hinduism, but with the mention of the “Diksha Sanskar” (dīkṣā saṃskār), or “Rite of Initiation,” the Sangh enters a new terrain. Dīkṣā saṃskār was Madan Mohan Malaviya’s variant on shuddhi; it was, at the time, the Mahasabha’s means of ritually formalizing the Hindu identity of untouchables. In Travancore in the same year, regional and national Sangh leaders, including Birla, publicly decried mass conversions of Depressed Classes— Ezhavas, in this case—to Christianity. Representing the Harijan Sevak Sangh, Rameshwari Nehru admonished Ezhava women to “remain loyal to Hinduism” and to prevent their men from becoming Christian (Desai 1936c). The undisguised engagement in religious rivalry was not confined to the aftermath of Ambedkar’s conversion announcement. In the 1940s in Punjab, another center of Christian mass movements, the Sangh provided legal support to members of the Depressed Classes who were fighting criminal cases in court. “As a result of such help given to the Harijans,” notes the regional report approvingly, “about 365 Harijan converts [to Christianity] rejoined Hinduism” (Thakkar 1946: 35). But for the ritual form, this could be an Arya Samaj report of a successful shuddhi. The Sangh here is counting untouchables won for Hinduism in a competitive arena by the extending of concrete support in the form of legal aid. Gandhi’s disavowal of the “blameworthy” methods of the Arya Samaj notwithstanding, his own organization was now, at times, acting in ways scarcely distinguishable from those championed by Shraddhanand. William Gould (2005: 113–14) notes that Muslim observers and British administrators perceived Gandhi’s and the Congress’s involvement in untouchable uplift from 1932 onward as reminiscent of the aggressive shuddhi campaigns of the 1920s; they “resembled movements of conversion more than simple campaigns for social reform.” What I have sought to demonstrate is that this was not merely a matter of perception or resemblance. If the labors of the Harijan Sevak Sangh “resembled” the Hinduizing efforts of the Arya Samaj

Deceptive Majority 158

or the Hindu Mahasabha, this was for good reasons. First, they sometimes took the same “hard Hinduization” forms—condemnations of conversion to Christianity and Islam, “reconversion,” diksha sanskar—or were taken up conjointly with the latter bodies. In north India, the Sangh appears to have collaborated at times with Malaviya’s Hindu Mahasabha as well as with the Arya Samaj (Sundar 1933: 7; Verma 1971: 142). Second, many Sangh activists were either themselves Arya Samajists or openly sympathized with the Arya Samaj. For instance, during his stay in the Sangh colony in Delhi overseen by N. R. Malkani, Gandhi discovered that “many of [the Harijan Sevaks] belonged to the Arya Samaj” and claimed allegiance to Swami Shraddhanand (Desai 1936a: 57). Most importantly, even in its “soft Hinduization” approach, the Harijan Sevak Sangh was ultimately helping to bring about the same fundamental transformation that the Samaj and the Mahasabha sought to realize: remapping the boundaries of the Hindu community so as to encompass the Depressed Classes, securing a majority in the nation to come.

The Mainstreaming of a Majoritarian Myth The Harijan Sevak Sangh carried on the legacy of Swami Shraddhanand’s Arya Samaj in other ways as well. Consider the following passages: The Muslim built the latrine…. It is only in towns with large or influential sections of Muslim population that the latrine becomes an essential part of the building. For instance, it is only in cities like Hyderabad, Bhopal, Delhi and Lucknow that most houses—even Hindu houses—would have latrines. The Muslims called the latrine a “JAI ZARUR” [sic] i.e. a place of necessity. Going to “Jungle” for easing would have no meaning in Arabian deserts and the custom of Pardah made privacy a necessity. The latrine in India has crept in. It has never been brought in as a part of our Hindu houses. The Bhangi is essentially a recent product of urban life, first created as an occupation by Moslems and later, in British rule, made into a hereditary caste. (Malkani 1965: 10, 25, 26, 137)

As we learned in the previous chapter, Swami Shraddhanand alleged that untouchability was a product of Muslim rule, and Arya Samaj prachāraks, in


Trustee Majoritarianism

an elaboration of this myth, claimed that Muslims introduced the enclosed toilet and thus manual scavenging to India. We might expect, then, that the passages above are lifted from an Arya Samaj tract. In fact, though, these are passages from Clean People and an Unclean Country, a book published in 1965 by the Harijan Sevak Sangh and authored by its then–vice president, Professor N. R. Malkani. Malkani was a close associate of Birla in operating the Sangh headquarters in Delhi from 1933 onward, who had by the time of Clean People served as member of Parliament (MP) for a decade and had chaired two parliamentary committees on the conditions of sanitation workers. His brother K. R. Malkani was a dedicated Rashtriya Swayamsevak Sangh (RSS) worker and chief editor of The Organiser from 1949 to the early 1970s (Jaffrelot 2007: 176). Like Shraddhanand and other Hindu nationalists who circulated the canard that Muslims introduced manual scavenging to India, N. R. Malkani supplies no actual evidence of the claim, and ignores the archeological and textual evidence of manual scavenging in India centuries before the advent of Islam. Instead, by suggesting a correspondence between demography and urban architecture (towns with more Muslims have more latrines) and positing differences in linguistic and defecatory habits (Muslims stool indoors in what they call a “place of necessity” whereas Hindus ease themselves outdoors in what they call the “jungle”), Malkani offers a kind of speciously ethnographic argument for historical origins. The cultural difference in toilet practices Malkani ascribes here is hung on the geographic contrast between “Arabian deserts” and the Indian “Jungle,” mapping the Muslim on the former and the Hindu on the latter in a familiar Hindu nationalist tactic of figuring Muslims as foreign. The social location of Malkani’s assertions is unambiguous; before Muslim rule, “our Hindu houses” were innocent of latrines, Bhangis, and Pardah. Remarkably, the myth of the Muslim invention of manual scavenging has become, in postcolonial India, a widely accepted idea among the sanitation labor castes and even in some elite circles (see, for example, Khator 2017). Malkani’s traffic in the trope undoubtedly bears some responsibility for its diffusion, as well as for its respectability. Scholarship in Hindi and English continues to cite Malkani as an authority on this point (see, for example, Shyamlal 1997: 24, 38; Valmiki 2008: 150). Writers and scholars who would know better than to cite openly partisan sources like Arya Samajist polemicists have followed Malkani—wrapped as he is in the moral authority of a Gandhian

Deceptive Majority 160

social worker and lifelong Harijan Sevak Sangh activist, and the governmental legitimacy of a chairman of parliamentary committees on the conditions of sanitation workers—without compunction. If this demonstrates how a Hindu nationalist fabrication, and more generally Shraddhanand’s “hard” communalist approach to majoritarian inclusion, thrived and gained respectability in the Harijan Sevak Sangh, another aspect of Malkani’s writing illustrates the Sangh’s characteristic fusion of Arya Samajist and Gandhian methods. Malkani, like the grassroots Arya Samajist Amichand Sharma, had years of experience working among the sanitation labor castes, and knew well the degree to which their religious life reflected a thoroughgoing engagement with Islam. He presented this ethnographic knowledge in a manner that wove together the representational practices of Sharma and of Gandhi. Some Bhangis of Panjab are known as “Baleshahis” who are followers of the great Moslem disciple of Guru Nanak. Others call themselves “Lal Begis” i.e. followers of a Moslem Pir of that name. A few are followers of another Moslem Pir … and even adopt Moslem personal names. But in spite of these Moslem appellations or even modes of worship, all the Bhangis are and have remained as staunch Hindus throughout the vicissitudes that Hinduism experienced during Moslem rule. (Malkani 1965: 131)

Like Sharma, Malkani acknowledges the prevalence of Islamic practices and orientations among the sanitation labor castes, but denies their significance. And like Gandhi, Malkani declares the Depressed Classes Hindu irrespective of their beliefs, practices, and self-appellations. He even reproduces Gandhi’s favored locutions: the “Bhangis” are Hindu “in spite of” their Muslim names and modes of worship. In its leadership, its publications, and its practices, then, the Harijan Sevak Sangh synthesized the strategies of majoritarian inclusion modeled by the Arya Samaj, on the one hand, and Gandhi, on the other. Decidedly Gandhian elements of the Sangh were nomenclature (the institutional preference for “Bhangi” and “Harijan”), organizational hierarchy (the policy of appointment rather than election for Sangh boards, with the Depressed Classes categorically ruled out of leadership), and approach to labor disputes (the Sangh generally sought to mediate with municipalities on behalf of sanitation workers and dissuaded the latter from striking (see, for example, Sastri 1933a; Verma 1971: 209). In all of this the sanitation labor castes found themselves cast in the role

161 Trustee Majoritarianism

of Uka in Gandhi’s paradigmatic script of Hindu reform, with its distinctive structure of agency. They were patiently to perform their “sacred duty” of sanitation work while “citizens … raise[d] their voice on behalf of them” and privileged caste Hindus schooled them in Hinduism. At the same time, Sangh leaders not infrequently mixed their roles, acting as trustees in the Gandhian manner, but also as missionaries in the mold of Shraddhanand. Rameshwari Nehru and Ghanshyamdas Birla exhorted the Depressed Classes not to convert to Christianity; the effort to persuade untouchables to identify as Hindu finds favorable mention in Sangh reports of dīkṣā saṃskār in UP and “reconversion” in Punjab. And it was in the persuasive mode that N. R. Malkani propagated fabrications borne of the Arya Samaj’s shuddhi campaigns: “The Muslim built the latrine” and the “Bhangis of today are those people who were warriors made captives after they fell to the enemy” (Malkani 1965: 10; Shyamlal 1997: 38). In the following chapter we will discover that this synthesis of modes of achieving an “indivisible family” of Hinduism, exemplified by the Harijan Sevak Sangh, spilled over into the Congress at large and characterized the process by which the first generation of sanitation labor caste politicians was mentored and encouraged to refashion their community. The Hinduizing project that played out in Lucknow, as we will see, featured both “hard” and “soft” forms. But crucial to understanding the transformations that followed, and why my interlocutors sometimes called it Gandhiji kā den, the “gift of Gandhiji,” is the “Constitution (Scheduled Castes) Order of 1950” by which the government of newly independent India named and recognized its untouchables. Article 17 of the Constitution—one of Ambedkar’s triumphs as chairman of the drafting committee—declared that “Untouchability is abolished.” The Constitution further established measures designed to protect and promote the interests of those who had suffered the structural exclusions of untouchability. For this system of compensatory discrimination to function, the state required means of determining who was entitled to it; the obvious choice was the category of “Scheduled Castes,” a successor of “Depressed Classes” in the language of the colonial state. But in the moment of the establishment of the new republic, the Congress government altered the contours of the category that it inherited. Whereas the Scheduled Castes of the late colonial dispensation were Dalits of any religion, including Muslims, the Scheduled Castes Order of 1950 inaugurated a new regime of recognition. It declared that “no person who professes a religion different from the Hindu religion shall be deemed to be a member of a Scheduled Caste.”

Deceptive Majority 162

So deeply has the post–Poona Pact common sense of the transhistorically Hindu untouchable pervaded the public sphere that the 1950 Order has attracted relatively little scrutiny; to a great many people, apparently, the decree seems a simple reflection of social facts rather than a technique by which those facts were quite recently generated. Scholars who have taken up the subject observe that the religious restriction of the 1950 Order contravenes the Constitution’s guarantee of religious freedom to all citizens. As Tanweer Fazal (2017: 21–23) puts it, the decree involves a “flouting of the principle of secularism” motivated, in part, by “a majoritarian anxiety” on the part of Hindu legislators. In Dieter Conrad’s (2007: 216) assessment, the Order represents a major step in “the legal consolidation of the majority community—legal Hindutva, as it were.” This it certainly was, and in this capacity cemented the framing terms of the Poona Pact—still shifting and contingent when they were pulled together in 1932—into the foundations of postcolonial law. I want to suggest, in other words, that the newly restrictive definition of the Scheduled Castes represented the culmination of two decades of assiduous labors by Gandhi and his followers in the Congress and the Harijan Sevak Sangh to enclose the Depressed Classes within the narrative of Hinduism. If, as I have argued, Gandhi’s signal approach to the inclusion of untouchables in the nation as Hindus hinged on a monological assertion of nominative power over a population defined as morally and politically non-agentive, then the Scheduled Castes Order of 1950 was its apotheosis.

5 Hinduization and Its Discontents Valmiki Comes to Lucknow

Deepak asks me to come home for his grandmother’s terahī, a rite and meal held on the thirteenth (terahī) day after her death. At the event, I hear one of Deepak’s relatives call it not a terahī but a chālīsī—a ‘fortieth.’ He then corrects himself. Deepak explains that Muslims commemorate the dead on the fortieth day, whereas Hindus do it on the thirteenth. His relative had momentarily forgotten that their life cycle rites, which until recently followed the Muslim pattern, were now conducted in the style of Hindus. We sit in Deepak’s house in a circle around a small fire while an Arya Samaj officiant—who turns out to be Balmiki like everyone else in the room—conducts the havan ceremony, reading Sanskrit prayers from a book. After the ritual, Deepak and I repair to a side room and I notice a portrait hanging on the wall, an old, somewhat blurry photograph of a tall and imposing one-eyed man dressed in the green robes of a Muslim faqir. I ask if the portrait is of a holy man of some sort. “That’s my grandfather,” Deepak replies. “Muhammad Pahalwan.” When India gained independence in 1947, the sanitation labor castes of Lucknow and its neighboring districts neither referred to themselves as Balmiki nor were referred to as such by others. They had predominantly musalmānā nām—names in a Muslim style—names such as Rukhsana, Nazira, and Allah Rakhi for women; and Nabbu, Ramzanu, and Anwar for men. “We were all Lal Begis,” explained a man born in 1923. “When I was eight, ten, twelve years old, there was not even mention of the name Valmiki.” The Lal Begis of Awadh entered neither Hindu temples nor—with a few exceptions—mosques. Religious life centered on the home, the neighborhood Lal Beg shrine, and Sufi centers like Dewa Sharif, where the panchayat of the 583 clan convened, and the tomb of Zohra Bibi in Rudauli, in the annual festival of which the Lal Begis performed a key ritual function (Nevill 1904: 53–54). Brahmins played no role, and Muslims a significant role, in Lal Begi sacraments: officiating

Deceptive Majority 164

at burials, rendering halal the sacrifice to Lal Beg, and reciting the fātiḥa at nuptials and innumerable domestic rites. Meat was invariably served at Lal Begi weddings. Among Lal Begi women the Ramadan fast was widely observed; women and men both carried replicas of Karbala (tāziya) in the Muharram processions for which Lucknow and the Awadh countryside have long been renowned. A little over sixty years later, in 2011–12, I meet not a single person in Lucknow who refers to himself or herself as Lal Begi—at least not initially, and not in mixed company. Most men and many women in the community call themselves Balmiki, and it is by this name that non-Dalit residents of Lucknow—when intending to convey respect, at least—refer to the sanitation labor castes. Hinduānā or Hindu-sounding names predominate: women’s names like Shanti, Sunita, and Asha; men’s names like Ganesh, Vinod, and Ram Kumar. The city’s Hindu temples, while still largely disconnected from community religious life, are now accessible, and each of the larger bastīs now has its own Rishi Valmiki temple. Burials remain common, but cremation is on the rise. Brahmins now preside, not infrequently, at weddings, and at wedding feasts vegetarian preparations are competing with the traditional biryani and mutton curry. Participation in Muharram remains popular, but no one fasts for Ramadan anymore. “Ab zamāna badal gayā, bhaiya [The era has changed, brother],” a great-grandmother tells me, speaking of how she no longer keeps the Islamic fast that she did in her childhood. “That age has passed [Zamāna nahīṅ hai voh].” These are, by any reckoning, significant transformations. Take Deepak and his grandfather Muhammad. If personal names are taken as transparent indices of religious identity—as they all too frequently are, by everyone from police to employers to anthropologists—then the current generation of Balmikis would appear to belong to an entirely different religion than their grandparents did at the time of independence. Yet this has transpired in the absence of a mass conversion—nothing like Ambedkar’s Buddhist movement or the Meenakshipuram conversions to Islam has taken place—or even the kind of ritual embrace of Hinduism that the Arya Samaj promoted in shuddhi and the RSS and Vishwa Hindu Parisad (VHP) have revived in ghar wapsi or “homecoming” events. How, then, did the transnomination of this entire community take place, and what, if not unambiguous badges of religious belonging, can musalmānā and hinduānā names be taken to signify?

165 Hinduization and Its Discontents

Should we consider these transformations a variety of Sanskritization? The turn toward vegetarianism and cremation in Lucknow does bring to mind M. N. Srinivas’s well-known term for collective upward mobility through the emulation of the practices of brahmins and similarly advantaged castes, just as it brings to mind Ambedkar’s mimetic theory of caste endogamy that anticipated Srinivas’s term (Rege 2013: 99–105; Srinivas 1996: 88). Nicolas Jaoul (2011) characterizes the adoption of Valmiki by the sanitation labor caste as “politically engineered Sanskritization,” a felicitous move insofar as it draws attention to the political mechanisms by which the movement was advanced. But the Sanskritization paradigm presumes the Hindu or proto-Hindu status of its protagonists as they begin their “upward” climb, whereas Lal Begis, as we have seen, regarded themselves and were regarded by others as a qaum not just “below” but apart from Hindus. For this and other reasons, Sanskritization misses the mark (Lee 2015: 3–4). Acknowledging that the developments in question involved a rupture in religious identification—yet without conversion as it is generally understood—Bhagwan Das (2007) and Vijay Prashad (2000) describe them as “Hinduization.” Combining this insight with Jaoul’s we might fairly characterize the Valmiki movement as an instance of politically engineered Hinduization. This chapter, in tracking the career of Valmiki in Awadh from 1947 to the present, illustrates both politically engineered Hinduization and the nonAmbedkarite, Harijan politics of which it was an integral part. Methodologically, we shift here from archives to oral history; our narrative is based on interviews with (more than a hundred) participants in and witnesses to the events described. Our focus follows the shift in the leadership of the Hinduizing project from advantaged caste actors to members of the sanitation labor castes themselves. In terms of scale, we narrow here from the all-India canvas of the preceding chapters to Lucknow and its adjacent districts—the territory of the 583, the Chuhra clan that predominates in central UP—in order to examine how majoritarian inclusion played out in a specific regional context.

From Cantonment to City By all accounts, it was in 1947 that Rishi Valmiki, and with him the prospect of a Hindu pahchān for the sanitation labor castes, arrived in the city of Lucknow. His vehicle was Kanhaiya Lal, a resident of the cantonment. The sanitation labor castes of Lucknow have long had two divisions, shaharwāle and kanṭwāle, those of the city and those of the cantonment. This separation stretches back, as so much does in Lucknow, to the revolt of 1857.

Deceptive Majority 166

In the years following the insurrection the colonial state not only razed to the ground two fifths of the old city of Lucknow but relocated and expanded its military installation to an immense campus just southeast of the city. The new cantonment, though spatially proximate, was segregated from the city in myriad ways, including administratively (Oldenburg 2010: 33, 48–55). In domains like sanitation, the parallel administrations of city and cantonment recruited differently: the former locally, following nawabi precedent, and the latter through networks of mobile military service labor that extended, web-like, from cantonment to cantonment, often bypassing whole regions in between. This late nineteenth century configuration has remained, at least insofar as the distribution of sanitation laborers is concerned, largely undisturbed ever since. The shaharwāle, then, understand themselves as locals, natives of Lucknow and its hinterland. With deep roots in the Awadh countryside, they speak Awadhi alongside Hindi and can point to ancestors who worked for the nawabs and the landed gentry. To a great degree the shaharwāle are co-extensive with the 583. Marriages are local: wedding processions rarely travel beyond Bara Banki or Sitapur, and often remain within Lucknow itself. The cantonment faction, on the other hand, has been mobile since at least colonial times. The marriage processions of the kanṭwāle trace the vectors of old labor migrations: they journey to Kanpur, Kolkata, Delhi, Meerut, Amritsar, Sagar, Jaipur. Rarely do they marry their caste fellows in the city; indeed, something of the aloofness of British cantonment culture lingers in the hint of hauteur with which some kanṭwāle regard their city counterparts, the coolness of cosmopolitans toward rustic cousins. Those kanṭwāle not employed in sanitation often find employment in the railways, strengthening their links to a transregional caste network. News travels along this network, as do holy men, ideas, and rumors. It comes as no surprise, then, that the person universally credited with inaugurating the Valmiki movement in Lucknow was a man not of the city but of the cantonment: Kanhaiya Lal Balmiki.1 Like most kanṭwāle families, 1

The Kanhaiya Lal Balmiki of our narrative should not be confused with Kanhaiya Lal Balmiki of Bulandshahr, UP, who was an MP (Lok Sabha) from 1952 to 1962. The two Kanhaiya Lals shared a great deal in common besides their name: both were Arya Samajists and Congressmen, part of the first generation of Balmiki politicians in UP. They were also friends. The Bulandshahr Kanhaiya Lal visited our protagonist at his home in Lucknow cantonment on several occasions; members of their respective families ultimately intermarried.

167 Hinduization and Its Discontents

Kanhaiya Lal’s had migrated from elsewhere; his parents came from Bareilly. Despite almost all schools in the 1920s being closed to Depressed Class children, Kanhaiya Lal’s parents managed to arrange for tutors to instruct him in Urdu and Hindi. He did some sanitation work as a child but primarily worked as a khānsāma (cook) and manager in the cantonment mess. In his free time he earned a reputation as a formidable pahalwān (wrestler), lāṭhībāz (cudgel fighter), and boxer. As his daughter told me, Kanhaiya Lal was also known for his height, fair skin, and striking appearance, allegedly inherited from his mother who “looked just like a Pathan.” According to his grandsons, Kanhaiya Lal often repeated the story of the first time he had heard the Ramayana and decided to become a performer. As a boy, the story went, Kanhaiya Lal accompanied his father to the edge of a field—as close as Dalits were allowed to approach—in which an all-night Ramayana recitation was taking place. Rapt in the performance, Kanhaiya Lal closed his eyes, imagining himself to be in the court of Ram itself—until his father, thinking him asleep, shouted at him to wake up. Kanhaiya Lal then resolved to learn the Ramayana himself, and after years of self-guided study memorized the Radhe Shyam version of the epic and began to perform it, accompanying himself on the harmonium, impressing listeners and often surprising those who learned that he was Dalit. It was these performances that brought him to the attention of the Lucknow cantonment chapter of the Arya Samaj. “They were all Baniyas,” Kanhaiya Lal’s daughter told me of the Arya Samajists who began to invite Kanhaiya Lal to Samaj functions. “Slowly, slowly, slowly, slowly, he began to join them…. The meaning of Arya Samaj was: caste [jāt pāt] does not matter here.” The first individual of the sanitation labor castes to join the Lucknow Arya Samaj, Kanhaiya Lal adopted the organization’s practices and other signs of affiliation: he attended meetings regularly, wore the sacred thread (janeu) and the brahminical hair-tuft (choti), and introduced vegetarianism and Vedic ritual at home. Kanhaiya Lal also kept company with peripatetic sadhus from his caste, including those of the lineage of Khak Shah Baba, said to have been one of the interlocutors of the Arya Samajist Amichand Sharma, author of Śrī Bālmīki Prakāś. Khak Shah Baba was buried not far from Lucknow in Sitapur District. While the relationship of this order of sadhus to the Arya Samaj is complex, one element of the complexity is the brotherhood’s reproduction in its own devotional literature of extensive passages of Sharma’s Śrī Bālmīki Prakāś,

Deceptive Majority 168

without acknowledgement of the source (for example, Bauhat 1996). It is unclear whether Kanhaiya Lal first encountered the message of Rishi Valmiki’s link to the sanitation labor castes from the privileged caste Arya Samajists of Lucknow cantonment or from the sadhus of Khak Shah Baba’s lineage who, going to or coming from their founding figure’s ashram in Sitapur, brought news to the cantonment of the Valmiki movement flourishing in Delhi, Punjab, and other points west. In either case, the two contexts reinforced one another. When his grandson told me that Kanhaiya Lal had instituted Friday night Ramayana recitations at their home followed by prayers to Valmiki, I asked him if the prayers ever contained the name of their composer. “I didn’t know then who it was,” he told me, “but in the praise verses [stūtī] there was this refrain: amī chand ardās uchāre, kripa dījo sāmi balmīk guru antaryāmī [Amichand utters this entreaty, give us grace, Lord Balmik, all-knowing guru].” Amichand Sharma’s Hinduizing text, then, had finally reached Lucknow, whether as a handful of verses or the complete tract. In addition to Amichand’s verses, Kanhaiya Lal also began to compose and perform his own bhajans in praise of Valmiki, verses that highlighted the sanitation labor castes’ newfound origins at the heart of Hindu tradition. One verse that people still remember is this: Lal Begi nām chhoṛo, riṣiyoṅ ki santān ho Abandon the name Lal Begi, you are the descendants of sages2

It was a call to repudiate Islamicate associations (indexed by the Lal Begi title) and embrace the narrative of a lofty Hindu genealogy—a pithy condensation of the central message of Śrī Bālmīki Prakāś. Yet, while adopting the Arya Samajist narrative, Kanhaiya Lal’s verses also subverted it on a key point: the sanitation labor castes here were not the descendants of Rishi Valmiki’s “untouchable” disciples, as Sharma would have it, but rather descendants by blood. This deft reworking of the message bore the implication that either Valmiki was a Chuhṛā or the sanitation labor castes were brahmins, either of which posed a problem for brahminical ideology. Upon independence Kanhaiya Lal’s acquaintances in the Arya Samaj recommended him for a job in the new government’s Madh Nishedh Vibhag (Abstinence Department) as a prachārak (propagandist, preacher). It was no small thing, as a Dalit, to secure a government job in a sector unrelated to 2

The second part could also be imperative: “Be the descendants of sages.” I am grateful to Frances Pritchett for pointing out this potential meaning.

169 Hinduization and Its Discontents

sanitation labor in 1947. In the Abstinence Department, Kanhaiya Lal’s circle expanded swiftly in two directions. On the one hand, he met and grew close with some of the leading lights of the Congress, including Ghanshyamdas Birla, Govind Ballabh Pant (then chief minister of the United Provinces, soon to be chief minister of Uttar Pradesh) and Chandra Bhanu Gupta (who would be UP’s chief minister thrice in the 1960s). On the other, his work—spreading propaganda against drink among the poor of Lucknow—led him to the bastīs of the shaharwāle, the 583s, his caste fellows in the city. Kanhaiya Lal went from neighborhood to neighborhood, convening caste meetings at which he persuaded his brethren to take vows against the consumption of liquor and to constitute committees to enforce these pledges. Alongside this message of abstinence—his official work—he also advocated a turn to Hinduism. As one of his earliest shaharwāle admirers recalled: “The Arya Samaj people told [Kanhaiya Lal], ‘These people will all become Muslim, so you go and persuade them that they are Hindu and that they should remain Hindu.’ So he came to our neighborhoods, held meetings, explained to us that we were Hindu.” He quickly developed a following among a set of young shaharwāle men. Govind Prasad, one of a triumvirate of shaharwāle reformers that would work closely with Kanhaiya Lal in the decades to come, described his mentor’s appeal as grounded in a new ethic of self-respect, manifest in his proud bearing and manner of speech. Govind Prasad told me, [Kanhaiya Lal] lived among the English. Now let’s say a policeman, a mere constable, shouts to my father, “Come here!” Well my father would fold his hands and with fear and trembling stammer, “Master, tell me.” But this Kanhaiya Lal, when even a high level police officer spoke to him he would coolly reply, “Yes what is it?” It makes a difference when you’re accustomed to dealing with [military] officers every day.

The revolutionary comportment that Kanhaiya Lal modeled—his pathbreaking hexis (Bourdieu 2013 [1972]: 82–94)—combined with his extraordinary message to win him both admirers and enemies. On the one hand, a group of kanṭwāle elders convened a panchayat that briefly excommunicated the upstart reformer. On the other, he galvanized a group of shaharwāle youth prepared to join him in intensive community reform. For Govind Prasad and others, “He became our leader [netā].” In October of 1950 Kanhaiya Lal organized Lucknow’s first Valmiki Jayanti—the celebration of Valmiki’s birth—in a field in the cantonment

Deceptive Majority 170

near his home. Govind Ballabh Pant, the chief minister of UP, attended and addressed the gathering as its chief guest. In 1952 independent India’s first elections were held; at Pant’s prompting, the Congress fielded Kanhaiya Lal as its candidate for the UP state assembly from the reserved constituency of Pihani in Hardoi District. He won the election, becoming one of the first Balmiki members of the legislative assembly (MLAs), and continuing to win elections as the Congress candidate for five consecutive terms, until 1974. In 1956, Kanhaiya Lal built Lucknow’s first Valmiki temple, in Baṛī Lāl Kurti in the cantonment. Ghanshyamdas Birla provided him the funds for the temple, as a plaque therein memorializes. This established a precedent. Over the next fifty years, another ten Valmiki temples and statues would be erected in the city, almost all funded and inaugurated by privileged caste Hindu politicians of the Congress and, later, the Bharatiya Janata Party (BJP): mayors, cabinet ministers, governors, and even a prime minister.3

Hailed as Harijan At the inaugural Valmiki Jayanti in 1950, Kanhaiya Lal’s political mentor and patron Govind Ballabh Pant delivered a speech addressing the sanitation labor castes assembled in the cantonment field. Lalta Prasad, another of the triumvirate of shaharwāle reformists attracted to Kanhaiya Lal, was present for the occasion. Sixty years after the event, he recalled Pant’s speech to me in this way: Pant ji said, “Look, you people should not abandon sanitation work.” Pant ji said this, these words at least I remember. Also, “You people, a lot of the young generation go about saying ‘leave this work, leave this work of sanitation.’ But if you leave this work you won’t be able to make such swift progress, other work won’t immediately be available. You all are Balmikis. But you people don’t revere Valmiki ji honorably. You people get intoxicated. Stop doing this. It’s true, isn’t it? You should study. Valmiki ji was a learned man. You people, too, should become learned if you’re going to write ‘Balmiki’ [as your title]. 3

Former prime minister Atal Bihari Vajpayee paid for the land on which the Nirala Nagar Valmiki temple was erected and laid the foundation stone for the Valmiki temple at the entrance of the King George’s Medical College (KGMC) servants’ quarters. Other figures who have supplied funds and/or inaugurated Valmiki temples and statues in Lucknow include former UP chief ministers Kalyan Singh and Ram Prakash Gupta, former UP cabinet minister and member of parliament Lalji Tandon, mayor Dinesh Sharma, and former UP governors Motilal Vora and Suraj Bhan. Bhan, unlike the others, was Dalit.


Hinduization and Its Discontents

Study at least enough that you can educate others, and stop getting intoxicated, and give up gambling, and don’t tell lies.” Pant ji spoke very well.

While this speech is not included in Pant’s fifteen-volume Selected Works, Lalta Prasad’s recollection of its contents is entirely consistent with the chief minister’s public remarks on Harijan matters on other occasions (for example, before the UP Legislative Assembly, Pant 2000b), and echoes a speech that he gave at a “Harijan welfare worker’s camp” (very likely a Harijan Sevak Sangh event) north of Lucknow three months earlier, in July 1950, where he said: An agitation promoting a class war for uplift of Harijans is not in the interest of Harijans and I am glad that Harijans have chosen the way shown by Mahatma Gandhi for their uplift…. I do not like the idea of Harijans giving up their indigenous professions and vocations. While they should not be compelled by anyone to do a particular job, it would not be in the interest of their economic well-being if they tried to give up their indigenous professions. It is also necessary for Harijans to receive higher education so that they may possess qualifications for higher services and professions…. But education should not result in an aversion for manual labour. (Pant 2000a: 322–23)

More than Gandhi, Pant stressed the value of “Harijans” obtaining education, and made clear that he wanted to see some of them attain high posts in government administration. Pant also embraced—rhetorically at least—the official criminalization of the practice of untouchability, whereas Gandhi warned that such measures would hinder the genuine reform of Hindu society. With these two exceptions, Pant’s hortatory faithfully replicated Gandhi’s: untouchables must not agitate for rights but cooperate with Congress reformers dedicated to their “uplift.” They should continue performing their ancestral professions of manual labor while undertaking moral reform within the community—giving up gambling, liquor, and “lying.” Consistent with Gandhi’s advice to “Harijan correspondents” and the habits that the Harijan Sevak Sangh sought to cultivate in the recipients of its beneficence, Pant prescribed to his untouchable Congress dependents a program of political obedience, labor quiescence, and internal reform. This was the program that Kanhaiya Lal, as Pant’s protégé—and more broadly, as a Congress Harijan—was invited to undertake. Inhabiting the role of a Harijan leader in the party of Mahatma Gandhi afforded privileged access to the power of the state, even as it meant adopting the vocabulary and operating within the constraints of a Gandhian social imaginary. There was,

Deceptive Majority 172

however, another opening in the horizon of possibilities stretched before the first generation of Congress Balmikis. As we saw in the previous chapter, the Congress in UP inherited not only Gandhi’s approach to the untouchability question but also that of Swami Shraddhanand’s Arya Samaj. This double inheritance was not only a matter of the Samaj and the Congress in UP overlapping extensively in personnel and programmatic activity (Gould 2005), nor only a matter of widespread, Arya Samaj–stoked, anti-Muslim sentiment that followed Partition and was relatively normalized in the UP Congress over which Pant presided (M. Hasan 2007; Z. Hasan 1996). In addition to these factors, what decisively shaped Harijan politics in the years after independence was the Hindu nationalist strategy—articulated by Swami Shraddhanand, popularized by the Arya Samaj, and eased into Congress respectability by Birla, N. R. Malkani, and the Harijan Sevak Sangh—of framing the plight of the sanitation labor castes (and untouchables generally) as a product of Muslim dominance, and thus of hardline Hinduization as a means of untouchable liberation. The receptivity of a large swath of the Congress establishment in UP to this formula after 1947 meant that Congress Balmikis could expand the scope of their activities beyond the limited domain prescribed by Gandhi precisely insofar as this expansion adopted this framework. That is, the Gandhian demand that Harijans limit themselves to internal reform could be circumvented, and the more ambitious effort to reform the casteism of the dominant could be undertaken, so long as the dominant could be identified as Muslim. As we will see, this afforded Congress Balmikis considerably greater room to maneuver in the effort to eliminate public forms of untouchability than the strictly Gandhian program did, even as it committed them to a communalist politics. The requirements of political obedience and labor quiescence, meanwhile, remained firm. An early indication of how Kanhaiya Lal would negotiate this space is found in an article he wrote, published in the Hindi journal Uṭṭhān (Uplift) on December 25, 1950. Entitled “Mehtar Jāti kā Durbhāgya tathā Uskī Vivaśtā” (The Mehtar Caste’s Misfortune and Helplessness), the article catalogs a number of “atrocities” or caste-based crimes—abductions, threats, a rape, and a murder—perpetrated on women and men of the sanitation labor castes in 1950, and appeals, in the name of the nation, for an end to untouchability. Kanhaiya Lal’s language both conforms to and departs from Gandhian orthodoxy. On the one hand, he refuses Gandhi’s nomenclature for the sanitation labor castes, preferring, as did so many of Gandhi’s interlocutors


Hinduization and Its Discontents

from these castes, Mehtar over Bhangi. On the other hand, in Gandhian fashion, he frames the article as an effort to change hearts—“I hope that this will also have an effect on the cruel and selfish hearts of our country”—and deploys the language of penance (prāyaśchit): “[I]f a solution is not quickly found to the problems of the Mehtar caste it will be a great mistake of the people of this nation, for which they will have to undergo great penance.” Kanhaiya Lal also draws a comparison between the inclusion of the Mehtar in the nation and the inclusion of untouchables in the Hindu community of bhakti narratives: “Remember that in the same way that in the Dwapar Yug the Pandavas’ sacrifice was incomplete until our Swapach Maharaj (Guru Balmiki) was served food, so likewise the work of building this nation [rāṣṭr] cannot be complete without our uplift [utthān].” This formulation echoes both Shraddhanand and Gandhi; moreover, it references the very story from the bhakti canon that Amichand Sharma made use of in Śrī Bālmīki Prakāś: the story of the inclusion of Balmik the Chandal in the Pandavs’ sacrifice, provoking Draupadi’s disgust (see chapter 3). Kanhaiya Lal thus makes explicit the analogy implicit in the earlier Arya Samajist’s work: nationalists are like the Pandavas, who suddenly discover that they need their erstwhile despised other in order to achieve a political goal. Kanhaiya Lal does not stop, though, with appeals to the nation’s conscience intoned in the key of bhakti. He also reports his own efforts, and those of his caste fellows, to actively combat atrocities of a certain sort: those perpetrated by Muslims. While he describes some of the perpetrators of caste crimes against Mehtars simply as “gunḍe” (thugs, rowdies) without specifying their religion, he characterizes others as “musalmān gunḍe” (Muslim thugs) or simply as Muslims. Anonymous crimes are left to the conscience of the nation, but the atrocities committed by Muslims—specifically, their harassment and sexual exploitation of Mehtar women—provoke Kanhaiya Lal and some of his caste fellows to fight back: to extract their women from Muslim homes and to organize scavengers’ strikes against private Muslim employers (though not against the municipality). We will return to the controversy over Muslim men and Mehtar women later. Here, the point is that Kanhaiya Lal, in his 1950 article, makes use of both modes of public engagement available to a UP Congress Harijan in this period: appeals for compassion in the idiom of bhakti, and calls for solidary action against predatory Muslims. Kanhaiya Lal died in 2004, before I undertook this research, and a flood in his daughter’s house destroyed her collection of his correspondence and

Deceptive Majority 174

other documents, including the handwritten journals he had maintained for decades. A fuller sense of the political environment in which he operated, though, may be obtained by turning to his close colleague Narain Din Balmiki, who served with Kanhaiya Lal as member of the UP Legislative Assembly from 1952 to 1957 before becoming MP (representing Shahjahanpur) from 1957 to 1962. With Kanhaiya Lal, Narain Din sought to reform the sanitation labor castes while improving the structural conditions in which they lived. Narain Din also shared with Kanhaiya Lal the patronage of Birla and Pant, and his correspondence with these and other Congress luminaries, preserved by his son and daughters, throw light on the Congress’s relations with the first generation of leaders from the sanitation labor castes. Born in a village in Allahabad District, Narain Din worked with his father aboard steamships in the Bay of Bengal. He joined nationalist politics in Rangoon; after its evacuation in World War II he returned to Allahabad and developed strong ties with the Tandon and Nehru families. In the sanitation workers’ strike of 1946 Narain Din mediated between sanitation workers and the provincial Congress administration, earning him the favor of Pant. Upon being elected MLA in 1952 he began living in Lucknow and in 1957 he brought his family as well. He founded the Hind Sweepers’ Sevak Samaj (HSSS), an all-India organization that received 80 percent of its operating costs as grants-in-aid from the national government, and whose official patrons were Pant and, upon Pant’s death, Prime Minister Lal Bahadur Shastri. Through the HSSS, Narain Din opened and operated schools and vocational training centers for the sanitation labor castes and other Dalits; his efforts were particularly directed at helping youth enter occupations other than sanitation. Coretta Scott King, during her 1959 visit to India with her husband, Martin Luther King, Jr., participated in an HSSS function at Narain Din’s invitation. In Narain Din’s correspondence with Congress leaders it is clear that he took every opportunity to draw attention to the untouchability issue and to solicit support for efforts his HSSS was undertaking for sanitation labor caste youth. The replies he received were consistently congratulatory: Indian vice president Zakir Husain, for example, wrote in 1965 to convey “my best wishes for the success of the Camp organized by Hind Sweepers’ Sevak Samaj on the occasion of Kumbh-Mela at Allahabad.” In a 1962 letter Ghanshyamdas Birla encouraged Narain Din’s social service by enclosing a check. Govind Ballabh Pant, serving at the time as India’s home minister, wrote a Hindi missive in August 1957 that applauded his protégé’s efforts while underscoring the

175 Hinduization and Its Discontents

non-agitational path the Congress expected its Harijans to follow. “I am pleased to learn that the Hind Sweepers’ Sevak Samaj in its function gave pride of place to a peaceful, constructive program [śāntimay rachnātmak progrām]. I hope that this program will prove especially beneficial for our Bhangi brothers’ becoming self-reliant, and will find great encouragement. Yours affectionately, G. B. Pant” (emphasis added). Yet when it came to the specific requests that Narain Din made of Congress leaders, his correspondence reveals a long series of rebuffs and disappointments. The Food and Agriculture Minister declined to help the HSSS establish a piggery in Allahabad (an attempt, on Gandhian lines, to make a traditional caste occupation profitable in a changing economy). Prime Minister Nehru regretted that there were no funds available to support Narain Din’s proposal for improving work conditions for sanitation laborers. The Minister for Education could do nothing about “alleged hardships of parents of Scheduled Caste candidates in getting admission … in Delhi schools.” Prime Minister Morarji Desai, in a 1978 letter that captures the tone and substance of almost three decades of such correspondence, wrote: “There have been attempts to eliminate the obnoxiousness from the work that [sweepers and scavengers] have been doing. I would suggest that your Samaj should create a sense of awareness among them so that they look for alternative employment and not depend on what is considered hereditary work. With good wishes, Morarji Desai.” Affectionate in tone, Congress leaders met Narain Din’s pragmatic proposals for sanitation labor caste uplift with a familiar refrain: leave aside structural inequality, focus on internal reform. While the assistance Congress leaders extended to Narain Din and Kanhaiya Lal’s efforts to benefit the sanitation labor castes appears to have been meager, it was nonetheless costly. As chief minister of UP and later as Union home minister, Pant took firm action against labor strikes in various sectors and expected his political dependents to play their part. In the wake of the five-day central government employees’ strike of 1960, for example, Pant wrote: My dear Narain Din, I have received your letter of 29 July. I was glad to know from it that you contacted P. & T. [Posts and Telegraphs] employees before the strike and took an active part in dissuading them from taking part in it. The strike presented a grave threat to the orderly life of the country, and I fully appreciate your efforts and the efforts of those whose names you have mentioned. Yours sincerely, G. B. Pant

Deceptive Majority 176

On the question of labor strikes by sanitation workers in particular, Pant was a believer in Gandhi’s dictum that “a Bhangi may not give up his work even for a day.” When sanitation workers struck in Calcutta in the 1960s, a contingent of Lucknow shaharwāle youth was mobilized and sent by rail to Calcutta to replace the striking workers. A man now in his sixties, who was among the teenagers sent to break the Calcutta sweepers’ strike, told me that it was Kanhaiya Lal’s men who enlisted them. According to his children, Narain Din was ambivalent about the title “Balmiki” that his colleagues in the Congress encouraged him to use in his political career, and by which they often addressed him in their letters. This was largely on account of his belonging to the Hela community, a sanitation labor caste populous in Allahabad and elsewhere in southern and eastern UP, rather than the Chuhra caste with which the Valmiki movement was more closely associated. The majoritarian aspect of the Valmiki movement does not seem to have been the source of Narain Din’s unease with the title. Indeed, like Kanhaiya Lal, Narain Din in his writing advanced the thesis that Muslims were responsible for the abject condition of the sanitation labor castes. In an undated pamphlet entitled Hela-Halalkhor, Narain Din attempted for his own caste (referred to in some regions as Halalkhor as well as Hela) what the Arya Samaj had done for Chuhras in Śrī Bālmīki Prakāś: to provide a respectable Hindu genealogy and distance it from Islam. Citing a 1925 tract by one Pandit Devdatt Sharma of Ghazipur, Narain Din (n.d.: 4) asserts that the Hela caste “is descended from a Kshatriya lineage. 630 years ago, Alauddin Khilji put the caste into this disgusting occupation [yeh ghriṇit kārya].” The evidence given is that Hela gotra names resemble those of kshatriya gotras. The prevalence of the Perso-Arabic title Halalkhor does not indicate that the caste migrated from Arab lands, Narain Din reasons. Rather, “we are Indian [Bhāratīya] and our religion is Vedic,” but Muslims proffered a more palatable name than the Sanskrit-derived Hela, a term of disrespect [anādar]. Ultimately the tract turns to exhortation: “O Hela-Halalkhor, awake! This is not your ancestral profession.... The Ramayan is your holy book, take it up! The assembly of the solar [kshatriya] lineage is the Hela’s patrimony” (Narain Din n.d.: 5). In rejecting the ancestral connection to sanitation labor, Narain Din makes a departure from the strand of Hindu majoritarian overture that would insist on the modern sweepers’ descent from the brahminical Chandal. But he follows another major strand, represented in texts like Patitoddhār, that attempts simultaneously to flatter and communalize Dalits by suggesting


Hinduization and Its Discontents

that they were kshatriyas whose fierce resistance to Muslims led to their degradation. In Gandhi’s view, the role that Harijans were to play in the eradication of untouchability was auxiliary: the uncomplaining performance of traditional forms of labor, cooperation with penitent privileged caste Hindus, and internal reform in hygiene, morality, and religious practice. With Ambedkar outmaneuvered (in the Poona Pact) and politically marginalized (in subsequent elections), this powerful paradigm proliferated, largely unchallenged, in the political discourse of the Congress of the 1930s and 1940s, and assumed concrete institutional form in the organizational structure of the Harijan Sevak Sangh, a body that heavily overlapped with the Congress. When Kanhaiya Lal and Narain Din were fielded by the UP Congress to be its Balmiki face in 1952, they found themselves interpellated in a Gandhian discourse that, by that time, had achieved hegemonic status in the party. Whether they recognized themselves in this discourse or not, it was on the condition of inhabiting it that they were invited to access political power; asked to share the stage of history, they were handed a script prepared by Gandhi. Hailed as Harijans, it was the Harijan paradigm that they had, perforce, to adopt in order to tap the power and resources of the state.

The Triumvirate Takes Up Valmikism The coterie that formed around Kanhaiya Lal between 1947 and 1950, when his work as a prachārak for the Abstinence Department took him from Lucknow cantonment to Lucknow city, centered on three shaharwāle youth who would, in the next six decades, exercise an extraordinary influence on the lives of their caste fellows in Lucknow. I was fortunate to interview each of them on multiple occasions in 2011–12; all three passed away in the next three years. We have already quoted Govind Prasad, one of the three, on his attraction to Kanhaiya Lal’s bold comportment when he first heard the latter speak at a caste meeting in December 1947. Like Kanhaiya Lal, Govind Prasad, born in 1926, worked for British soldiers in his youth, washing dishes at a military factory and serving as personal assistant to an officer for seven years until the latter’s return to England upon Indian independence. Govind Prasad studied to eighth standard in a Christian missionary school. Between 1947 and 1953 he did sanitation work in the Provincial Hygiene Institute of King George’s Medical College (KGMC), after which he secured employment as a postman, a position he held until retirement.

Deceptive Majority 178

We have also heard from Lalta Prasad; it was his recollection of Pant’s Valmiki Jayanti speech that we quoted earlier. Born in 1923, Lalta Prasad studied to eighth standard and worked a variety of jobs: he did sanitation at the Provincial Hygiene Institute (with Govind Prasad), was a laboratory assistant at Isabella Thoburn College, and performed several kinds of technical labor in the railways, where eye injuries wrecked his vision. On account of his austere lifestyle and tireless labors as a promoter of education within the caste, he was generally known as Master Lalta Prasad, and also as “our Gandhi.” The third and youngest member of the shaharwāle reformist group, Achhe Lal, inherited Kanhaiya Lal’s mantle as a Balmiki “face” of the Congress. Born in 1932, Achhe Lal in 1950 secured employment as one of Lucknow’s sanitary supervisors, becoming the first member of the sanitation labor castes to hold a position in the municipality higher than that of sweeper. He became president of the UP Safai Mazdur Sangh (Sanitation Workers’ Union), in which capacity he engaged with several UP administrations and served as a member of statelevel and national-level committees on the issues of sanitation and scavenging (including the Malkani Committee). In 1985 the Congress appointed Achhe Lal a member of parliament (Rajya Sabha), after which he was appointed to the UP Legislative Council (Vidhan Sabha). All three spent their youth in the city but nurtured deep roots in the villages of Bara Banki. By the early 1950s, each of them held government jobs other than sanitation worker (postman, railway technician, sanitary supervisor), setting them apart from their parents’ generation and from most of their peers. As a group they were better educated than the majority of their caste fellows, and through Kanhaiya Lal (and to an extent Narain Din as well) they became better politically connected. Yet unlike Kanhaiya Lal or Narain Din, who would always remain outsiders to the shaharwāle, these three were 583s, the paradigmatic shaharwāle, members of the largest clan in Awadh of the most populous sanitation labor caste in South Asia. Their parents and siblings had Muslim-sounding names (Lalta Prasad himself was known by a musalmānā name until the Valmiki movement), revered Lal Beg, and held caste meetings in a Sufi shrine. They could appeal to their brethren in a way that their mentors, though caste fellows in a broad sense, could not. By 1950, Govind Prasad, Lalta Prasad, and Achhe Lal began to accompany Kanhaiya Lal on his rounds in the bastīs of the sanitation labor castes of Lucknow and its environs. They convened meetings where they pressed their caste fellows to send their children to school and cease drinking liquor. They

179 Hinduization and Its Discontents

performed Kanhaiya Lal’s bhajans, singing while Kanhaiya Lal played the harmonium. Munnu Balmiki, who witnessed some of these performances in his childhood, recalled to me the quartet’s message in this way: Munnu: “Abandon all of these bad habits. Our guru is Valmiki ji. Adopt him [unko apnā lo]. You people should not go for conversion [dharm parivartan]. Be Hindus, become Hindus, remain Hindus.” Joel: Wait, was it “become Hindus” or “remain Hindus”? Munnu: “Become.” Joel: Become? Munnu: “Become.” Yes. “And if you become Hindu then you will get benefits. From the government you will get this, you will get that.” The government certainly took care of them! Made them MLAs and MLCs4…. They also talked about Gandhi, “Gandhi did this, Gandhi did that.” There was a lot about Gandhi and Valmiki…. “Quit these dirty habits, study, educate your children, quit drinking liquor, don’t do bad things,” they said all this…. “Don’t convert to Islam, caste brothers! Don’t do all this, don’t keep Muslim names, otherwise you won’t get reservations, you won’t get any government benefits,” this is what they said.

This conjoining of abstinence, education, Hinduization, Gandhi, and government benefit, and the contrasting of this assemblage with liquor, bad habits, conversion, and Islam, was characteristic of the reformist message. There were other elements as well—regarding debt, the control of women’s sexuality, funerary equipment, meat, history, and Lal Beg—that the reformists articulated in new configurations, plotting them into a powerful evaluative scheme. To this ideology reformists gave the name Vālmīkiyat, that is, Valmikism. Lalta Prasad also called it Hinduat, or Hinduism. The following sections detail the propagation of Vālmīkiyat, and its contestation, in four key arenas.

Ishqbāzī: The Cohabitation Controversy In the majoritarian contract, the promise of Hindu belonging was tied to the repudiation of Muslims. This included the repudiation of sexual intimacy. 4

Recall that Kanhaiya Lal was an MLA (member of legislative assembly) and that Achhe Lal became an MLC (member of legislative council) after his stint as MP (member of parliament).

Deceptive Majority 180

From 1948 to the early 1950s, a controversy erupted among the Lal Begis of Lucknow as the reformists—Kanhaiya Lal, the shaharwāle trio, and a few other men—objected to the cohabitation of a number of women of their caste with Muslim men and sought, with Arya Samaj support, to marry the women instead to their caste fellows. We have already seen that Kanhaiya Lal raised this issue in his 1950 article in Utthān. Here let us try to piece together this episode using his article and the recollections of the triumvirate, alert to the silences that these accounts produce. Elsewhere I have characterized as intimate untouchability the distinctive relation formed between Dalit women (less often, men) and the privileged caste householders to whom they provide domestic sanitation labor—a relation structured by sharp economic and social asymmetry and avoidance practices animated by fears of pollution, while also suffused with affects brought into play by daily interaction in homes shielded from public view and the interdependence and mutual vulnerability entailed in one person’s cleaning the bodily expulsions and interior domestic spaces of another (Lee 2018: 6–9). Representations of this relation in Hindi and Urdu literature and in autoethnographic social science portray a scene as marked by degrees of warmth and mutual concern as it is shot through with hierarchical differentiation—an adunation of intimacy and dominance (I. Ahmad 2010; Ali 1940; Chakrabarty 2018; Chander 2015; Dikshit 1996; Nagar 2017; Valmiki 2010). I was compelled to temper my skepticism toward the depiction of affection in some of these representations when women and men who cleaned domestic latrines in Lucknow portrayed their relations with their employers in similar terms. The intimacy of intimate untouchability may take a variety of forms, many of which have little to do with sex—the secret-sharing, “gossiping” friendship of the matron of the house with her Mehtarani, whose touch she scrupulously avoids, is a prominent one (I. Ahmad 2010; Nagar 2017; Valmiki 2010). That said, sexual intimacy, too, is among the forms this relation takes. In a range of representations of the privileged caste, heterosexual male gaze (Hindu, Muslim, or otherwise), the Mehtarani appears as a sexually charged figure, a sign calling forth interpretants of desire. This correlates with the observation in human rights literature that, due to their position in the nexus of caste, class, and gender hierarchies, women engaged in domestic sanitation labor are uniquely vulnerable to advances from men in the homes they service (Irudayam S. J., Mangubhai, and Lee 2011). It also correlates with what the sociology of downward mobility tells us: that cohabitation and marriage are not altogether uncommon paths for this relation to take—indeed, they are a leading cause


Hinduization and Its Discontents

of what Shyamlal (1997) calls aspraśyīkaraṇ, that is, “untouchabilization” or “becoming untouchable.” In Lucknow as elsewhere, especially in earlier decades when domestic dry latrines were more common, it was not unheard of for a woman of the sanitation labor castes to take up residence with her employer. When the employer was Muslim—and his “pollution” concerns thus milder, on the whole, than those of Hindu householders—such cohabitation led, in some cases, to a quiet absorption of the woman into her employer’s family and community: the Mehtarani became Muslim. In other cases, involving either Muslim or Hindu men, cohabitation led to a scandal, but sanitation labor caste panchayats tended to permit the “offending” couple to marry on the condition that the man join the woman’s caste (“untouchabilization”), a procedure common enough to have formal rites (Crooke 1890: 38; Greeven 1894: 50–52; Shyamlal 1997). When men declined marriage, sanitation labor caste panchayats levied fines against them or refused their community labor services, while the women went on to marry within caste. Traditionally, that is, sanitation labor caste panchayats, unlike their advantaged caste counterparts, responded to sexual breaches of caste boundaries with neither violent retribution nor the physical or social death of the woman. This is not to say that Dalit women’s sexuality was not policed at all, but that “infractions” were treated with a flexibility quite unlike what obtained among the savarn (on other differences from the sexual norms of brahminical patriarchy, see Kolenda 2003: 3–78; Searle-Chatterjee 1981: 42–59, 68–73). The advocates of Valmikism sought to change this. In his 1950 article, Kanhaiya Lal wrote the following: Having toured most of the cities, towns, and villages of Uttar Pradesh, I have inspected the worrisome and pitiful condition of my caste. In several districts of the province women of my caste have been ruined [pathbraṣṭ] by Muslim rowdies [musalmān guṇḍe]. If I were to go into all of it, the topic would get out of hand. In Lucknow and Bara Banki alone, in the last six months, overcoming great difficulties we have obtained [barāmad kiyā] the following women of our caste from Muslims’ homes: [here five women are named in a numbered list]. In addition to these, there are more women of our caste who have been induced by Muslim rowdies to run away and live with them. These, too, we are planning to reclaim [punaḥ prāpt karne kā vichār kar rahe hain], though this work is not free of danger.

This portion of the article takes on the character of a crime report: the talk of “inspection” (avalokan), the numbered list of names, the peculiarly police-work

Deceptive Majority 182

phrasing of “obtaining” women (the conjunct verb barāmad karnā suggests the seizure of contraband), and the identification of the offenders as guṇḍe, a category of criminal. This is not accidental; police were indeed involved in the events summarized here and Kanhaiya Lal formulates his language accordingly. The account also resonates with one of the most polarizing tropes to circulate in Indian public discourse in the years following 1947: the kidnapping of Hindu (and Sikh) women by Muslims during Partition, and the effort by the state to “reclaim” such women from their captors in Pakistan (V. Das 2006; Datta 1999). Consistent with the terms of the discourse of kidnap and rescue, Kanhaiya Lal’s phrasing makes men the agentive subject—it is they who ruin, induce to run away, find, and reclaim—and women as their objects. The vocabulary and framing shift somewhat when we turn to the account of Lalta Prasad, the member of the shaharwāle triumvirate best known for his work as an educator. It was from Lalta Prasad that I first heard of the cohabitation controversy. We had been discussing the various changes he and his friends had helped bring about in the community when I asked him, for the first time, whether the Arya Samaj had supported these efforts. Later he would tell me a great deal about the Arya Samaj—for instance, that he and Achhe Lal and Govind Prasad were all, in the 1950s, dues-paying, janeu-wearing members of the Samaj—but the first thing that my question prompted him to relate was this: Lalta Prasad: There were some girls in our community—I may as well tell you the truth—there were some girls who ran off to the Muslims [musalmānon ke yahān bhāg gayī thīn]. The Muslims made them run off with them [musalmān bhagā le gaye the]. Joel: Did they marry? Lalta Prasad: They just made them run off; I mean, had an affair and made them run off [ishqbāzī maiṅ bhagā le gaye]. Now, the Arya Samajists, they were really sore about Muslims. Back then, even now…. So we told them how the Muslims had made our girls run off, and they, with police help, got our girls out, and brought them to their place and kept them, kept them in the rakṣāmaṇḍal…. Then some of them we got married into our community, and others were married into the Hindu community [kuchh log ham apne samāj mei le āye, shādī kar diyā, aur kuchh Hindu samāj mei]…. [One of the young women] didn’t like it around here so we got her married in [a town in western UP] and she settled there.


Hinduization and Its Discontents

Lalta Prasad’s account introduces ambiguity regarding the women’s volition: first he says that they ran off (bhāg gayī thin), but then that they were made to run off (bhagā le gaye the). Furthermore, his glossing of the situation as ishqbāzī—philandering, romancing, having affairs: a term that implies illicit but consensual relations—implicates the women’s agency more strongly than Kanhaiya Lal’s portrayal of them as “ruined” (pathbraṣṭ kī gayīn). Note as well that Lalta Prasad, quite unselfconsciously, refers to Hindus in contradistinction to the collective self—“some of them we got married into our community, and others were married into the Hindu community”—a resurfacing of the old common sense even in the discourse of one of those committed to Hinduization. Lalta Prasad’s narrative also complicates the trope of kidnapping and confinement, for here the women appear doubly kept: first by Muslims, but subsequently by the Arya Samaj (or the police?) in a rakṣāmaṇḍal (“place of safety”—about which more in a moment). But where did the women want to be? That one “didn’t like it around here” is suggestive. Yet we still know little about the women’s perspective on either their cohabitation or marriage among “their own.” It was Govind Prasad, another of the triumvirate, who explained to me about the “place of safety” in which the women were “kept.” This was the Dayānand Rakṣāmaṇḍal, an institution operated by the Lucknow Arya Samaj in Rakabganj in the old city for the first decade or so after independence. The Arya Samaj used this space to house “Hindu” women—my interlocutors mentioned servants, domestic sanitation workers, and sex workers—whom it had “rescued” from “Muslims,” in circumstances that were far from clear. It was, as Lalta Prasad later described it to me, “like a jail [jail jaisā].” Elderly, privileged caste Arya Samajists were alarmed when I asked them about the rakṣāmaṇḍal in interviews at their Rakabganj office and wanted to know where I had heard about it. Their characterization of the institution confirmed Lalta Prasad’s and Govind Prasad’s accounts. The rakṣāmaṇḍal’s dubious legality became a liability after the initial post-Partition years of high communalism subsided, and the institution was quietly brought to a close. Of the reformists, Govind Prasad’s account of the controversy was by far the most detailed, and also the most conflicted. Unlike his associates, Govind Prasad, as he reflected on the reform effort decades after it transpired, often described it in corrosively critical terms: “I am a deceiver [Makkār huṅ],”

Deceptive Majority 184

“We did all kinds of dishonest dealings [Ye sab beīmānī ham karte the].” Proud of some of the transformations that he and his comrades wrought—above all the revolution in education—he found others, in retrospect, troubling. His narrative of the struggle over women’s cohabitation with Muslims oscillates between auto-critique and self-justification, between sympathy and outrage toward the women in question: The women of our homes, of our caste, they used to go out into society to labor. Some of them went out and stayed. It’s like this: any woman, if all day every day you’re going to forty or fifty houses and cleaning the filth, collecting excrement by hand and carrying it away, well—if you live there [instead of working there] then you don’t have to do all this. So would you rather live there or live here? So when we started this social reform group, well—what is social reform? Isn’t this part of it? Isn’t this a part of it that there’s a woman who’s up and become a bībī [lady, madam, wife (to her employer)], left two children behind, and she’s sitting there [that is, luxuriating, not working], and her children are here wandering around hungry and naked, and none of us have a house worthy of the name? So those women whom the Muslims had locked up at their places, we went with the police and took them, forcibly dragged them by their hands and feet, and filed a court case against them. But where to keep them? For the duration of the case, when we scarcely have huts to live in ourselves, where should we keep them? So the Arya Samaj kept them in Rakabganj … where they used to lock up women. Then we would get a boy [of our caste] ready and get them married. That’s the sort of thing we were doing. We didn’t care for our lives! Brother, death is better than dishonor [beizzatī].

Govind Prasad then explained that the new marriages were conducted by Arya Samajists on their Rakabganj premises according to their rites. Before the weddings, the Arya Samaj conducted shuddhi ceremonies to purify the women after their stay with Muslims. He then backtracked to the matter of the legal proceedings the reformists had initiated against the women. The lawyer who represented the Arya Samaj in its legal affairs, himself a member of the Rakabganj Arya Samaj, handled the cases for the reformists free of charge. In cases where the Muslim man and Mehtar woman were simply living together, the reformists won the case easily enough, separating the couple and marrying the woman off. But in other cases the ishqbāzī had been formalized as marriage.

185 Hinduization and Its Discontents

In one such instance the city magistrate arbitrating the case chastised Govind Prasad and his associates, saying: “They had a court marriage, according to the law they’re both adults…. You have no right to bring them away.” [The magistrate] did right by the law. But we leapt up and said, “What is this?” We made a big scene, [saying,] “We’ll abandon our profession, we’ll die, send us to jail! What law says that a man fighting for his welfare should be sent to jail?”

Undeterred by this setback, the young reformers sought to advance their cause by other means. According to Govind Prasad, it was a supportive word from Chief Minister Pant himself, secured by Kanhaiya Lal, that enabled them to again have the married couple arrested and separated, and this time they were able to marry the woman off within the caste. When I asked whether the women themselves opposed the reformists, Govind Prasad replied, “They didn’t just oppose us. Inside the court, in front of the magistrate, they took off their shoes and hurled them at us.” In this account, then, the women appear fully agentive; their cohabitation with Muslim employers is an attempt to secure upward class mobility and relief from stigmatized labor. Without disregarding the unequal and coercive dimensions of the cohabiting relation—he maintains that the Muslims kept the women “locked up”—Govind Prasad acknowledges a complex social field in which Muslim men represent, for Dalit women, one pathway out of caste oppression. But for the purveyors of Vālmīkiyat, this long-standing state of affairs would not do. The old order excluded untouchable men from participation in the patriarchal competition for izzat (honor, shame); what was foundational to privileged caste manhood was structurally denied its ontological other. But in the new regime untouchable men were invited to join the game, to aspire to izzat and to adopt its gender ideology (Paik 2014: 179). With its rakṣāmaṇḍal, the Arya Samaj offered sanitation labor caste men a means to assert control over their women’s sexuality; indeed, to lay claim to the women as theirs, rather than always potentially someone else’s. But these means were made available, and the invitation to accrue izzat extended, only insofar as the Dalit men’s efforts targeted Muslim antagonists and could be channeled into communal, rather than caste or class, conflict. The rakṣāmaṇḍal did not offer its “refuge” to domestic workers sexually exploited by Hindu employers. For the reformers, becoming Hindu was partly accomplished by relocating the women of their caste from “Muslim” to “Hindu” spaces. This assertion

Deceptive Majority 186

of gendered control was not only bodily and spatial but also nominative and representational. The women at the center of the cohabitation controversy, like most 583s of that generation, had musalmānā names. The reformists and their Arya Samaj supporters, as part of their intervention, renamed them. Rukhsana became Rama Devi, Sadiqa became Shanti, and so on.5 Not only this, but Kanhaiya Lal, in his 1950 article, published their (new) names. By any measure, to print in a magazine the names of “ruined” women (in a numbered list) was an extraordinary act. This move was noteworthy, I would argue, in three ways. First, it constituted a kind of performative speech act in Austin’s sense, transforming Lal Begi women into Hindu women by publicly naming them as such under felicitous conditions (namely Kanhaiya Lal’s unique authority as a Congressman and expert on his own caste). Second, it was an assertion of representative power in the Gandhian style of nomination without persuasion; irrespective of how Rukhsana might have construed her pahchān, Kanhaiya Lal represented her in tones agreeable to the Hindu reformist imagination—as Rama Devi. Third, it signaled a moment of patriarchal arrival: an inaugural performance of Balmiki men’s possession of izzat by shaming women.

The Repudiation of Lal Beg Since the 1920s, the Arya Samaj had sought to alienate the sanitation labor castes from one axial figure of their pantheon, Lal Beg, while encouraging the identification of another, Balmik, with the Sanskrit poet and brahmin sage Valmiki. Some Arya Samajists, like Amichand Sharma, tried to sideline Lal Beg by subordination to Balmik, urging that “Lal Beg was simply another disciple of Shri Balmiki ji…. [Therefore] those [of you] who call yourselves Lal Begiyas [Lal Begis] are making a mistake” (A. Sharma 1928: 21). Others took a more aggressive approach, circulating the idea that Lal Beg had actually been a Muslim nawab who raped Mehtar women until their men murdered him (and then took to worshipping the site of the murder) (Prashad 2000: 95–96). This narrative template for the transcribing of caste grievances (here rape by dominant caste men) onto a communal canvas foreshadowed the politics we have been discussing, the politics of the rakśāmaṇḍal. The assault on Lal Beg in Lucknow featured elements of these approaches, but also developed its own distinctive trajectory. The campaign began with the


These are, of course, pseudonyms.


Hinduization and Its Discontents

verses—some composed by Amichand Sharma, others by Kanhaiya Lal—that the reformists sang, including the one discussed earlier: Lal Begi nām chhoṛo, rishiyon ki santān ho Abandon the name Lal Begi, you are the descendants of sages

Here Valmiki (the sage referred to) was figured in contrastive opposition to Lal Beg; the reformers held that to embrace the idea of descent from Valmiki was, perforce, to abandon allegiance to Lal Beg. The new caste guru would not complement the old, but replace him; the model was not accretive, but iconoclastic. Consequently, while half of the message (“you are the descendants of sages”) was intended to flatter and encourage pride, there was no escaping the fact that the other half of the message constituted a condemnation of tradition. In response, shaharwāle in Lucknow and Bara Banki composed verses of their own that inverted the attack and ridiculed Valmiki and his advocates. For example: Netwā āye, netwā āye, hamkā jhāṇṭ mīki batāran “Leaders” have come, “leaders” have come, they’re saying our pubic hairs are “mīki”

Mīki is a nonsense word that, conjoined with jhāṇṭ (pubic hair), rhymes with Bālmīki. The quip thus corrupts and parodies reformist speech; the refrain “You all are not Lal Begi, you are Balmiki” becomes “You are not Lal Begi, your pubic hair is mīki.” I heard several versions of this witticism from elderly interlocutors. Quips and slogans aside, arguments were brought to bear on the matter. The advocates of Hinduization developed two contradictory lines of reasoning regarding Lal Beg. One is perhaps best summed up by Lalta Prasad: “Lal Beg, you see, was called a Muslim. And we people, well, we are Hindu. So it behooves us to revere Balmik ji, not to revere Lal Beg ji. This is what we preached.” This approach, jettisoning the complexities and contradictions of ancestral religion like so much dead weight, accommodated a religious field now neatly and agonistically bifurcated into Hindu and Muslim by simultaneously resignifying Lal Beg as Muslim (whereas previously he was “ours”) and “we people” as Hindu (whereas previously we were neither Hindu nor Muslim). This line of reasoning found little of value in the ancestors and their ways; elders who clung to Lal Beg were figured, in contradistinction to the reformers themselves, as uneducated, and their traditional practices as superstition.

Deceptive Majority 188

The other argument was that Lal Beg was not Muslim at all; rather, he was a Hindu deity in Muslim guise, a ruse by the ancestors to escape persecution by Muslim overlords. This idea figures the ancestors as subaltern manipulators of signs for whom dissimulation was a means to preserve tradition under oppressive conditions. Such a premise treats tradition as a domain for interpretation sensitive to historical conditions, and thus invites an engagement with the past both more generous and more sophisticated than its reduction to superstition. That said, in its particulars this argument, like the other, accepted the idea that the world was transhistorically cleft into a Hindu–Muslim binary to one or the other of whose elements all social facts could be assigned. Thus one of the key “proofs” offered for Lal Beg’s being “really” Hindu rather than Muslim was that traditional Lal Beg shrines featured a red flag—red being a “Hindu” color. Another was that the propitiation of Lal Beg traditionally involved an offering of bhāng (a preparation from the cannabis plant), a “Hindu” rather than a “Muslim” libation—again, in a dyadic social imagination. In either case, whether reimagined as Muslim or as crypto-Hindu, Lal Beg was to be dislodged from his status as the caste’s preeminent object of reverence as a matter of practice. Both lines of reasoning put forth by the reformers were intended to produce the same effect: the end of the ritual reverencing of Lal Beg. Consider, again, Lalta Prasad: “We boycotted him [Lal Beg]. [We said,] ‘Whoever reveres or worships him, or at weddings builds that structure with five laḍḍus [the shrine], you must desist.’” As the advocates of Hinduization gained followers, then, the physical destruction of Lal Beg shrines began. These shrines typically consisted of a square mud platform crowned by five domes and often indented with niches for oil lamps, and were located just outside of Lal Begi homes. Phulmati, a woman born around 1950, recalled to me how men in her community who had joined the reformist wave demolished some of the shrines in the Bara Banki village where she spent her childhood before moving to Lucknow. “They [the demolishers] said, ‘This belongs to the Muslim community. We are Hindu. This [shrine] is Muslim.’ With this idea they did it. They said, ‘These five domes have gone over to the Muslims. It is Muslim religion.’” Bande Lal, born in a Bara Banki village in 1933, described to me how he destroyed his family’s Lal Beg shrine. He grew close to the reformist circle in the late 1940s and 1950s and sometimes joined Kanhaiya Lal and the shaharwāle triumvirate in the meetings they convened and at Valmiki Jayanti functions. In narrating these times, Bande Lal dwelt on his sister’s wedding, when he

189 Hinduization and Its Discontents

announced to his family his opposition to the traditional cock sacrifice to Lal Beg. Bande Lal: So my mother scolded me, saying “Arre, this is Lal Beg Sahib! Arre, he must be worshipped! Otherwise something bad will happen. Let us do this.” So I said, “Whatever needs to be done, I will do. But this I will not do.” And I started digging it [the Lal Beg shrine] up. Joel: Digging it up? Bande Lal: Yes, I started digging it up, uprooting it. Joel: You removed it? Bande Lal: Yes, but then my mother started beating me. Joel: Beating you? Slapping you? Bande Lal: (laughing) Slapping, yes. What to do? So I backed off. But when she left the village, I threw it into the pond…. Mother went off to buy some things in another village, and I saw no one was at home, so I dug it up, threw it in the pond, earth and all. Afterward my brothers, everyone told me that I shouldn’t have, and [when she returned, mother] abused me roundly, called me names, insulted me, but her biggest worry was that this fellow has dug it up and thrown it away—god forbid something happens to him! (laughs)…. [But] nothing happened to me.

Far from suffering the retributive misfortune that his mother feared, Bande Lal, buoyed by the networks of Arya Samaj and Congress patronage that his friendship with the reformists made available, prospered. First a sweeper at Lucknow University, he got a better job as a sweeper in a scientific research institute, and then was promoted to laboratory attendant—still a fourth class employee, but free from the stigma of bearing the broom. His family did not rebuild the shrine. Extirpation was only one of the ways in which Lal Beg shrines disappeared as the Valmiki movement spread; more often, they were not uprooted but simply neglected. As Lal Begis began to adopt Valmīkiyat, many simply ceased to perform the periodic maintenance required to maintain the integrity of their shrines. Slowly, monsoon rains dissolved the domes and niches; erosion rendered the platforms indistinguishable from the earth over which they had risen.

Deceptive Majority 190

The repudiation of Lal Beg meant not only the abandonment of his shrines but the cessation of his rites, notably the cock sacrifice to which he was traditionally entitled during wedding celebrations and at his annual feast on Jamghaṭ (the day following Diwali). Consider the following conversation, in which members of a family of seven sisters described to me how their household Hinduized. Zulekha (third sister): Our father [who died in 2000] began to object that we could not remain in two things, we would have to go for one—either become Lal Begi or become Balmiki. He started to say that we should all be [the latter]. Sanjay (Zulekha’s son): This much was certain: from Lal Begis we developed [develop karke] and became Balmikis. We were Balmikis, we just didn’t know it. [… From the time of grandfather’s change of heart] the cock was no longer sacrificed over there at the shrine. We killed it here at home. Joel: Any special reason? Safana (sixth sister): Yes, the reason was this, that ever since Valmiki Jayanti started to be celebrated here in the neighborhood, our father from that time found out that we people are Balmikis. And Lal Begis became others. So he started saying that we are Balmikis, we will remain Balmikis, we will not be Lal Begis. Since then he ended it [the cock sacrifice….] Naziran (eldest sister): Father began saying, “No, I am Hindu, and why should we have the cock killed for Lal Beg? We will kill it, but just ordinarily [not ritually], we will not have it killed [by the Muslim sacrificer].” After that we were not allowed to get the cock killed for Lal Beg.… He saw that our Valmiki ji was a Hindu…. “When our guru is a Hindu, then why should we have a Muslim kill the cock, why should we have him recite [the fātiḥa]? We are Hindu, so even if we do eat the cock on Jamghaṭ, we will not get it killed by a Muslim.” This is what he said.

The words that the daughters and grandson attribute to their late forebear illustrate the kind of categorical reshuffling that followed acceptance of the arguments in support of Valmīkiyat. The father concedes to tradition on one point: the family will continue to eat fowl on Jamghaṭ, along with their caste fellows. But in the new dispensation the flesh will no longer be consecrated by a Muslim officiant who recites Quranic verses; slaughtered not in the halal fashion but “ordinarily,” the feast will cease to bear the signs of Islamic sacrality.

191 Hinduization and Its Discontents

The father of the seven sisters was far from alone in forbidding his family to ritually reverence Lal Beg. As the reformers grew in influence, an increasing number of families stopped performing the heretofore de rigeur cock sacrifices at weddings and Jamghaṭ. That the youngest member of the family, Sanjay, plotted this change in a discourse of development or progress—“from Lal Begis we developed and became Balmikis”—is significant. In the next section we will see why.

Naming-Education “Originally my name was Muslim.” This utterance, or a variation on it, was so frequently given voice in my interviews with Lucknow Balmikis born before the 1980s that I learned to expect it. Many of my interlocutors’ stories of renaming resonated with that of Rishi Kumar, a poet, song-writer, and retired hospital sanitation worker born in the 1940s: Rishi Kumar: Originally my name was Muslim. Hasanu. That was the name I was given at first…. Joel: Your parents, sisters, brothers, everyone called you by this name? Rishi Kumar: Earlier they called me by this name. Now everyone knows me as Rishi Kumar. Joel: Did you change it yourself? Rishi Kumar: No no, my name, in truth, changed on account of my studies. When I was studying, that’s when my new name got written [likhā gayā]. Joel: Do you remember how it came about? Rishi Kumar: Yes, my father had it written. My father said, “His name is not good. In the future it will cause problems, give him trouble. In studying, in going other places, it will cause him difficulties.” So he renamed me according to his own thinking. There was an MLA, one Kanhaiya Lal Balmiki. He had a brother named Rishi Kumar. I was renamed after him.… My name was changed in 1955. It was when my name was recorded at school. When admissions happened, the form was filled, at the time of filling the form my name was written as Rishi Kumar.

Deceptive Majority 192

Rishi Kumar was not singled out for this treatment. His parents changed their names as well: My father’s original name was Rahim. Later, when we came here from the village, he took the name Ram Das.… My mother’s name was Hafiza, but afterward it was changed to Ganga … because it would be difficult for her to go to temples; when she would say “My name is Hafiza,” they wouldn’t easily let her enter the temple.

Between the late 1940s and the 1970s, members of the sanitation labor castes in Awadh ceased to use musalmānā names, and adopted hinduānā names in their place, in great numbers. The nominative practices of virtually the entire 583 clan transformed in this period. With regard to his mother, Rishi Kumar mentions the “difficulty” (dikkat) of entering Hindu temples bearing an apparently Muslim name. This is relatively rarely cited as a factor in transnomination. A more common context for adopting hinduānā names was the obtaining of government employment. In many accounts, women and men changed their names precisely when they applied for positions as sanitation workers in the municipality, at the university, or in government hospitals. Rishi Kumar hints at this when he remarks that Rahim became Ram Das “when we came here from the village.” For his parents, as for many of the 583s who urbanized in the middle decades of the twentieth century, leaving the village and joining government service were simultaneous events. Relatives or caste fellows who had migrated to Lucknow earlier and already held government jobs advised them to apply for work with hinduānā names. Having musalmānā names, especially in the years immediately following independence and Partition, was seen as a liability, a potential source of “difficulty” (dikkat) or “trouble” (pareśānī), for fourth class employees in government departments where anti-Muslim sentiment sometimes ran unchecked. Transnomination at the moment of securing government employment was, for a period, so widespread that it gave rise to the term “sarvis nām” (service name), the hinduānā name used for employment purposes, as distinct from “ghar kā nām” (home name), the often musalmānā name used among family and community. More than any other, though, it was the context of Rishi Kumar’s own renaming—school admission—that prevailed in my interlocutors’ accounts. The reformists consistently stressed the practical value of the adoption of hinduānā names—both given names/forenames and the surname Balmiki— in securing admission to government schools and exemption from school fees.


Hinduization and Its Discontents

Education, and the governmental financial assistance that brought education potentially within reach, were themes rarely absent from the appeals the reformists made to their caste fellows. As Master Lalta Prasad put it, their message had as its refrain the following: “Brothers, we people are not Lal Begi but Balmiki. We should write ‘Balmiki’ [as our surname]. And by doing this we will get exemptions from fees, our daughters and sons will get fee exemptions. If we write ‘Lal Begi’ then they will not get scholarships and fee exemptions.” He then described a particular incident from around 1960: I had this friend. There was a Gītā Vidyālay [a government-supported school in our neighborhood, run by prominent local Congressmen]. They had a column on the application for admission asking what your religion is. Isn’t it? So my friend wrote “Lal Begi.” He submitted the forms. Then [the admissions officer] said, “Kindly pay the full fees.” My friend replied, “Why?” So [the admissions officer] said, “You are Lal Begi, Muslim. Therefore you will have to pay full fees.” So he came and told me about it, and told Govind Prasad, and we advised him, “You’ve made a mistake, bring the forms to us, we’ll fill out a new one.” So the second form we filled out, and we wrote “Balmiki” and in brackets “Hindu.” Meaning Scheduled Caste.… Then he did not have to pay fees.

Awareness of the legal recognition of the category of Scheduled Caste, and of state programs designed to address caste-based inequality by creating educational and other opportunities for Scheduled Castes, arrived in the bastīs of Lucknow piecemeal in the years after the Constitution took effect. Lalta Prasad’s friend knew that under Congress rule, his caste status should entitle his children to a fee exemption in school, yet he remained unaware of the religious exclusions that restricted that entitlement. Having naively identified himself according to his community’s traditional self-description—as Lal Begi by religion—he found himself confronted by the new regime of recognition: “[N]o person who professes a religion different from the Hindu religion shall be deemed to be a member of a Scheduled Caste.” The admissions officer understood Lal Begis to be Muslim, or at least to be more akin to Muslims than to the other recognized religious communities, and accordingly declared the family ineligible for financial relief. Since even modest school fees were beyond the means of Lalta Prasad’s friend, as they were for most sanitation labor caste families, securing his children’s education meant formally relocating his family vis-à-vis the state’s regime of recognition. It meant becoming—or at least professing to be—Hindu.

Deceptive Majority 194

We met Gauhar Lal, born in 1939, in chapter 2, where he elaborated the folk theory of mimesis “as the king, so the people.” A retired accountant and one of the most highly educated of the shaharwāle of his generation, Gauhar Lal told the following story of this own experience accessing education at a government school in Lucknow: Gauhar Lal: I filled out the scholarship form for intermediate [eleventh and twelfth standard]. On the form, for religion, for caste I wrote Mehtar. And I was disqualified for scholarship. I didn’t get it. They said, “This is not allowed.” So the principal called me and said, “You’ve written Mehtar as your caste, why didn’t you write Balmiki?” I said, “What is Balmiki? I don’t know Balmiki.”6 He said, “This has been recognized, from the government this has been already recognized, that only Balmikis will obtain scholarships.” So I got a magistrate, an acquaintance, to make out a certificate saying, “Now I am Balmiki.” Understand? … I was eighteen. So when I got the Balmiki certificate, and again sent the form, then I got the scholarship.

I heard many such stories. Sometimes only the children’s surname or title was changed (from Lal Begi or Mehtar to Balmiki), other times the given name was changed as well. Not infrequently, the parents changed their names at the same time, in order to avoid scrutiny by admissions staff, or embarrassment to their children, on account of the mixture of musalmānā and hinduānā names in the same family. Ratan Lal “Sadhu” (born in 1948), for instance, told me: When I went to school, one of our community leaders said to me, “Your father’s name is Ali. This will not do, your classmates will laugh at you…. So in place of Ali, write Alok Nath.” … So then—my father’s real name [aslī nām] is Ali—then for school papers I wrote that my father’s name was Alok Nath. And to this day he’s known as Alok Nath.

From a strictly legal point of view, the advice that the reformists dispensed to their caste fellows was based on a half-truth. Balmiki was, to be sure, recognized as a Scheduled Caste in UP (and several other states), according to the Constitution (Scheduled Castes) Order of 1950. The reformists were thus quite right to emphasize the recognition that the title Balmiki promised. But Lal Begi, too, was (and remains) a Scheduled Caste; insofar as the state was concerned, Lal Begis were Hindu, belonged on the Schedule and were entitled to the same benefits as every other Scheduled Caste. The much-reiterated 6

What is italicized here marks what Gauhar Lal said in English.


Hinduization and Its Discontents

contention that identifying oneself as Lal Begi disqualified one from Scheduled Caste benefits was thus technically incorrect. Yet in practice, as is demonstrated in Lalta Prasad’s friend’s example and many others like it, the reformers’ advice was entirely sound. School principals and admissions officers—the local actors through whom the new regime of recognition was mediated—clearly were under the impression that Lal Begi was not a Scheduled Caste, such that nominative Hinduization proved to be very much the pragmatic means of accessing education that the reformists presented it to be.

A Tale of Two Cauldrons: The Meat Controversy There are degs and there are bhagaunās. Dictionary entries suggest that the two are similar in appearance and identical in function: both are cauldrons used in cooking large quantities of food over an open fire. Both can be wrought from a variety of metals, though bhagaunās today are usually aluminum whereas degs tend to be made of heavier metals like brass and copper. The only necessary structural distinction is that degs are rounded whereas bhagaunās are cylindrical with a flat bottom (see Figures 5.1 and 5.2). But with the penetration into everyday life of the bifurcation paradigm (Gottschalk 2013), the tendency to designate all South Asian phenomena according to a binary categorical schema of Hindu and Muslim, the deg and bhagaunā have become, in the bastīs of Lucknow and the villages of Awadh, mutually opposed indexes of communal allegiance. I was mystified, at first, by how frequently my interlocutors in Lucknow brought up degs and bhagaunās when describing the community’s adoption of Valmīkiyat. I had assumed that the primary material traces of the Hinduizing turn would be physical objects of worship like Lal Beg shrines and Valmiki temples. What surprised me was that it was cooking vessels, more than shrines, that people wanted to talk about. Several months into my stay in Lucknow I began to understand why. Nuptials and obsequies provide arguably the most important, and certainly the most frequent, occasions for caste sociability. The feasting of caste fellows that is de rigeur at weddings and death rites takes place, for a majority of the community, in the narrow lanes of the bastī itself, or on wasteland immediately adjacent to the bastī, or, in the village, on whatever land is vacant in the Mehtar quarter. The relatively recent practice of renting halls or lawns for weddings remains, for most, economically out of reach. While most members of the community in the city now live in pakkā homes—brick, rather than mud and thatch—kitchens are tiny. Professional catering, also for economic

Deceptive Majority 196

Figure 5.1 Deg Source: Drawing by Nikhil Mayur.

reasons, is unfeasible. Therefore the preparation of the feasts that accompany weddings and deaths generally occurs in the same place as the events themselves, starting two or three hours before the meal is to begin but often continuing concurrently, and is usually done by male relatives working with cauldrons balanced on bricks over an open wood fire. Around these steaming aromatic vats women and men gather, warming themselves if it is winter, helping or heckling the cooks. The site and process of food preparation, as much as the feast itself, creates a sensorially rich nucleus of sociability within the caste. By all accounts, this arrangement has been in place for a long time: longer than pakkā houses, longer than migration to the city. When reformists advocated a change in cauldron, then, they were seizing on a material centerpiece of community life enmeshed in a web of affective bonds formed over generations. As we will see, what was at stake was not only the kind of cooking vessel but what food it would contain, whether or not liquor would be served, and the entire form and content of nuptials, obsequies, and other life cycle rituals. In many of my interlocutors’ accounts, all of these objects of controversy concentrated in the sign of the deg and the bhagaunā.

197 Hinduization and Its Discontents

Figure 5.2 Bhagaunā Source: Drawing by Nikhil Mayur.

Of the shaharwāle reformers, it was Achhe Lal—the union leader and, eventually, MP—who led the charge against using degs and serving meat at weddings and other caste gatherings. As he put it to me, We raised our voice against it. About meat, about flesh, we said that it should not be made at weddings and such [śādī-vādī]. Because it incurs great expense. One point was that it is expensive, another was that it is just not a good thing. Some eat [meat], others don’t. So we said, cook vegetarian: pūrī, ālū, kachaurī, make dishes like this. They’re less expensive to make.… In our caste we used degs. We had to borrow them from Muslims, we had to rent them. Muslims used to make every kind of meat in these—buffalo meat was cooked in them, cow flesh too. Well our people used to eat all of this, a lot. So the group of us, we said, “When you borrow their [Muslims’] vessels, well— they throw everything in there!” We said, “Get bhagaunās—there are these aluminum bhagaunās—get them, and prepare food in the bhagaunās.” … Getting back to the deg controversy, veg or non-veg [kacchā–pakkā]. We said “Make kacchā [that is, vegetarian]!” We said, “These Muslim brothers, when they pray namāz, recite the fātiḥa [over the food], do you know what they’re saying? Brothers, if you want to recite the fātiḥa, do it yourself! You already do the cooking and pay for the whole thing yourself, so why not also fold your

Deceptive Majority 198

hands and pray whatever it is you want to pray? When Muslims do it, we don’t know in whose name the fātiḥa is recited.”

Achhe Lal here adduces a number of reasons for the reformers’ campaign against meat and degs: meat dishes are costlier to prepare; potentially not all guests eat meat (“Some eat, others don’t”); and degs are used by Muslims promiscuously (“they throw everything in there”) and to prepare beef. Importantly, for Achhe Lal the deg/meat controversy also includes the custom of Muslims blessing the food served at community functions by reciting the fātiḥa over it. This practice he impeaches by invoking the doctrine that prayers are best uttered by the person for whose benefit the prayers are intended and in his own language—a tenet of the Protestant semiotic ideology we will discuss in the next chapter. Here it can be briefly noted that Achhe Lal and his colleagues applied this doctrine unevenly: the Arabic spoken by Muslims invited to bless food at 583 weddings was rejected for being incomprehensible to its sponsors, while the Vedic Sanskrit recited by Arya Samajists at their ritual functions—attended and sometimes sponsored by the reformist triumvirate, all of whom joined the Arya Samaj but none of whom knew Sanskrit—escaped this criticism. That serving only meat at weddings produced exclusions was a problem that the reformers felt acutely. As Lalta Prasad put it, “We invited everyone to join us and eat. Muslims would eat, but Hindus would not eat, saying, ‘Buffalo meat is made in those [degs], and we are Hindu, so how can we eat?’ It wounded us deeply that guests would come to our door and leave without eating or drinking.” A great many of the reformists’ caste fellows, until recently, did not perceive this to be a problem at all. This was because in most sanitation labor caste weddings, almost nobody outside the caste, least of all a Hindu, was expected or imagined to attend—one’s wedding guests were precisely one’s caste fellows. This was, of course, a consequence of social structure; the homogeneity of wedding and funeral gatherings was, and to a considerable extent remains, an effect of untouchability. What was exceptional was the situation of the triumvirate and a few of their supporters, who from the 1950s onward nurtured friendships with fellow Arya Samajists, Congressmen, and other privileged caste Hindus and had occasion to invite such individuals to weddings in their families. The concern that other kinds of meat were cooked in degs was likewise peculiar to the reformers. Achhe Lal admits as much when he notes of beef and buffalo meat that “our people used to eat all of this, a lot.” That beef


Hinduization and Its Discontents

and buffalo meat should be negatively valued, and further that one should take interest in what is cooked in a vessel at other times, were elements of an ideology of purity and pollution that the reformists, channeling the concerns of their Arya Samaj and Congress patrons, had to introduce to, and inculcate in, their caste fellows. Lalta Prasad, in his telling, stressed that these ideas were part of a Hinduizing effort: What we preached [prachār karte the] was Hinduism [Hinduat]. Meaning: “You all are Hindu, you should do the circumambulations [a Hindu wedding rite], you should celebrate nuptials with Hindu rites and practices.” And the main issue was this, that we people used the degs of the Muslims, and among the Muslims, in degs, the Muslims cook buffalo meat. So we forbade this, saying “They make buffalo meat in these; stop using these [degs]. And don’t get Muslims to cook.”

Govind Prasad, on the other hand, emphasized economic reasons, above all else, for advocating reform in wedding fare. In his account, it was the need to liberate the caste from chronic indebtedness to moneylenders that motivated the reformers to promote vegetarian dishes in place of meat. In order to effect the change they desired, the reformists had to create an alternative to the degs that Muslim patrons loaned and rented out for sanitation labor caste weddings, funerals, and other events. Their first strategy was to rent out bhagaunās from Hindu merchants. Lalta Prasad recounted this venture: So we thought, why don’t we approach some Hindus, saying, “Kindly give us [that is, rent to us] some cooking vessels [bartan].” Then we’ll use those to cook so that everyone can eat. But the Hindus declined. They said, “You people [tum log] eat meat and such [mās-vās].” We approached a number of them: Sonal Chand Aggarwal had vessels, Kashi Lal the cloth merchant had some, several people, several Hindus, had vessels for rental. But they didn’t give them.

This having failed, the reformists created a bartan samiti—a cooperative society for cooking utensils—whose members all contributed funds with which bhagaunās and other cooking implements were purchased and made available for any member, or caste fellow for whom a member would vouch, to borrow under certain conditions. Lalta Prasad was voted president of the samiti, Govind Prasad the secretary. Their circle, which included the small but growing number of their caste fellows who had secured government

Deceptive Majority 200

employment in fields other than sanitation, were the samiti’s early members. Within a few years the samiti grew to have one hundred members (at which point membership was capped) and amassed sufficient bhagaunās to supply multiple clients on a single date. Their rates were competitive. But while increasing numbers of the shaharwāle made the switch from degs to bhagaunās and were persuaded of the benefits of at least some of the changes that the reformists championed, others resented the multipronged assault on tradition and drew a line at the attempt to transform wedding fare. Since reformists, by their own and others’ accounts, made weddings their primary occasions for the haranguing and entreaty of their caste fellows, it was only to be expected that weddings would become the rallying point for resistance to reform. In the 1950s, a series of quarrels between the reformist faction and traditionalists culminated in a major conflict, triggered by a wedding, in which the former brought a suit against the latter and both sides aired their grievances before Lucknow’s city magistrate. The few people old enough to remember it call it the Lal Begi–Balmiki muqadma—the “affair” or “case” over Lal Begi and Balmiki—a name it earned because, while its immediate cause was the clash over meat at weddings, it quickly assumed the proportions of a referendum on the entire Hinduizing project, a show of strength between the partisans of Lal Beg and the apostles of Valmiki, a reckoning over the future of the caste’s pahchān. Accounts vary as to how, precisely, the conflict erupted, though the basic contours are clear enough. A kinsman of one Amin Chaudhury hosted a wedding. This kinsman, who had recently come under the influence of the reformist group, broke with 583 tradition by neither inviting a Muslim to recite the fātiḥa at the wedding feast nor serving meat and liquor. Amin Chaudhury took issue with his kinsman over these innovations and the dispute turned rancorous: in some accounts the reform-minded host insulted Amin Chaudhury before the assembled guests, in others Amin Chaudhury and his coterie forced the kinsman to eat meat and drink liquor, and then publicly paraded him drunk in an attempt to discredit the Balmiki faction. Perhaps both occurred. What is unanimous in participants’ recounting is that the fight began over the question of serving meat at weddings. Achhe Lal, Lalta Prasad, and Govind Prasad were all named as litigants on the side of the reform-minded kinsman; opposing them were Amin Chaudhury and four or five other caste fellows, all somewhat older than the reformist group (and all long deceased by the time of my research). The members of

201 Hinduization and Its Discontents

both parties were 583s. The reformers took the counsel, again free of charge, of the same Arya Samajist lawyer who represented them in the cohabitation cases; Amin Chaudhury and his fellows had a lawyer remembered by some as Muslim, by others as a Hindu Kayasth. It was a civil rather than criminal suit and was heard and arbitrated by the city magistrate (or in one account his deputy). Those who remember the Lal Begi–Balmiki muqadma recall that each side argued for its interpretation of the caste’s religious standing by reference to ritual practice construed as custom. Operating under the long shadow of colonial anthropology and its taxonomic norms, the parties sought to establish an essential nature for the caste by reading ritual practice as evidence. The Arya Samajist lawyer thus posed questions like, “Do you consult pandits to determine auspicious wedding dates?” and “By what rites are your weddings conducted?”—questions that could have been lifted from the enumerator’s manual of the 1911 census. Answers to queries of this sort could not, even in this framework of judgment, definitively settle the matter because the caste was genuinely divided, the reformist camp having begun to consult pandits and adopt Hindu wedding rites like the seven circumambulations (sāt phere) while the traditionalists did not consult pandits and conducted wedding rites in a largely musalmānā fashion. Amin Chaudhury’s faction explained to the magistrate that Lal Beg was “our ancestral object of worship [purkhayātam pujātru]” to whom a cock sacrifice was customarily due at weddings, upon which the caste traditionally feasted on the sacrificed bird. The reformers acknowledged the practice, but rehearsed their objections to it as superstitious and uneconomical. The decisive blow to the Lal Begi faction came, according to some accounts at least, when the reformists, through their lawyer, revealed the content of a prenuptial ritual, Behervānī, that the 583s traditionally conducted clandestinely, “out in the jungle” (the literal meaning of Behervānī). This rite centered on the sacrifice of a piglet.7 In Achhe Lal’s recollection, in which the proceedings were imbued with considerable drama, it happened like this:


The sacrifice of a piglet to a local Dalit goddess is widely attested as a prenuptial ritual observed by the Chuhra caste across north India. In some regions, like western UP, this sacrifice is not concealed and the flesh of the sacrificial victim is consumed. In Awadh, by contrast, the sacrifice is performed secretly and the flesh is not eaten, but given away to another Dalit caste or simply discarded in the wild.

Deceptive Majority 202

So then [our lawyer] said [to Amin Chaudhury], “Pray tell me this: in these parts, do you people sacrifice swine? Prior to weddings, in an act of worship known as Behervani, do you people not sacrifice piglets?” And he replied, “Yes. … So the magistrate in his judgment declared, “These people cannot possibly be Muslim!”

Despite the implication of (Achhe Lal’s recollection of) this declaration by the magistrate, it is not clear that Amin Chaudhury and his group were arguing that the Lal Begis were Muslims per se. They evidently objected to being classed as Balmiki and Hindu, but whether they asserted a Muslim identity or simply presented themselves as Lal Begi is not certain. The reformists, though, clearly articulated, in a discourse comprehensible to the state, a position that the state was already predisposed to support—that the sanitation labor castes should be classed as Hindu. The magistrate ruled in the reformist group’s favor. This being a civil case, official consequences for Amin Chaudhury and his group were minimally damaging—they may have paid a fine. The caste as a whole made no immediate or unanimous change in response to the judgment; the conflict over meat and degs continued to simmer for decades at 583 weddings and funerals. But the ruling did demonstrate that the topography of pahchān had shifted, that longstanding habits of community self-perception and other-recognition were now at odds with power, that Valmiki afforded Dalits a certain purchase in the new regime that figures like Lal Beg did not. The judgment also tilted the landscape of authority within the caste by exhibiting the fluency of the purveyors of Hinduization, and the stammering of the older caste leadership, in the language of the postcolonial state. The ruling indicated where the future of the caste must lie, if it was to prosper in independent India. The members of the defeated party in the Lal Begi–Balmiki muqadma had, as I mentioned, all died years before I began my research. Eventually, though, I was able to locate the son of one of Amin Chaudhury’s fellow litigants, who was ten or twelve years old at the time of the case and recounted to me his memories of the controversy and its aftermath. His account, unsurprisingly, inverted major assumptions of the reformists’ narratives. In his recollection, for example, it was vegetarian wedding fare, which required multiple preparations and thus multiple bhagaunās, that was economically onerous, whereas the traditional single-dish meat preparation was affordable. His narrative of a series of obsequies provides a glimpse into how the traditionalists sought to retrench their increasingly beleaguered position in the years following the muqadma.

203 Hinduization and Its Discontents

My aunt and uncle said to us, “When we die, this is the food we want served, this one [Lal Beg] is who we want honored. When we are not here, you all [of the next generation] must make this food.” That is what they said. Well, those people objected, [saying,] “Do not prepare meat!” So when uncle died my aunt called all of us boys—she called Lallu, she called me …[,] the caste was there in numbers [for the obsequies] and she had us make biryani and so forth [meat dishes]. Then these leaders [netā log] made a hullabaloo that this was wrong, that we had put poison in the food, all kinds of things, and they wrote a report [that] everyone was dying [from the poison].8 So that no one would eat this food.

When his aunt died, he and his cousins, in accordance with her wishes, bore her corpse to the burial ground on a charpoy (the Lal Begi and musalmānā way), rather than on the disposable bamboo-and-cloth bier used by Hindus (titki). The reformers, he said, blocked kinsmen from participating in the funeral procession. When I asked him why the reformers objected to meat and charpoys, he noted yet another locus of conflict—pilgrimage to Sufi shrines, a practice important to many 583s and denounced by the advocates of Vālmīkiyat—and concluded with a theological reflection: I don’t know [why they objected]. Now here I am. I have always gone to Ajmer and such places [Sufi shrines]. I still go. That is the way we always have done things. Ram and Rahim are one and the same. What is separate? So, brother, now I even go to those places, to temples, to Bajrang Bali. He is Ali, after all. Whether you call Him Bajrang Bali or Ali, it’s the same thing. But those people say, “No. Do only Hindu ritual.”

His was a mournful narrative, a story of families divided, of death observances vitiated by controversy, of community lost. At only one point in our long conversation did his tone change to one of mirth. Having worked his way up from sweeper to clerk over a forty-year career at a government research institute, he retired some years ago with a pension. With the initial installment of his pension, he told me with a chuckle, he made a trip to Jaipur, purchased for himself a 3,500-rupee deg, and brought it home. “I make my daliya and everything in it. Uṛad dāl doesn’t boil down properly in a bhagaunā. But it cooks nicely in a deg.” 8

Govind Prasad, in a separate interview, confirmed that the reformists had indeed spread rumors of poisoning as a strategy in the campaign against meat.

Deceptive Majority 204

Beyond Internal Reform It is essential to bear in mind that the dissemination, contestation, and normalization of Vālmīkiyat within the caste took place concurrently with struggles by the sanitation labor castes to improve the conditions of their existence with respect to society at large. At the same time that reformers and traditionalists were clashing over what foods to serve at weddings, the caste as a whole was also engaged in a campaign to obtain decent housing, to overcome discrimination in schools, to improve dehumanizing labor conditions, and to combat the everyday humiliation of untouchability practices in public space. For our purposes, the struggle to reform society at large and its concurrence with the campaign of internal caste reform are essential for two reasons. First, many of the individuals who organized sanitation labor caste struggles against entrenched public forms of discrimination and exploitation in Lucknow were precisely the leaders of the internal caste reform movement. Their successes in the former domain earned them enormous credit among their caste fellows and buttressed their cause in the latter. Second, at the hands of these leaders and their Congress and Arya Samaj patrons, the campaign against casteism took on a communalist hue, giving substance to the notion that untouchable liberation lies with hardline Hinduization. We have discussed education in relation to naming practices: for a generation of 583s, musalmānā names were dropped and hinduānā names assumed at the threshold of the institutions of modern education. This is a telling statement about the Dalit experience of postcolonial Indian secularism. At the same time, taking a step back, it must be recognized that the sanitation labor castes’ attending school at all, under whatever conditions, constituted a revolutionary break with the past. When Balmiki reformers guided their brethren in the ways of the new regime, advising them to change their own and their children’s names in order to smooth the passage through school admissions and secure exemptions from prohibitive fees, they helped give rise to the first generation of the sanitation labor castes to be formally educated, bringing about what Shailaja Paik (2014) has described as a tectonic shift in the structure of Indian society and a change of immense practical benefit to the individuals and families involved. When Master Lalta Prasad, despite the handicap of near-blindness from eye injuries incurred as a railway worker, provided free, evening tutoring sessions to hundreds of youth from his caste over a span of decades, he contributed to a transformation of lives that earned him goodwill even among those wounded by his withering scorn for tradition.

205 Hinduization and Its Discontents

Consider housing. Until the early 1960s, in keeping with widespread patterns of environmental casteism, the Lal Begis of Lucknow lived in conditions both insecure and insalubrious (Lee 2017). As Govind Prasad put it, “We all lived next to latrines, defecating grounds [bampalus-vampalus], along open drains and sewers [nāle-vāle]. Some lived behind the mansions [where they worked]. Mostly along drains. It was filthy.” Homes were built of mud and whatever materials were available. In the great flood of 1960, the swollen Gomti wiped out two of the largest of these precarious Lal Begi settlements and a fire devastated another. The reformist group, emboldened by their links to power, organized their caste fellows to squat on a just-completed government housing scheme intended for others (now Paper Mill Colony), and then, making good on Achhe Lal’s leverage with municipal government and Kanhaiya Lal’s and Narain Din’s standing with then chief minister C. B. Gupta, got the arrangement regularized. Unlike the servants’ quarters and sewer-bank shantytowns to which urban Lal Begis had been hitherto confined, this settlement had pakkā houses, enclosed toilets, and other amenities. Following this success, the reformists orchestrated further canny advances onto government land and housing in other parts of the city, gains that translated into relatively secure and dignified living conditions for large numbers of their caste fellows between the 1960s and 1980s. Here again, the partisans of Valmiki delivered concrete, meaningful, and desired change, buttressing their status as caste leaders and reinforcing the idea that the politics of representation that they modeled, whatever else one might say about it, worked. Until the 1960s in Lucknow (and until the 1990s in parts of the surrounding districts), untouchability practices in public space were the norm. Lal Begis were prevented from taking water from public wells or taps themselves; they had, rather, to request “touchables” to pour water, from above, into their waiting vessels. Shopkeepers kept kaṭorīs (small, shallow bowls) of water at their counters with which to rinse coins offered by Dalit customers before putting them in the cash box. And so on. Article 17 of the Constitution, brought into effect in 1950, rendered such practices offenses against the law, yet they continued in Lucknow, as in most places, until challenged by politically organized Dalits. In the 1960s, the reformist triumvirate mobilized their caste fellows to stage a series of confrontations in Lucknow intended to eliminate two public forms of untouchability that they found particularly demeaning: exclusion from chai stalls and the denial of barbers’ services. Again with the counsel of the Arya Samaj lawyer, and now also with the active support of the

Deceptive Majority 206

Congress-appointed city magistrate, the reformers and their coterie in 1965 requested chai from a chai stall proprietor in Chowk, and when they were, as usual, denied, they filed a criminal case against him. After initial setbacks, the chai vendor was arrested; eventually a compromise was reached in which he opened his stall to Dalits and apologized and the criminal charges were dropped. The reformers next took action against a chai vendor in Aliganj, and this time the threat of arrest sufficed. Using similar tactics, the Balmiki leaders approached select Lucknow barbers, who, as a class, refused to shave or cut the hair of men of the sanitation labor castes. In Babuganj an Arya Samajist police inspector beat up a recalcitrant barber. As word of this and the arrest of the chai vendor circulated, the barbers of Lucknow began to comply with the law. In mid-twentieth century Lucknow, Hindus and Muslims both normatively practiced untouchability with the sanitation labor castes. Yet there were differences of degree. As we have seen, Muslims rented degs and other cooking utensils to Lal Begis without compunction, and ate when invited to Lal Begi weddings, whereas not even the teetotalling, vegetarian, janeu-wearing reformists could persuade Hindus to loan them bhagaunās or eat at their feasts. Where Hindu chai vendors uniformly denied service to the sanitation labor castes, Muslim chai vendors were split: some denied service, others offered service, though usually in a separate set of cups than those used for the general public. Achhe Lal, who played a leading role in the chai vendor and barber campaigns, held that with regard to untouchability practices in Lucknow in the 1960s, “Hindus practiced casteism much more, and Muslims did it less. [Hinduoṅ maiṅ to zyāda zāt pāt thī, Musalmānon maiṅ kam thī.]” That this distinction is drawn by Achhe Lal is significant, given his role in deciding which barbers and chai vendors would provide the staging ground for the legal confrontations over untouchability. The chai stalls in Chowk and Aliganj that the reformers selected for their action were owned and operated by Muslims; the proprietor arrested was one Siddiqi. Likewise it was on Muslim barbers that the reformers brought to bear the threat of police action. The anti-untouchability campaign that Lucknow Balmiki leaders led, justifiably confident of local Arya Samaj and Congress support, was waged as though only Muslims practiced untouchability. That the reformists correctly assessed the conditions under which they could rely upon Congress and Arya Samaj backing in improving their caste’s structural situation is supported by an example of what happened when those conditions were not met. A facet of the reformist movement on which we

207 Hinduization and Its Discontents

have touched only glancingly was debt. The advocates of Vālmīkiyat generally promoted economization—especially the reduction of expenditure on alcohol and on weddings and other community events—and sought to reduce the community’s crippling indebtedness to moneylenders. While critics accused some Balmiki leaders of hypocrisy in this domain, there was one reformer, Shiva Rao, whose commitment to the liberation of the sanitation labor castes from debt was, by all accounts, much more than rhetorical. Shiva Rao, a worldly bachelor who hailed from outside UP—some say he came from Maharashtra, others from Madhya Pradesh—was an intimate of Kanhaiya Lal (in whose home he lived), Narain Din, and the shaharwāle triumvirate. In addition to participating in many of the reform initiatives we have discussed, Shiva Rao also studied patterns of debt among his caste fellows in Lucknow, identified moneylenders whose exploitation of Balmikis could be demonstrated to be illegal, and filed cases against them. Shortly thereafter, in 1966, he was shot and killed by unknown gunmen in Paper Mill Colony. The presumed antagonists in this confrontation over the Balmikis’ standing in society, unlike in the untouchability cases or in the cohabitation controversy, were Hindu mahājans, moneylenders. The sanitation labor castes demanded justice, but their leaders were unable, this time, to summon outside aid. Their fellow Arya Samajists declined to assist. Narain Din (by this time ex-MP) and Kanhaiya Lal (still MLA) turned in vain to their Congress benefactors. The murder of Shiva Rao was left unsolved; no arrests were ever made. By the early 1970s, then, the limitations of Harijan politics, as well as its benefits, were manifest to the sanitation labor castes of Lucknow. Twenty years of vigorous reformist activity by the first generation of Congress Balmikis had shown what a program of political obedience, labor quiescence, and internal reform could accomplish, what it could not, and at what price. The efforts of Kanhaiya Lal, Narain Din, the shaharwāle triumvirate, and their like-minded comrades reaped the community undeniable gains in literacy and education, in housing, and in the curtailing of public forms of untouchability. The means by which these changes were achieved made clear that in postcolonial India’s political mainstream—that is, in the Congress—untouchable advancement hinged on Hinduization. As Ratan Lal “Sadhu” put it to me, “After Independence, we reaped great benefits from becoming Balmiki. The profit that accrued to us upon becoming Hindu [Hindu ban karke hamko jo fāydā milā], well, it was myriad: many kinds of [government] benefits and schemes, as well as political gains and economic improvement.” Hinduization, in turn,

Deceptive Majority 208

implied alienation from Muslims, whether in the agonistic mode of the chai vendor, barber, and cohabitation cases, which draped the dominant caste– Dalit axis of conflict in Muslim–Hindu garb, or in the cool severance of longstanding Dalit–Muslim ties, the quiet dismemberment of the economy of degs, fātiḥa reciters, wedding invitations, dargah patronage, and so on. The first two decades of independence demonstrated that the communalist substrate of the Congress’s de facto Harijan policy—Shraddhanand’s gift to Congress Balmikis, their means of circumventing Gandhi’s insistence on purely internal reform—was at least as resilient as Gandhi’s paradigm of the Harijan as a meek and grateful bhakta. One implication of all this was that for the generation whose names were changed upon school admission—the first generation of Lal Begis, that is, to be taught that they were Hindu—the elements of Vālmīkiyat appeared not as jumbled components of a risible ideology, as they had appeared to some of their parents, but as a stable constellation of ideas whose validity was confirmed by experience. That vegetarianism went hand in hand with progress, Valmiki with government jobs, and education with Hinduism, was for this generation less a contentious assertion than a social fact, reflected in the increasingly disparate fortunes of reformist and traditionalist families among the 583. Likewise, the signs Lal Beg, Islam, meat and (in ironic company with Islam) liquor were soldered, first by discursive and then by experiential repetition, to the objects backwardness, illiteracy, debt, and the humiliatory past. This is why today one often hears things like, “Our family is all educated, none of us eats meat.” Or, about a notorious drunkard, or a Bara Banki rustic who has yet to shed his Awadhi pronunciation, “You should ask him about Lal Beg.” This is why people’s utterance of the musalmānā names of their own parents or older siblings is so often accompanied by looks of embarrassment and nervous laughter.

Maze kī Bāt Let me conclude this chapter and this section of the book by suggesting, in the words of Govind Prasad, one direction that the third and final section will take. This may appear to run counter to much of what this chapter has argued; yet as we will see it is much more complicated than mere contradiction. As the preceding narrative should have made clear, Govind Prasad played no small role in the Hinduization of the sanitation labor castes of Lucknow, in the transformation of Lal Begis into Balmikis, of Anwar and Ayesha into Shankar


Hinduization and Its Discontents

and Shanti. As my account should also have suggested, Govind Prasad, in our many discussions in the last year of his life, expressed a profound ambivalence toward the reformist enterprise in which he was so deeply involved. One of his favorite themes, to which he returned again and again, was the history of missionary involvement with his community over the centuries. In his words, “there were three kinds of missionaries who approached us: Muslim, Christian, and Hindu.” He would then describe each with sympathy and a touch of critique: compassionate Sufis in the Mughal past, concerned British Protestants in the colonial era, and—the group he knew best, having been a card-carrying member—the helpful, if majoritarian and communalist, Arya Samajists. Once, after a discourse on this well-worn theme, Govind Prasad then lowered his voice to a gravelly whisper. “Maze kī bāt yeh hai,” he said, “ki hamne sab ko dhokhā diyā.” The interesting thing is this, that we deceived every one of them.

III Semiotics of the Oppressed

6 Victory to Valmiki Declamatory Religion and the Wages of Inclusion

Dates and Gestures I was walking along the main gully of a Balmiki neighborhood when a fiftyyear-old woman in a dark colored sari, sitting alone on a charpoy, gave me an impassive look, patted the empty space next to her, and said in a tone that would brook no refusal, “Baiṭho.” Sit. I sat. Soon we were joined by two more women of her age, one wielding a bamboo staff freshly whittled, at one end, to a sharp point. This sharpened staff was soon to be thrust into a tightly ringed bundle of dried and split date palm fronds to make a laggā jhāṛū, a long-handled broom. A short distance away sat the elderly Shyam Mehboob in his skullcap, sunning his arthritic knees in front of his house. None of the three women around me wore a bindī, burqā, sindūr, or other conventional markers of Hindu or Muslim femininity. The women asked me if I had any pull at the municipality—could I get them regularized, their status converted from contract laborers to permanent municipal employees? No one in their families, they explained, had pakkī naukrī—a permanent position—and only permanent employees enjoyed any kind of security. I replied honestly that I had no influence at the municipality, and asked whether their children were in a position to help support them financially. One woman had eight children, six of whom were alive. Do they live here, I asked, or have they moved away? She laughed at my question. “They’re all in England,” said the woman next to me on the charpoy. “They send her money from there,” added the woman with the bamboo staff. “Yes,” concluded the first woman, “and my husband lives in Saudi!” The three laughed at their joke and my naïveté. In fact, all six of the woman’s surviving children live in the bastī, and all work as sanitation laborers. Some young men wandered over and joined the conversation. It was autumn, a season of Hindu festivals as well as, this year, Muharram. There were

Deceptive Majority 214

also caste-specific events upcoming: Valmiki Jayanti, the commemoration of the birth of Rishi Valmiki, was only two weeks away, and shortly thereafter would follow Jamghaṭ, the day on which, according to a colonial source, Lal Beg used to be ritually venerated. Curious what my interlocutors might have to say about the various imminent festivals, I asked, “So, are there any tyauhārs (holidays, festivals) coming up?” Immediately one of the young men replied, “Yes, Valmiki Jayanti!” “When is it?” “On the first.” “No, friend,” interjected another young man. “It’s on the eleventh.” “It’s on the first of October,” insisted the first. “It’s on the eleventh,” repeated the second. “Check your calendar again.” While the young men argued over the holiday’s timing, I turned to the women, who had not yet said anything about tyauhārs. “Diwali is coming,” remarked the mother of six. The three women watched me—was I imagining it?—with a kind of evaluative and expectant air. What is it they wanted me to say? I wanted to ask if they had any memories of the festival of Lal Beg, but people usually dismissed such questions, telling me all of that had ceased long ago. “And the following day is … Jamghaṭ, yes?” I asked. “Yes,” said the woman with the bamboo staff, smiling slightly. “On that day–” She then switched semiotic tracks from speech to gesture. Raising her straightened right hand before her, she moved it through the air as though it were a knife sawing flesh. The pantomime lasted several beats and appeared, as far as she was concerned, to complete the syntagmatic chain she had begun verbally. Then she added, “Cocks will be cut. Liquor will flow.” Struck by the incongruity of her reply with everything I had heard about the at-least-half-century-old abandonment of Lal Beg, and also by her marked choice of gesture—an icon in the Peircian sense, a sign that conveyed by mimetic resemblance an Islamic manner of performing animal sacrifice—I asked a confused and half-formed question. “In every home?” “In every home,” answered the mother of six. “Cocks and liquor.”

Two Tyauhārs That the present is necessarily haunted by the past is an idea embedded in the north Indian vernacular: in everyday Hindi and Urdu, bhūt means both “the past” and “ghost.”

215 Victory to Valmiki

This and the next chapter are situated in the present, yet are pervaded by the bhūt of what came before. Memories of the Lal Begi–Balmiki muqadma, the forcible reclaiming of “Balmiki” women from Muslim men, and the bitter feuds over funerary and nuptial rites that roiled the community in the 1950s and 1960s, live on in minds, bodies, acts, and utterances in the Balmiki bastīs of Lucknow today. The unfinished project of majoritarian inclusion continues to both stall and advance, contending with old and new challenges by reconstituting methods from the past. In the aggressive new Hinduization efforts called ghar wāpsī (homecoming) by the hard-right Vishwa Hindu Parishad (VHP) and Rashtriya Swayamsevak Sangh (RSS), launched in 2014, the outlines of the last century’s shuddhi campaigns are unmistakably visible— phantoms returned to stalk the earth again. This and the next chapter are ethnographic explorations of the haunting of the present by the particular past this book has described. We will consider ways in which a century of struggles over the pahchān of the sanitation labor castes structures Dalit politics and religious life now, and the ways in which this history shapes intimate habits of speech, gesture, and comportment. We will examine the fruits of Hinduization, as well as the furtive persistence of older ways: what Govind Prasad, at the end of the last chapter, referred to when he said, [W]e deceived every one of them. In the conversation recounted above, the gesticulant middle-aged women and garrulous young men spoke to me of two tyauhārs—again, holidays or festivals—with strong structural parallels. Valmiki Jayanti is the annual celebration of Rishi Valmiki, the god, guru, or progenitor from whom the sanitation labor castes now overwhelmingly take their name and pahchān. Jamghaṭ is—or was—the annual propitiation of Lal Beg, the prophet, guru, or preceptor from whom the sanitation labor castes overwhelmingly took their name and pahchān in the colonial period. Yet the apparent homology between these annual ritual observances of the community’s past and present tutelar, held within days of each other in autumn, seemed to be in tension with the manner in which the two were discussed: Valmiki Jayanti rolled readily and volubly off the tongues of the young men, whereas the women practiced indirection, not uttering the name of Jamghaṭ themselves but speaking of the events of “that day” only after I named it. Different kinds of actions, moreover, were invoked with respect to the two anticipated tyauhārs. “Check your calendar again,” said the youth—a fitting exhortation for a ritual event whose temporal location has been a subject of debate. Organizers of Valmiki Jayanti have argued both sides of the question “Would it be more politically efficacious to fix the date according to the Gregorian calendar or the Hindu?”

Deceptive Majority 216

Meanwhile “Cocks will be cut, liquor will flow”—and the mimetic gesture preceding it—seemed to invoke efficacious action of a different order. On the one hand, a government holiday printed on a calendar; on the other, blood and intoxicants: these signs of the sacred are suggestive of the distinct semiotic ideologies in which the two tyauhārs are implicated. In this final section of the book, I offer a detailed account and interpretation of these two tyauhārs, ritual events thick with significance for Dalit religion and the majoritarian project that seeks to encompass it. In this chapter it is Valmiki Jayanti that I describe, narrating my experience of the annual festival in 2011. The analysis attends to the particular semiotic forms at the center of the day’s celebrations: the jhānkī—a public procession of youth costumed as Hindu heroes and deities—and the kāryakram—an associational form, often translated as a “function,” organized around speech-making. Alternating with description and exposition of these key features of Valmiki Jayanti I propose arguments of three kinds, each related to a different literature. One considers how acts of government recognition extended to the sanitation labor castes in Valmiki Jayanti illuminate the peculiar face of secularism that the postcolonial state shows to its Dalit citizens. Another tracks a failure of recognition, reflecting on what it means when performances of Hindu belonging are interpreted otherwise. A third addresses the assumptions about proper ways of signifying religious subjectivity that undergird the speech genres and ritual forms that make Valmiki Jayanti what it is.

Declamatory Religion and the Ethic of Publicity In 1886, the Reverend Andrew Gordon published Our India Mission, a firsthand account of thirty years of missionary work by American Presbyterians in Punjab. In it, Gordon describes in extensive detail the early years of the “mass movements” wherein Dalit communities—primarily Chuhras and Megs, glossed by missionaries as “aborigines” and “low tribes”—took to Christianity. Gordon devotes a great deal of attention to the initial converts in this “religious awakening” and the suffering they endure upon embracing Christianity. One such figure is Pipo, “the first man who stood up and openly declared himself a believer” in a village near Sialkot. This bold act leads to brutal consequences; the Muslim landlords for whom Pipo’s caste fellows work as agricultural laborers have Pipo beaten nearly to death and his family is denied access to wells and food from shops. (While acknowledging that the landlords are motivated by the apprehension that the Dalits ‘would cease to work for them on the Sabbath in

217 Victory to Valmiki

the event of their becoming Christians,” Gordon, in keeping with conventions of missionary writing critiqued by Viswanath (2014), portrays this as religious persecution rather than the control of Dalit labor.) When Gordon and his fellow missionaries make another journey to the village some months later, Pipo made us a visit secretly at night. This poor, lone, persecuted believer, Pipo, seemed greatly depressed by the bitter opposition of the Muhammadans, and especially by the fact that the rest of [his caste fellows] had yielded to it and left him alone. He asked us whether it was not possible for him to be a Christian in his heart only, without showing it outwardly. In reply to this, we taught him the absolute necessity of confessing Christ before the world, without doing which we could not expect Him to acknowledge us in the great day. (Gordon 1886: 205)

Later in the same text, Gordon writes of an early convert named Kanaya, who is not only physically assaulted but also excommunicated by his caste fellows— under pressure from the landlords—and prevented from seeing his wife and children. The missionaries with whom Kanaya takes refuge arrange a secret night meeting with Kanaya and his wife, Ramdei, outside the village. At the meeting the missionaries implore Ramdei to become Christian and rejoin her husband, to which she replies: “Let Kanaya come over to us, and if he must be a Christian, let him keep it to himself—shut up in his own heart—and not speak of it to others; he can then come [back to our village], and we will dwell together as in former days.” “This, Ramdei, would be impossible,” said [missionary] Scott—Kanaya emphatically affirming the same thing; “to do as you wish would be the same as to deny Christ; and by so doing we would prove ourselves utterly unworthy of Him.” (Gordon 1886: 258)

In these and other instances in his account, Gordon contrasts two ways of religious being-in-the-world. One, proposed by Dalits like Pipo and Ramdei, is for the subject to be Christian “in his heart only” while tactically concealing religious commitments from violence-purveying landlords and fearful caste fellows. The other, insisted upon by Protestant missionaries, is for the subject to “openly declare” his belief, “showing it outwardly” and “confessing Christ before the world” without regard for the consequences. Of course we are dependent here on Gordon’s representations of the words and actions of Pipo and Ramdei, representations structured in part by conventions of missiological narrative; but while we cannot know in what precise terms these historical

Deceptive Majority 218

Dalits made their case, the repeated occurrence in this and other missionary texts of the plea for a more clandestine Christianity suggests that requests of this sort were indeed made. What Gordon and his fellow missionaries advocate here belongs to a constellation of normative ideas of proper and improper modes of relating to the sacred and signifying religious subjectivity—a semiotic ideology—that Robert Yelle (2012) and Webb Keane (2007), among others, have identified as an unacknowledged, distinctly Protestant strain in the genealogy of global modernity. Initially articulated in Protestant diatribes against Catholics in the Reformation, this semiotic ideology was subsequently reproduced in polemics directed at Hindus, Muslims, Buddhists, and indeed whomever Protestant missionaries encountered in the mission field. The ideology regards with suspicion or hostility a host of semiotic forms—among them collective chanting, “vain” repetition in prayer, “empty” ritual gesture, ornate and ambiguous language—that tend, in this way of thinking, to misattribute agency to human action and deny the transcendence of God. In their place it valorizes such modes of signifying as plain, unambiguous speech and sincere, spontaneous prayer. As Keane and Yelle have demonstrated, the Protestant semiotic ideology privileges the written word over oral recitation, scripture over custom, prose over verse, the referential function of language over the pragmatic, and the dematerialized semantic content of signs over their sensuous materiality. To the condemnation of the latter element and the prescription of the former in each of these binary pairs, the purveyors of the ideology brought that spirit of militant iconoclasm and moral urgency in which the Protestant critique of Catholicism was borne. Religious communities across the globe responded to Protestant criticisms in part by emphasizing precisely those semiotic forms in their traditions that the ascendant ideology esteemed; that every “religion” came to be seen as possessing a “holy book” in the nineteenth century is but the most well-known aspect of this transfiguration of traditions in conformity with Protestant priorities. Meanwhile, apparently outside the religious sphere, the structure of the Protestant critique (and more than a little of its affect) featured centrally in the empiricism of Francis Bacon—whose avowedly iconoclastic assault on the “idols” of language was foundational for scientific discourse—and went on to be a staple of modernist critiques of tradition and, in a twist unanticipated by its pious early advocates, secular critiques of religion in general. Thus what originated as a set of normative claims mobilized in a factional fight among Christians became, without most

219 Victory to Valmiki

of us recognizing it, the prevailing semiotic ideology of modernity (Yelle 2012; Keane 2007). Ritual gesture is viewed with distrust in this ideology. Certain gestures, to be sure, retained significance in Protestant ritual practice throughout the centuries of criticism of “empty ritual” in Catholicism and beyond— kneeling for prayer, for example. On the whole, though, bodily movement in ritual suffered a tremendous devaluation from the Reformation onward, in keeping with the overall dematerialization and disembodiment of religious practice. Where it remained important, as in some ecstatic sects, biblical (that is, textual) precedents were found and justifications were made that upheld God’s sovereign agency—the use of ritual gesture as though it were efficacious in its own right was strictly for Papists and the heathen. Even the gestural sign of “making the cross” was subject to the accusation of superstition, if imagined to protect or bless the gesturer when not accompanied by sincere devotion in the heart. A seventeenth century Protestant guide to the faithful makes the case against gesture: “It much glorifieth God to worship him, according to the glory of his wisdom and goodness, and it dishonoureth him to be worshipped ignorantly and carnaly, with spells and mimical irrational actions, as if he were less wise than serious grave understanding men” (Baxter 1673, cited in Yelle 2012: 104). Importantly the kinds of gesture censured here are “mimical,” which is to say they are icons, representing what they represent by way of likeness or resemblance, in the manner of a mime (“making the cross” is an icon). Icons, bodied forth “carnaly,” correspond here to ignorance and irrationality and dishonor God; in contrast, what glorifies God, as we learn later in the same passage, are (relatively disembodied) words, that is, symbols rather than icons. In particular, the kind of words valorized here come in the form of “a well composed, serious, rational and reverent address … to God” (Baxter 1673, cited in Yelle 2012: 104). While this text is directed against Catholics, another seventeenth century polemic devalues ritual gesture by identifying it with the practices of brahmins. “These Bramanes as they discharge their Ministeriall function … straine their bodies into certaine mimicall gestures…. They must neuer reade of the booke deliuerred to Bremaw [Brahma], but it must be by a kinde of singing, and quauering of the voice” (Lord 1630, cited in Yelle 2012: 104). Here iconic gesture is aligned with distinctive styles of recitation of Sanskrit verse; these semiotic forms involve bodily “straine” and vocal “quauering,” qualities that throw into relief the gross corporeality of the action, a corporeality figured as inherently morally suspect (in addition

Deceptive Majority 220

to its smacking of chicanery) and contrasted with the solemn dignity of what Protestants often called “plain speech.” Secrecy, too, fares poorly in the Protestant ideology. It was a staple of Reformist polemic that the Catholic priesthood abused their congregations’ ignorance of Latin to cover up the Church’s false and self-serving interpretations of scripture. Echoing such critiques, no less a champion of the human sciences than Max Müller likened the brahmins’ exclusive control over Sanskrit religious knowledge to the Catholic clergy’s monopoly on Latin, and sought to expose the secrets thus concealed with the bright light of the printed, translated text (Yelle 2012: 76). But it was not only that recondite sacred languages harbored secrets; within vernacular tongues, as well, a whole range of speech genres and literary styles nurtured ambiguity and polyvalence, and thus the danger of falsehood and corruption. To cut through this thicket of deceptive language the fearless Protestant wielded the sword of direct, clear, univocal utterances— of “plain speech.” (In more secularized accounts, the wielder of this sword was the Müllerian Orientalist scholar, or the Baconian experimental scientist.) Protestants cited Second Corinthians 3:12–13 as the biblical mandate for their preferred linguistic style: “Seeing then that we have such hope, we use great plainness of speech: and not as Moses, which put a veil over his face” (cited in Yelle 2012: 23). Importantly, the clarity of Christian plain speech is contrasted with the “veil” of secretive language and the concealment of the sacred that Paul ascribes to Judaism. Here we glimpse a yet older stratum of the Protestant animus against secrets—one directed against Christianity’s own Jewish origins. Three of the Gospels have it that when Jesus died on the cross, the curtain (or “veil”) of the temple in Jerusalem was torn from top to bottom, exposing the holy of holies—heretofore hidden from the people—and baldly contradicting the idea implicit in temple ritual that what is sacred must also be, to some degree, secret. In the new dispensation, all was to be revealed. Foreshadowing the violence of Protestant iconoclasm, veils were now figured as objects to be torn. “For nothing is secret,” says Jesus in the Gospel of Luke, “that shall not be made manifest, neither any thing hid, that shall not be known and come abroad.” Elements of the Protestant ideology were thus quite ancient; yet it was not until the Reformation and the global spread of missions that it was articulated as a coherent set of valuations against which the semiotic repertoires of the entire world would come to be judged. What has been written about the global ascendance of the Protestant semiotic ideology has insightfully emphasized the path it prescribes from


Victory to Valmiki

formal, oral, and sensuously material ways of relating to the sacred—which become sinful (in overtly Protestant accounts) or childish and backward (in secular modernist accounts)—to a dematerialized, “spiritual,” and textual mode of being. What I wish to add to these observations is that, along with these other moves, purveyors of the Protestant ideology also charted a moral trajectory from secrecy to publicity, insisting on the renunciation of ambiguity and concealment and on the necessity, instead, of declaiming one’s faith in public, of “confessing Christ before the world,” as missionaries demanded of native converts in nineteenth century India. Protestant polemic linked secrecy in general to the devious obfuscations attributed to brahmins, the Jewish priesthood, and the Catholic clergy. Concealment of religious allegiance in particular was figured as a willful betrayal of Christ—a crime freighted with terrible moral weight in the stories of Peter and Judas in the Gospels. To escape the sinful conditions associated with secrecy, the proper Protestant subject was called upon to make unambiguous, public averments of faith: verbal utterances (and sometimes written texts) that were to be witnessed—whether by friendly co-religionists or hostile others—and consistent from context to context. In the collective recitation of the Credo in the church service, in the statement “I am Christian” before the census enumerator, university admissions officer, dominant caste landlord, or militant anti-Christian mob, and in announcements of religious allegiance in yet other contexts, the publicity of the declamations—their subjection to scrutiny from multiple angles—was to be a guarantee of their truth. Publicity itself, in Protestant narratives, began to accrue value as a potential sign of moral virtue. As with other entailments of the Protestant semiotic ideology, the incitement to declamation, with its abhorrence of secrecy and valorization of publicity, found its way into the practices not only of those who accepted the missionaries’ message but also of their detractors and opponents, and ultimately society at large. For instance, the missionary practice of public preaching in bazaars—a distinctively Protestant innovation that began to transform the character of South Asian public space in the mid-nineteenth century—was taken up by Muslims, Arya Samajists, and others who, while contradicting missionaries’ doctrinal claims, adopted the assumptions implicit in bazaar preaching as a semiotic form. Among these unspoken premises was the idea that one’s religious allegiance was a thing to proclaim verbally in public; another was that every individual, even among the riff-raff in the market, possessed such an allegiance and the capacity to alter it. Much of the

Deceptive Majority 222

capillary spread of the new norm, though, occurred not through the agency of missionaries at all, but through operations of colonial rule in which, arguably, Protestant conceptions were embedded. The colonial census, for example, now well-known as a provocation to identitarian thinking, compelled colonial subjects to make unambiguous spoken averments of religious allegiance in response to the question “What is your religion?” Silence or demurral was not to be countenanced (any more than was multiple affiliation)—the state demanded speech. Colonial subjects were required not only to have a religion (a single religion) but to declare it in a verbal formulation before witnesses (the enumerator as the state’s representative, as well as neighbors, onlookers, passers-by, and, in the prevailingly rural conditions of the colonial census, the village headman before whose home and under whose watchful eye census enumeration often occurred). The census, along with other state operations that raised the religion question, thus induced and ultimately helped normalize the kind of public speech act we are describing—a variation on the utterance “I am Christian” (“I am Hindu,” “I am Muslim,” and so on)—among a vast swath of the population, beyond the comparatively small circle of those directly touched by missionaries. The semiotic repertoire thus encouraged—a way of signifying one’s religious status through clear, univalent, public utterances— we might call the declamatory mode. Gandhi’s views on secrecy and publicity might be taken as an index of the degree to which the Protestant semiotic ideology, increasingly in secular garb, had taken hold in global thought by the early twentieth century. In Young India in 1925, Gandhi (quoted in Bailey 1991: 3) wrote, “Truthfulness is the masterkey. Do not lie under any circumstances whatsoever, keep nothing secret, take your teachers and your elders into your confidence and make a clean breast of everything to them.” From any other historical figure such moralizing might seem trite. In Gandhi’s case, though, the absolutist injunction against concealment (“keep nothing secret”) and in favor of moral transparency (“make a clean breast of everything”) conveys force due to its consonance with the extraordinary, perhaps historically unprecedented, publicity with which Gandhi’s everyday life was carried out. In letters, speeches, articles, newspaper interviews, and his autobiography, Gandhi wrote with often startling frankness about matters conventionally sequestered from public view: political disagreement among colleagues, as well as such intimate biographical details as diet, bowel movements, marital discord, and efforts to overcome sexual desire. Making a clean breast of everything—declaiming in words his beliefs,

223 Victory to Valmiki

struggles, and shortcomings—was his approach not only to his “teachers and … elders” but to the world at large. If political actors as a class were suspected of duplicity, Gandhi set himself apart by making all aspects of his life available for public inspection. William Mazzarella (2010: 2) observes that the Mahatma was “a pathbreaking innovator in techniques of mass publicity.” Part of what is so striking about Gandhi’s innovation is the moral value it grants to the transparent, unveiled, publicized life. Probity, here, is achieved by means of self-exposure—by means, indeed, of a more or less constant series of verbal utterances announcing one’s inner state.1 Gandhian practice forges, to a greater degree than even the most prolific missionary author of self-probing letters-from-the-field, an alignment of ethicality and publicity. The Mahatma’s semiotic style teaches that if the morally corrupt take refuge in secrecy, the good confess their truth before the world. Lest such ethical valuations of signifying practices appear rather too historically widespread and broadly accepted to be attached to particular ideological formations—who does not believe that “secrets and lies” go together, or that transparency encourages ethical behavior?—it may be helpful to recall contrasting examples. “Your faithfulness in secret is better than your faithfulness in the open,” states a verse composed, under conditions of political repression, by the sixth Shi’i Imam Jafar al-Sadiq in the eighth century CE. The verse continues, “and your worship in the reign of falsehood while your Imam is hidden and you are fearful of the enemy … is better than worship when the truth is manifest” (Clark 2005: 52–53). Or more simply, as another Shi’i tradition (hadith) has it, “God likes to be worshipped in secret” (Clark 2005: 48). In taqiyya, the Shi’i doctrine of tactical dissimulation, we find the articulation of a semiotic ideology that is in some ways a perfect inversion of that we have been describing. Here it is the secret that is valorized, legitimate, and aligned with the ethical, while publicity is viewed as morally suspect. Silence and demurral—and emphatically not unambiguous public declamations of religious allegiance—are accorded value; another Shi’i hadith maintains that the greater ultimate reward is stored up for those “who hold their tongues” (Clark 2005: 53). I raise the example of taqiyya—which will be considered at some length in the next chapter, in relation to the rites of Lal Beg—as it is 1

With respect to his manner of indefatigably publicizing intimate details of his life, Gandhi’s presaging of norms of twenty-first century electronic “social media” here is notable.

Deceptive Majority 224

both useful as a foil for the Protestant semiotic ideology and relevant to the cultural and historical context of the sanitation labor castes in north India. But one could consider a number of semiotic ideologies that throw into relief the provincial rather than universal genealogy of modern moral assessments of religious signification. The curtain in the Jewish temple, mentioned earlier, points to a multitude of traditions in which secrecy necessarily attends the sacred and in which exposure to public inspection (of sacred words, acts, or persons, outside of ritually delimited contexts), far from guaranteeing virtue, would bring about defilement or catastrophe. Esoteric traditions, of course, tend to endow secrecy as such with value, as do several types of social and political formation, from religious hierocracies to military intelligence networks to trade guilds to underground revolutionary movements. Within Christianity itself one can point to ideological currents pulling away from the declamatory mode of “confessing Christ before the world”—for example, the injunction in the Gospel of Matthew: “But whenever you pray, go into your room and shut the door and pray to your Father who is in secret; and your Father who sees in secret will reward you.” As with other targets of the Protestant semiotic ideology, though, traditions and social formations favoring a legitimate secrecy have on the whole seen their position eroded in modernity. Georg Simmel (1950: 336) observed the unprecedented shift in his time, noting that in the nineteenth century “publicity invaded the affairs of state to such an extent that, by now, governments officially publish facts without whose secrecy [earlier] no regime seemed even possible.” The precipitous decline of secret societies and their credibility over the twentieth century, and the degree of transparency—unthinkable in previous centuries—now expected of religious institutions by state and society are further indices of the taking root of a moral objection to secrecy as such. At the same time, as a host of scholars have noted, a subjectivity characterized by a reflexive, individuated awareness of possessing a singular religious “identity” and a readiness to present oneself in terms of it—arguably strange to much of human history—has become a global norm. This norm takes sustenance from the verbal declarations of unambiguous religious allegiance required of subjects by religious authorities and the modern state. It is another matter that what I am calling the declamatory mode has come to be contrasted not only with “veiled” languages and semiotic styles but also, in a further Protestant schism, with the kind of sober reserve often associated, in the present, with the more staid Protestant denominations, with middle


Victory to Valmiki

class religion in general, and with influential strands of secular liberalism. For advocates of this semiotic style, highly visible and audible public displays of religious identity are distasteful if not detrimental to society. But it must be observed that a plea for reserve is by no means a call for secrecy, ambiguity, or concealment, nor is it a challenge to the normativity of verbal declarations of religious allegiance before the state. The most non-demonstrative Protestants, along with the most committed secularists, still dutifully report their religion to the census taker. The volume, visibility, and relation to public space of markers of religious identity come up for objection, not the necessity for unambiguous religious self-identification as such (that is, self-identification in terms of singular, mutually exclusive religious categories, including “atheist,” “secularist”). Moreover, while certain forms of self-publicizing are disparaged as flamboyant or self-serving, the presumption that publicity (especially figured as “transparency”) works ultimately to ensure ethics is not fundamentally questioned by the advocates of reserve. We might designate this further distinction, then, as one between the declamatory mode of religious signification and the merely declarative. In tension with one another on some axes of comparison, the two are jointly rooted in Protestant criticism of Catholics and other religious others, and share a common disapprobation of secrecy. To note all of this is to observe that Protestant assessments of semiotic forms, while sharing with one another sufficient discursive overlap and historical effect as to warrant description as a coherent ideological formation, are by no means monolithic, and have generated and continue to generate new fractures, off-shoots, and hybrids (Keane 2007: 37–81). Nor were the semiotic ideologies targeted by Protestants in India monolithic. The Protestant missionaries who, like Andrew Gordon, found themselves wrangling with Dalit women and men in north India over whether a clandestine Christianity was Christianity at all were steadfast advocates of the declamatory mode, while their interlocutors apparently saw in secrecy a legitimate alternative. But the story I mean to tell in this and the following chapter does not, as so many accounts of the colonial or missionary encounter do, flatten the multipolar structure of relations that actually obtained in colonial India into a simple binary. I am indeed arguing, with Yelle (2012), that the missionary encounter in north India was (among other things) a collision of semiotic ideologies. But there were more parties to this clash than either missionary accounts, which tend to reduce the world to a dichotomy of Christian– heathen, or postcolonial accounts, which sometimes do the same with the dyad

Deceptive Majority 226

of colonizer–colonized, suggest. Punjab and the United Provinces in which Protestant missionaries and Dalits interacted in the nineteenth century were highly heterogeneous in terms of religious communities and normative ideas regarding religious signification. Dalits—and particularly the sanitation labor castes, whose distinctive forms of labor put them into more intimate domestic contact with elites than did other subordinated castes—were acquainted with the traditions of the Sunni, Shi’i, Rajput, Brahmin, and Sikh landowners for whom they worked. Lal Begi folklore makes evident that the sanitation labor castes were aware of how the religious communities among whom they lived related to the sacred and displayed religious belonging, and how these semiotic styles diverged from one another. Thus, despite the missionary tendency to lump everything foreign under the category of “heathen,” what the missionaries actually encountered was not a single or monolithic counterpoint to their own semiotic ideology, but several distinct ideologies, some of which shared principles with their own (Protestantism’s valorization of iconoclastic publicity, for instance, resembles that of normative Sunnism) while others diverged dramatically. In their encounter with Dalits like the Pipo and Ramdei of Gordon’s account, the missionaries described a semiotic ideology redolent of the Shi’i doctrine of taqiyya, a doctrine with which Dalits, especially in regions where Shi’i landlords and political institutions predominated, would have been familiar. To do justice to the historical and ethnographic evidence at hand, it is crucial to acknowledge that Dalits had the resources of multiple religious traditions on which to draw in formulating their position. They were not, as earlier generations of social scientists characterized them, “low caste Hindus” or tokens of “popular Hinduism” whose imaginations were obediently circumscribed to that tradition’s ideas, but rather intimate outsiders to the religious formations of all of the dominant groups, acquainted with their traditions but from a distance. Far from a dichotomous confrontation between Protestantism and Hinduism, then, the missionary encounter in Punjab and the United Provinces was a multipolar collision of socioreligious formations, the heterogeneity of which must be understood in order to make sense of how signifying practices were transformed. We will consider possible sources of Pipo and Ramdei’s ideas in the following chapter. Here, the focus is on the ideology of signification on display in the annual Valmiki festivals that have been gaining prominence in urban north India since they began in the 1950s. The semiotic ideology implicit in Valmiki Jayanti echoes what Protestant missionaries urged on their Dalit

227 Victory to Valmiki

interlocutors in the nineteenth century, and what Arya Samaj and Congress patrons of the sanitation labor castes, adopting and adapting the ideology for their own purposes, further encouraged. In the nineteenth century, the religious practices of the sanitation labor castes included neither processions through public space, living tableaus of caste members costumed as Hindu deities, public speeches with explicit averments of Hindu identity, nor the Vedic fire sacrifice (homā). Today, the putative god (bhagwān) or guru of the same community is celebrated with all of these. The point is not only the novelty of this assemblage of practices in the religious life of the sanitation labor castes—not only, that is, that Valmiki Jayanti celebrations were invented out of whole cloth in the mid-twentieth century. More fundamentally the point is that what underlies these collective ritual forms—the mode of relating to the sacred and signifying religious allegiance, the very way of religious beingin-the-world—bears no resemblance to that of the community’s religious past. The public ritual forms of Valmiki Jayanti and their rupture from earlier Dalit traditions are, I believe, symptomatic of a broader trajectory in religious ways of life in India, not only among the sanitation labor castes but across a whole spectrum of society. This transformation, closely linked to the politicization of Hinduism and what Sudipta Kaviraj (2010) describes as the shift from “thick” to “thin” religion, may be illuminated afresh if we acknowledge it as part of the ascendance, globally, of declamatory religion. The whole cluster of ritual celebrations of the new preceptor of the sanitation labor castes may be understood as a collective public declaration of Hindu identity, an act of confessing Valmiki before the world.

Valmiki on Parade: The Jhānkī Before giving an account of the Valmiki Jayanti processions of 2011, a brief recapitulation of the history of Valmiki Jayanti in Lucknow is in order. In 1950, Kanhaiya Lal Balmiki organized and hosted the first Valmiki Jayanti in Lucknow in a field near a sanitation labor colony in the Lucknow cantonment. The chief minister of UP, Govind Ballabh Pant, was the chief guest. From the cantonment the annual practice spread to the city, where the shaharwāle triumvirate of Achhe Lal, Govind Prasad, and “Master” Lalta Prasad organized the event. Initially it was held in a field at KGMC where many sanitation workers lived, and later in each of the major Lal Begi/Balmiki bastīs in the city. Since the late 1990s the various independent Valmiki Jayanti celebrations have come together at the Valmiki statue at Parivartan Chowk

Deceptive Majority 228

(Transformation Square) in the center of the city for a single function. The Nagar Nigam (municipal corporation) provides the bulk of the funds for Valmiki Jayanti—an early clue to the kind of secularism practiced by the municipal government. Valmiki Jayanti was declared a holiday in the 1980s, but it is a restricted—as opposed to gazetted—holiday, which is to say that government offices, schools, and so on, remain open, but government servants have the option to take the day off work without penalty. The date of Valmiki Jayanti has shifted over the decades; at times it has been a fixed date in the month of October, at other times a date fixed vis-à-vis the Indic lunar calendar. As a state holiday, the date of the Jayanti is śarad pūrṇimā, which ordinarily falls in October. In practice virtually no one observes Valmiki Jayanti outside of the sanitation labor castes and, to the limited extent they are compelled to, the municipal authorities for whom they work. Two semiotic forms, as I have said, dominate the events of Valmiki Jayanti in Lucknow. One is the jhāṅkī. The jhāṅkī (plural jhāṅkiyāṅ) is a procession of horses, chariots, musicians, and youth made up and dressed as heroes, villains, gods, and goddesses of the Hindu pantheon. This genre of sacred spectacle has both Vaishnava Hindu and plebeian roots; the name derives from the verb jhāṅknā—to cast a glance, to peep—and historically denoted both a pageant in honor of Krishna and a “peep-show” (McGregor 1993: 392; Platts 2004 [1884]: 401). What follows is a description of the jhāṅkiyāṅ I accompanied through Lucknow in 2011, lightly modified from fieldnotes.

* It is the full moon of the month of Śarad—the eleventh of October. Valmiki Jayanti begins with a havan ceremony at the Valmiki statue in Lal Bagh and a series of kāryakrams across the city. The long afternoon hours, though, are filled with the movement through the capillaries and arteries of the city of a dozen processions, one or two from each of the major Balmiki bastīs, toward the Valmiki statue at Parivartan Chowk, the central point of the map of contemporary Lucknow. At this great traffic circle, where the ancient poetsage, seated half-lotus on a concrete plinth, looks across the road to a midstride Subhas Chandra Bose, the processions will converge in the evening for a collective commemoration of Valmiki in spectacle, song, and speech. I go to one of the larger Balmiki bastīs to observe preparations, approaching by the back route, through Mirganj’s maze of alleyways. In the rear of the bastī


Victory to Valmiki

there are no signs of a festival underway. The few people I meet there have no plans to participate in the day’s Valmiki Jayanti celebrations. Saleeman, a sweeper at KGMC nearing retirement, is shelling peas with her neighbor Meena. Gesturing languidly toward the front of the bastī, Saleeman tells me that it is “[t]hose people over there [udhar ke log]” who do all the “pageants and processions [jhāṅkiyāṅ aur julūs].” At the front of the bastī, two jhāṅkiyāṅ are being organized. Each jhāṅkī is sponsored by an individual or small group, a man or men of the community who solicit contributions from their neighbors (and thus caste fellows, the residents of this bastī being almost all Balmiki) and also spend their own money. This year one jhāṅkī is sponsored by a BJP man and the other by a stalwart of the Congress. Their relationship, however, is neighborly, and while preparations for the two jhāṅkiyāṅ are conducted separately, their procession from the bastī to Parivartan Chowk is a joint affair, whose splendor is understood to reflect not on this or that faction but on the bastī as a whole. The jhāṅkī sponsored by the Congress man is running late. In the dimly lit main room of the sponsor’s house a hired make-up artist is painting the faces of neighborhood children—blue for Ram and Lakshman, white for Sita, red for Hanuman—while other children are still being dressed for their roles. The jhāṅkī sponsored by the BJP man, meanwhile, is already fully costumed, painted, and armed. In the main room of this sponsor’s home—one of the two air-conditioned rooms in the bastī—I find myself face to face with the entire cast of the Ramayan, most bearing weapons. The role players are local Balmiki youth, somewhat older than their prepubescent counterparts in the Congress jhāṅkī. A lanky twenty-year-old stands in a corner wearing a spiked helmet and holding a tall spear. He seems out of place iconographically, so I ask an organizer and am told that he is Rana Pratap—that is, the sixteenth century Rajput ruler of Mewar whose opposition to the Mughals has made him a favored symbol of Hindu militancy. I ask the young man about his role; he tells me he does not know who Rana Pratap was or when he lived. The youth playing Ravan asks me to take a photo of the whole ensemble, and I oblige. On the wall of the room, just above the heads of this lively Hindu tableau, hangs a framed photograph of the dargah of Moinuddin Chishti in Ajmer. In this community, I am reminded, even BJP supporters acknowledge the power of Muslim shrines. Two brass bands have arrived in the front of the bastī. Acquaintances tell me that they are apne log, our people, in the expansive sense often used by

Deceptive Majority 230

Balmikis to refer to other castes involved in sanitation. Virtually all of the brass bands in Lucknow belong to the Dhanuks, the other sanitation labor caste in the region. One of the bands, in loose formation in front of the Valmiki temple, breaks into a merry, anarchic march—word has come that the mayor’s convoy is arriving. While the Congress jhāṅkī scrambles to get ready, the BJP jhāṅkī files out of its sponsor’s home and crowds into the temple. Since the lanes of the bastī are too narrow for cars, the mayor—a poised young professor of commerce whose bright saffron kurta advertises his BJP allegiance—and his entourage of armed guards, local Balmiki leaders (netās), journalists and photographers arrive on foot. Escorted into the temple by the Balmiki leaders, the mayor poses for photographs: inclining prayerfully before the image of Rishi Valmiki, lifting a coconut for a simple puja at the Shiva lingam, standing with the decorated troupe of Ramayan characters. He departs again to another raucous brass serenade. The BJP jhāṅkī makes its final preparations. Horses are led before the temple for several of the characters to mount. Ten-headed Ravan, who has no equestrian experience, repeatedly tries and fails to mount his steed, to the great amusement of the crowd. Still the Congress jhāṅkī is not ready. Since they both represent the bastī, the two jhāṅkiyāṅ cannot depart without each other by general agreement. Two hours pass. The musicians bake in their uniforms. At a samosa stall I sip chai and take notes while Rana Pratap, leaning on his spear in the shade nearby, downs a packet of pān masālā. Slowly the afternoon heat loses its edge. Finally the Congress jhāṅkī is ready and the procession begins. A troupe of drummers—young men of the bastī—lead the way from the Balmiki bastī through the main street of the adjacent, mixed-caste but largely Mallah2 neighborhood spread along the bank of the Gomti. After the drummers follow brass bands interspersed with the BJP jhāṅkī, the decorated youth now sitting in horse-drawn chariots wrapped in gold and silver paper and bedecked with plastic bouquets. Ravan, now at ease on his mount, rides ahead, enjoying the villain’s role, waving his plastic sword and leering at bystanders lining the road. Next comes the chariot of characters associated with Valmiki’s hermitage: Sita, her sons Lav and Kush, and their white-bearded host and protector Rishi Valmiki, all ensconced in thrones flanked by golden lions. These youth hold 2

A “Backward” caste (in UP, although in other states they are a Scheduled Caste) traditionally associated with fishing and ferrying.

231 Victory to Valmiki

their somber-faced poses as though embodying the popular iconography of these figures in posters and statuary. The similarly statuesque heroes’ chariot— Ram, Lakshman, Hanuman and his vānar senā (monkey army)—follows, and after another brass band, a chariot bearing Shiva, Parvati, their son Ganesh, and his consort Lakshmi brings up the rear. After a short interval the Congress jhāṅkī follows, in similar order, except that their caravan also includes a chariot dedicated to the popular and peaceable twentieth century holy man, Shirdi Sai Baba. Perhaps this is the Congress sponsor’s answer to the BJP jhāṅkī’s militant Rana Pratap. Both sides of the road are lined with onlookers, silent and unsmiling. Women waiting for their turn at a public water pump turn and watch, middle class homeowners observe from their porches. I am reminded of other processions I have accompanied in other north Indian cities—Ambedkar Jayanti in Kanpur, marches in Patna and Delhi protesting violence against Dalit women—and the looks of cool antipathy with which those were met. But this procession, oriented as it is to the display of Hindu piety, is of a fundamentally different kind than those, and I am eager to learn what spectators think of it. Further along the route, in a low-rent commercial area of Daliganj, I leave my place among the processors to mingle with the street-side onlookers and hear their impressions of the jhāṅkī. At the doorway of a petty goods shop I ask the shop owner, who identifies himself as Baniya by caste,3 what the procession is all about. “It’s like this,” he tells me. “This is a Jayanti. Of the guru of these jamādārs.4 Balmiki. He was a jamādār. It’s a government holiday, actually.” At the intersection with the Daliganj Bridge the procession has to ascend a steep slope while simultaneously crossing one lane of heavy vehicular traffic and merging with another. A middle-aged woman with hennaed hair watches the exertions of the horses, the trepidation of the chariot passengers, and the risky, ad hoc efforts of Balmiki men to direct traffic. What is this [yeh kyā hai?], I ask her. “A procession [julūs] of the jamādārs.” “For what occasion [Kis khushī maiṅ]?” 3


An umbrella for several “Forward Caste” communities associated with commerce and notionally identical to the vaishya category of the classical four-fold varna system. Another name for “sweepers” (see chapter 1).

Deceptive Majority 232

“It’s his/their birth [Janam hai unkā].” “Whose?” “The jamādārs.” It’s their birth-festival day [Un kā janam divas hai].” The first couple of chariots make it over the slope and through the traffic. I join three men at a pān stall by the bridge and ask them, “What is this?” Without speaking, the first man makes the sweeping-a-broom gesture with his hand. “Broom-people [Jhāṛū-wāle],” he explains. “Jamādārs. They have some baba [old man, saint] … Balmiki. It’s his birth anniversary.” “Do you celebrate it?” “No, no one else celebrates it, only those people.” “Are you Hindu?” “No, I’m Muslim. But this fellow’s Hindu,” he says, motioning to one of his companions. I ask the companion: “Are you celebrating Valmiki Jayanti?” “No, we don’t celebrate it.” These and other comments from onlookers correspond with what I have heard from a wide pool of non-Balmiki interlocutors whom I have asked about Valmiki Jayanti in recent days—the procession is peculiar to “Mehtars,” “Harijans” and “broom-people”; though the deities displayed are unmistakably Hindu, no Hindus claim the parade as their own. Twilight falls as the processors manage to complete the merge at the Daliganj bridge in Lucknow’s rush hour. No sooner is the hurdle cleared than bright lights and brass ebullience from the direction of Daliganj announce that the jhāṅkī from the Vivek Nagar bastī is close behind. Our group slows, theirs accelerates, and soon the jhāṅkiyāṅ of the two bastīs fuse. The Vivek Nagar jhāṅkī is rather less grand than the one I am accompanying—instead of horsedrawn chariots, their youth sit on plastic chairs tied to plywood platforms secured to the seats of bicycle rickshaws. Behind each child stands poster art of the god or hero he or she represents. These youth, too, have been coached to maintain the pose associated with the deity’s popular iconography. As we cross the river and turn westward, our ranks swell: battered flatbed municipal trucks, commandeered for the evening, transport women, children, and men from the city’s Balmiki bastīs, while others come by motorcycle, bicycle, or foot. We arrive at Parivartan Chowk, the heavily trafficked roundabout that is the terminus for jhāṅkiyāṅ from bastīs all across Lucknow, and merge with a crush of horses, chariots, trucks, rickshaws, brass bands, costumed children, processors, and spectators. On all sides of the roundabout panḍāls have been erected, bamboo-and-cloth booths sponsored by associations of Balmiki men


Victory to Valmiki

whose volunteers distribute leaf-bowls of sweet steaming halwā, rice with chhole, ālū-pūrī, and fruit salad. Along with Balmikis, the city’s non-Balmiki poor throng the panḍāls. Against all odds, vehicular traffic continues to crawl through the multitude, making liberal use of the horn. Above, dozens of roadside hoardings and banners strung across the road announce Valmiki Jayanti with Hindi text, drawings of Rishi Valmiki, and photos of members of Balmiki voluntary associations. Multiple amplification systems—those of the jhāṅkiyāṅ as well as those of the more elaborate of the panḍāls—compete for sonic dominance. In this deafening soundscape the mood is cheerful; women and men mingle, socialize, laugh, and eat. The last procession to arrive is the much-anticipated Husseinabad jhāṅkī, with its impressive train of eight camels. Shortly after their arrival the final program of Valmiki Jayanti begins, a kāryakram held in the park adjacent to Parivartan Chowk, at which the mayor and prominent members of the Balmiki community honor one another and make speeches in praise of Rishi Valmiki (about which more anon). After about two hours of speeches the crowd melts away—on foot, on motorcycles, in shared rickshaws. As with festivals celebrated by other communities, so too at the end of Valmiki Jayanti the expanse of park and road where the event took place is all but carpeted in trampled, wet, food-soiled bowls of stitched leaves and crushed plastic water cups. My friend Anand, whose mother and uncles all work as municipal sweepers, gestures at the expanse of waste and says, “Tomorrow morning, who are the people who will be cleaning this all up?”

Genealogies of the Jhānkī: Spatial Strategies, Political Arrival, and the Limits of Recognition One of my contentions is that Valmiki Jayanti, in ways discontinuous with the sensibility of Dalit religion in earlier times, exemplifies the declamatory mode of religious being-in-the-world that the Protestant semiotic ideology, in both its religious and secularized guises, helped usher into predominance in modernity. This line of argument I will resume later in the chapter, after considering the kāryakram form. Here I would like to make another kind of interpretation. I propose that Valmiki Jayanti, taken as a whole, is a performance of what Lisa Mitchell (2014: 521) calls “political arrival”—that is, the successful wresting of recognition from the state—of the sanitation labor castes. It is a staging of the inclusion in the nation of the erstwhile excluded Balmikis, a ritual reenactment of the

Deceptive Majority 234

bargain struck by mid-twentieth century sanitation labor caste leaders with their Congress and Arya Samaj patrons. Yet as the foregoing account should also suggest, political arrival has its limits, and state recognition does not guarantee recognition by others—neighbors, ostensible co-religionists, society at large. As a semiotic form, the jhāṅkī draws together a tangle of ritual and political antecedents. One that has attracted scholarly attention in the UP context is the urban procession of politically organized Dalits, of which Ambedkar Jayanti is the paradigmatic case and Ravidas Jayanti and Mahaparinirvan Divas (Ambedkar’s death anniversary) are also examples (Beth 2005; Jaoul 2007; Lee 2008; Lynch 1969, 1981, 2012; for outside of UP see Patwardhan 2011; Rao 2009). Scholarship has treated these processions primarily as “strategies of popular assertion” that target regnant, hierarchical norms in the distribution and use of public space for disruption and overthrow (Jaoul 2007: 174). Thus Sarah Beth (2005: 400–01), building on Owen Lynch’s (1981) landmark analysis of the 1978 Ambedkar Jayanti in Agra and the riots that followed, describes such processions as a Dalit “infiltration” of zones of urban territory traditionally monopolized by the privileged castes, as an “invasion and reclaiming of the street” by subalterns asserting their rights as citizens. Beth helpfully distinguishes three stages, at once spatial and temporal, in the history of Dalit processions—an initial phase, confined to the space of the Dalit bastī; a confrontational phase in which Dalits process through privileged caste neighborhoods and often face retaliation; and the “civic” phase marked by state recognition and its concomitants: legitimate access to central public space, preauthorization of routes, police control, and containment of various sorts. In fact the latter two stages are not so neatly distinguishable, as Nicolas Jaoul’s (2007) rich account of Dalit processions in Kanpur and its hinterland— where even state-approved processions on government land are marked by low level violence, police intimidation, and constant standoffs between Dalit activists and local authorities—makes clear. In any case, what these accounts collectively argue is that Dalit processions constitute a novel assertion of equal rights of access to public space—not, importantly, in the capacity of laborers, in which capacity Dalits have long had access, but as citizens and co-owners— and that retaliation by the privileged castes signals resentment and resistance to this contestation of their spatial dominance. Valmiki Jayanti can be understood in these terms. For Dalits to march en masse through posh neighborhoods like Hazratganj and to arrest traffic at


Victory to Valmiki

Parivartan Chowk during Lucknow’s evening rush hour is certainly to assert rights through a politics of spatial occupation. The people of the city are indeed compelled to acknowledge—in a manner not seen before the advent of Dalit processions—the Balmikis’ legitimate right to occupy public space as a rights-bearing collective, as people who belong, rather than as menials whose presence in public space rests on their cleansing it of other people’s waste. Yet the grounds of the assertion of rights in Valmiki Jayanti, and the nature of the rights-bearing collective it conjures, are distinct from those of the Dalit processions on the Ambedkar Jayanti model. Whereas in Ambedkar Jayanti, demonstrators emphasize their status as Dalits and as citizens—that is to say, as people whose structural negativity can be converted into positive political content by democracy and struggle (Khare 1984: 2; Rao 2009: xii)—in Valmiki Jayanti, participants display themselves primarily as Hindus, as people who seek to convert their historical stigma into positive religious value. This brings into relief one of the most crucial, and underexplored, differences between Ambedkarite politics and the politics of Dalits patronized by Congress— “Harijan politics,” as we have been calling it—and the difference between their respective relationships with the state. From its inception, Valmiki Jayanti, in presenting the sanitation castes as Hindu bhaktās, conformed to the paradigm of untouchable inclusion in the nation favored by Gandhians and Arya Samajists in the Congress. For this reason, Valmiki Jayanti has enjoyed state recognition of one degree or another since 1950, whereas Ambedkar Jayanti has gained state recognition only after bloody struggle. That is, Ambedkar Jayanti was declared a state holiday in UP only after nine Dalits were killed, scores injured, and violence between Dalits on the one side and police and privileged castes on the other reached such a pitch that the army was called in to restore order in Agra (Lynch 1981). When mobilized under the organizing principle of Hindu pahchān, Dalits have found ready accommodation from the state—the chief minister appears as guest, Valmiki Jayanti is declared a public holiday, the municipal authority covers expenses. But when mobilized on the basis of a history of oppression, Dalits have attracted the surveillance and at times violence of the state. The Hindu basis of the spatial and recognitive claim asserted by the sanitation castes also explains why Valmiki Jayanti in Lucknow does not elicit the overt hostility that Ambedkar Jayanti processions routinely face from the caste Hindu public. Ambedkar Jayanti, and the autonomous Dalit politics it represents, remind the privileged castes of the slow erosion of their traditional

Deceptive Majority 236

grounds of dominance. Until the ascendancy of the Dalit-led Bahujan Samaj Party (BSP) to power in UP, this kind of procession was the target of assaults by bricks and stones (Lynch 1981: 1952) and country-made bombs (Jaoul 2007: 181); now it is more often greeted by subtler signs of dislike.5 In contrast, the spectators of Valmiki Jayanti, as we have seen, appear unagitated by the proceedings. Bystanders are quick to assert the irremediable otherness of the “broom-people” passing by with their “baba Balmiki,” but no one seems to find their jhāṅkī provocative; there are no physical attacks on Valmiki Jayanti parades. If the Dalit occupation of public space as an autonomous collective of rights-bearing citizens is seen as threatening by some caste Hindus, the Dalit occupation of the very same public space dressed up as a dependent subset of the Hindu body politic gives no cause for alarm. Perhaps, then, the more apt comparison for the Valmiki Jayanti jhāṅkī is not with Dalit processions of the Ambedkar Jayanti type, characteristic of the 1960s to the present, but rather with Ramlila and Holi processions of the 1920s and 1930s, Hindu claims on public space in which subordinate caste yet “touchable” laborers—what Nandini Gooptu (2001: 192) calls “the shudra labouring poor”—played significant roles for the first time. It was in these decades that a number of innovations, reflecting the activism of nationalist and Hindu majoritarian organizations and the popularization of their messages, were introduced to existing Hindu festivals, altering their semiotic form (Freitag 1989; Gooptu 2001: 213–34). Public processions featuring “very large crowds of people wielding staffs, flags, swords and other arms … imparting an aura of triumphant and aggressive expansionism to Hinduism” were foremost among these innovations, and came to be major aspects of Holi and Ramlila (Gooptu 2001: 191–92). The urban shudra poor—Kurmis, Ahirs, Gujjars, Khatiks, and other traditionally agriculturalist castes recently arrived in the towns of UP—performed in these new processions in numbers, attempting to impress and even outdo the Hindu commercial classes who were their economic overlords and the processions’ main sponsors (Freitag 1989: 231). “In order to assert themselves, and faced with social and economic marginalization as well as overt and growing upper-class disdain or prejudices, it became relevant for the shudra poor to claim their own pivotal role in the history of Hinduism” 5

The situation in Maharashtra is similar; for accounts of the lethal violence with which Ambedkarite processions have been met there, see, among many others, Anupama Rao (2009: ch. 5) and Anand Patwardhan (2011).

237 Victory to Valmiki

(Gooptu 2001: 196). The Arya Samaj’s myth of the Muslim invention of untouchability, modified for a shudra audience, provided a narrative through which the newly Hinduizing castes could do this. Arya Samajists circulated the idea that the debased status of virtually all subordinated castes was the result of Muslim oppression following these castes’ militant defense of Hinduism in the age of Islamic political expansion. Accepting the invitation to majoritarian militancy, the new participants in Hindu processions publicly displayed themselves as protectors of Hindu religion—declaiming their Hindu status, and embodying the declamation in spectacle. Thus, for example, Mallahs in 1924 formed a Khūnī Dal (Band of Blood) to defy, at any cost, the court injunction won by Muslims restraining the annual Ramlila procession from playing music before an Allahabad mosque (Gooptu 2001: 197, 232). The Valmiki Jayanti jhāṅkī echoes these processions of the 1920s and 1930s in significant ways. In terms of formal features, the jhāṅkī, like the Ramlila procession in particular, is a mobile pageant of armed characters from the Ramayan narrative through the streets of the city. The jhāṅkī also shares with the earlier generation of processions the deep involvement of the Arya Samaj. Relatedly, the Valmiki Jayanti jhāṅkī reveals the same operational logic of majoritarian inclusion: the vertical expansion of the boundaries of the imagined Hindu community in order to maximize a horizontal break with, and consolidate a majority opposed to, Muslims. As the frontier of Hindu society expanded “downward” over the twentieth century, what was newly thinkable in the 1920s for borderline untouchable castes like Mallahs and Khatiks became, by the first Valmiki Jayanti in Lucknow in 1950, imaginable even for the most abject of “untouchable” castes. In a delicate two-step with Hindu nationalists, both the shudra poor and, later, the sanitation castes, took up semiotic forms that publicly broadcast claims to an honorable Hindu status—despite, indeed in the face of, the collective lived experience of brahminical contempt. Yet this parallel, too, has its limitations. The shudra poor and Balmikis drew from an overlapping set of modern myths circulated by the Arya Samaj, but whereas the former asserted their claim to dignified Hindu status primarily by emphasizing their allegedly kshatriya past, the latter stressed instead their putative link with Valmiki. If the touchable poor were defenders of the faith, the Balmikis would be progenitors of the faith, those who sang the story of Ram before Ram was even born (among the miracles attributed to Valmiki is that he is said to have composed the Ramayan centuries before the events which it describes). The difference partly reflects the gulf between the traditional

Deceptive Majority 238

occupations of the two groups. In the 1920s and 1930s, agrarian shudras of UP could look back on a history of employment, albeit a sporadic one, in the temporary fighting forces of the preceding centuries and as armed retainers of local landlords (Gooptu 2001; Kolff 2002); the figure of the lāṭhī-wielding Ahir could thus be transposed from the village setting to the urban Holi parade while still resonating with both spectator and actor. The sanitation labor castes, by contrast, bring to the Valmiki Jayanti jhāṅkī their tradition of performance as musicians, especially drummers, in processional rituals, most frequently weddings. To link a tradition of processional musicianship with Rishi Valmiki requires imagination: Valmiki is known as poet (kavi) rather than a musician, and the distance between the poet and the musician in brahminical social ontology is vast. The music of ritual procession is particularly closely tied to untouchable and other lowstatus groups, such that to play processional music is almost always, from a brahminical perspective, to perform one’s own stigma (Booth 2005; Raheja 1988). To escape this quandary—that is, to elide the distinction between highstatus poet and low-status musician—both Hindu nationalists and Balmikis introduce the term Ram batānewālā, the announcer of Ram. Through this device Valmiki, with his metered verse, and Balmikis, with their drums and brass bands, are jostled into the same category. Musicians and militants generate different kinds of spectacles. While the laboring shudra poor of the early twentieth century and the sanitation labor castes of today both proclaim their Hindu-ness through iconic representations of the cast of the Ramayan, their performances resonate differently and produce different kinds of effects. In the 1920s and 1930s, the Muslims of UP’s towns correctly perceived, in the swelling of the army of Ram by the sword-wielding shudra poor, imminent danger (Gould 2005). As the landlord’s retainers brought their skills with weapons from the village to the urban religious procession, they embodied militancy, conveying a readiness to do violence by way of signs whose history was widely recognized. In Valmiki Jayanti, on the other hand, the display of arms by people known to be professional sweepers and musicians elicits neither the feeling of intimidation one might expect of minorities in response to majoritarian display, nor the signs of admiration one might expect from a gratified majority. Even the performers themselves seem uninterested in the militant potentiality of their roles; it is with buffoonery, not menace, that the Balmiki Ravan swings his aluminum sword through the air, while Shiva and Ganesh ignore their trident and axe. The grave


Victory to Valmiki

comportment of most of the decorated Balmiki youth suggests piety, if not weary obedience. Thus the muscular display of Hindu virility associated with the more communalized of Ramlila and Holi processions does not find quite the echo in Valmiki Jayanti that their formal parallelism and genealogical links might lead one to expect. All of this may help explain why the Valmiki Jayanti jhāṅkī, even as it succeeds in publicizing the political arrival of Balmikis, seems to stumble as a declamatory enactment of Hindu pahchān. If its audiences are both the state and the nonBalmiki public, only the former, it seems, is persuaded by the performance. Let us set aside the state for the moment. Where the organizers of Valmiki Jayanti, from Kanhaiya Lal Balmiki in 1950 to a whole class of Balmiki leaders in 2011, would like the processors to be seen and heard as Hindus, as descendants of the poet who discursively produced Ram Raj and in this sense founded Hinduism, what ordinary people of Lucknow appear to see and hear are jamādārs and jhāṛū-wāle, sweepers and broom-people, celebrating their caste-specific baba. The call for Hindu solidarity implicit in a procession of the characters of the Ramayan through the city is met not with Hindu solidarity, but with the categorical affirmation of difference. “No, we don’t celebrate it.” “[N]o one else celebrates it, only those people.” It is “a procession of the jamādārs.” The contrast here between reaction solicited and reaction achieved can be highlighted by comparing the Valmiki Jayanti jhāṅkī with another of its antecedents: the jhāṅkī in Rās Lila performances in Vrindavan. In these popular Hindu devotional dramas based on the stories of Krishna and the gopīs, the term jhāṅkī denotes a tableau vivant of divine and quasi-divine characters that punctuates and often concludes a performance. While the actors—who are all brahmin boys—hold their poses, audience members come on stage and perform acts of devotion before them, offering flowers, touching feet, and so on (Hawley 1981: 14–18). Yet in Valmiki Jayanti, similar displays of actors dressed up as iconographically unmistakable divine beings result not in devotional gestures but in assertions of alterity. In Peircian terms, we have in the jhāṅkī a sign or representamen whose interpretants—and, therefore, working backward, objects—depend to no small degree on the social location of the interpreter.6 Balmiki leaders clearly 6

As Kockelman (2006: 82–83, 2007: 378) observes, the Peircian semiotic object is often not a concrete thing but rather a kind of projection back from the various interpretants to which a sign gives rise.

Deceptive Majority 240

intend the jhāṅkī-sign to convey to other minds the idea-object “those people are Hindus,” yet every non-Balmiki observer I ask interprets the same jhāṅkīsign with the idea-object “those people are jamādārs/sweepers/Harijans/ broom-people.” The point is not that Valmiki Jayanti jhāṅkiyāṅ are polysemic, which of course they are, but that the different interpretants of the sign tend to part ways along caste divisions and to animate opposing ideologies. This is what Vološinov (1973: 23) calls the “social multiaccentuality of the ideological sign”—the capacity of the sign to incite divergent or even antipodal interpretants in different classes of people in a given sign community. Vološinov reminds us that polysemy, the open-endedness of semiosis, is a matter not simply of differences of interpretation, but of such differences structured by and instrumental to class struggle. What often goes unrecognized in the South Asian context is that channels of semiosis often diverge precisely along caste lines. Castes are not only endogamous groups, occupational categories, and so on, but also competing schools of interpretation within the broader sign community. As a bid to secure societal recognition of the sanitation labor castes as Hindus, the Valmiki Jayanti jhāṅkī falters and runs aground. The carefully constructed sign, despite its formal congruence with antecedents of devotional display that in other contexts elicit acts of Hindu solidarity or even reverence, is severed from its intended object as it leaves the bastī, crossing the line of touchability and being subject to the interpretive community of the street. The promises of Hindu majoritarianism notwithstanding, the implicit assertion that Balmikis are simply Hindus and should be seen as such has few takers.

Valmiki in Oratory: The Kāryakram The kāryakram is a form of associational practice often translated as “function,” in the sense of a “public ceremony; a large or formal social event; an organized social gathering” (OED Online 2020a). Perhaps its generality, or global familiarity, or inherent dullness has shielded the kāryakram from analytical attention. Yet its ubiquity in South Asian public life, and the importance attached to it by subaltern groups in particular, should give us pause. The banality of the kāryakram may help conceal the work that it does, as a form, to the content it mediates, and to the social collectives that make use of it. Like the English function, the kāryakram is a labile form, adaptable to suit a great variety of purposes: the school anniversary, the book release, the political party meeting, the NGO seminar. Meaning a sequence (kram) of actions


Victory to Valmiki

(kārya), kāryakram, taken literally, could describe virtually any occurrence. In practice, though, the kāryakram almost always denotes an associational form constituted by a distinct set of practices (speechmaking, honoring with garlands, applause), spatial arrangements (division of stage and audience, facing one another in chairs), speech registers (formal), technology (microphone, sound equipment), division of labor (presider, chief guest, secretary), and procedural norms (order of speakers, and so on). The presence of chairs is not incidental—people who squat or sit on the ground do not belong at a kāryakram. Unlike a mass rally or workers’ demonstration (associational forms that share other properties with the kāryakram) the kāryakram requires of its participants modes of comportment associated with urbanity and economic comfort. Though derived from the more or less secular associational life of bourgeois civil society, the kāryakram readily accommodates religious content. Praise of holy men and shouts of victory to gods are standard speech acts at Valmiki Jayanti kāryakrams, as we will see in a moment. In the following account of three Valmiki Jayanti functions in Lucknow in 2011, by drawing attention to the kāryakram’s semiotic form, I hope to unsettle the deceptive familiarity of the public “function” and suggest what assumptions may underlie the form itself.

* In Lal Bagh in central Lucknow, under a concrete roof in a traffic island circumscribed by a busy three-way intersection, sits a bronze statue of Rishi Valmiki erected in 1993. Right hand poised to write, the sage gazes contemplatively at the manuscript—presumably of the Ramayana—sitting on his lap. The centrality of the written text in this image seems apposite on the morning of Valmiki Jayanti, when the traffic circle is all but engulfed by banners and hoardings announcing in Hindi the date, time, location, organizers, and sponsors of the celebrations planned in honor of the rishi, along with messages such as “a hearty welcome to all of you devotees [āp sabhī bhaktoṅ kā hārdik swāgat hai].” Even more than images of Valmiki, it is Devanagari text that predominates in the decoration of the area. At about eight in the morning I arrive at the traffic circle and am greeted by about two dozen Balmiki men gathered at the base of the statue. These are core members of the Uttar Pradesh Balmiki Seva Samaj, one of dozens of

Deceptive Majority 242

voluntary associations sponsoring Jayanti events today. This group is convened by the charismatic Jiya Lal, an energetic, white-haired senior netā (leader) in the informal politics of the municipal sanitation department, who sits me down next to him. Nearby a group of qawwāls—performers of Sufi devotional music—lay out their instruments on a carpet. At the base of the Valmiki statue a pandit—a genuine pandit, I am told, brahmin by birth—dressed in pure white sets up a rusty metal basin at the base of the statue, neatly stacks kindling in it, and sets it aflame to inaugurate a havan, a ceremony in which oblations are poured into a fire. It is a rite with ancient Vedic roots, popularized in the twentieth century by the Arya Samaj. Jiya Lal and his comrades address the officiant respectfully as “pandit ji” and touch his feet, while he marks their foreheads and mine with red tilak. The pandit launches into the recitation of Sanskrit prayers, his voice electronically amplified and broadcast over the roar of traffic. I strain to follow his Sanskrit until, a few minutes later, he wraps up with the popular Hindi devotional song “Om Jai Jagdīś Hare,” followed by a clinching sequence of call-and-response victory cries. Victory be to Lord Vishnu! (Viṣṇu bhagwān kī jai ho!) Victory be to Lord Shiva! (Śiva bhagwān kī jai ho!) Victory be to Maharishi Valmiki! (Mahāriṣi vālmīki kī jai ho!) Victory be to all the Gods and Goddesses! (Sārī devī devtāoṅ kī jai ho!) Victory be to Mother India! (Bhārat mātā kī jai ho!) Victory be to the Eternal Religion [Hinduism]! (Sanātan dharm kī jai ho!)

After distributing prasād of fruit and sweetcakes, collecting his earnings and giving a smiling namaste to the qawwāls, the pandit departs. Taking his place, a middle-aged woman in a crisp pink sari, Jiya Lal’s sister Shyama, sits at the base of the Valmiki statue and begins applying tilak to newcomers. Strictly speaking the havan is now finished, but the rapidly growing crowd, which now includes women and children in numbers, seems more interested in the post-havan function— the kāryakram—that is just beginning. A podium is placed by the statue, and red plastic chairs are set up between it and the qawwāls seated on their carpet. Standing at the podium, Jiya Lal energetically congratulates his caste fellows for turning out for the event. Other men in his association follow, delivering short, formulaic felicitation speeches, usually concluding with shouts of victory to Valmiki. The qawwāls are invited to perform bhajans, Hindu devotional songs, and they oblige. These four musicians profuse the

243 Victory to Valmiki

somatic and sartorial signs of north Indian Muslim musicians—long hennaed hair, beaded necklaces, loose white kurta-pajamas with black vests and green scarves—and thus look quite like the qawwāls that perform in Sufi dargahs. Yet like several other qawwālī bands in Lucknow, the performers here at Lal Bagh are all Balmikis—a band based in the bastī where Jiya Lal lives. When a small motorcade pulls up to the traffic circle, Jiya Lal seizes the microphone to announce that the mayor—whom he calls “the messiah [masīhā] of Lucknow,” “our elder brother,” and “the savior of the Valmiki community”—has arrived. Accompanied by his armed escort, the young professor-politician emerges and coolly receives several dozen garlands of marigolds from the Balmiki men whom Jiya Lal calls in sequence to welcome him—Hari Balmiki, Ramesh Balmiki, Gopal Balmiki, the surname ringing like a refrain. Some of the men I know, and it occurs to me that in other social contexts they have introduced themselves not with the Balmiki surname, but with titles that index no obvious caste provenance. In today’s kāryakram, though, everyone is emphatically Balmiki. Followed by news photographers and television crews, the mayor ascends a steel stepladder and garlands the Valmiki statue with marigolds. He is then given a seat—next to me, to my surprise—while Jiya Lal at the podium continues his adulatory introduction. The pragmatic netā is not all praise, though, and takes the occasion to remind the mayor to fulfill his promise of getting the ugly electric transformer that looms over the Valmiki statue removed. He then calls the mayor to the podium to address the gathering. As the young politician rises he hands me the computer printout he was skimming during Jiya Lal’s speech—a brief article on Rishi Valmiki from the general knowledge Hindi website The mayor’s speech begins with the customarily solemn naming of dignitaries present: again, a catalog of Balmiki men. After outlining a number of projects beneficial to sanitation workers that his administration plans to implement, he turns to the subject of Rishi Valmiki: Regarding the Balmiki community [samāj] I always have some special remarks to make. If there had been no Maharishi Valmiki, then who, today, would speak of Lord Ram? And if there had been no Ram, Ravan would not have been checked. And if Ravan had not been checked, then vice, atrocity, and egoism [anāchār, atyāchār, ahaṃkār] would still pervade the earth. So if you consider it, if anyone has done the work of eradicating vice, atrocity and egoism from the earth, it was Maharishi Valmiki, through the medium

Deceptive Majority 244

of that perfect specimen of humanity Ram, who did it…. So, whenever I talk about the Balmiki community, I say, “The people of the Balmiki community are the most excellent [śreṣṭh] of all!” If caste is to be analyzed, well, the caste system shouldn’t exist, who knows why it was started. But looking at it from a different angle, in comparison with other people the Balmiki community is the most excellent.

In support of his assertion the mayor points out that thieves are not be found among the Balmikis, and the Balmikis are hard-working. This has ultimately to do with the fact that “you all are his [Valmiki’s] offspring [santān].” I am struck by the conformity of the mayor’s message to the major premises of the now eighty-year-old Arya Samajist tract Śrī Bālmīki Prakāś. The valorization of the Chuhra caste is accomplished precisely on the grounds of their putative genealogical link with Rishi Valmiki, whose value, in turn, derives entirely from his propagation of the story of Ram. By way of two rapid substitutions, the sanitation labor castes are invited to identify with Ram—the martial deity favored by north Indian Hindu nationalists (Fuller 2004)—and his project of “checking” Ravan and “eradicating” vice. The beingmost-excellent (sab se śreṣṭh honā) of the Balmikis rests on this identification with Ram. The mayor concludes his speech to more applause, more garlanding, and more photographs. A brass band that has arrived in the meantime erupts into ebullient heterophony as the mayor and his armed escort drive away.

* My next stop is the UP state headquarters of the Congress Party, set back among the breezy bungalows and leafy gardens of Mall Avenue. Here another Valmiki Jayanti celebration is underway. My friend Anand, whose brother-inlaw aspires to be the new face of Balmiki youth in the Congress, whisks me into the great white bungalow that houses the headquarters. In a high-ceilinged room about fifty people—men on the left, women on the right—sit on white cushions and mattresses while a minor Congress functionary stands facing them at a microphone in the middle. About two thirds of the people present are Balmikis whom I recognize. The speaker regrets that the intended chief guest of the function, Dr. Rita Bahuguna Joshi—president of the UP Congress Committee, professor of history and former mayor of Allahabad—cannot


Victory to Valmiki

attend on account of being called suddenly to Jhansi by no less a personage than Rahul Gandhi, General Secretary of the Indian National Congress. Joshi’s absence, the failure of any high functionary to put in an appearance, and disappointment at the event’s low turnout are the themes of this speaker and several more who follow him at the microphone. Then one speaker, a portly, tilaked Congressman addressed by all as Pandit ji, says the following: Our elders created the varna system…. They explained that the head is the brahmin. It’s the brahmin. The arms are the kshatriyas. The stomach is the vaishyas. And these feet are the shudras…. And Valmik, Valmik was called a shudra. So it is they whom I honor first! It is they whose feet I touch! If anyone on this earth is great, then it is the Balmiki community that is the greatest [Agar is dhartī par koi baṛā hai, to sab se baṛā Bālmik samāj hai]!

Scattered applause ensues. Pandit ji continues, saying that everyone, in infancy, helplessly produces mal—“dirt,” a common euphemism for excrement—and mothers clean this dirt. And this dirt makes the mother and father of every child a Balmiki! [more applause] They take such care, they clean and remove our filth [gandagī], and they do so swiftly. We should honor their commitment. They are our god [Ve hamārā devtā haiṅ].

I observe in myself a wave of incredulity that these Gandhian tropes of the 1930s are actually being put forth and applauded in a mixed-caste political gathering in 2011. The metaphor of the scavenger as society’s mother, lovingly removing its excrement and warranting, on those grounds, universal respect, is one Gandhi repeatedly invoked in the first half of the twentieth century. But that was the better part of a century ago, before the universal franchise, the spread of an Ambedkarite discourse on dignity and human rights, and the rise of a Dalit woman to the post of chief minister of UP, not to mention the legal abolition of manual scavenging by the Government of India. Gandhi’s defense of varṇāśramadharma and his sacralization of scavenging are something of an embarrassment for many admirers of Gandhi today. I find it remarkable that these ideas could remain valid currency in a forum like this in a political center like Lucknow. But perhaps what is remarkable is my naïveté. The Congressman goes on in this vein, and the applause continues. The next speaker, too, takes

Deceptive Majority 246

up the Balmikis-as-mother trope, and later a Congresswoman repeats the idea that we should honor Balmikis because they are the feet of society.

* On the roughly triangular strip of land between the outer wall of the city’s sports stadium, a liquor store, and the Parivartan Chowk traffic circle stands a statue of Rishi Valmiki, installed in 2000. In 2009 the strip was renamed Valmiki Park, and yesterday the municipality bestowed on its modest iron entrance the title Lav Kush Dwar (Lav Kush Gate), after the sons of Rama whose early years were spent in Valmiki’s hermitage. In this park, on a bamboo-and-plywood stage, after the convergence of all the Balmiki bastīs’ processions at Parivartan Chowk at sundown, the decorated youth of the various jhāṅkiyāṅ assemble into a collective tableau facing their relatives, friends, neighbors, and caste fellows taking their seats opposite on red plastic chairs. Munnu Balmiki, the union leader, stands on the stage calling prominent members of the Lucknow Balmiki community to join him. A band of Balmiki youth plays popular Hindi film songs. A girl in high school sings an older, more poignant Hindi film song. As the park fills to capacity the speeches begin. Most are by prominent Balmiki men. There are felicitations for Valmiki Jayanti, exhortations to educate the community’s children. Even Achhe Lal Balmiki, the octogenarian former MP who almost never attends public functions or leaves his home (see chapter 6), is present and delivers a brief speech. One of the few women invited on stage is Tara Balmiki, the social activist. In her speech she recites a poem of her own composition that raises critical, anti-caste themes that have not been aired at the day’s kāryakrams thus far. Its refrain is: Nahīṅ chāhie aisī saṅskriti jo hamko majbūr kareṅ Narak banāe merī zindagī insānoṅ se dūr kareṅ We do not need this culture—this culture that renders us powerless That makes my life hell; that separates person from person.

An Ambedkarite, Tara Balmiki ordinarily criticizes her caste fellows for having embraced Rishi Valmiki and warns them of the dangers of the Hindu majoritarian politics associated with the Valmiki movement. So I am more than a little astonished when she concludes her short speech with a victory cry to Maharishi Valmiki.


Victory to Valmiki

Munnu Balmiki invites the mayor, who has just arrived (his day, like mine, seems to be a series of Valmiki Jayanti kāryakrams), to honor the Balmiki youth dressed as the cast of the Ramayan. He obliges, garlanding a young man in Rishi Valmiki costume. An enormous, twenty-kilogram mega-garland is hauled onstage and held over the shoulders of the mayor. Media cameras flash. Munnu and others compel me to join the mayor under the garland. The mayor delivers a speech. I am made to deliver a speech. There are so many speeches that I lose count. All of it is amplified. As I will learn the next day, the victory cries to Rishi Valmiki were so loud they were heard by students at Lucknow University a kilometer away.

Declamatory Hinduism When asked to give a speech after the mayor, I acquiesced with the intention of making gracious but brief remarks. Standing before the microphone, though, I found myself suddenly voluble; I spoke longer than I meant to, revealing more of myself than intended, and I noticed myself repeating phrases and replicating cadences of speakers who had preceded me. Why was this? And what brought Tara Balmiki, consistent critic of her community’s embrace of Valmiki, to close her oration with a cry of “Victory to Valmiki”? Why did men who ordinarily introduce themselves as Hari Kumar and Raju Chaudhury become, on this day, Hari Balmiki and Raju Balmiki—and that, too, broadcast over loudspeakers and boldly printed on banners and hoardings? It is a foundational insight of linguistic anthropology that, as humans, the way we speak is governed by the conventions of the speech genres through which our daily discourse moves. To be a socially competent adult is to modulate one’s utterances—often profoundly—to suit the unspoken requirements of context; how we speak to the traffic cop who has pulled us over for speeding necessarily and consequentially differs from how we order a drink, deliver a sermon, banter with colleagues, or murmur in our lover’s ear. To a greater degree than many of us would like to admit, it is the genre, not the person, that speaks. Called to the stage, participants in Valmiki Jayanti, myself included, adopted with more or less volition the formal register and high lexicon of the speech genre native to the kāryakram form; we found ourselves not merely speaking but speechifying. It is surely one of the most distinctive features of the kāryakram that it turns ordinary people into orators, usually orators with a relatively fixed and stereotyped repertoire. At the microphone I felt the genre exercising its subtle force on me as I became long-winded and drawn to bookish,

Deceptive Majority 248

Sanskritized word choices, especially those that sounded in speeches before mine. Something similar seemed to happen to Tara Balmiki. “Victory to Rishi Valmiki” was not only inscribed on banners and hoardings across the city but also reverberated through the soundscape around us; it brought to a decisive, optimistic, and rhetorically satisfying conclusion the speeches of virtually every speaker of the evening. Rather than an abandonment of her critical position, then, Tara Balmiki’s victory cry signaled a kind of acquiescence to the rhythms of the speech genre, a yielding to the soundscape. She and I were in the grip, as it were, of the kāryakram—under the influence of the semiotic form, with its characteristic oratorical demands. The influence of the form flows deeper than oratorical style. Consider the orientation of the kāryakram to mass-mediated publicity. The television, print, and online news media were invited and present at each one of the Valmiki Jayanti functions: at Lal Bagh, Congress headquarters, and Parivartan Chowk. Photographs were staged and speeches delivered with an eye to how they would appear in the newspaper the next day. The organizers knew that Valmiki Jayanti was unlikely to excite huge media interest—no one was surprised that the Hindi papers relegated the story to the third or fourth page of local news, while Urdu and English papers did not report it—but the degree of coverage was less important than the fact of coverage. What mattered was that the day’s events be broadcast to an audience beyond those immediately present—that the function be presented before the news-consuming public. This is characteristic of the kāryakram as a semiotic form. A kāryakram without media is not complete; the performance must be at least potentially available for general inspection. Charles Taylor draws a distinction between the “topical common space” in which face-to-face communities routinely gather—the locus of the religious service, the theater performance, the school assembly—and the “metatopical common space” in which a public, that imagined community whose members are unlikely ever to meet most of their fellows, and which constitutes itself through the reflexive circulation of discourse, carries out its peculiar existence (Anderson 1991; Taylor 2007; Warner 2002). In the kāryakram we find a performance in a topical common space that self-consciously directs itself to a metatopical common space. Through the kāryakram form, face-to-face communities lay claim not only to the traffic circle or conference hall where the event occurs, but also to television airtime and space on the newspaper page. These remain the signature venues in which various literate publics, as


Victory to Valmiki

well as the nation—that is, the national public, paradigm for all other publics— narrate and imagine themselves. The kāryakram, I would suggest, is a form through which face-to-face communities perform their membership in these broader publics: civil society, the middle class, the nation. This accounts for the high stakes organizers attach to the potential evaluative judgments of others, those not in the face-to-face community: the acute self-consciousness over what might seem trivial matters of procedural correctness, order of speakers, and so on, and why questions like “How did you like our function?” and “Do you think we pulled it off?” proliferate after a kāryakram. This may also help explain why the competence to participate in a kāryakram appropriately— knowledge of proper comportment, where to sit, how to interact with a television news crew, all of which requires prior experience—often seems to function as a badge of middle class status and “arrival” in civil society. The competence required in such performances of membership is thus that of representation in the modes encouraged and modeled by the broader publics whose affirmation is sought. Publicity is the purpose. Like the jhāṅkī, the kāryakram as semiotic form is fundamentally oriented toward publicity through self-display. But whereas the jhāṅkī is directed at the topical common space of the neighborhood and city center, the kāryakram seeks also to claim territory in the metatopical common space of the nation. It is also a more textual form; if the jhāṅkī says “We, too, are Hindus” through visual image, the kāryakram elaborates the message in textual image (the writing on banners, hoardings) and, above all, speech. The sacred, in the kāryakram, is not to be found in gesture or in silence. The early morning havan at Lal Bagh, to be sure, retains the meaningful gestures of oblations into the fire; the pandit’s movements here are understood to have sacred significance. To some degree, the postures assumed by Balmiki youth in the jhāṅkī might also be seen as the sacred embodied in gesture. These matter, but it should be recalled that these are relatively recent additions to Valmiki Jayanti celebrations. The kāryakram—the oldest and most consistent manifestation of Valmiki Jayanti, the form it took in 1950 when Govind Ballabh Pant was the chief guest and has retained ever since—signifies the sacred not through gesture but emphatically through words. The name of the guru and the names of gods are to be written and displayed for all to see, and uttered and amplified for all to hear. The sacred, in the semiotic ideology that seems to animate the kāryakram form, must be textual and must be publicly declaimed. In this, and in Valmiki Jayanti as a whole, we may discern echoes of

Deceptive Majority 250

the exhortations of Protestant missionaries like Andrew Gordon that Dalits, and indeed anyone wishing to be free from the morally perilous conditions ascribed to “veiled” religions and traditions of secrecy, must publish the glad tidings of their religious allegiance, “confessing Christ before the world” in clear and unambiguous speech.

* In discussing the jhāṅkī I argued that, as an effort to win social recognition as Hindus, Valmiki Jayanti runs aground on the shoals of enduring caste prejudice. On the other hand, as an enactment of political arrival, the Valmiki Jayanti kāryakram is comparatively successful. It is true that the function held at the Congress headquarters proved, for those among Lucknow Balmikis who lay their hopes with the Congress, an embarrassing exposure of the low priority the party leadership accords the community (this was equally embarrassingly evident in the Congress’s Valmiki Jayanti kāryakram in 2019). But the kāryakrams in the morning at Lal Bagh and the evening at Parivartan Chowk succeeded by any number of criteria as demonstrations of the political influence and acumen of a circle of Balmiki netās. The kāryakrams dramatized that Balmiki leaders can command resources of the municipal authority; obtain from the mayor of the city his time, attention, and specific promises to address specific grievances; secure television and newspaper coverage of their events in which they and their supporters are named. At the evening function at Parivartan Chowk, even as the BJP mayor was giving a speech inside the park, local leaders of the rival Samajwadi Party (SP) and Congress put in appearances at the panḍāls just outside the park. They were eager to meet Balmikis, to shake hands, to be seen at Valmiki Jayanti. The state’s recognition of Balmikis as a political constituency worth patronizing, and its leaders as political agents, finds expression in Valmiki Jayanti. Political arrival, though, is a matter of degree (Mitchell 2014: 521). The kind of agency exercised by Balmiki netās on Valmiki Jayanti is shaped and circumscribed in particular ways by six decades of Harijan politics. If the first generation of Congress Balmikis adopted, arguably by necessity, a program of political obedience, labor quiescence, and internal reform, the political practices formed in that generation have not entirely disappeared. This is especially apparent in the case of labor quiescence.

251 Victory to Valmiki

Two days before Valmiki Jayanti, three municipal sanitation workers died of gas inhalation while cleaning Lucknow’s sewers. Almost all sewer cleaners in Lucknow, as elsewhere in India, are Dalit; not atypically, the three deaths in Lucknow, beyond a short notice in the local section of the newspaper, provoked little public reaction (cf. S. Anand 2007; Ramaswamy 2005). Over the course of Valmiki Jayanti, in the functions at Lal Bagh, at Congress headquarters, and at Parivartan Chowk, Balmiki netās made demands of the municipal government: Valmiki Park should be beautified, its new Lav Kush Gate should be repainted, the electric transformer that looms over the Valmiki statue in Lal Bagh should be relocated. But at the end of the day, reflecting on all the speeches I had heard, I realized that not a single speaker had mentioned the sanitation workers who died in the sewer two days earlier. On matters of life, death, and livelihood—the avoidable death of sanitation workers, the routine disavowal of workers who die on the job by the municipal authority, and the privatization of the labor sector in which the majority of the community is employed—nothing had been said.

Figure 7.1 Sohan Lal (who goes by another name in this book) Source: Portrait by Sanya Darapuri

7 Lal Beg Underground Taqiyya, Ethical Secrecy, and the Pleasure of Dissimulation

Another Time, Another Place Lal Beg appears to be dead. That is, a multitude of signs are available that elicit the interpretant “Lal Beg is dead. The tradition of Dalit religious autonomy he represented is a forgotten thing of the past.” Such signs include scholarly exposition. In his book Bālmīki Jayantī aur Bhangī Jātiyāṅ, Bhagwan Das (2007: 18–24) provides a valuable account of the Lal Begi religion. Das’s account is entirely in the past habitual tense—it describes what the sanitation labor castes used to do. In his social history of the sanitation labor castes in Delhi, Vijay Prashad learns of a mid-twentieth century text written by a Dalit activist who reclaims the spirit of the oral traditions of the nineteenth century, valorizing Lal Beg alongside the radical Chandal Balmik. Prashad (2000: 99) notes that the last copy of the text was lost in the 1960s, and proposes that we understand “that loss as a metaphor for the disappearance of the tradition of the radical Bala Shah [that is, Lal Beg] who boldly castigated the powerful.” Other signs of the prophet’s death include the complete absence of the term “Lal Begi” in public fora on the problems faced by the community formerly known by that name. At rallies, marches, seminars, and conferences addressing manual scavenging and caste discrimination against sanitation workers, one hears nothing of the antinomian prophet or of the sanitation labor castes as his followers (or even ex-followers). Blank looks and expressions of non-comprehension are also such signs. In my years working with Dalit advocacy organizations I sometimes asked co-workers about this figure about whom I was reading in the colonial archive. “Lal who?” was often the response. When I elaborated, my colleagues in the Dalit rights movement cautioned me that Balmikis, far from retaining any tradition of religious difference, were on the whole

Deceptive Majority 254

enthusiastic exponents of Hinduism; Ambedkar’s critical Buddhism, they told me, was largely lost on the sanitation labor castes, who had fallen for the seductions of majoritarian inclusion—“fascinating Hindutva,” in Badri’s Narayan’s (2009) evocative phrase—harder than had other major Dalit castes. If they had produced some red-clad saint or Sufi back in the Mughal period, there was no sign of him now. Secrecy is a semiotic strategy. One of its operations is the temporal and spatial displacement of the secret, a displacement capable of endless regress. Jeanne Favret-Saada demonstrates this in her study of witchcraft in rural France in the 1970s, a landmark in the ethnography of dissimulation. To the researcher, peasants in the Bocage speak of witchcraft as always located in “the old days,” an averment that many a journalist and folklorist takes “as the basis for firm statements about the forthcoming disappearance of witchcraft (and this has been going on for more than a century) without thinking that a discourse on the past is perhaps not quite the same as a past event” (FavretSaada 1980: 64, fn. 1). Relatedly, practitioners of witchcraft are spoken of as always elsewhere—such charlatans as still pretend to supernatural power, Bocage residents assured Favret-Saada in her early days of fieldwork, are not to be found in our village, though perhaps you might find them in some neighboring département (Favret-Saada 1980: 31–38). Likewise, in my initial forays into the research that led to this book, I found Lal Beg securely situated in the past and in the rural. A host of signs, as noted above, pointed to the pastness of Lal Beg. The expert literature seemed particularly decisive—“Lalbeg [sic] as a reference of identification was abandoned,” writes anthropologist Rama Sharma (1995: 141) of Delhi in the 1930s, adding, with an equal sense of finality, that the old prophet’s shrines “were replaced by Balmiki temples.” Early conversations in Lucknow were similar. When I asked Tara Balmiki, for whose Dalit feminist organization I had worked as a volunteer in previous years, about the Lal Begi tradition, she recalled that her grandmother had married a Lal Begi after the death of her first husband, a Mazhabi Sikh. The Lal Begi husband then gave Tara’s aunt the musalmānā name of Allah Rakhi—a telltale sign, in Tara’s account, of the “half-Muslim half-Hindu” character of Lal Begis, whom she thought of, in any case, as people of her grandparents’ generation. Tara’s husband chimed in to say that if there were still Lal Begis somewhere, they were most likely in the dehāt, the rural hinterland, not in the city. In another early discussion, a prominent Balmiki entrepreneur of Panipat, north of Delhi, told me that there

255 Lal Beg Underground

was no one left in that town who could tell me about Lal Beg. Then, during a tour of the bastī in which he spent his childhood, he showed me the gleaming, white-tiled Valmiki temple that he had sponsored there, and explained that the temple stood on what had once been the site of the neighborhood Lal Beg shrine. “I had it demolished,” he told me matter-of-factly, “and then I had this built.” The rupture was stark and complete. Strata of earth bore witness to Lal Beg’s consignment to a place beneath the foundations of the Hindu present. Lal Beg’s pastness was an archaeological fact. For all of its utility as a means of maintaining secrecy, the perpetually receding horizon of spatial and temporal displacement—witches are “back then” and “somewhere else”; Lal Beg is another time, another place—remains susceptible to puncture. This is due to several factors. One, as we will see, is that even when the secret is the collective property of a bounded community, not everyone in that community may be invested, or equally invested, in its concealment. It is also due to the internal structure of the secret as a social fact. As Georg Simmel reminds us, the secret is animated by an internal structural tension between concealment and revelation. What gives secrets their force in human affairs is their ever-present capacity to be revealed; what introduces the secret into historical significance is its release from concealment to revelation, or its partial disclosure such that the presence of a secret is made known, ushering in a struggle over the conditions of its further exposure. Paradoxically, in order to exercise its power as a secret, the secret must reveal itself, at least in part. Thus what Simmel (1950: 333–34) calls the “fascination of betrayal”— the consciousness “that one holds the power of surprises, turns of fate, joy, destruction”—perpetually tempts the possessor of the secret and strains against the simultaneous drive to maintain concealment. After the tour of his childhood bastī with the Valmiki temple built atop a razed Lal Beg shrine, the Panipat entrepreneur and I repaired to the modest brick home of one of his caste fellows nearby, where, over cups of chai, we continued to discuss caste tutelars old and new. Our host, a young husband and father, after quietly listening to our conversation for some time, led us to a corner of his one-room home and had us squat before a low shelf built into the wall. He slid open the shelf’s front panel. Half visible in the darkness at the back of the shelf was a triangular niche constructed out of two-and-ahalf bricks, before which lay matches, globular remnants of candles, and faded marigolds. We light candles here every night, he told us. Perhaps out of a desire not to contradict further his educated and prosperous caste fellow, our host

Deceptive Majority 256

said little else. But it was clear that he wanted us to know that he and his family maintained a shrine for Lal Beg. Back in Lucknow, I found my way to one of the larger sanitation labor caste bastīs in the city and began talking with a recently retired government servant—let us call him Daulat Ram—whom I met sitting on a bench next to the bastī’s Valmiki temple. This was one of my first conversations not mediated by colleagues in the Dalit rights organization. Daulat Ram spoke at length about Rishi Valmiki, author of the Ramayan, from whom he said his community was descended and after whom the neighborhood—Valmikipurva (though most people, he admitted, still call the bastī by an older, pejorative title)—is named. An hour or so into the conversation I ventured to ask about Lal Beg. Daulat Ram smiled and asked where I had heard the name. I explained that I had read it in books by Bhagwan Das and others. Anticipating the usual invocation of the rural past, I asked whether there were people, somewhere, in the present, who were believers or followers (mānnewāle) of Lal Beg. “Ham sab haiṅ (We all are),” he replied. I must have looked startled; taking me by the hand, Daulat Ram led me down a gully to the very back of the bastī, as far as one could get from the main road. On the ground abutting the exterior brick wall of one of his caste fellow’s homes was a raised, square platform of dried earth crowned by five rough domes. Noting that the central dome was partly missing, Daulat Ram replaced it with a broken chunk of brick that he found on the ground in the gully. “Ḍhāī īṅṭh [Two-and-a-half bricks],” he said. Whereas the triangular niche of Lal Beg shrines in Punjab and western UP uses the bricks intact, the five-domed Lal Beg shrines of central and eastern UP are made by breaking the bricks in half, rounding them, and affixing them atop a square platform. With evident pleasure, Daulat Ram told me that I should be present for Jamghaṭ, the day when cocks are sacrificed to Lal Beg and bhāng is consumed. He told me I should see the Lal Beg shrines in the other sweeper colonies of Lucknow as well, and gave me directions to one nearby. All of this was more than a little unexpected. Everything I had read and been told until this time supported a narrative in which the “religion of the sweepers” attested in the nineteenth century was abandoned in the mid-twentieth and its prophet forgotten. Now, the young man in Panipat and Daulat Ram in Lucknow were suggesting that Lal Beg’s death had been staged. This was not simply a matter of contradictions between two discourses competing for public acceptance, as conflicting interpretations of human action are generally—and often rightly—assumed to be. Those who acknowledged being mānnewāle

257 Lal Beg Underground

of Lal Beg in the present did not contradict their caste fellows who insisted on his pastness; rather, they yielded to the dominance of the latter discourse, appearing to accept the position of their own discourse as subordinate and clandestine. “We revere him,” Daulat Ram said of Lal Beg, “but we revere him”—and here he used the English word—“underground.” How does one revere a prophet underground? If, by “underground,” Daulat Ram meant to suggest that the rites of Lal Beg were a secret, then why was he telling me—and not merely telling me, which could be taken as a revelation of the presence of a secret while still concealing its content, but inviting me to participate? The fact of Lal Beg’s perdurance in the present, it seemed, needed both to be concealed and contained, yet also to be circulated and shared. Was this determined entirely by context—by the social location of the speaker and listener, the where and when of the communication? My apparent foreignness and externality to local caste hierarchy no doubt mattered, for example. Other questions were raised by the modes of signification by which Daulat Ram and the young man of Panipat initially disclosed Lal Beg’s presence: in the moment, both privileged non-verbal over verbal signs, and neither uttered the name of the prophet. These initial signs of an “underground” ritual practice seemed to hint at a semiotic ideology at some distance from that of declamatory publicity characteristic of (Protestant) modernity and exemplified by the “overground”—so to speak—celebration of Valmiki. Of what might this ideology consist, and what ethical considerations inform it? How does concealment relate to signifying practices associated with active dissimulation? The slight smile that accompanied Daulat Ram’s discussion of Lal Beg raised further questions: What affects are brought into play by “underground” religion or its revelation? What part does pleasure play in secrecy? In the previous chapter I spoke of women who told me that blood and liquor would flow “in every home” on Jamghaṭ, the day after the lamp festival of Diwali. Daulat Ram, too, mentioned Jamghaṭ as the occasion for an annual propitiation of Lal Beg. A colonial account of “the ritual of Lal Beg” gives it the same date (Greeven 1894: 38). In a convergence suggestive of a broad affinity among anti-gods, Jamghaṭ is known elsewhere in India as Balipratipada, and it is on this day that the legendary ashura king Bali—enemy of Vishnu and the brahminical order—returns to earth and is, for one day, honored. In the weeks leading up to Jamghaṭ, even as Daulat Ram and others directed me to the Lal Beg shrines tucked into corners of the Balmiki bastīs of Lucknow, others in the community assured me that the spilling of blood and bhāng by

Deceptive Majority 258

which the prophet “used to be” propitiated was a thing of the past: “Now that is all finished.” When I mentioned the Lal Beg shrines of which I was newly aware, these were acknowledged but woven into the accommodating narrative of spatial and temporal remoteness—they were feeble holdouts of a dying generation, uneducated, backward-looking, and attached to their village roots. In these conversations I was told that while a handful of the old guard might still talk about Lal Beg and maintain the occasional shrine, nothing would actually happen on Jamghaṭ—nothing, that is, but the pleasures associated with a rare day off work, as it is the one day during the demanding Diwali season that municipal sweepers are released from duty. Until the day finally arrived, then, I did not know what to expect. The following is a lightly edited excerpt from my fieldnotes of Jamghaṭ in 2011.

The Rites of Lal Beg Early in the morning I find the bastī more bustling than usual, no doubt largely on account of this day being a rare holiday from sanitation work in the municipality, hospitals, government offices, and universities. Children in the main gully greet me with their usual effusive salutations, but today alongside “Namaste!” and “Salām Elekum!” they add, “Happy Diwali!” Jamghaṭ does not have its own special verbal greeting. The Valmiki temple is open but not busy. While I chat with the temple caretaker and his neighbors sitting on the stoop, a young woman enters the temple and conducts puja on her own. Selfservice is the typical mode of worship at the Valmiki temple; there is no priest or pujari. I proceed to the back of the bastī. Near the high wall separating the back of the Balmiki colony from the gardens and bungalows of an adjacent middle class neighborhood is the home of Buggan, whose family built and maintain the Lal Beg shrine Daulat Ram first pointed out to me some months ago. The shrine has been moved and improved since then. It now sits adjacent to Buggan’s doorway, on a two-tiered, ziggurat-like, square platform of concrete, atop which rise five well-sculpted domes. The structure is freshly whitewashed, and nestled between the domes is a necklace of fresh marigolds. Buggan’s teenage daughters have designed a rangolī of black, red, and yellow powder on the ground next to the shrine. Buggan, who invited me to come this morning, tells me I am early. Born in 1958 and raised in his mother’s village in Bara Banki, Buggan came to Lucknow with his parents when they sold their mules and small plot of land

259 Lal Beg Underground

to work in the city as sweepers—his mother in the municipality, his father in a beer brewery. Buggan rebuilt the family shrine at their new home in the bastī using earth transported from the village original. Though Buggan’s family maintains the shrine, it is available for everyone in the bastī to use—whether for Jamghaṭ or for the sacrifice to Lal Beg that, Buggan tells me, remains part of many weddings. Not far away sits Shyam Mehboob, Buggan’s elderly neighbor, the one man in the bastī who routinely wears a skullcap, a sartorial sign widely assumed to signify Muslim-ness. His sons invite me into their house for chai to pass the time before the action starts. Sukhdev, the younger son, offers an interpretation of the five domes of the Lal Beg shrine that I have not heard before. Some identify the domes with the pāṅch pīr, five great Sufis of legend; others maintain that they symbolize the five Pandava brothers of the Mahabharata. Sukhdev suggests that they represent the five elements (tattvās) of ether, wind, fire, earth, and water. His older brother Rajan advises epistemological modesty: “We can speculate that they are thus. But what they are, we do not know.” Half an hour later, I rejoin Buggan in the gully next to his home, where he chats with a man holding a steel plate in one hand and a rooster in the other. They tell me that the kāṭnewālā (“cutter”) is on his way. In their theorization of sacrifice, Hubert and Mauss (1964) distinguish the sacrificer, the individual who performs the sacrifice, from the sacrifier (in French, sacrificiant) the person for whose benefit the sacrifice is conducted. Most people I have asked do not know the name of the sacrificer who is presently en route, but tell me that he is Muslim and lives somewhere nearby. Buggan says his name is Taj; his neighbor thinks it is Sarfaraz. The cutter is sighted coming down the main gully. A slim, clean-shaven man in his forties, thinning grey hair neatly combed, wearing a blue shirt and pressed trousers, his appearance strikes me as that of a government clerk on his day off. Except for the knife in his right hand. Approaching us almost jauntily, he exchanges a few brief words with Buggan and begins to prepare the ritual space. Doffing his sandals, he empties a polythene bag full of earth to the left of the shrine and shapes it into a mound with a depression at its center. He fills a small clay cup with dirt in which to plant incense sticks, rolls up his sleeves, and nods to Buggan, whose sacrifice will be the first of the day. Barefoot, Buggan and one of his daughters place before the shrine stainless steel plates heaped with garlands of marigolds, a box of incense sticks, piles

Deceptive Majority 260

of malīdā (shredded roṭī mixed with sugar), sesame sweets, and an oil lamp fashioned out of dough, as well as a steel pitcher (loṭā) full of bhāng sherbet: sweetened water topped by a thick, leafy layer of cannabis. The sacrificer, squatting, empties the box of incense sticks and lights the entire bundle. Buggan brings a cock from his house and squats next to the sacrificer, who waves the bundle of smoking incense over the bird and over the shrine several times. He then places one of the marigold garlands over the cock’s neck and, with Buggan keeping the tense bird restrained, touches its beak to the closest of the five domes. He removes the garland, mumbles something almost inaudibly—later he will explain to me that it is three repetitions of “Bismillah Allahu Akbar”— and blows a quick burst of air onto the cock’s head. Taking the bird from Buggan, the sacrificer secures its feet and wings with his feet and knees, pulls its head back, and applies the humble kitchen knife to the cock’s neck in a sawing motion, holding the bird steady as it spasms and as its blood, after spraying the whitewashed wall, flows into the pre-formed depression in the pile of earth. A cup of chai appears before my face and I realize that Buggan’s neighbor Ramesh is offering it to me, his arms extended over the now dense crowd of women and children with whom I am watching the sacrifice. I thank him and take a sip. Buggan and the cutter—sacrifier and sacrificer—remain silent and still as the life drains out of the bird. Once the cock is truly dead, the sacrificer wipes the blood from his blade onto the nearest dome of the shrine, marking it a deep maroon, and hands the feathery mass back to Buggan, who takes it into his house where it will be plucked and cooked. The next man in the queue steps forward with his plate, pitcher, and cock. The procedure begins again. By this time a queue of men and children bearing birds, plates of malīdā, and pitchers of bhāng winds through much of the gully. At first women seem to be onlookers rather than participants, observing from neighboring balconies, roofs, and charpoy, but before long Phulmati—whose husband is Daulat Ram—appears with a rooster under her arm. After Phulmati several women join the queue as bearers of sacrificial victims. The collective affect in the gully seems cheerful, at time raucous. A cock escapes its captor and runs squawking about, prompting general laughter until it is again caught. In the space immediately encircling sacrifier, victim, and cutter, a degree of solemnity prevails; but some joking penetrates even here. Seeing the difficulty the sacrificer seems to be having with one of the birds, one person and then another comments loudly on the bluntness of his blade. An alternative is

261 Lal Beg Underground

produced from a nearby household, then another. Before long four knives lie before the shrine. The cutter gives each of them a try and finally settles on one. The maroon stains on the whitewashed shrine steadily expand. What is offered at the shrine varies somewhat between clients; everyone offers a cock, malīdā, and sherbet, but not everyone’s sherbet contains bhāng, and some of the poorer clients forego the candles, incense, sugar batāśe, and so on. The consumable offerings are treated in the manner of prasād and tabarruk, the redounding gift of popular religion in South Asia: a fraction is placed (or spilled, in the case of sherbet) on the shrine and the rest is returned to the sacrifier to be distributed to the family at home (Pinkney 2013). As a donation to the ritual specialist, some clients place eleven rupees on the corner of the shrine, others twenty-one. The cutter periodically clears these off the shrine and puts them in his pocket. I lose count after about a dozen sacrifices. When there is a midday lull in the traffic the cutter abruptly slips on his sandals and starts off for somewhere. I hasten to accompany him.He introduces himself as Muhammad Yaqub, a Pathan by caste (he volunteers this—I do not ask), and tells me that he has been doing this work for the Lal Begis, as he calls them, for several years. He also invites me to his home in a neighboring quarter to discuss the day’s events at leisure after all the cutting business is over. At this moment we bump into the very client to whose house we are rushing—it seems Yaqub does house calls—who has decided, despite his earlier arrangement with Yaqub, not to sacrifice at home but at Buggan’s shrine. So with the client and his bird we retrace our steps and squat again at Buggan’s shrine, where more clients arrive and it becomes apparent that Yaqub will be occupied well into the afternoon. Friends in other neighborhoods have insisted that I spend at least part of Jamghaṭ with them, so I take leave of Yaqub, Buggan, and his neighbors. Leaving the bastī I see that the rubbish depots are now covered by a blanket of bird feathers. Some residents are preparing the sacrificed birds for the evening feast, while others—women and men—squat in circles gambling on a card game called lakṛī (wood). I catch a tempo across the river and soon reach the bastī on the edge of the old city’s largest drain, where a Balmiki youth who recognizes me says “Andar ho rahā hai! Jāie [It is going on inside! Proceed].” Something about his utterance hooks my brain—the vagueness of the referent gives me pause, it seems strangely familiar—but a few steps later my attention is drawn to acquaintances gathered around the shrines in this bastī. There are two outdoor Lal Beg shrines here. The first, which abuts the front

Deceptive Majority 262

wall of a home much like Buggan’s, has already seen a good deal of action, judging from the pool of blood at its base and the heaps of marigolds, roses, malīdā, batāśe, and so on, covering its five domes. It seems the morning’s sacrifices took place there, whereas now the activity has shifted to the second shrine, at the base of a tree in a wide spot in the gully, where I meet my friend Rishi Kumar, the poet and retired KGMC sweeper we met in chapter 5, who introduces me to the man who performs the sacrifices for this bastī. Aftab, also known as Lallan, squats next to the shrine wearing a checkered lungī, threadbare kurta and very thick spectacles. He is seventy or eighty years old, he tells me, and has been serving this function for the neighborhood since he was thirteen. He lives in a part of the old city that he says “is so interior you will never be able to find it,” but has spent most of his working life as a roadside barber (hajām) at the gate of the dargāh of Shah Meena Shah. Clients appear while we are chatting and Aftab resumes his work. His method differs from that of Yaqub; Aftab restricts his labor to ending the life of the bird, while the client is left to herself to waft the incense, garland the cock, light candles, and so on. For some clients, Aftab dips his thumb in the blood of the sacrificial victim and applies it, in the manner of tilak, to the client’s forehead. Though I offer no bird, Aftab spontaneously marks my brow with blood as well. Warm and viscous, this tactile sign makes me start. It communicates sensuously and indexically the animal life just taken, the act of violence by which the forces of sacrifice are released. In marking me with blood Aftab also interpellates me as sacrifier (of sorts), hailing me in a role I did not until that moment see as my own. Someone in a nearby house begins playing Hindi film music at very high volume; Bollywood is now the soundtrack to the sacrifices. One client, dressed in garments of pure white, silently performs a complete prostration before the shrine. Another client seems drunk and quarrels with Aftab. Mahesh Balmiki appears and we greet each other warmly. Mahesh is a staunch Ambedkarite and rationalist whom I met most recently at a funeral in Bara Banki, where he delivered to his caste fellows an extensive discourse critical of Hinduism and “superstition” of all sorts. I notice with surprise that Mahesh carries a rooster under his arm. He approaches the Lal Beg shrine, prays, gives the cock to Aftab to sacrifice, receives the blood tilak, and returns to his home with the carcass. The rites take on a somewhat different color in each bastī. In Karim Ganj, the sacrificer is not an outsider to the community, but a Lal Begi/Balmiki man

263 Lal Beg Underground

who has formally converted to Islam. Seeking to imbue the rites with Islamic pedagogical value, he recites the fātiḥa over the victim not sotto voce, like Yaqub and Aftab, but audibly and clearly, explaining the meaning to his caste fellows along the way. In Swati Nagar, sacrifices are performed by a team of Muslim musicians—some call them dafālīs, others qawwāls—who add their music to the proceedings. Asha, owner and caretaker of the only remaining outdoor Lal Beg shrine in that bastī, tells me that only two families sacrificed here in the morning, a dramatic decline from the previous year. It seems that most families in Swati Nagar now propitiate Lal Beg in the privacy of their homes, while some have stopped performing the rites altogether. She also tells me that on one occasion when a cock was unavailable on Jamghaṭ, a neighbor sacrificed a cat instead. In each of the bastīs the day concludes with a feast. At twilight in the Yar Ganj bastī in the old city the air is permeated with the aroma of cooking meat. I am the guest of my friend Safana, her nephew, and six sisters for a dinner of goat curry; while their family stopped sacrificing cocks to Lal Beg two years ago, they continue the tradition of a meaty celebratory meal. We are just about to eat when the disembodied voice of the state intrudes. Safana’s work supervisor has shown up outside the house and is calling out to her from the street, saying that she needs to report to Chowk immediately for extra sweeping duty. Safana refuses: this is her day off, it is dark outside, it is dinner time, and she has a guest. The supervisor insists: he has information that the city’s sanitary inspector will be making a tour of Chowk in the morning, and the whole area needs to be spanking clean. Tense words are exchanged and ultimately the supervisor leaves, having extracted a promise that Safana will report to work by five in the morning rather than the usual six. The day’s festivities conclude under the cloud of this reminder of the relentless control of Dalit labor by others.

Gestures and Things Rarely Named Lal Beg, then, is not as dead as he appears to be. Yet the discourse of his being in the past—literally buried-and-gone—remains vibrant; it is espoused, as we will see, even by some of the people who nurture Lal Beg shrines in the bosom of their homes. Appearing to be dead—as with appearing to be anything—is an active semiotic process, a profusing of signs tending to elicit a particular interpretant. The absence of Lal Beg from general knowledge is not just a passive matter of people not having heard of him; it is also that, among the sanitation labor castes and those who take an interest in them, a narrative of

Deceptive Majority 264

the prophet’s recession into the past enjoys a robust circulation. This discourse circulates in the same bastīs where an equally robust profusion of signs, especially on Jamghaṭ but also on other occasions, indicates living devotion to the old caste prophet in the present. And one hears neither talk of contradiction nor accusations of hypocrisy. How does this hold together? I suggest that we begin by considering the kind of signs involved. Note that the signs of the prophet’s pastness conform to what we ordinarily think of as a discourse; while they do include such material indexes as Valmiki temples built atop former Lal Beg shrines, they primarily consist of such manifestly discursive entities as spoken narratives and, to a lesser extent, written texts (like the scholarly accounts mentioned earlier). The content of the discourse— that part, at least, which could be summarized as “we used to offer sacrifice in thralldom to supernatural beings, but now we are enlightened”—conforms to that liberatory narrative of disenchantment that is a hallmark of both Protestantism and secular modernity, and gives the discourse the feel of familiarity and inevitability (Keane 2007; Yelle 2012). Whether we celebrate the ending or mourn it, this is a story we have heard before, one we know how to represent. Meanwhile its form—wordy, quotable, logocentric—likewise gives the narrative of Lal Beg’s pastness a legibility to modern eyes, and a great attraction for academic purposes. Transcribed interviews make for such convincing data; messages wrapped in verbal signs and therefore iterable on the printed page so readily satisfy the empirical needs of social science. The signs of Lal Beg’s enduring vitality, on the other hand, tend to be of a different order. For one thing, very little is actually said about the rites on Jamghaṭ. The semiotic forms that predominate tend to be unwordy, unquotable, even logophobic. When the boy at the entrance of KGMC servants’ quarters told me that andar ho rahā hai—it is going on—something tugged at my memory. Later I realized what it was. In the days leading up to Jamghaṭ, and on the day itself, almost nobody described the central event using a word for “sacrifice.” The obvious choices in the north Indian vernacular are the Sanskrit-derived bali and the Arabic-derived qurbān, both common terms widely understood and used to refer to the killing of animals in a ritual context. With very few exceptions, my interlocutors used neither term to denote the sacrifice of the cock. Nor was there an alternative technical term in circulation. Instead, most people spoke of the rite, if they spoke of it at all, without giving it a name. When I pleaded for elaboration, some described it pragmatically, as murghā chaṛhānā (raising up the cock [that is, onto the shrine]) or

265 Lal Beg Underground

murghā kāṭnā (cutting the cock). Some used the indefinite pronoun voh (it). Many, like the boy at the entrance of the bastī, signaled “it” even more ambiguously, by means of a subjectless verb. Grammatically there is no “it” in andar ho rahā hai; the literal meaning is “going on inside.” What was going on could be anything; in the absence not only of a subject but even of a pronoun, one needed considerable previous knowledge of context to follow the boy’s meaning. There is nothing particularly unusual or marked about absent subjects in the grammar of everyday Hindi–Urdu; what is noteworthy is that this quotidian construction, which formally secures ambiguity, was among the favored means of signifying the cock sacrifice. At the beginning of the last chapter I spoke of a woman with a bamboo staff who signified the rites of Lal Beg to me not by speech but by sawing the air with her right hand as though she were cutting a sacrificial victim’s neck. This pantomime, it turns out, was not peculiarly hers. In conversations in the Balmiki bastīs of Lucknow in the days before and after Jamghaṭ, I noticed that many of my interlocutors deployed the same gestural sign. For many, indeed, it was this silent movement of the hand, rather than any verbal formulation, that did the denotative work of a word like bali, qurbān, or sacrifice. It was as though speech hesitated to enter this particular domain of meaning. Approaching it, the speaker was compelled to switch tracks. Here gesture, with its charged silence, was the appropriate semiotic vehicle, while words were resorted to, if at all, only reluctantly and out of compassion for an obtuse, word-dependent anthropologist. The word, famously, is the preeminent example of that class of signs Peirce designates as symbols: signs that signify objects on the grounds of (more or less arbitrary) convention. Neither ghoṛā nor Pferd nor horse has a necessary or natural connection to the equine mammal signified by these words; the connection, rather, is upheld by general agreement among speakers of, respectively, Hindi, German, and English. Gesture, on the other hand, can belong to any of Peirce’s three sign types. Some gestures, like the shaking or nodding of the head to indicate agreement or disagreement or attentiveness, are symbols just as words are, since their significance depends on cultural convention. Other gestures, notably pointing at something to draw attention to it, function as indexes, as they signify their objects by physical contiguity. But the kind of gesture by which Jamghaṭ was signified to me by the woman with the bamboo staff, along with many of her caste fellows, was manifestly an icon. The pantomime of sacrificial violence represents its object by likeness

Deceptive Majority 266

or resemblance: the movement of the hand resembles the sawing of the knife. The sacrificer’s actual cutting of the victim’s neck, along with the uttering of Bismillah that precedes it (for most of the sacrificers, at least), are also icons, insofar as they represent by likeness ancient and routinely reenacted acts of ritual animal slaughter in early Islam. My interlocutors’ silent sawing gesture is thus a representation of the cutter’s slaughter of a cock on Jamghaṭ, which is itself a representation of Islamic sacrificial orthopraxy—an icon of an icon. The cutter’s movement of the knife, it should also be noted, is a ritually efficacious sign in that it renders the meat of the victim halal. In the marked preference for gesture over speech in the particular domain of the cock sacrifice to Lal Beg, then, what begins to come into view is a patterned avoidance of symbolization in favor of iconicity. This could be interpreted as evidence in support of Valentine Daniel’s (1984: 216) hypothesis that “in South Asia the iconic function occupies a place of privilege in the construction of reality.” This may well be, though the problem at hand demands explanation not at the level of South Asia as a civilizational whole but rather at the level of particular South Asian communities in particular circumstances. Adapting Daniel’s suggestion for our context, could it be that, for those who continue to revere Lal Beg, the iconic function occupies a place of privilege in action of grave and ultimate import—in the domain of the sacred, and in matters of preservation of the community—whereas the symbolic function suffices for action aimed at less consequential ends? Could a system that assigns greater moral value to (iconic) gestures than to words help account for the copresence of a bouquet of silent signs of the prophet’s vitality alongside a bundle of verbal signs of his deadness—is it that only one of these bundles of signs carries real moral weight? The logocentrism that is second nature to the scholarly enterprise and to modernity itself may make it difficult to imagine such a devaluation of the symbolic function. Recall that gestures indicating or effecting sacred action—gestures like those that proliferate on Jamghaṭ—are precisely what come under censure in the Protestant semiotic ideology, with its contempt for “certaine mimicall gestures” and “mimical irrational actions” by which God is “worshipped ignorantly and carnaly” (Yelle 2012: 104). Indeed Jamghaṭ features several of the targets of Protestant critiques of semiotic forms: mimetic gesture, indiscernible whispers, allegedly sacred speech whose referential meaning participants do not understand. These signifying practices, in Protestant thinking, were to be swept away, along with the “veiled” religions

267 Lal Beg Underground

with which they were associated, by the unambiguous plain speech of Christians. It was very much in this spirit that Protestant missionaries working among the sanitation labor castes in the colonial period, as well as some of their counterparts in the colonial bureaucracy, produced their portrayals of Lal Begi religion and ways of life. The secretive language and gestures of Lal Begi religious practice were, in these accounts, part of their condition of thralldom to unreal beings and empty semiotic forms, thralldom from which Christianity promised liberation (Gordon 1886; Greeven 1894; Griswold 1934; Youngson 1896). Missionaries invited new Christians to exemplify the achievement of good through publicizing what was once secret; converts who had earlier worked as Lal Begi “priests” were called upon to expose the fraudulence of the clandestine rituals over which they had once officiated (Griswold 1934). It will not do to try to apprehend Lal Begi ideas about signification and secrecy using only the terms bequeathed us by the Protestant semiotic ideology. One would like to understand the tradition on its own terms. But how does one go about apprehending the principles of a logophobic practice? When the very point of an ensemble of signifying practices seems to be that words cannot be trusted to handle serious matters, how does one seek further understanding? Where else might one turn for exegesis?

Sacred Secrets, Ethical Secrecy “God likes to be worshipped in secret.” So states a Shi’i Imami hadith, that is, a tradition attributed to the twelve Imams, political and spiritual leaders of the Shi’a in the first three centuries of Islam (Clark 2005: 48). A hadith attributed to Jafar al-Sadiq, the sixth Imam, holds that “[y]our faithfulness in secret is better than your faithfulness in the open, and your worship in the reign of falsehood (bāṭil) while your Imam is hidden and you are fearful of the enemy … is better than worship when the truth is manifest” (Clark 2005: 52–53). These verses pertain to the Shi’i doctrine of taqiyya, defined by scholars of Shi’ism as “defensive ambiguity” (Mariuma 2014: 105) and “precautionary concealment or dissimulation of one’s true religious identity and beliefs under adverse circumstances that endangered one’s life or property” (Daftary 2010: 48). Taqiyya was initially developed as a doctrine by the fifth and sixth Imams during a period of intensive political repression of the Shi’a by first the Umayyad and then the ‘Abbasid Caliphates, persecution in response to which the Imams themselves,

Deceptive Majority 268

though well known as Islamic scholars, attempted to conceal from authorities their leadership of the Shi’a (Dakake 2006: 331–41; Kohlberg 1975: 396). As Lynne Clark (2005: 46) points out, taqiyya has two primary applications: one as a legal-political dispensation intended to ensure the safety of the believer and survival of the community, the other as a principle of esotericism wherein conditions are established for the transmission of sacred secrets. Historically the importance of the legal-political application of the doctrine has waxed and waned in correlation with the political fortunes of the Shi’a—the need for tactical dissimulation being, in theory at least, much less when persecution is absent or the polity is itself Shi’i (Daftary 2010; Kohlberg 1975). As esoteric principle, meanwhile, taqiyya has more steadily enjoyed the status of an article of faith (Clark 2005; Dakake 2006). In both of its primary applications, the doctrine of taqiyya advances a semiotic ideology wherein what is sacred is to be shielded from, rather than exposed to, general scrutiny. Secrecy attends the sacred, whereas publicity, associated with wielders of political power, is morally suspect. Modes of signification favoring the protection of secrets are highly prized; Shi’i lore is rich in stories of the Imams’ use of polyvalence, circumlocution, artful gesture, and meaningful silence to protect the Shi’i faithful from detection and persecution (Dakake 2006: 339–45). Such semiotic forms are to be mastered by ordinary Shi’a as well, when the political context calls for it. Where taqiyya is necessary, Shi’a should scarcely be noticed by their politically dominant neighbors: “[W]hen people see them, they do not recognize them, and when they are absent from them, they do not miss them,” as an Imami hadith has it (Dakake 2006: 338). The formulation here—to be seen but not recognized— captures the semiotic operation at the very heart of taqiyya. To invite seeing without recognizing—to cultivate misrecognition, to pass—is to produce signs calibrated to induce divergent responses (interpretants) depending on the social location of the interpreter, signs that reveal to co-sharers in the secret while concealing from others. A number of theoretical and historical problems have kept the doctrine of taqiyya from receiving the scholarly attention, outside of the specialist field of Shi’i history, that it warrants. There is, first of all, the epistemological conundrum that taqiyya poses: how can the veracity of claims about identity be assessed when the concealment of truth is, under certain conditions but not others, precisely the goal? A famous case is that of the thirteenth century scholar Nasir al-Din Muhammad al-Tusi, who wrote treatises elaborating the


Lal Beg Underground

views of two rival Shi’i sects (Ithna’ashari and Ismaili) at two different times in his life. This fueled speculation on both sides that one or the other treatise was a product of taqiyya, composed in duress under conditions more like house arrest than patronage (Daftary 2010: 52). Because successful taqiyya covers its own tracks, modern scholars find themselves on shifting sands in trying to assess the rival claims—when the author may be intentionally deceiving us, to what evidence do we turn? The epistemological problem is one that taqiyya shares with esoteric traditions more broadly; as Hugh Urban (1998: 209) puts it, with a Bengali tantric tradition in mind, “how can one study or say anything intelligent at all about a religious tradition that practices active dissimulation, that is, a religious tradition that deliberately obfuscates its teaching and intentionally conceals itself from outsiders?” The epistemological problem is real enough, but what has arguably done more to dampen the study of taqiyya is its reputation, developed over several centuries of Western representations of Islam, as a license for indiscriminate lying. Orientalist descriptions on the whole took a dim view of taqiyya, figuring it less as considered doctrine and more as simple pretext. Twentieth century accounts, including some used by the American military, often cite Ignatius Goldziher’s 1906 study, which says of taqiyya that “in its use as frivolous hypocrisy, this has influenced, in the ethical sense, the general mind of Islam in Persia very badly….” (Goldziher 1906, cited in Mariuma 2014: 91). Popular disparagement of Islam in Europe and the United States makes extensive use of the idea that Muslims (not just Shi’a) are enjoined by their faith to practice “holy deception” or “stealth jihad,” as taqiyya is glossed in that literature (Mariuma 2014: 98–99). There is, as well, a long history of Sunni polemics directed against the Shi’a for their alleged abuse of taqiyya, though the debate over the doctrine within Islamic discursive traditions is largely ignored in popular Western representations. I suggest, though, that taqiyya’s bad reputation has to do not only with the kind of representational violence now widely recognized as foundational to many Orientalist and neo-imperialist accounts of Islam, but also, underlying such accounts but extending beyond them as well, with the pervasiveness in modernity of an unstated, deeply seated moral unease with any practice of truth concealment, an aversion to secrecy tout court. Simmel (1950: 330–32) appears to be contending with just such an aversion when, in his analysis of secrecy, he goes to considerable pains to establish the ethical neutrality of his subject. The unrecognized prevalence of a shared moral sensibility that, ultimately,

Deceptive Majority 270

good is achieved by exposing the hidden rather than the other way around would help explain why, even in scholarship liberated from the distortions and anti-Islamic prejudice of many earlier accounts, taqiyya remains a relatively unpopular topic, and why it has not been significantly tapped as a conceptual resource for social theory. The unease with concealment, moreover, is particularly acute on the question of religious allegiance or what has come to be known as identity. This would help explain why those scholars who do take the doctrine seriously are drawn to its esoteric rather than legal-political application. That people might hide the content of a mystical principle still holds a certain romantic appeal; that people might actually conceal who they are seems an affront to modern values—that anyone should live, or have to live, in a manner other than openly, for all to see, with complete identitarian transparency, provokes disquiet if not indignation. What I am suggesting is that the scholarly and popular reception of taqiyya, by both hostile and sympathetic observers, reflects the pervasiveness of the Protestant semiotic ideology in modernity. We cannot shake the sense that the good must be achieved through publicity, that one’s religious or identitarian status should be proclaimed before the world, and that deliberate secret-keeping smells of bad ethics. But if encumbrances such as these have hampered the analysis of taqiyya thus far, they need not continue to do so. The epistemological problem posed by dissimulation, while it does make unusually intensive demands on the interpretive process, is different not in kind but in degree from the epistemological challenge of studying any human practice. Deceit, after all, is essential to human interaction, from the little white lies of which everyday politeness consists to the great collective self-deceptions that enable us to ignore burning contradictions in cultural life (Bailey 1991; Bourdieu 2013 [1972]; Simmel 1950). That taqiyya is more self-aware than some practices of truth concealment does not make it impossible to analyze. The Orientalist stereotype of taqiyya as “frivolous hypocrisy,” for its part, can be addressed by critically attending to the conditions of knowledge production under which it emerged, circulated, and acquired “authority.” The most difficult obstacle to a clear-eyed study of taqiyya, perhaps, is acknowledging as parochial the moral assumptions about secrecy and publicity that pass as universal in modernity. In theory, though, just as with other ideological axioms we hold dear, we should be able, having acknowledged them, to suspend the moral assessments they animate for the purposes of analysis. If we do this, we may be able to perceive

271 Lal Beg Underground

in the doctrine of taqiyya something other than what recent centuries of Protestant hegemony have prepared us to perceive. One of the primary Qur’anic verses from which theorists of taqiyya derive its legitimacy is the 106th verse of Surah an-Nahl: “Whoever expresses disbelief in Allah after having believed [will suffer greatly]—except him who is compelled, his heart being still at peace in belief.” Key here are two points: first, that taqiyya is an exception to the general rule, not a rule unto itself; and, second, that the exception hinges on conditions of coercion. The word translated as “compelled,” often also translated as “forced,” is the Arabic verb uk’riha, the noun form of which is ik’rāha, compulsion or coercion. Commentarial traditions (tafsīr) on the verse cite a number of cases of early Muslims who suffered and in some cases died at the hands of non-Muslims for refusing to renounce their faith; the Prophet’s praise of these individuals establishes a paradigm of noble martyrdom quite like that of Christianity. At the same time, though, the verse and its commentarial tradition acknowledge that sometimes the harm courted by an adamantine refusal to recant— what is, in actual social life, a rather extreme commitment to the perfect correspondence of intention and utterance, object and sign—is too great, and that the faith is better served by tactical retreat. Such is the case of the nearmartyr Ammar bin-Yasir, whose parents were brutally murdered and whose own insincere renunciation of faith before his persecutors spared his family further loss of life. Citing hadith in which the Prophet grants his blessing to bin-Yasir’s decision, commentators create space for an alternative paradigm to that of martyrdom. In broad strokes, what emerges here is an ethics of religious self-disclosure sensitive to context and alert to the ways in which coercion structures social life. Rather than a single universal rule to be applied to all people in all circumstances, a legitimate norm and a legitimate exception are both delineated, and political compulsion is acknowledged as a factor necessary to consider when making and assessing moral choices. Questions that further refine this rough outline— In which contexts is taqiyya appropriate? Who should and should not practice it, and under what kinds of compulsion? Is it ever obligatory?—have figured centrally in debates in Shi’i literature in the thirteen centuries since the doctrine was formulated. A number of Shi’i scholars, considering possible situations in which taqiyya could lead to the loss of life, opined that it should never be practiced in such cases. The influential tenth century authority al-Sheikh al-Muhid held that “taqiyya should be practiced only if it is known

Deceptive Majority 272

by a necessary knowledge (bi’l-darūra) or if it is probable (qawiya fi’l-zann) that its abandonment would cause harm to the true religion or to the believers in this world” (Kohlberg 1975: 400). In support of this view are stories of the Imams directing one group of Shi’a to practice taqiyya while having another group, which was relatively freer from persecution at the time, publicly debate their non-Shi’a detractors (Kohlberg 1975: 400). The incipient sociological analysis of such narratives is developed in later Shi’i works that distinguish between the disempowered, for whom taqiyya is more necessary, and persons of authority, who are in a stronger position to weather persecution and should thus let their religious allegiance be known (Kohlberg 1975: 401). In all of this taqiyya takes shape not as a universal dictate, but as a situational practice whose exercise necessitates ethical reasoning. Taqiyya theory, as we might refer to the body of scholarship that elaborates the doctrine, is attuned to social structure and to the differential effects of persecution on different categories of person. It takes these into account when considering potential outcomes and assessing risks faced by individuals and communities, believers and non-believers. What taqiyya brings to “the enlargement of the universe of human discourse” (C. Geertz 1973: 14), then, is a theory of ethical secrecy. Truth concealment functions here as a means of saving lives or, more often, minimizing harm and enabling vulnerable groups to collectively survive structural violence. Not an endorsement of the kinds of deceit practiced by ruling groups in pursuit of power, taqiyya theory discriminates between the deceptions of the dominant and the subterfuges of the subjugated, insisting that the differing intentions and effects of the two, and not just their shared semiotic form, determine their ethical value. In this, too, taqiyya theory contrasts with the Protestant ideology, which assesses all semiotic practice according to a uniform standard. Whereas the Protestant ideology locates ethics in the universally applicable—which, it must be remembered, is a guiding principle for many a liberatory project, even as it is also leveraged against many a claim for collective rights—taqiyya insists that a true ethics must acknowledge that the stakes of self-disclosure can differ radically between landlord and laborer, colonizer and colonized, majority and minority. Semiotic practices of truth concealment that warrant moral condemnation when taken up by one group in one context may be ethically sound when taken up by another group in another context. Dissimulation by subaltern groups under conditions of oppression has long been, as James Scott (1987, 2012) characterizes it, a “weapon of the weak,” an “infrapolitical” tactic that dares not announce itself as political,


Lal Beg Underground

even though its political consequences can be profound. But what the modern commitment to the morality of publicity has perhaps disabled us from seeing is that dissimulation of this sort may be more than a survival tactic—that it may be, simultaneously, an ethical practice. What taqiyya theory provides is a demonstration of the kind of ethical reasoning that can undergird subaltern practices of secrecy and deception. Pradeep Bachhan told me about the time when Hindu militants came to his village in Bara Banki district. It was in the early 1990s, during the mobilization to destroy a sixteenth century mosque associated with the Mughal emperor Babur and to replace it with a temple to the Hindu god Rama. In this campaign the RSS employed the strategy—entirely in keeping with the logic and practice of majoritarian inclusion earlier developed by the Arya Samaj—of reaching out to Dalit communities along the way, interdining with them, and enlisting them in the ranks of volunteers (kār sevaks) heading to Ayodhya to demolish the mosque and build the temple. Pradeep’s village was on one of the routes through which RSS activists were traveling to Ayodhya. Pradeep told me, The RSS people invited us to meetings. And they gave us plastic containers and said “We will come on this day, you please pack some puris for us.” So our people were very happy. “These people who practiced untouchability, who threw their leftovers [jūṭhan] at us, today they’re eating our puris!” … The puris-vuris were given, they were packed and given.

The RSS volunteers were pleased with their success and moved on. In their time in the village they were not shown the imambara that Pradeep’s family keeps behind their home, where his caste fellows store the sacred paraphernalia with which they processed—and continue, today, to process—with Muslims on Muharram. Nor were they shown the several Lal Beg shrines in the village. Nor were they told that Pradeep’s parents’ names were Ramzanu and Hasina. When the day came for new recruits to set out for Ayodhya, Pradeep explained, “I didn’t go. From our community nobody went. Not a single Balmiki went.” A story such as this appears, from one angle, to confirm the stereotype of Balmikis as dupes of Hindutva, unable to see the cynical use to which the RSS is putting them and materially contributing to the Hindu majoritarian project. From a secularist point of view the failure to confront the RSS for its assault on the pluralistic fabric of society is unforgiveable. From an Ambedkarite point of view perhaps the worst aspect is the agreeability of Balmikis to feed, rather than challenge, those who until the day before treated them with disgust. There are

Deceptive Majority 274

indeed compelling reasons to advocate for a more frontal, explicit, and vocal— in a word, declamatory—politics of resistance to religious majoritarianism. Yet it must also be acknowledged that for Pradeep and his caste fellows in the impoverished Lal Begi bastī of a village in Bara Banki district, teams of RSS volunteers en route to Ayodhya presented a real potential danger, should they be provoked by the remonstrations of critical Dalits. The path of apparent acquiescence—of signaling agreement with Hindu nationalists while concealing enduring attachments to Muslim neighbors and Lal Begi ways of life, of seeming interested yet failing to show up for acts of majoritarian violence—enabled the sanitation labor castes of Pradeep’s village to contribute precious little (some puris) to the anti-Muslim cause while evading harassment or worse at the hands of the RSS. However dissatisfying such a politics might be from some points of view, it does reflect a concern with the minimization of harm, a concern attentive to the differential capacity of differently socially positioned groups to weather persecution. Ethical reasoning of this sort informs many of the acts of truth concealment with which this book is concerned. For the sanitation labor castes, to present one’s daughter to the school admissions officer as Shanti when she has always been known as Khadija, or one’s son as Alok when his name is Ali, is to employ a subterfuge that secures Scheduled Caste status, thus financial relief from school fees, and thus access to education—a subterfuge, that is, that works to mitigate the structural harm of untouchability. As a would-be first-generation entrepreneur, to apply for a loan as Jagdish, son of Sitaram, is to avoid the extra questioning and likely denial of loan that would ensue from admitting that one’s father’s name was actually Ahmad—presenting one’s father as Hindu is a concealment of truth designed to make possible occupational mobility from sweeper to shopkeeper, again toward minimizing the harm of untouchability. By bringing taqiyya theory to bear on certain practices of the sanitation labor castes I do not mean to suggest that the latter think of their profession of Hinduism to the state as a deployment of taqiyya. Nor is it my argument that in the Lal Begi tradition we can discern the traces of a Shi’i sect whose signature mode of protection was all too successful. There are communities that fit this description—Ismaili Shi’i groups in western India that dissimulated as Hindus when their leaders came under Mughal persecution, who came to be categorized as Hindu in the colonial period, and who either embraced or rejected their nearly forgotten Shi’i heritage in the identitarian struggles that ensued—as the work of Shafique Virani (2011) and Dominique-Sila Khan (2004) documents.


Lal Beg Underground

But Khan’s (2004: 59, 71) suggestion that Lal Begis be counted among such groups, while intriguing, is speculative. Marshaling evidence for such a claim is difficult enough with respect to more literate communities like the Guptis and Nizar Panthis, whose practice of taqiyya left at least some trace in the written record; while the Lal Begi oral tradition does feature taqiyya-like tropes, as we will discuss shortly, this is not sufficient reason to posit direct lineal descent from Shi’i forbears. What I am suggesting, rather, is that taqiyya theory, like the Protestant semiotic ideology, was among the discourses historically and culturally available to the sanitation labor castes in north India, a discourse that, whether named or not, would have supplied its concepts to their practice. The socioreligious landscape of colonial north India, as I argued in the previous chapter, was multipolar, and the sanitation labor castes were acquainted, as intimate outsiders, with the ideas and practices of Sunni, Shi’i, Sikh, Brahmin, and Rajput landlords as well as of Christian missionaries. In Awadh, where the Shi’a enjoyed comparatively high representation in the landlord class and where the policies and pageantry of the nawabs helped popularize a kind of cultural Shi’ism throughout the region, Shi’i ideas and practices were a staple of public discourse, known to a broad swath of society that included the sweepers and servants employed by Shi’i masters. Nineteenth century Lucknow witnessed a series of public controversies over the applicability of taqiyya in a Shi’i polity; local ‘ulema disputed visiting scholars and the nawabs over a host of practices for which individuals and groups invoked dissimulation—from defying authority and cursing the first three caliphs during Muharram to playing music and drinking wine—and made judgments as to their propriety (Cole 1988: 141–42, 154, 241–42). Taqiyya was in the air; to live in colonial Awadh was to be familiar with the concept and the ethical debates it invariably provoked. The sanitation labor castes of the nineteenth century, then, especially in regions of disproportionate Shi’i influence like Awadh, would have been acquainted with taqiyya, and may well have drawn on the doctrine in reflecting on their collective religious life. My argument, however, does not rest on whether Lal Begis explicitly derived their tactics from the Shi’i repertoire. Whether developed in dialogue with the Shi’i elements of their cultural milieu or not, key practices of the sanitation labor castes reflect a semiotic ideology, and an ethical orientation, quite like that of taqiyya. Taqiyya theory, then, as an elaboration of subaltern thought rare for its explicitness and remarkable for its resemblance to what we are describing, may, in considering religious life in the Dalit bastīs of postcolonial north India, be good to think with.

Deceptive Majority 276

Ethics of Disclosure? If secrecy can be ethical, what, then, are the ethics of revealing secrets? Ethnography is an expository genre in the full sense of the word—it exposes what it seeks to interpret, putting out into public view things which otherwise often go unnoticed, unremarked, even hidden. Ethnographies derive no small part of their sense of purpose from the unstated idea that the good is accomplished by exposure, a key assumption of Protestantism, Gandhian publicity, and positivist social science that, as this chapter has sought to demonstrate, runs athwart the valorization of secrecy at the heart of taqiyya theory and subaltern practices of surreption. Is there, then, a responsible manner in which to attempt an ethnography of secrecy? Is not such an effort a violation of the very ethics I have been seeking to elucidate? Like Hugh Urban (1998: 218), I think there is no single resolution to this conundrum, as the ethical implications of publishing secrets differ enormously depending on context. One strategy is what Audra Simpson calls “ethnographic refusal,” a principled resistance to telling “everything,” a resistance that “acknowledges the asymmetrical power relations that inform the research and writing about native lives and politics” and in which “the goals and aspirations of those we talk to inform the methods and the shape of our theorizing and analysis” (Simpson 2014: 98, 105). Simpson’s strategy, importantly, is not to abandon ethnography—to reveal nothing—but rather to write with scrupulous awareness of the advantages and risks of ethnographic representation in the real world, and to set limits to her portrayal with careful attention to her interlocutors’ own practices of refusal. This is essentially the approach I have tried to adopt in this book. Aspects of my method are standard practice in the discipline: with the exception of three individuals, now deceased, who insisted that I not give them a pseudonym when publishing my research (the reformist “triumvirate” of chapter 5), each of my interlocutors is protected from identification by layers of anonymity—personal names, location names (at the bastī level), and other potential indicators of identity have been changed. It is worth noting that this kind of systematic anonymizing illustrates well the dictionary definition of subterfuge—“Something designed to conceal one’s true attitude, identity, etc.; a pretence, a smokescreen” (OED Online 2020c). Considering as well the ethical principles underlying pseudonym use and related research protocols—particularly the weighing of potential risks to vulnerable populations—we have, in current anthropological practice, taqiyya by another name.

277 Lal Beg Underground

But this only addresses part of the problem. Even if my work does not put any one individual at risk of exposure, is it not still a betrayal of sorts to disclose the “underground” rites, musalmānā names, and enduring attachments to older, often disavowed ways of life that my interlocutors shared with me? To answer this I must clarify two points about research practice that will lead us back to Daulat Ram and Simmel. While conducting fieldwork I was consistently forthright about the nature of my research and my intention to publish it; my interlocutors knew me and introduced me to others as someone writing a book about their community’s past and present. They also knew that I would use pseudonyms for informants in any publications that might result from the research. It was with this foreknowledge that many, like Daulat Ram, took me by the hand, showed me Lal Beg shrines, narrated their recollections, and invited me to participate in the rites of Jamghaṭ. Many others, of course, said that Lal Beg was the past and “all of that is finished now”; but even they did not ask me to refrain from writing about it. This is in contradistinction to the way my interlocutors spoke to me of another, truly clandestine category of rite to which I was once invited to participate. Except to say that it did not run counter to any of the claims I have made in this book, this is an experience about which I will not write. My account, like Simpson’s, is attentive to my interlocutors’ refusal, in certain key domains, to permit ethnographic representation. I am not here to tell “everything.” Following my interlocutors’ cues, it turns out, is precisely what led me to write about the ongoing, if “underground,” tradition of Lal Beg, and all of the musalmānā lifeways that go with it. It was Daulat Ram, Phulmati, the young father in Panipat, the woman with the bamboo staff in Lucknow, and many others among my acquaintances who, knowing very well what I was about, brought me before the shrines. To reflect on their practice is to recognize that many among the sanitation labor castes would welcome a book about the old caste prophet, a book that takes seriously those who continue to revere him. The question of the ethicality of lifting the veil of spatial and temporal displacement behind which Lal Beg has been concealed, then, begins to take on a different aspect: that of taking sides in an intracommunity dispute. The wounds of the decades-long conflict between proponents and opponents of Hinduization among the sanitation labor castes of Lucknow continue to hurt, especially for those whose cherished lifeways have been discursively wrought into indexes of backwardness. That this camp would like to see their prophet written about sympathetically, and would not mind puncturing, along the

Deceptive Majority 278

way, the pretensions of some of their caste fellows to a smart Hindu modernity unblemished by strange shrines and cousins with Muslim names, is a truth I must acknowledge. I have sought all along to present fairly both sides of this dispute; nonetheless, as often happens with ethnography, deciding what to include in the monograph means, perforce, aligning more with the wishes of one faction than another. To write about Jamghaṭ at all is to take the reader where Daulat Ram took me and where some of his more prosperous neighbors did not, to heed the gestures of the woman with the bamboo staff more than the voluble insistence of those who would bury Lal Beg in the past. But the choice to disclose is not only a matter of factionalism. It has also to do, I think, with the very nature of secrecy, and the kind of power it grants. The secret, as Simmel observed, smolders in the breast of its keeper. At the core of the secret is a structural tension between the attraction of concealment and the temptation of betrayal, which is “like the fascination of an abyss, of giving oneself away” (Simmel 1950: 334). The secret’s tension “is dissolved in the moment of its revelation. This moment constitutes the acme in the development of the secret; all of its charms are once more gathered in it and brought to a climax” (Simmel 1950: 333). Revelation can transform the social world; to disclose what is concealed is thus to exercise a kind of power. Simmel compares the power wielded by the keeper of the secret to that of a person of financial means. The feeling of power which accompanies the possession of money becomes concentrated for the dissipator, most completely and sensuously, in the very instant in which he lets this power out of his hands. The secret, too, is full of the consciousness that it can be betrayed; that one holds the power of surprises, turns of fate, joy, destruction—if only, perhaps, of self-destruction. (Simmel 1950: 333–34)

The power over secrets—over the where and when and how much of their revelation, if they are to be revealed at all—is one of the few forms of power the sanitation labor castes in north India can be said to enjoy. Like the cultivation of misrecognition—inviting others to see but not recognize, distributing signs meant to put a different object in the mind of the interpreter than that in the mind of signer—the exercise of the power bestowed by secrecy is the exercise of one of the few modes of agency not radically curtailed for Dalits by the structural violence of untouchability. The art of the timely reveal, as much as the techniques of concealment, is a weapon in the subaltern arsenal, a tactic

279 Lal Beg Underground

in the semiotics of the oppressed. It dawned on me only gradually that, in guiding me to the places they did, my interlocutors were, in a canny exercise of representational power, making use of my guarantee of pseudonyms and role as writer-of-a-book to reveal, from behind a screen of scholarly anonymity, a secret that was fairly well burning to be told.

Degrees of Disavowability I often met with the sanitary supervisor Munnu Balmiki at his home in Adampur, one of the Balmiki bastīs of Lucknow. He comes from a reformist family; his uncle was close with the shaharwāle triumvirate, and Munnu grew up listening to Kanhaiya Lal’s bhajans in praise of Valmiki and diatribes against Lal Beg. In our interviews Munnu told me how in his youth he, along with his uncle and the shaharwāle triumvirate, had convinced their relatives and the whole bastī to stop performing the cock sacrifice and feast of Lal Beg. “We brought it to an end [Hamne band karāyā].” I asked him if he had agreed with the various rationales that his fellow reformers had offered in persuading their caste fellows to Hinduize. “I had to [Mānnā hī paṛā],” he replied. The conversation then turned to Jamghaṭ, which had transpired just days earlier. “We had a cock cut [that is, sacrificed],” Munnu volunteered. I mentioned that I thought I had seen him that day with a rooster. “I was defeathering it. We do it for a feast and drink.” I signaled awareness that some reformist families continue to prepare a feast but without all the ritual and the Muslim officiant; surely this was what Munnu’s family was doing. “No,” he said, and then clarified that the cock I had seen him defeathering on the recent Jamghaṭ was sacrificed by none other than a learned Muslim, a maulāna. “The maulāna cut it. After reciting the Bismillah. We get the Bismillah done by him only. Look, the cutting business, you see, there is halāl and harām, and the cutting work, only a Muslim can do it.” Not two minutes earlier, Munnu had declared that the rites of Lal Beg ceased some thirty–forty years ago as the result of his own and fellow reformists’ efforts. Now he was telling me that on Jamghaṭ just a few days before, his family had feasted on a cock sacrificed and rendered halal by a Muslim officiant. Seeing my perplexity, Munnu smiled as he added, “I had to offer it [Chaṛhānā hī paṛā]. It’s an old tradition. We sustained it [nibhāyī].” Both when speaking of his agreement with the reformist program, and when acknowledging his family’s sustaining of the old tradition, Munnu used

Deceptive Majority 280

a verbal construction that conveys action done under compulsion, literally that which “fell on me to do.” The formulation absolves the subject of responsibility for the action, or rather suspends the question of responsibility, leaving uncertain the subject’s will or intention. It is a simple and common feature of Hindi–Urdu, but one that, importantly, renders more or less disavowable the action that “fell on” the subject to do. Another day at Munnu’s house, during a particularly long and rich interview in which he again emphasized the finality of the Hinduization of the community, Munnu led me to a room in the back that I had never seen. His elderly mother gave us a piercing look as we entered and I felt like an unwitting intruder. On the walls of the back room were shelves on which were propped framed photograph of the Sufi dargāhs Ajmer Sharif and Dewa Sharif, chromolithographs of Kali and other goddesses, and posters of the ka’aba at Mecca and the tomb of the prophet in Medina. Beneath all this, in the corner of the room stood a large, square, earthen platform crowned by five domes. In front of this Lal Beg shrine incense was burning. I have often wondered at Munnu’s choice to show me the shrine hidden inside his home. It seems clear that he wanted me to know that there was more going on here than his own and others’ accounts of the Hinduization of the sanitation labor castes. Though he never retracted, revised, or otherwise qualified his narrative of the eradication of Lal Beg and the triumph of Valmiki, he nonetheless invited me to glimpse, within his own home, the relatively clandestine continuation of exactly what he was declaring to be no more. One thing that becomes clear from this is that it is not only those who resisted the community’s Hinduization who, it seems, would like to see the veil over Lal Beg at least partially lifted. I know from other conversations with Munnu that, despite his continuation in the political path charted by his uncle and other reformists, he has come to regard this path—the program of political obedience, labor quiescence, and Hinduizing internal reform urged on the sanitation labor castes by Gandhi, Govind Ballabh Pant, and other Congress patrons—as, in his words, “slow poison.” Munnu’s jaded perspective converges in some ways with that of Govind Prasad, the most bitter of the reformist triumvirate, whose regret over some of the transformations he once championed perhaps motivated his hinting to me that “we deceived every one of them.” In saying and showing these things, did Munnu and Govind Prasad seek to signal to those outside the community, through me, that the sanitation labor castes have not been so thoroughly and obediently Hinduized as they

281 Lal Beg Underground

are often portrayed to be? Knowing me to be an academic with close ties to Ambedkarite and activist circles, did they mean anonymously to convey to Dalits elsewhere that the apparently credulous succumbing of Balmiki leaders to Hindutva was never sincere? My interactions with Munnu Balmiki raise another matter that needs addressing. Thus far, we have spoken rather loosely about secrecy, truth concealment, dissimulation, and their interrelation. Yet there are clearly distinctions to be drawn between the kinds of practices I am describing: the young father in Panipat who showed me and his entrepreneur caste fellow the Lal Beg hidden in his shelf never claimed that reverence of the old prophet was a thing of the past. Munnu did. We might, beginning with these two examples, sketch a typology of forms of dissimulation and plot them along a scale. Simmel’s distinction between the error and the lie provides a starting point. The lie consists in the fact that the liar hides his true idea from the other. Its specific nature is not exhaustively characterized by the fact that the person lied-to has a false conception about the topic or object; this the lie shares with common error. What is specific is that he is kept deceived about the private opinion of the liar. (Simmel 1950: 312)

What is helpful here is the attention to the semiotic structure of deception: errors are mistakes of interpretation (failures to identify the object corresponding to the sign) made by the interpreter, whereas lies are mistakes of interpretation willfully elicited in the interpreter by the sign-maker. We may then focus our attention on the practices of the sign-maker, where what is distinct about dissimulation properly resides, rather than searching vainly for criteria in the interpreter’s relation to a given state of affairs or to “truth.” Simmel’s “error,” then, can be the cutting off point of a spectrum of forms of dissimulation, each of which would fall under his over-broad category of “lie” (the moral charge of which we may, with the earlier discussion of taqiyya in mind, bracket and set aside). At the end of the spectrum nearest to the error is mere concealment. At the far end is what we might call the hard lie or (ironically) the verifiable lie: a deception that can be irrefutably demonstrated to be such, a lie from which, once caught, one cannot escape. What distinguishes practices from one another along this scale are degrees of risk and disavowability. Concealment is what the young father in Panipat practiced; his Lal Beg was decidedly hidden, but his hiding of the shrine did not run counter to any assertions or other signs he put forth regarding his pahchān (at least in the time

Deceptive Majority 282

we spent together). Concealment was the method of the women who uttered the name of Diwali but not of Jamghaṭ, waiting to see if I knew of the unspoken holiday and only then telling me of its flows of blood and bhāng. Concealment well describes what many, perhaps most of my interlocutors practice with respect to their enduring attachments to Lal Begi modes of sociality and ritual practice. Here Lal Beg, Jamghaṭ, and musalmānā names and traditions are not spoken of; nor, however, are they denied or repudiated. Recall from chapter 1 that even the census enumerator’s suspicion-laden “You’re not, for instance, Lal Begi, are you?” elicited from the man in the pink tee-shirt periphrasis but not denial. Mahesh Balmiki, the Ambedkarite and outspoken critic of superstition who surprised me at Jamghaṭ by having a rooster sacrificed by Aftab the cutter, presents a somewhat different situation. In his case, the ritual practice he undertakes discreetly in the bastī is indeed at odds with utterances he makes in other contexts. Yet one can imagine, without much difficulty, a resolution of this tension that would not require an admission of deceit, let alone “lying.” Not wishing to seem rude or dogmatic I did not press Mahesh to clarify how the cock sacrifice squared with the rationalist, anti-ritual ideology that he espoused, but several “rule-and-exception” rationales could have been reasonably applied. A step further along the spectrum is a practice exemplified by Munnu. Here, again, there is concealment, but the contradiction between what is concealed and what is declaimed—and “declaimed” is the proper word here, Munnu being among the impressive speech-makers of Valmiki Jayanti—is sharper. Yet even here, the potential accusation of hypocrisy is held at bay by Munnu’s “I had to,” a scarcely noticeable yet scrupulous linguistic positioning of himself as not entirely committed to either of the conflicting signs with which he presented me—a holding open of the indeterminacy, in both cases, of the sign–object relation. Munnu’s semiotic practice takes greater risk, and is less readily denied, than that of either Mahesh or those who simply conceal. Yet it remains, ultimately, disavowable; if needed, he could still embrace one discourse at the expense of the other. Further still along our spectrum, yet decidedly shy of the hard lie, are practices of active subterfuge. We will turn to two examples of subterfuge— one in legend, one in the built environment of Lucknow—in a moment. Finally, we come to the far end of our scale of dissimulation: the verifiable lie, the deception that cannot be disowned. This is an exceedingly rare thing,


Lal Beg Underground

a chimera almost, in the signifying practices I am trying to describe. The capacity of dissimulation to work—the efficacy of this tactic in protecting its users—rests precisely on its disavowability. From mere concealment to active subterfuge, each practice has a “safe” interpretation, an explanation that satisfies the demands of the dominant discourse without completely beggaring the rule of non-contradiction. This is why, in a scene of genuine dissimulation, the researcher seeks discursive affirmation in vain. To practice deception in a way that cannot be disavowed is reckless, a violation of the code. Nobody says, “I am dissimulating.” Before turning to the two examples of subterfuge, we must acknowledge one last set of questions that my interactions with Munnu raises. When, observing my confusion, Munnu said, “I had to offer it. It’s an old tradition,” what accounts for the smile that flickered on his face and, discernible in my audio recording of the interview, the mirth that entered into his tone? Why is it that the lady with the bamboo staff smiled when I uttered the name of Jamghaṭ, and that Daulat Ram did the same when I spoke, unsolicited, the name of Lal Beg? Can something as evanescent as a slight smile constitute data? Yet is it not precisely through evanescent signs—gestures and silences and tones of voice—rather than in more capturable words, that clandestine objects are necessarily conveyed? One clue, perhaps, might be provided by Govind Prasad, with whose remark about the community having deceived missionaries of all persuasions I concluded chapter 6. Of all the ways that Govind Prasad could have framed his assertion, what he actually said was maze kī bāt yeh hai …(“the interesting thing is …”). Mazā, the word I am giving here as “interesting,” more properly means pleasurable, or even tasty, delightful. A savory snack and a lovers’ tryst have as much a claim to mazā as does a juicy bit of discourse. What Govind Prasad may be pointing us to here—again, as with other linguistic forms considered in this chapter, in the kind of everyday Hindi–Urdu that does not draw attention to itself—is the cluster of positive emotions that, along with the more obvious fear of discovery, give secrecy its affective charge. Especially given the structural violence that deprives Dalits of many of the sources of satisfaction other communities often enjoy, dissimulation has its pleasures.

The Great Red God Several times in my fieldwork I heard a story about the Mughal emperor Aurangzeb. Though I heard the story from a number of Balmiki men in

Deceptive Majority 284

different contexts, its content was remarkably consistent from telling to telling. One version was narrated to me by Rishi Kumar, the poet and retired sanitation worker who introduced me to the sacrificer Aftab during Jamghaṭ. It was in fact on Jamghaṭ itself, over cups of tea during a lull in the sacrifices, that Rishi Kumar told me his version of the tale: Our ancestors lived in tribes in the jungles on account of persecution from staunch [kaṭṭhar] Hindus…. [Our ancestors] lived like adivasis, like banjaras, in dwellings in the jungle…. Then one time when Aurangzeb—what I’m telling you is what I heard from the tongues of my elders; my father and grandfather often mentioned this, they would tell me when I would sit and listen— Aurangzeb used to kill people for conversion. Whoever did not convert, he would kill them. Just hearing the word “Hindu” he would murder people. Well, when our ancestors considered that they had already barely survived [the aforementioned persecution from Hindus] with their caste and religion, they thought, how will we escape this disaster? So the day that he was to come to their region, these people—our ancestors—thought, “Well we certainly can’t fight with Aurangzeb…. So why don’t we make a fool out of him? Why don’t we dupe him [Kyoṅ nahīṅ is ko bevakūf banāyā jae]? We’ll manage to keep doing what we’ve been doing, and at the same time he’ll let us go! Our caste will survive [Hamārī jāt bhī bach jāe].” So these people consulted with each other and decided, “Why don’t we build something that, when Aurangzeb sees it, will cause him to let us go?” And then and there, at the base of a tree, they built a little platform and put five domes on it. Earthen domes…. [They also planted a red flag and assembled materials for the traditional Lal Begi cock sacrifice. Finally Aurangzeb] arrived and asked, “What’s going on?” So they said, “We’re doing pūjā.” “What kind of pūjā?” They replied, “We’re honoring our pir.” So he said, “Which pir is that?” So then at that very moment they thought, “What can we tell him? What name should we give to this thing we’ve made?” And they said, spontaneously, “This is our Lal Beg.”

Rishi Kumar emphasizes that he is channeling the words of his elders, suggesting the possibility of an unbroken chain of oral transmission originating in the time of the story’s action, that is, the latter half of the seventeenth century. In support of such a possibility there is, in several of the Lal Begi oral traditions collected in the late colonial period, the trope of a Delhi king whose wrath the sanitation labor castes narrowly escape. There is even, in those sources, a scrape with Aurangzeb: it was a Chuhra man or pair of brothers who, according to legend, stole from the Delhi fort the severed head of the ninth Sikh guru Tegh Bahadur, executed by Aurangzeb, and brought it to Guru Gobind Singh.

285 Lal Beg Underground

Since persecution by Aurangzeb is key to taqiyya narratives elsewhere—the historically attested persecution of an Ismaili Shi’i leader and his followers by Aurangzeb did lead the community now known as Guptis to take up protective dissimulation—scholars hoping to find in the Lal Begis an underground sect of Shi’a might see in Rishi Kumar’s story a suggestive parallel (Khan 2004; Virani 2011). Yet there is no mention, in the Lal Begi songs and stories transcribed in the colonial period, of the central element of the foregoing story: the invention of Lal Beg as a means to evade oppression by Aurangzeb. The motif of the angry Delhi king in Lal Begi legend emerges, instead, from an act of perceived ingratitude and the violation of a food taboo: King Sikander, pleased with his loyal servant Jiwan, bestows on him a fine steed and is piqued when he later discovers that Jiwan and his caste fellows have eaten it. Even in the story of the rescued guru’s head, Aurangzeb is an incidental figure and the question of the hero’s religious affiliation is not raised; primarily it is a tale of Lal Begi bravery. In two key respects, Rishi Kumar’s story bears the imprint of the Arya Samaj’s project of majoritarian inclusion. In its choice of Aurangzeb as the villain and in its foregrounding of the trope of forced conversion, the story resembles other narratives generated and circulated by Arya Samaj prachāraks in the 1920s and 1930s as part of their effort to woo the Depressed Classes. That this could be the case and that Rishi Kumar’s contention that he heard it from his father and grandfather could be true is possible when we consider local family histories: Rishi Kumar’s father and grandfather were, in fact, part of Kanhaiya Lal Balmiki’s circle of intimates, and thus deeply engaged with the Hinduizing discourse of the Valmiki movement. Yet the means by which the ancestors contend with Aurangzeb— subterfuge and dissimulation—are not to be found in Hindu majoritarian propaganda of the early twentieth century. The genealogies that the Arya Samaj promoted among the subaltern castes were militant; present degradation was a consequence of the untouchables’ ancestors’ military opposition to Muslim kings, their muscular defiance of forced conversion. Hindu nationalists did not advocate dissembling as a way of contending with Muslims. Moreover, if we look closely at Rishi Kumar’s tale, we find that the ancestors do not carry out their act in the name of the Hindu community, which is mentioned not sympathetically (let alone as “our own”) but as an earlier wave of oppressors. Instead, the ancestors take the action they take in the name of the caste community: hamārī jāt bhī bach jāe, “our caste will survive.”

Deceptive Majority 286

While it may be impossible to reconstruct a genealogy of the narrative with certainty, it seems likely that it reflects a blend of early twentieth century Hindu majoritarian mythmaking and Lal Begi oral tradition. Irrespective of origins, the verve with which the story is told suggests that it continues to be a generative site of meaning for the community today. I propose that we read the tale as a parable of subaltern tactics that speaks as much to the present of the sanitation labor castes, and their relation to the state, as it does to the past. The height of the narrative in Rishi Kumar’s telling is the internal dialogue of the ancestors as they prepare for an encounter with their overlord—a textbook example of a “hidden transcript” (Scott 1987). Far from the ears of their oppressors, the ancestors improvise a subterfuge that will enable them, for the moment at least, to escape persecution. They will stage a religious performance that appears to conform to that of their rulers while still allowing for the continuation, more or less surreptitious, of the group’s autonomous tradition (“We’ll manage to keep doing what we’ve been doing, and at the same time he’ll let us go”). The centerpiece of the deception is an ambiguous sign. Seeing the five-domed shrine, Aurangzeb perceives its object to be some obscure Sufi pir and, for his interpretant, lets the Lal Begis off the hook. For the Lal Begis the object of their sign remains their tutelary deity—Rishi Kumar does not elaborate, but the ancestors are clear that their worship at the shrine is continuous with “what we’ve been doing”—and their interpretant, presumably, is a sigh of relief. And perhaps something more, something like pleasure. When the ancestors go beyond talk of survival and deploy the language of “dupe” and “make a fool out of,” an unmistakable note of delight enters the narrative. Like a page from a manual of taqiyya, the story valorizes the skillful manufacture of ambiguous signs as a technique enabling a vulnerable community to evade persecution by the politically dominant. The ancestors are upheld as canny semioticians. In a game of wits with the emperor, it is they who prove the more nimble handlers of the sign–object relation; they are the ones who determine what object Aurangzeb will take the shrine-sign to represent, and, accordingly, what action he will take. Given the radical asymmetry of power in which this encounter takes place, the ancestors—or rather, those who tell their story—have cause to be delighted. Not only do they manage to survive the ordeal with their traditions intact, but their doing so exposes their oppressors, despite all the structural advantages they enjoy, as bevakūf, dupes or fools.

287 Lal Beg Underground

Rishi Kumar’s story situates the taqiyya-like tactic in the past, in a time when the political domination of Muslim overlords made the apparent adoption of musalmānā forms and practices an astute maneuver. Our other example of subterfuge, in which the Lal Beg shrine again plays a starring role, is very much in the present. There stands, at an intersection of roads not far from one of the Balmiki bastīs of Lucknow, a tall, square, concrete platform painted red, atop which rise five copper domes—one in each corner and the fifth, slightly taller, in the center. This is the largest Lal Beg shrine in the city. Most Lal Beg shrines, as the foregoing chapters have intimated, have been demolished, abandoned to dissolve in monsoon rains, or concealed in closets and back rooms. The few that remain, placed far from public streets in the rear of Balmiki bastīs, are, as they have been since at least the colonial period, made of earth. Yet this one stands outside the bastī proper, at an intersection in a mixed-caste, mixed-class neighborhood, and is both far larger and made of more durable materials than virtually all of its counterparts. There is a further difference. On this Lal Beg shrine, facing the central dome from one side of the platform, sits a small, white, marble representation of Nandi, the bull that serves as the vehicle (vāhana) of the Hindu god Shiva. It is a well-known rule of Hindu iconography that Nandi faces Shiva; in countless Hindu temples, Nandi can be seen in the courtyard or space immediately preceding the space where the image of Shiva, usually in the form of a lingam or stylized phallic shaft, sits. In the shrine in Lucknow of which we are speaking, Nandi faces the central dome of the Lal Beg shrine’s five copper domes. With the inclusion of the Nandi figurine, the non-Balmiki passerby—or more precisely, anyone unfamiliar with Lal Beg and his iconography—is invited to interpret this central dome as a Shiva lingam (albeit a somewhat peculiar one, boxed in as it is by the four subsidiary domes). The Lal Beg shrine elicits interpretation as a Shiva shrine. When I asked pedestrians and shopkeepers what they made of this shrine, most answered that it was a Mahadev temple. Mahadev (“great god”) is a name of Shiva. Some even—like the poet Shah Nasir almost two centuries earlier, who mistook a Lal Beg shrine for a Sufi’s grave—paused before it to utter a prayer. Members of the Balmiki family who built and maintain the shrine— like the ancestors of Rishi Kumar’s story—do not disabuse visiting outsiders of their initial assumption. Rather they facilitate it. When asked, family members say that it is a shrine of “Lal Mahadev.” This name immediately

Deceptive Majority 288

assures the addressee that the object is indeed a form of Shiva (Mahadev), one to which has been added the specification Lal (“Red” or “Dear”), in the kind of localizing naming practice to which maintainers of small shrines are entitled. Yet by referring to the shrine in a way strongly suggestive of a Shaivite Hindu status the caretakers do not necessarily betray or close out the old caste tutelary. The “Lal” of Lal Beg is, of course, retained, and the substitution of Beg for Mahadev—the replacement of a Turkish/Uzbek/Mughal honorific by a Sanskrit one—while it does brings into play a different set of indexical associations, maintains the attitude of reverence shared by both terms. Thus, Lal Mahadev is not implausible as an alternative title for Lal Beg, just as it sounds reasonable as a local variant of Shiva. The name, like the structure to which it is linked, is an ambiguous sign, composed so as to suggest (at least) two distinct objects in the minds of (at least) two distinct categories of passersby: one sees and hears a triumphant Hindu god, the other sees and hears an antinomian Dalit prophet. Either interpretation is sustainable, just as either interpretation is disavowable. If the genius of the imagined ancestors was to improvise a name both faithful to autonomous tradition yet cast in an Islamic idiom to placate Muslim overlords, those who repeat the story today are no less adroit in naming themselves and their guru to a postcolonial state whose promise of remedial care is conditional on the profession of Hinduism. But perhaps all this talk of disavowability and calculated ambiguity misses the point. We might, instead, consider how Lal Mahadev—the great red god—is a capacious title, how its semantic breadth can encompass both Shiva and Lal Beg, how the shrine might signify, more than anything, the monist sensibility that pervades much of Dalit religion, indeed much of religion in South Asia. One traditionalist resolution to the Lal Beg/Valmiki conflict is to collapse the two, to suggest that ultimately they, as well as other figures of the Lal Begi pantheon like Khwaja Jhaumpra, are ultimately one and the same. As Anand Vivek Taneja (2018: 152–53) observes, there are Balmikis in Delhi who equate Lal Beg with the Panch Pir of Sufi legend, and see both as identical with the jinn that haunt the medieval ruins of Delhi. The refrain from the Lal Begi liturgical songs transcribed in the colonial period that remains central among Lal Begi/Balmiki sadhus of today is bolo, mominoṅ, vohī ek—“say, o believers, that is the one.” In a theology thus inclined to isomorphism, perhaps Lal Beg, in a sense, is Shiva. And Shiva is Lal Beg. Perhaps, given the monist flavor of Lal Begi religion, the two belong in the same shrine.


Lal Beg Underground

Let us hold open the door to such an interpretation. At the same time, let us heed the words of Bhairav Lal, the eldest of the siblings that maintain the shrine. Bhairav Lal taught me a great deal in the conversations we had, usually over tea at his preferred neighborhood snack shop, spread out over the period of my research. His guidance was crucial, for instance, in my becoming aware of and then learning to navigate the local terrain of secrets, half-secrets, and common knowledge. I recall once, when I broached a topic I thought not to be sensitive, he raised his finger and said disapprovingly, “This is not to be discussed in mixed company [Yeh general mai bāt nahīṅ karnī hai].” Like so many of his caste fellows, Bhairav Lal has endured mistreatment by those who employed him in sanitation labor, and has been ill-used by the community’s political patrons. As a boy he was among the youth of his caste sent with brooms from Lucknow to Calcutta to break a sweepers’ strike there, an act organized by Kanhaiya Lal Balmiki, the champion of the Valmiki movement, apparently at the behest of his Congress benefactors. In our several interviews, Bhairav Lal spoke cautiously and relatively little about the shrine he and his family maintain. On a recent visit, though, this changed. It was six years after our first meeting, and two years since we had last seen each other when Bhairav Lal and I unexpectedly met in a lane in his bastī. Much had changed since we had last met: several of our mutual friends had died and Bhairav Lal, now in his seventies, had himself only narrowly survived a heart attack. I had completed my PhD and, on this visit, brought copies of the dissertation to share with those of my interlocutors who were interested in it. Moving more slowly than in the past, Bhairav Lal greeted me warmly, sat me down and told a grandson to fetch us chai. After we checked in with each other on health and family, he began, without my having asked, to talk about Lal Beg, Jamghaṭ, and his family shrine. “Our guru ji,” he said, “has disappeared with the coming of Valmiki [Hamāre gurujī ghāyab ho gaye haiṅ Vālmiki ke āne se].” And yet, on the most recent Jamghaṭ, “it” had been performed at the shrine. Jamghaṭ, he continued, was when “all the tantrics, and all the ghosts, spirits, and living corpses [bhūt, pret, aur betāl] gather together.” Lal Beg remained in good company among the anti-gods. Bhairav Lal then remarked on the beauty and the pakkā-ness, solidity of form, of his family’s red shrine, and continued, “Under compulsion, we added the image of Nandi ji. And the trident. Showing the form of Shankar [that is, Shiva], we attracted people. People understood that this was an image of Shankar ji.”

Deceptive Majority 290

He added, “We made fools of them all. [Sab ko bevakūf banāyā.]” “Who?” I asked, stalling for time to scribble his words into my notebook, “Which people?” I wanted to make sure I understood properly who, in Bhairav Lal’s startlingly and unprecedentedly direct remarks, was being made bevakūf. “People of other castes. People of the Hindu religion. They come on Shivratri [a festival in honor of Shiva], there’s a long queue. They come thinking ‘this is a lingam,’ ‘this is Shiva.’ But our people only come [at other times],” he added, specifying the monsoon month of Savan and, of course, Jamghaṭ. The categorical contrast between “our people” and “people of the Hindu religion” rolled off Bhairav Lal’s tongue with intuitive ease in 2017, roughly a century after it began to give way to a sociological common sense underwritten by Hindu majoritarianism, and nearly seventy years after that new common sense became instituted in the law of the postcolonial state. Bhairav Lal’s utterance, which uncharacteristically avows an intention to deceive, also features some of the very phrasing attributed to the Lal Begi ancestors in Rishi Kumar’s narrative. So why don’t we make a fool out of him? They said, contemplating the folly of courting open conflict with a political authority interested in their assimilation to the dominant religion. Why don’t we dupe him? We’ll manage to keep doing what we’ve been doing, and at the same time he’ll let us go.


One morning I am taking turns reading aloud an Edgar Allen Poe story with the ten-year-old son of my friends Jagdish and Sunaina, when a ghost of sorts—a bhūt—turns up at their door. Recall that in Hindi–Urdu, the word bhūt denotes both “ghost” and “the past.” Answering the knock, Jagdish opens the door and begins conversing with some persons in the gully, apparently strangers. Then a change comes over Jagdish’s voice, and he calls me over. A smiling team of two men and a woman stand outside the door, bearing clipboards and ID badges suggestive of an NGO. They are checking in with residents of the bastī, they tell us, to confirm whether their homes have actually received promised connections to the municipal sewer line currently being laid, the construction of which has thoroughly torn up the ground in this warren of lanes, rending them all but impassable. Jagdish confirms that their home has acquired a sewer connection. But he, in turn, has a question for the man in front. “Voh kya likhā hai āpke shirt pe?” Jagdish asks. What’s that written on your shirt? The man laughs nervously. He wears his white polo shirt inside out, so that the name and logo of his organization, printed over the breast pocket, are difficult to make out. But not too difficult. Reading the faint print backwards, I speak the words that caught Jagdish’s eye: Harijan Sevak Sangh. “Āīye,” says Jagdish, inviting the three inside. As he and his colleagues enter my friends’ modest home, the man explains that they have worn their shirts inside out ever since the state government banned the use of the word “Harijan” for official work. Despite its partially concealed name, though, he assures us, the Harijan Sevak Sangh is alive and well. I am surprised. How is it that this institution of the 1930s, rejected as disingenuous by Ambedkar within months of its inception, made partly redundant by the birth of the Ministry of Welfare upon Indian independence, and rendered increasingly regressive-sounding by subsequent decades of Dalit and backward caste political assertion, still exists? Moreover, having scrutinized in libraries and archives the Sangh’s founding documents and internal correspondence from the 1930s and 1940s, how could I have overlooked the question of its continuation?



Jagdish serves the trio glasses of water, while Sunaina offers a plate of onion pakore that she fried just minutes ago. Thus begins the silent social and ethical test of which each of us in the room is acutely, watchfully aware, though we all take pains not to indicate it. Unlike some privileged caste guests my friends have hosted, the Sangh employees drink the water and eat the pakore their Dalit hosts have served. This bodes well, and in the ensuing conversation, even when Jagdish is explaining his strong objections to the term “Harijan” and urging the Sangh members to desist in their use of it, the mood is cordial. The Sangh employees make no promises to change, but they listen respectfully. When I ask about the Sangh’s ongoing activities, the senior member of the trio is forthcoming. Hygiene and education remain at the center of their work. They helped orchestrate sanitation at the recent Kumbh Mela, where the president of India and chief minister of UP lauded their efforts. The Sangh is on Twitter and Facebook. It continues to receive 90 percent of its operating budget from the government. When I ask if religious bhajans and recitations remain part of the Sangh’s activities, the senior of the three is delighted with the question, and invites us to the satsang session—a form of Hindu spiritual discourse—that is part of their weekly programming. One of the key arguments of chapter 4—that it was not only overtly majoritarian organizations like the Arya Samaj that conjoined efforts to improve Dalits’ material conditions with attempts to Hinduize them through explicit religious instruction, but that this was also the approach of Gandhi’s Harijan Sevak Sangh, notwithstanding the Mahatma’s efforts to depict it as above the communal fray—thus appears to hold true today, eighty years later. The structure of agency encoded in the Sangh’s organizational form also remains true to Gandhi’s vision: it persists as a body of privileged caste Hindus acting unilaterally as “trustees” for the objects of their penitent service, irrespective of what the latter may think of what is being done “for them.” Funded and applauded by an ostensibly secular state, the Sangh continues to enjoy legitimacy as a social service organization somehow untainted by its religious practice at the same time that Muslim and Christian social service organizations—even those that eschew religious activity in their operations— are being rhetorically delegitimized and materially disenfranchised by the state. In this respect the Sangh may be read as diagnostic of the character of secularism Dalits have experienced in postcolonial India. It is now seven years after the bulk of the research for this book was completed. In Lucknow again (and Delhi) for a new chapter of life and study, I am confronted daily with palpable social and political effects of the processes the book has



traced. I thought that much of what I had written in the book was history— chapters 2 to 5, after all, traverse the period from 1880 to 1950, which is surely the “past”—yet the events of that period seem to be unfolding all over again, the discourse of that time resounding afresh in speeches and headlines. The ghost of Swami Shraddhanand now speaks in the voice of the BJP leadership currently enjoying the largest electoral victory it has ever won (in June 2019), buoyed in part by a rise in support among Dalit voters. Shraddhanand’s thencontroversial formula for majoritarian inclusion is now an unquestioned axiom of mainstream politics, while the fear of small Hindu numbers he inculcated reverberates with new vigor in a recrudescence of concern with “the Muslim birthrate” and conversion through “love Jihad.” The majoritarian anxiety the swami so vividly performed remains as much a provocation to anti-Muslim sentiment—and a goad to cynical gestures toward Dalit welfare—now as it was in the 1920s. The ghosts of Gandhi and the first generation of Harijan Sevak Sangh luminaries are clearly audible in the current administration’s campaign for Swachh Bharat, “Clean India.” Prime Minister Narendra Modi’s description of sanitation work as “spiritual” (adhyātmak) and his recent act of washing the feet of five sanitation workers at the Kumbh Mela while not acknowledging, let alone conceding to, their demands for a living wage and the mechanization of sewer cleaning, reproduces with stunning fidelity the vision articulated by Gandhi in the 1930s and 1940s, from his insistence that “Bhangis may not go on strike” to his valorization of scavenging as “a sacred duty.” Like someone trapped in a Faulkner novel, I begin to sense that the past, far from being dead, is not even past. What manner of bhūt are we dealing with? What kinds of ghosts are these that, not content to haunt faded memories and gloomy archives, and resistant to countless attempted exorcisms by anti-caste critics and advocates of a genuinely secular state, parade about in the bright light of day—albeit, at times, with shirts inside out? Let me draw this book to a close by revisiting, seven years later, some of the people, places, and events with which this study is concerned. As with the Harijan Sevak Sangh team with the inverted polo shirts, these encounters may remind us of some of the book’s key arguments, alert us to their ongoing ramifications, and raise further questions.

* If the activities of the BJP regime currently (in 2019) in power illustrate the ongoing character of the project of Hindu majoritarian inclusion in



both of the distinct styles outlined in chapters 3 and 4 (the Arya Samaj’s missionary persuasion and Gandhi’s monological nomination), other recent developments throw light on why that project seems perpetually unable to achieve its goal. Chapter 5 charted the career of Hinduization in the bastīs of Lucknow from the late 1940s to the present. Crucial to the success of reformist Balmikis, I emphasized, was their capacity to leverage their Congress and Arya Samaj networks to deliver concrete benefits to their community such as school admissions, municipal jobs, and regularized housing. One highly visible sign of the political patronage that these reformists could muster were the Valmiki temples that national-level politicians from Ghanshyamdas Birla in the 1950s to Atal Bihari Vajpayee in the 1990s funded or inaugurated in sanitation labor caste bastīs. Emblematic of the majoritarian contract, these temples announced both the Dalit pledge “We profess to be Hindu” and the state’s counter-pledge “And thus we support you.” Inhabitants of the poor and precarious neighborhoods where these temples were located gained a measure of security from these temples, an assurance of political backing written in brick and mortar. So, at least, it seemed. I squeeze into a shared autorickshaw bound for King George’s Medical College (now University) in Lucknow, excited to visit after a long hiatus friends in the bastī known as KGMC Servants’ Quarters. This is the home of Rishi Kumar (formerly Hasanu), the poet and retired hospital sweeper who told the Aurangzeb story, and of Mahesh Balmiki, the Ambedkarite critic of superstition who offered a cock to Lal Beg on Jamghaṭ, and of other friends. At the turn to the lane where the bastī begins I climb out of the auto and find myself standing before an enormous padlocked gate. Beyond it, where KGMC Servants’ Quarters once stood, now stretches the flattened wreckage of hundreds of homes. Even the Valmiki temple at the front of the bastī, which boasted a commemorative plaque laid by no less a personage than then–prime minister Atal Bihari Vajpayee, is gone, replaced by a pile of dirt and rubble. Visiting friends in other neighborhoods I learn that the residents of KGMC were given notice that their bastī was slated for demolition; there were meetings and protests but Balmiki leaders were unable to prevent it. While the razing of the bastī was anticipated, the bulldozing of the Valmiki temple took most people by surprise. Both the KGMC administration and BJP-led state government deny having ordered the temple’s demolition. A protest is organized. As with Valmiki Jayanti, so with the protest against his temple’s destruction: the only people present at the demonstration are Balmikis. In Delhi, meanwhile, a temple dedicated to Guru Ravidas—the early modern



poet-guru of the Chamar community, revered by many Dalits while also claimed as a Hindu saint by Hindu nationalists—is destroyed by government order on account of its being on forest department land; with the Ravidas temple, too, Dalit protests erupt, while Hindu nationalists keep silent. One cannot help but be struck by the dissonance between these developments and the concern, among proponents of Hindutva, with temple desecration. Violence done to Hindu temples in Indian history is a cherished theme of Hindu nationalist discourse, a provocation to moral outrage that plays a crucial role in mass mobilizations and representations of the past. Yet here, upon the actual demolition of temples, we find no outrage, no exhortations to save Hinduism, no calls for retribution. With effort, Balmiki politicians manage to elicit a promise from Lucknow’s former mayor that action will be taken on the matter; on the whole, though, the response from political patrons is muted. This relative indifference, in the absence of a Muslim iconoclast responsible for the destruction, indicates what the Hindutva discourse on temple destruction distorts: in fact, it has never been animated by a concern for the disrespecting of houses of worship as such—historical instances of Jain, Buddhist, and brahminical kings attacking and looting one another’s sacred sites over a millennium of interreligious conflict figure nowhere in Hindu nationalist representations of history (Thapar 2005). For proponents of Hindutva, temple desecration only matters when Muslims are the desecrators. The relative indifference to the bulldozing of the Valmiki and Ravidas temples also indexes the contradiction at the heart of the project of majoritarian inclusion, the contradiction that Shraddhanand identified but was never able to resolve: Hindu nationalists need Dalits to be Hindu, but do not actually cultivate anything like fellow-feeling toward their ostensible co-religionists, and indeed often cannot stomach the lived implications of their newly shared pahchān. An embodied conviction in savarn supremacy and revulsion for the abject Dalit, widely though not universally cultivated among the privileged castes, has proven as resilient as the majoritarian compulsion to shared identity. A purely calculative majoritarianism would impel Hindu nationalists to rally in defense of demolished Ravidas and Valmiki temples, but affective realities get in the way—ghṛṇā (disgust), not sahānubhūti (fellow-feeling), often prevails in the embodied dispositions of the privileged castes toward Dalits, making it difficult for even the most driven proponents of Hindutva to mobilize savarn support for Dalit causes. In any case, the solicitousness that Hindu nationalists have at times exhibited for Valmiki and Ravidas has never extended to anything like popular Hindu devotion for these figures.



The parties most interested in the land on which KGMC Servants’ Quarters and its Valmiki temple stood are the government-run medical college and the state itself. In earlier decades, the Congress and the Arya Samaj in Lucknow supported Balmikis in prosecuting Muslim chai vendors and barbers for practicing untouchability, yet did nothing when Balmikis sought their help in obtaining justice for one of their leaders who was murdered while investigating Hindu moneylenders. The BJP, in power at the municipal, state, and national levels at the time of the Valmiki temple’s demolition, appears to be following the same script. With no Muslim available to be cast in the role of antagonist, the ruling party shows minimal interest in taking up Balmiki complaints of injustice. The good faith of the Hindu nationalist promise of savarn solidarity for those Dalit communities who acquiesce to the Hindu label may be measured by the leveling of Rishi Kumar’s neighborhood.

* In chapter 6, I argued that the sanitation labor castes, in step with many communities in South Asia, have increasingly adopted what I called declamatory religion, a mode of being-in-the-world that prizes unambiguous public averments of religious self-identification, a style rooted partly in the Protestant semiotic ideology that underwrites many an assessment of what is “modern” and which is significantly at odds with earlier forms of Dalit religion. I illustrated this argument through an analysis of the semiotic forms of the jhānkī (procession of youth costumed as Hindu gods) and kāryakram (function) at Valmiki Jayanti, the annual celebration of Rishi Valmiki. I also contended that Valmiki Jayanti is a performance of Balmiki “political arrival”—a display of gains made by the sanitation labor castes such as the capacity to dominate public space and access political power—that reveals, as well, the limits of those gains. Seven years later, does any of this hold? Again I participate in Valmiki Jayanti from the early-morning havan at Lal Bagh to the jhāṅkiyāṅ of Balmiki youth processing through the city dressed as the cast of the Ramayana to the cacophonous kāryakram at Parivartan Chowk at night. Again I mingle with the crowd of onlookers lining the streets of Lucknow. When I ask three women at a street corner in Daliganj what is going on, one replies, “You’ve heard of Jamadars? They take out this procession once a year.” Further down the street three young men tell me that this is Balmiki Jayanti. “Who is Balmiki?” I ask. “Broom-people [jhāḍūwāle]. Jamadars.” “Do you



celebrate it?” “No, we’re Muslims.” “Do Hindus celebrate it?” “No, just these people.” An impressively mustachioed customer at a pān stall states: “This is a procession taken out on the full moon of Sharada.” “By Hindus?” “Yes.” “Do you celebrate it?” “Mostly Balmikis celebrate it. I am a brahmin. We celebrate Navaratri. Different communities celebrate different things.” The expenditure on spectacle is even more pronounced this year than in 2011: there are now professional dancers, a new troop of men in gold-andwhite uniforms carrying electrically illuminated parasols, and more camels in the Husseinabad jhāṅkī than ever before (eleven). The young man playing Ravan this year gives as lively an equestrian show as his predecessor, but this time—since “smart phones” have begun to appear even in these economically disadvantaged bastīs—the performance is periodically frozen so that the ten-headed demon-king can shoot selfies with friends. At the culminating kāryakram at Valmiki Park, the organizers repeatedly admonish their caste fellows in the audience to move in an orderly fashion and to sit properly in their chairs, a reminder of the anxiety over the performance of middle class comportment characteristic of the kāryakram as a semiotic form, and indicative of its orientation toward a public beyond that immediately present. From the middle of the audience I watch Munnu Balmiki deliver from the bamboo stage a heavily amplified speech in praise of the chief guest. The latter is the same saffron-clad professor-politician who graced the event in 2011, when he was hailed as “messiah of Lucknow” and “savior of the Valmikis,” though he is noticeably thicker now, and more exalted, having risen from mayor of a city of 2.8 million to deputy chief minister of a state of 204 million. Munnu reminds the audience that the Valmiki statue we are sitting next to is a gift of Dr. Sharma, and that he also promised ten lakh (one million) rupees for the beautification of Lav Kush Dwar, the gate that serves as entrance to the park. This rings a bell—I recall now that the then-mayor made this promise at this very event in 2011. The man sitting next to me has a similar recollection. “Arre, this is that same BJP mayor who promised ten lakh years ago. Why do they invite such people?” As Sharma takes the stage to begin his speech, my neighbor adds, “These bastards just talk and talk. Thieves.” In his speech the poised politician informs us that “the Hindu religion is unified on account of you Balmikis”—a succinct acknowledgment of the indispensability of the sanitation labor castes to the Hindu majoritarian project, and a pithy synopsis of several chapters of this book—and congratulates them on the achievements of the sage after whom they are named. He concludes by



promising to build a Valmiki University [vishvavidyālay] somewhere in India. Does anyone believe this? Munnu Balmiki’s nephew seizes a microphone and shouts “Dr. Dinesh Sharma zindābād [may he long live]!” This conventional rallying cry invites the crowd to respond in unison, “Zindābād, zindābād!” Embarrassingly enough for the organizers, though, this crowd makes no response at all.

* The declamatory mode of religious being produces unmistakable effects on the tenor and texture of everyday life—on soundscapes, visual geographies, habits of speech, political imaginations, norms of self-presentation, and the understandings of self that such norms encourage and naturalize. The ascendance of the declamatory mode in religious practice in South Asia in general—and in particular among the sanitation labor castes—is transforming subaltern religion, and transforming, more broadly, the quality of social life. In chapter 7, though, I  suggested that there is another story to be told of contemporary Dalit religious life, one that does not negate the rise of the declamatory, but points to a different, older mode of religious being that endures furtively alongside or, to follow our guide Daulat Ram, underneath it. Endures? Persists? Even in the matter of word choice I find myself replicating the narrative of the inevitable demise of Lal Beg and his ilk, the modern teleology that spells doom for any rustic prophet or deity unwilling to either fall in line as a proper Muslim or proper Hindu, or to reinvent itself as kitsch or commodity, or to enunciate a liberatory ideology sufficiently resemblant to that of Ambedkar or Phule. And is the teleology misplaced? Under the kinds of ongoing pressures this book has described—pressures exerted by organizations urging a militant majoritarianism as well as by a state that requires the profession of Hinduism in exchange for access to ameliorative discrimination, in addition to the more friendly pressures of an emancipatory Dalit politics that repudiates as “superstitious” premodern prophets and rites like Jamghaṭ—is it not possible that the Lal Begi tradition, already diminished from what it was in the nineteenth century, will soon be extinguished? Certainly this is the expectation I meet, in the seven years since my primary stint of doctoral fieldwork, whenever I present my research. Intrigued by the history of Dalit religious alterity, sympathetic listeners express incredulity that it continues in the present, and confidence that whatever cock sacrifices to Lal Beg I may have witnessed in 2011, such a tradition can only be a vestige. Living and teaching thousands of miles from Lucknow, when I read the news and

299 Epilogue

scholarship from north India, so much of which emphasizes the robustness of the forces that make the Lal Begi tradition seem impossibly archaic, I begin to doubt my own findings. I question whether, in my choice of interviewees, I gave disproportionate attention to the elderly. In my enthusiasm for the wisdom of the buzurg (venerable), did I neglect the youth, and thus miss the possibility that the traditions I am describing are in fact collapsing with the passing of Daulat Ram’s generation? Or perhaps my doubts, and the certitude of those who insist that in times like ours a tradition like that of Lal Beg must be dead or dying, are the effect of an ensemble of semiotic practices that foster precisely such interpretations: the discursive displacement of the secret onto the past and the rural, the agreeable acquiescence to Hindu names and iconography, the indoor relocation of Lal Beg shrines, gestures, silences, circumlocutions, subjectless verbs, slight smiles, a thousand scarcely noticeable signs. Such an ensemble, cultivated over generations and valorized in lore, corresponds to taqiyya, reflecting both its ethical reasoning—its concern with minimizing the harm to which structurally vulnerable groups are disproportionately subject while maintaining their traditions in adverse conditions—and its semiotic ideology, which locates the sacred in the secret. It is Jamghaṭ again, but much has changed. Daulat Ram has died, as have all three of the reformist triumvirate. Seven years ago I spent much of Jamghaṭ at the Lal Beg shrines in KGMC Servants’ Quarters—now the shrines, like the bastī, have been bulldozed, and those who lived and worshipped there are scattered across the city. Wondering if these endings suggest a metaphor, I head to Adampur, Daulat Ram’s bastī, to see whether the rites of Lal Beg, too, have ceased. As I turn into the bastī a crowd of two dozen or so young men are striding energetically to the open area in front of the Valmiki temple. They quickly form a circle and get down to business: a cock fight. Given the circumstances, it is a clever idea for a diversion: it seems the wait time at the Lal Beg shrine at the back of the bastī is quite long, and since nearly everyone in the queue carries a rooster, a cock fight seemed just the thing. Buggan—next to whose home the shrine sits, and whom we met in the last chapter—sees me on the edge of the gladiatorial ring, shakes my hand, gestures with his head at the back of the bastī, and says, “Ho rahā hai”—[it] is happening. Ho rahā hai: the very verbal formation, simple, subjectless, ambiguous, that I slowly learned to discern as the preferred means by which Lal Begis denote, with scrupulous disavowability, their annual rites. I proceed to the crowded rear of the bastī



where I meet acquaintances old and new, most bearing roosters, plates of malīdā, and vessels of bhāng. This year the cutters are Muhammad Zameer, a middle-aged property dealer who hands me his business card, and his son Yunus, in blue jeans, tee-shirt and camouflage vest, who nods at me. Between sacrificing cocks at the five-domed shrine and reciting the fātihah, the father chats with the families who come to the shrine. “Zameer brother!” calls out Rakesh Balmiki as he enters the area from a side alley. “Ah, Rakesh brother! All well?” Zameer knows many people by name, and they seem to know him as well. Children address him as Zameer Mama (uncle), and twenty-something Anand tells me that Zameer Uncle helped him get admission into college. The young man who played Ravan in the recent Valmiki Jayanti swaggers into the lane, a rooster under each arm, grinning at me as he maneuvers into the queue. Ravan’s arrival makes me suddenly aware of the large number of young people waiting for their moment at the Lal Beg shrine. People of all ages are here with their offerings, from preadolescent girls and boys to their grandparents. This is not a rite only for the old. Nor are there signs that overall observance is declining. In Vivek Nagar seven years ago, Asha, who maintains the only outdoor Lal Beg shrine in her bastī, regretted that only two families had come to her shrine that year to make the sacrifice on Jamghaṭ. When I arrive at her place this time, four families are waiting with their cocks and other paraphernalia, and when the cutters arrive—here a team of professional musicians, Shah Rukh Dhol Master and his partner Naseer, who belong to the Muslim dafālī community—more families turn up to make sacrifice. Shah Rukh gives me his business card and apologizes to all for being late. They have been in Vivek Nagar already for three hours, he says, doing sacrifices at shrines inside people’s homes. Asha asks the dafālīs to stay and play some songs after the cutting, but they demur, saying they have many more house calls yet to make. One of the women waiting with a rooster teasingly accuses the musicians of kanjūsī, of miserliness with their time and talent. Banter continues in this vein, with pauses as Shah Rukh and Naseer recite the fātihah and drain the cocks’ blood at the green-painted, five-domed shrine. Meanwhile Asha, who works at a school, tells me that she has known these dafālīs since her childhood, as Shah Rukh’s father and her own father were friends; in fact Shah Rukh’s father used to do the Jamghaṭ sacrifices for her family and bastī before Shah Rukh came of age. Is this a dying tradition? Is it even diminishing? Asha’s children, who decorate the shrine each year before Jamghaṭ with silver foil and colorful powdered rangoli, are learning about Lal Beg Baba from their mother. “Is mei

301 Epilogue

sacāī dikhtī hai,” she says of the shrine. The truth can be seen in it. In Adampur as well the youth seem full participants in the rites with their elders; while there are certainly reform-minded families that stay away, I see no indication that the youth who observe Jamghaṭ are embarrassed to be involved in it or that their numbers are decreasing. We have seen that the Arya Samaj sought to drive a wedge between the sanitation labor castes and their Muslim neighbors, with whose cultural traditions their own significantly overlapped, by propagating the myth of the Muslim invention of manual scavenging, encouraging identification with a martial Hindu tradition ranged against a Muslim other, reframing existing caste-based tensions between Muslims and Dalits as religious discord, and instigating new local-level conflicts over women, property, and land. Between the Arya Samaj and the RSS and BJP who subsequently adopted these strategies, we can track a century of Hindu nationalist efforts to convert Lal Begis with names like Ramzanu and Allah Rakhi into militant Hindus prepared to enact violence on Muslims. Most analyses declare the success of Hindutva in coopting (non-Ambedkarite) Dalits—especially Balmikis—and point to the participation of Balmiki youth in several incidents of communal violence in Delhi and Meerut as evidence of the same (Prashad 2000). Those incidents have drawn warranted scholarly attention; but what I am pointing out, alongside Jaoul (2012) and Ring (2006), is that when structural conditions favor conflict and political patrons encourage majoritarian violence, it is the everyday production of neighborliness and the non-commission of violence that demands our analytical attention. It is not only the spectacular realization of Hindu majoritarian aims that has decisively shaped the quality of social life in postcolonial India, but equally or more so the unspectacular thwarting and subversion of such aims—the furtive enervation of Hindutva by those often too hastily portrayed as its footsoldiers. Taking into account the last ten decades of determined efforts to communalize the sanitation labor castes, what is astonishing is not that Balmiki youth in one region have been enlisted in majoritarian street violence, but that Balmikis in most of India have not. Indeed what is remarkable is that behind the hypervisible and hyperaudible signs of the success of majoritarian inclusion, those aspects of the Lal Begi tradition that fostered neighborly interdependence and a sense of overlapping sacrality between Lal Begis/Balmikis and Muslims—aspects of the tradition that several generations of Hindutva activists have sought to uproot—are quietly thriving. These are findings that require more subtle, ethnographically grounded and semiotically attuned concepts than “identity” and “recognition.”



The experience-near, north Indian vernacular concept of pahchān, I have maintained, offers analytical leverage on the actually occurring politics of caste and religion in a way that these ideas of global anglophone social thought do not. It is not merely that the frameworks of “identity politics” and the “politics of recognition” press our story into a narrative mold cast in Euro-American contexts whose universality may be questioned; more fundamentally, the singular and transparent “identities” such frameworks presume are projections—continuously reaffirmed and reified by the declamatory mode of religious being, but projections nonetheless—of a semiotic ideology that, while exceedingly influential globally and in contemporary north India, is not the only salient schema for relating selves to others and to the sacred. To analyze politics through pahchān is to not fall for the overvaluation of the symbol—indeed the word—that modernity inculcates in those who would be its subjects, but rather to be alive to the whole panoply of iconic and indexical signification as well, and to the value the latter may hold in non-dominant ideologies. In an account conceptually guided by pahchān, it is possible to see how the majoritarian project can enjoy triumph and suffer defraudment at one and the same time—these processes develop along divergent semiotic channels; the adept semiotician communicates pahchān differently by means of different classes of chihn (signs).

* Finally let us return to the beginning. Early in this study I described a pattern of utterances wherein some of my interlocutors unselfconsciously referred to Hindus as others. I adduced evidence that these contrastive deployments of “Hindu” have a genealogy in the songs, prayers, and legends of the Lal Begi tradition, which figured Lal Begis as a qaum or ummat distinct from and, in some contexts, agonistically opposed to Hindus, Muslims, and Sikhs. Often, though, my interlocutors’ casual contrasting of themselves with Hindus was then contradicted by avowals of Hindu-ness, an affirmation of the regnant common sense. Were there contexts in which, when the contrastive utterances of the Lal Begi tradition are pointed out, the speaker would not retreat behind the official order of things? I also described, in the opening chapter, the way in which census enumerators exercise nominative authority over the caste and labor designations and Hindu-ness of the enumerated to which the latter, for the most part, show signs of acquiescence, even as they simultaneously profuse disavowable signs of alterity. But how representative were these encounters?



Ultimately, was all of this “just” a Lucknow story? Does the old prophet have underground followers outside of Awadh? These questions are on my mind as I ride the sleek, crisply air-conditioned Delhi Metro to a Valmiki ashram in a central location of the nation’s capital. If Lucknow figures in popular representations as a provincial town steeped in Islamicate nostalgia and syncretic practices—and thus hospitable, in a sense, to the kind of traditions this book has documented—Delhi has the reputation of a global megacity hurtling into the future, a future increasingly painted saffron (the preferred color of Hindu nationalist aesthetics). Besides, by all accounts the sanitation labor castes were Hinduized in Delhi a full two decades before the Valmiki movement arrived in Lucknow—more reason to doubt that our story will have traction here. Through an arched gate I enter the simple, dusty, neem tree-shaded ashram and pay my respects at the samādhī of the founder of this order of itinerant ascetics of the sanitation labor castes, Khak Shah Baba (whom we encountered briefly in chapter 5). Gulab Singh and Jaidev Singh, two knowledgeable laypersons, sit me down with them and take turns with me on a bubbling hookah as they tell me stories of their order and community. Forty-something Gulab Singh, whose upbringing in rural Haryana richly colors his Hindi, starts with Guru Balmik Swamy, author of the Ramayana, whom he describes as the light of the world (vishwā jyoti). Thus far it is the familiar Hinduized caste narrative. Then he tells me that Balmik also originated the community’s liturgical refrain vohī ek—that is the one. “The Supreme Soul [paramātma] is one. Not two, not three, four, five— no! You must have noticed, in India many gods can be seen, yes? Vishnu, Mahender, Shiva, Hanuman. Hindu people, they worship them. But our holy men told us, ‘There is only one Supreme Soul.’” It is a sharp contrast he draws between Hindu polytheists and “our holy men” insisting on the singularity of the divine. He then elaborates on the Supreme Soul. Balmik, he explains, is one of its avatars. So are Khwaja Jhaumpra, Jesus, and Lal Beg. Taking a deep pull on the hookah, I listen as Gulab Singh launches into the cycle of Lal Beg legends: the prophet’s miraculous birth, dining with the sweepers of Delhi, resurrection of the three-legged horse, conflict with the royal qazi. Gulab proceeds to the story of Balmik manifesting himself as a rotting corpse in Kashi: a story well attested in the oral traditions documented in the colonial period but abandoned—or so I had thought— after the Arya Samaj’s linking of the Chandal Balmik with Rishi Valmiki. Faithful to nineteenth century accounts of the story, Gulab Singh’s version pits Balmik against “the Hindus” of Kashi. Surprised by the unambiguity of the



distinctions he continues to draw, I ask him how he responds when the census enumerator comes knocking. “In [the census], well, we write Hindu religion [Us mai phir Hindu dharm likhte haiṅ].” “Okay [Achhā].” “In that, because, in the government record—in the record entry we are counted as Hindus. But that’s not how our heart feels [Lekin ham is tarah kā man nahīṅ karte], we don’t believe that [ham nahīṅ mānte haiṅ is cīz ko].” Gulab Singh then recites a verse from memory: Hindu, turkī, varṅā, varṅī Zāt-pāt se hameiṅ kyā karnī Makkā dwārkā ek hī melā Guru ke mat se karnā melā Hindu, Turk, varna distinctions, Caste and such—what’s all that to us? Mecca, Dwarka—they’re two of a kind Our path is that of the Guru

This closely resembles the colonial-era Lal Begi oral traditions we encountered in chapter 2: the distance from both Hindu and “Turk,” the appearance of spatial indexes of Islam (Mecca) and Hinduism (Dwarka) to which a Lal Begi third is contrasted. But outside of the archive, this is the first time I am hearing such words uttered aloud. I plead for exegesis. Gulab elaborates on the first half of the verse: “You know about varna, right? Like in Hindu society: brahmin, kshatriya, vaishya, shudra. We have nothing to do with all that [In se koī lenā-denā nahīṅ hai].” With a mock tone of mock arrogance he says, “I am a brahmin!” and then, in his usual voice, “We don’t harbor such feelings [Is tarah kā bhāv nahīṅ rakhte].” This is a layered statement; given the semantic density of the deceptively simple-sounding bhāv and rakhnā, it could also be translated as “We don’t cultivate such affective dispositions.” “Now, Mecca and Dwarka,” he continues, proceeding to the second half of the verse. “They’re the same thing. You know Dwarka, right? In Gujarat. [Important] for the Hindus. But we are to follow the tradition [mat] of our Guru. The path that our Guru Maharaj showed us—the one of humanism [mānavtā]—only that is our path. That is our ideology [vicārdhārā].” Whether from the tobacco or the discourse, my head is spinning. Contrastive deployments of “Hindu” are flying thick and fast. Oral traditions I have

305 Epilogue

understood to be out of circulation for a century are on the tip of Gulab Singh’s tongue. My interlocutor is making boldly explicit what a symphony of silence and evasion seemed to be suggesting during the census operations in Lucknow. And all of this in the heart of Delhi, whose early history of Arya Samaj activism and present reputation for a fast-paced, Hindutva-shaded futurism make it a far less intuitive host for a tradition like that of Lal Beg. Our story is not, then, “just” about Lucknow, or “just” about Lucknow and the surrounding districts of central UP, dotted as they are with Lal Beg shrines. Nor is it just about shrines, or rites, or occasional unguarded utterances suggestive of a subtle sense of alterity. Gulab Singh is elaborating Lal Begi religious autonomy as a tradition that is—when provided favorable conditions for expression—as discursively vibrant as it is alive in ritual and material practices; he propounds a tradition guided by ethical ideals couched in a critique of the hierarchical practices and affective dispositions normatively cultivated by the dominant. What do we really know of Dalit religion? Before Ravidas was interpellated in early twentieth century narratives as either a gentle Hindu bhakta desirous of brahmin leadership or a proto-Ambedkarite revolutionary, how did followers of the Lodi-period Chamar poet-guru conceive of themselves and their relations to other socioreligious formations in north India? Daya Pawar writes of Bhādava, a deity whose rescue of Mahars from dominant caste fury over an incident of transgressive eating neatly mirrors the central Lal Beg myth. What do we know of Bhādava and of the ways in which Mahars understood their deities and practices vis-à-vis those of other communities in the period prior to enumerative politics, before Ambedkar was compelled by Gandhi to “accept the position that you are Hindus … in spite of your statement to the contrary”? How many Dalit prophets, gurus, and anti-gods remain concealed in archives and oral traditions—or perhaps even in the contemporary cover of plain sight? What manner of subaltern religious formations might reveal themselves— illuminating, in the process, a new cartography of South Asia’s socioreligious landscape—once we desist in the projection of the categories of a recently manufactured, state-supported, majoritarian common sense onto earlier periods, and onto people in the present whose apparent acquiescence to the pahchān urged by the dominant may be not the end, but the beginning of the story? Some days after the interview with Gulab Singh, I am walking with my two children through the ruined grandeur of a Sultanate-period mosque in south Delhi. As we navigate the lanes adjoining the mosque, I notice a small brick shrine, an empty triangle set on a low platform against the outer wall of



a modest home. Further along in the same neighborhood is another, and then a third. My children and I sit down on a platform near the third and strike up a conversation with a slim henna-haired man, Azad, perhaps in his fifties, who lives there and tells me that he works for the municipality. When I ask about the triangular shrine, he says it is “Mahadev”—the great god. “And Durga, too, yes?” I ask, gesturing at the small, free-standing, clay goddess figurine that has been set at an angle in front of the shadowed niche. “Yes, that too.” Though we are 550 kilometers from Lucknow and I have never seen this neighborhood or met this man before, our conversation is beginning to feel familiar. I think of Bhairav Lal’s five-domed shrine in Lucknow, and the polysemy—indeed, the knowing solicitation of divergent interpretants by differently socially located viewers—put into play by his addition of the Nandi figurine. I wonder whether the free-standing, apparently Hindu Durga image that here supplements an otherwise aniconic shrine may be serving a similar function—if, that is, my own interpretation of this simple brick construction is not misplaced. “When is the puja for this great god?” I ask. “Diwali.” “Ah, I see. Is it, by chance, the second day of Diwali, the day after the firecrackers? Is it on Jamghaṭ?” Azad smiles. “If you’re free, you could come,” he says. “Bring the children.”


Adcock, C. S. 2014. The Limits of Tolerance: Indian Secularism and the Politics of Religious Freedom. New York: Oxford University Press. Ahmad, Imtiaz. 2010. “Can There Be a Category Called Dalit Muslims?” In Dalit Assertion in Society, Literature and History, edited by I. Ahmad and S. B. Upadhyay, pp. 239–53. Hyderabad: Orient Blackswan. Ahmad, Nazir, ed. 2008. Kalam-e Bullhe Shah. Islamabad: Akademi-e Adabiyat-e Pakistan. Ahmed, Mirza Ghulam. 1904. “Our Attitude towards the British Government.” The Review of Religions 3(3): 83–99. ———. 1984. Malfuzat, Vol. 3. London: Mubarak A Saqi. Aktor, Mikael. 2010. “Untouchability in Brahminical Law Books.” In From Stigma to Assertion: Untouchability, Identity and Politics in Early and Modern India, edited by M. Aktor and R. Deliège, pp. 31–63. Copehnhagen: Museum Tusculanum Press. Ali, Ahmed. 1940. Twilight in Delhi. London: Hogarth Press. Aloysius, G. 1998. Religion as Emancipatory Identity: A Buddhist Movement among the Tamils under Colonialism. New Delhi: New Age International. Alter, James P. 1986. In the Doab and Rohilkhand: North Indian Christianity, 1815–1915. Delhi: ISPCK. Ambedkar, Bhimrao Ramji. 1990. “Who Were the Shudras? How They Came to Be the Fourth Varna in the Hindu-Aryan Society. ” In Dr. Babasaheb Ambedkar: Writings and Speeches, Vol. 7. Bombay: Government of Maharashtra. ———. 1991. Dr. Babasaheb Ambedkar Writings and Speeches, Vol. 9. Bombay: Education Department, Government of Maharashtra. ———. 1992. Dr. Babasaheb Ambedkar: Writing and Speeches, Vol. 5. Bombay: Education Department, Government of Maharashtra. ———. 2005. Dr. Babasaheb Ambedkar Writings and Speeches, Vol. 2. Mumbai: Government of Maharashtra. ———. 2009. What Congress and Gandhi Have Done to the Untouchables. Delhi: Gautam Book Centre.

References 308

———. 2013 [1936]. Annihilation of Caste: The Annotated Critical Edition. New Delhi: Navayana. Anand, Mulk Raj. 1935. Untouchable. Bombay: Kutub Popular. Anand, S. 2007. “Dying in the Gutters.” Tehelka Magazine, December 8. Anderson, Benedict. 1991. Imagined Communities: Reflections on the Origin and Spread of Nationalism. New York: Verso. Appadurai, Arjun. 1993. “Number in the Colonial Imagination.” In Orientalism and the Postcolonial Predicament: Perspectives on South Asia, edited by C. A. Breckenridge and P. v. d. Veer, pp. 314–49. Philadelphia: University of Pennsylvania Press. ———. 1988. “Putting Hierarchy in Its Place.” Cultural Anthropology 3(1): 36–49. ———. 2006. Fear of Small Numbers: An Essay on the Geography of Anger. Durham: Duke University Press. Asad, Talal. 1993. Genealogies of Religion: Discipline and Reasons of Power in Christianity and Islam. Baltimore: Johns Hopkins University Press. Azad, Muhammad Husain. 1907 [1880]. Ab-i Hayat. Lahore: Naval Kishore. Bailey, F. G. 1991. The Prevalence of Deceit. Ithaca: Cornell University Press. Baines, J. A. 1893. Census of India 1891, General Report. London: Eyre and Spottiswoode. Bandyopadhyay, Sekhar. 1997. Caste, Protest and Identity in Colonial India: The Namasudras of Bengal, 1872–1947. Richmond, Surrey: Curzon. Basu, Amrita. 1996. “Mass Movement or Elite Conspiracy? The Puzzle of Hindu Nationalism.” In Making India Hindu: Religion, Community, and the Politics of Democracy in India, edited by D. Ludden, pp. 55–80. Delhi: Oxford University Press. Bauhat, Nausheri Lal. 1996. Maharishi Valmiki Gyan Dipak Prakash. Modinagar: Maharishi Valmiki Mandir tatha Ashram. Bauman, Chad and Richard Fox Young, eds. 2014. Constructing Indian Christianities: Culture, Conversion and Caste. New Delhi: Routledge India. Baxter, Richard. 1673. A Christian Directory, or, a Summ of Practical Theologie, and Cases of Conscience. London: Robert White. Bayly, Susan. 1999. Caste, Society and Politics in India: From the Eighteenth Century to the Modern Age. Cambridge: Cambridge University Press. Bellamy, Carla. 2011. The Powerful Ephemeral: Everyday Healing in an Ambiguously Islamic Place. Berkeley: University of California Press. Beth, Sarah. 2005. “Taking to the Streets: Dalit Mela and the Public Performance of Dalit Cultural Identity.” Contemporary South Asia 14(4): 397–410.



Birla, Ghanshyam Das. 1953. In the Shadow of the Mahatma: A Personal Memoir. Bombay: Orient Longmans. Bīrūnī, Muḥammad ibn Aḥmad. 1983. Alberuni’s India: An Account of the Religion, Philosophy, Literature, Geography, Chronology, Astronomy, Customs, Laws and Astrology of India about A.D. 1030. Translated by E. Sachau. New Delhi: Oriental Reprints. Booth, Gregory D. 2005. Brass Baja: Stories from the World of Indian Wedding Bands. New Delhi: Oxford University Press. Bourdieu, Pierre. 1977. “The Economics of Linguistic Exchanges.” Social Sciences Information 16(6): 645–68. ———. 1991. “Genesis and Structure of the Religious Field.” Comparative Social Research 13: 1–44. ———. 2013 [1972]. Outline of a Theory of Practice. Translated by R. Nice. Cambridge: Cambridge University Press. Briggs, George Weston. 1953. The Doms and Their Near Relations. Mysore: Wesley Press and Publishing House. Brown, Judith M. 1990. Gandhi: Prisoner of Hope. New Delhi: Oxford University Press. Brubaker, Rogers and Frederick Cooper. 2000. “Beyond ‘Identity’.” Theory and Society 29: 1–47. Burn, R. 1902. Census of India 1901, Volume XVI: N.-W. Provinces and Oudh, Part I, Report. Allahabad: Government Press. Butalia, Urvashi. 1998. The Other Side of Silence: Voices from the Partition of India. New Delhi: Penguin. Campbell, James M. 1885. Gazetteer of the Bombay Presidency, Vol. XVIII, Part I, Poona. Bombay: Government Central Press. Campbell, W. E. M. 1908. “Memo to Commissioner, Allahabad Division.” In Improvement in the Administration of the Sanitary Department, Vol. 36: U.P. State Archives. Certeau, Michael de. 2011. The Practice of Everyday Life. Translated by S. Rendall. Berkeley: University of California Press. Chakrabarty, Dipesh. 2002. Habitations of Modernity: Essays in the Wake of Subaltern Studies. Chicago: University of Chicago Press. ———. 2018. “The Dalit Body: A Reading for the Anthropocene.” In The Empire of Disgust: Prejudice, Discrimination, and Policy in India and the US, edited by Z. Hasan, A. Z. Huq, M. C. Nussbaum and V. Verma, pp. 1–20. New York: Oxford University Press.

References 310

Chakravarti, Uma. 2006. Everyday Lives, Everyday Histories: Beyond the Kings and Brahmanas of ‘Ancient’ India. New Delhi: Tulika Books. ———. 2019. Gendering Caste: Through a Feminist Lens. New Delhi: Sage. Chander, Krishan. 2015. “Kālū Bhangī.” In Intakhāb-E Afsāna. Lucknow: Uttar Pradesh Urdu Akademi. Chandramouli, C. 2011. Instruction Manual for Updating of Abridged Houselist and Filling Up of the Household Schedule. New Delhi: Office of the Registrar General and Census Commissioner, India, Ministry of Home Affairs, Government of India. Chatterjee, Partha. 1993. The Nation and Its Fragments: Colonial and Postcolonial Histories. Princeton: Princeton University Press. ———. 2006. “B. R. Ambedkar and Troubled Times of Citizenship.” In Political Ideas in Modern India, edited by V. R. Mehta and T. Pantham, pp. 73–90. New Delhi: Sage. Cherian, Divya. 2015. “Ordering Subjects: Merchants, the State, and Krishna Devotion in Eighteenth-Century Marwar.” Ph.D., History, Columbia University, New York. Christopher, K. W. 2015. “Cast(E)Ing Narrative: Caste, Class and Religion in Mulk Raj Anand’s Untouchable.” Interventions: International Journal of Postcolonial Studies 17(1): 64–81. Clark, Lynda. 2005. “The Rise and Decline of Taqiyya in Twelver Shi’ism.” In Reason and Inspiration in Islam: Essays in Honour of Hermann Landolt, edited by T. Lawson, pp. 46–63. London: I. B. Taurus. Cohn, Bernard. 1987. An Anthropologist among the Historians and Other Essays. Delhi: Oxford University Press. Cole, Juan. 1988. Roots of North Indian Shi’ism in Iran and Iraq: Religion and State in Awadh, 1722–1859. Berkeley: University of California Press. Conrad, Dieter. 2007. “The Personal Law Question and Hindu Nationalism.” In The Oxford India Hinduism Reader, edited by V. Dalmia and H. v. Stietencron, pp. 187–230. New Delhi: Oxford University Press. Cox, Jeffrey. 2008. The British Missionary Enterprise since 1700. New York: Routledge. Crooke, William. 1890. An Ethnographical Hand-Book for the Northwest Provinces and Oudh. Allahabad: North-Western Provinces and Oudh Government Press. Daftary, Farhad. 2010. “Religious Identity, Dissimulation and Assimilation: The Ismaili Experience.” In Living Islamic History, edited by Y. Suleiman, pp. 47–61. Edinburgh: Edinburgh University Press.

311 References

Dakake, Maria. 2006. “Hiding in Plain Sight: The Practical and Doctrinal Significance of Secrecy in Shi’ite Islam.” Journal of the American Academy of Religion 74(2): 324–55. Dalmia, Vasudha. 2007. “Introduction.” In The Oxford India Hinduism Reader, edited by V. Dalmia and H. v. Stietencron, pp. 1–23. Delhi: Oxford University Press. ———. 2010. The Nationalization of Hindu Traditions: Bhāratendu Hariśchandra and Nineteenth-Century Banaras. Delhi: Permanent Black. Daniel, E. Valentine. 1984. Fluid Signs: Being a Person the Tamil Way. Berkeley: University of California Press. ———. 2000. “The Arrogation of Being: Revisiting the Anthropology of Religion.” Macalester International 8(1): 171–91. Das, Bhagwan. 2007. Bālmīki Jayantī aur Bhangī Jātiyāṅ. Delhi: Gautam Book Center. ———. 2009. In Pursuit of Ambedkar. Translated by Isha. New Delhi: Navayana. Das, Veena. 2006. “The Figure of the Abducted Woman: The Citizen as Sexed.” In Political Theologies, edited by H. d. Vries and L. E. Sullivan, pp. 427–43. New York: Fordham University Press. Datta, Pradip Kumar. 1999. Carving Blocs: Communal Ideology in Early TwentiethCentury Bengal. New Delhi: Oxford University Press. Dehlawi, Sayyid Ahmad. 1898–1918. Farhang-e Asafiya. Delhi: National Academy. Deliege, Robert. 1993. “The Myths of Origin of the Indian Untouchables.” Man 28(3): 533–49. Desai, Mahadev. 1936a. “Weekly Letter: With Harijan Workers.” Harijan 4(8): 57–60. ———. 1936b. “Weekly Letter: A Talk with Rao Bahadur Rajah.” Harijan 4(7): 50–51. ———. 1936c. “Weekly Letter: The Temple Entry Campaign.” Harijan 4(16): 121–23. Dikshit, Madan. 1996. Morī Kī Īnṭ. Delhi: Shabdkar. Dirks, Nicholas B. 2001. Castes of Mind: Colonialism and the Making of Modern India. Princeton: Princeton University Press. Douglas, Mary. 2006. Purity and Danger: An Analysis of Concepts of Pollution and Taboo. New York: Routledge. Dube, Saurabh. 1998. Untouchable Pasts: Religion, Identity, and Power among a Central Indian Community, 1780–1950. Albany: State University of New York Press.

References 312

Dumont, Louis. 1980. Homo Hierarchicus: The Caste System and Its Implications. Translated by M. Sainsbury. Chicago: University of Chicago Press. Eaton, Richard. 1982. “The Political and Religious Authority of the Shrine of Baba Farid.” In Moral Conduct and Authority: The Place of Adab in South Asian Islam, edited by B. Metcalf, pp. 335–56. Berkeley: University of California Press. Fallon, S.W. 1879. A New Hindustani–English Dictionary. Banaras: Medical Hall Press. Favret-Saada, Jeanne. 1980. Deadly Words: Witchcraft in the Bocage. Translated by C. Cullen. Cambridge: Cambridge University Press. Fazal, Tanweer. 2017. “Scheduled Castes, Reservations and Religion: Revisiting a Juridical Debate.” Contributions to Indian Sociology 51(1): 1–24. Flood, Gavin. 1996. An Introduction to Hinduism. Cambridge: Cambridge University Press. Flueckiger, Joyce Burkhalter. 2006. Amma’s Healing Room: Gender and Vernacular Islam in South India. Bloomington: Indiana University Press. Freitag, Sandria B. 1989. Collective Action and Community: Public Arenas and the Emergence of Communalism in North India. Berkeley: University of California Press. Fukazawa, Hiroshi. 1991. The Medieval Deccan: Peasants, Social Systems and States, Sixteenth to Eighteenth Centuries. Delhi: Oxford University Press. Fuller, C. J. 2004. The Camphor Flame: Popular Hinduism and Society in India. Princeton: Princeton University Press. Gaborieau, Marc. 1972. “Muslims in the Hindu Kingdom of Nepal.” Contributions to Indian Sociology 6(1): 84–105. Gait, E. A. 1902a. Census of India, 1901, Vol. VI, the Lower Provinces of Bengal and Their Feudatories, Report. Calcutta: Bengal Secretariat Press. ———. 1902b. Census of India, 1901, Vol. VIa, Bengal, Part II, Tables. Calcutta: Bengal Secretariat Press. ———. 1913a. Census of India, 1911, Vol. I – India, Part I – Report. Calcutta: Superintendent Government Printing. ———. 1913b. Census of India, 1911, Vol. I - India, Part II - Tables. Calcutta: Superintendent Government Printing. Gandhi, M. K. 1925. “Hindu–Muslim Tension: Its Cause and Cure.” Young India 6(22): 173–83. ———. 1933a. “Students and Vacation.” Harijan 1(8): 5. ———. 1933b. “Disposal of Carcasses.” Harijan 1(6): 3. ———. 1934a. “Gandhi’s Alleppy Speech.” Harijan 1(51): 7–8.



———. 1934b. “Notes: A Sevak’s Difficulty.” Harijan 2(36): 288. ———. 1934c. “How They Should Act.” Harijan 2(10). ———. 1934d. “To Harijan Sceptics.” Harijan 1(52): 6–7. ———. 1936. “The Ideal Bhangi.” Harijan 4(42): 336. ———. 1946a. “Animal Sacrifice.” Harijan 10(20). ———. 1946b. “Left-Handed Compliment.” Harijan 10(23): 224. ———. 1946c. “Sweepers’ Strike.” Harijan 10(11): 96. ———. 1946d. “A Harijan’s Letter.” Harijan 10(14). ———. 1966a. The Collected Works of Mahatma Gandhi, Vol. 19. Ahmedabad: Government of India. ———. 1966b. The Collected Works of Mahatma Gandhi, Vol. 21. Ahmedabad: Government of India. ———. 1971. The Collected Works of Mahatma Gandhi, Vol. 48. Ahmedabad: Government of India. ———. 1972a. The Collected Works of Mahatma Gandhi, Vol. 50. Ahmedabad: Government of India. ———. 1972b. The Collected Works of Mahatma Gandhi, Vol. 49. Ahmedabad: Government of India. ———. 1972c. The Collected Works of Mahatma Gandhi, Vol. 51. Ahmedabad: Government of India. ———. 1972d. The Collected Works of Mahatma Gandhi, Vol. 52. Ahmedabad: Government of India. ———. 1973a. The Collected Works of Mahatma Gandhi, Vol. 56. Ahmedabad: Government of India. ———. 1973b. The Collected Works of Mahatma Gandhi, Vol. 54. Ahmedabad: Government of India. ———. 1976a. The Collected Works of Mahatma Gandhi, Vol. 64. Ahmedabad: Government of India. ———. 1976b. The Collected Works of Mahatma Gandhi, Vol. 66. Ahmedabad: Government of India. ———. 1981. The Collected Works of Mahatma Gandhi, Vol. 84. Ahmedabad: Government of India. ———. 1983. The Collected Works of Mahatma Gandhi, Vol. 87. Ahmedabad: Government of India. Gazetteer of the Gurdaspur District, 1891–92. 1892. Lahore: Civil and Military Gazette Press. Geertz, Clifford. 1973. The Interpretation of Cultures. New York: Basic Books.

References 314

Geertz, Hildred. 1975. “An Anthropology of Religion and Magic, I.” The Journal of Interdisciplinary History 6(1): 71–89. Ghassem-Fachandi, Parvis. 2012. Pogrom in Gujarat: Hindu Nationalism and Anti-Muslim Violence in India. Princeton: Princeton University Press. Goldman, Robert P. 1976. “Vālmīki and the Bhṛgu Connection.” Journal of the American Oriental Society 96(1): 97–101. Goldziher, Ignatius. 1906. “Das Prinzip Der Takija Im Islam.” Zeitschrift der Deutschen Morgendlandischen Gesselschaft 60: 213–26. Gooptu, Nandini. 2001. The Politics of the Urban Poor in Early Twentieth-Century India. Cambridge: Cambridge University Press. Gopalaswami, L. N. 1934. “Tamilnad Report for June and July, 1934.” Harijan 2(30): 239–40. Gordon, Andrew. 1886. Our India Mission: A Thirty Years’ History of the India Mission of the United Presbyterian Church of North America Together with Personal Reminiscences. Philadelphia: A. Gordon. Gottschalk, Peter. 2000. Beyond Hindu and Muslim: Multiple Identity in Narratives from Village India. New York: Oxford University Press. ———. 2013. Religion, Science, and Empire: Classifying Hinduism and Islam in British India. New York: Oxford University Press. Gould, William. 2005. Hindu Nationalism and the Language of Politics in Late Colonial India. Cambridge: Cambridge University Press. Graham, James Reid. 1943. “The Arya Samaj as a Reformation in Hinduism with Special Reference to Caste.” Ph.D. Dissertation, Religion, Yale University. Greeven, Richard. 1894. The Knights of the Broom: An Attempt to Collect and to Explain Some of the Ceremonies of the Sweepers in the Benares Division. Benares: Medical Hall Press. Griswold, Hervey DeWitt. 1934. Insights into Modern Hinduism. New York: Henry Holt and Company. ———. N.d. “The Admission of Village Christians to the Lord’s Supper.” (Burke Library, Union Theological Seminary, New York.) Gudavarthy, Ajay. 2019. “The Crisis in Dalit-Bahujan Politics.” The Hindu, Delhi, November 4, p. 10. Guha, Ranajit. 2010. The Small Voice of History. Delhi: Orient Blackswan. Guha, Sumit. 2013. Beyond Caste: Identity and Power in South Asia, Past and Present. Leiden: Brill. Gupta, Charu. 2016. The Gender of Caste: Representing Dalits in Print. Seattle: University of Washington Press.



Gupta, Dipankar. 2005. “Caste and Politics: Identity over System.” Annual Review of Anthropology 34: 409–27. Guru, Gopal. 2009a. “Archaeology of Untouchability.” Economic and Political Weekly 44(37): 49–56. ———, ed. 2009b. Humiliation: Claims and Context. New Delhi: Oxford University Press. Hacker, Paul. 1978. “Aspects of Neo-Hinduism as Contrasted with Surviving Traditional Hinduism.” In Kleine Schriften, edited by L. Schmidthausen, pp. 580–608. Wiesbaden: Franz Steiner. Hacking, Ian. 1985. “Making up People.” In Reconstructing Individualism, edited by H. M. Sosna and D. E. Wellbery, pp. 222–36. Stanford: Stanford University Press. Hansen, Thomas Blom. 1999. The Saffron Wave: Democracy and Hindu Nationalism in Modern India. New Delhi: Oxford University Press. Hasan, Mushirul. 1981. Communal and Pan-Islamic Trends in Colonial India. New Delhi: Manohar. ———. 2007. From Pluralism to Separatism: Qasbas in Colonial Awadh. Delhi: Oxford University Press. Hasan, Zoya. 1996. “Communal Mobilization and Changing Majority.” In Making India Hindu: Religion, Community, and the Politics of Democracy in India, edited by D. Ludden, pp. 81–97. Delhi: Oxford University Press. Hawley, John Stratton. 1981. At Play with Krishna: Pilgrimage Dramas from Brindavan. Princeton: Princeton University Press. ———. 1991. “Naming Hinduism.” The Wilson Quarterly 15(3): 20–34. Hegel, G. W. F. 2018. The Phenomenology of Spirit. Translated by T. Pinkard and M. Baur. Cambridge: Cambridge University Press. Hermansen, Marcia. 2008. “A Twentieth Century Indian Sufi Views Hinduism: The Case of Khwaja Hasan Nizami (1879–1955).” Comparative Islamic Studies 4(1–2): 157–79. Hill, Jane H. 2008. The Everyday Language of White Racism. West Sussex: WileyBlackwell. Hubert, Henri and Marcel Mauss. 1964. Sacrifice: Its Nature and Function. Translated by W. D. Halls. Chicago: University of Chicago Press. Ibbetson, Denzil. 1883. “Bala Shah – Lal Beg – Aronakarit.” Panjab Notes and Queries 1(2). ———. 1970. Panjab Castes, Being a Reprint of the Chapter on “the Races, Castes, and Tribes of the People” in the Report on the Census of the Panjab. Patiala: Languages Dept., Punjab.

References 316

Ilaiah, Kancha. 2005. Why I Am Not a Hindu: A Sudra Critique of Hindutva Philosophy, Culture and Political Economy. Kolkata: Bhaktal and Sen. Imad-ud-Din. 1885. A Mohammedan Brought to Christ: Being the Autobiography of the Rev. Imad-Ud-Din, D.D., of the Church Missionary Society’s Mission at Amritsar, North India, and Honorary Chaplain to the Lord Bishop of Lahore. Edited by C. M. Society. London: Church Missionary House, Salisbury Square. Irudayam S. J., Aloysius, Jayshree P. Mangubhai, and Joel G. Lee. 2011. Dalit Women Speak Out: Caste, Class and Gender Violence in India. Delhi: Zubaan. Jaffrelot, Christophe. 2005. Dr. Ambedkar and Untouchability: Analysing and Fighting Caste. Delhi: Permanent Black. ———, ed. 2007. Hindu Nationalism: A Reader. Ranikhet: Permanent Black. Jangam, Chinnaiah. 2017. Dalits and the Making of Modern India. New Delhi: Oxford University Press. Jaoul, Nicolas. 2007. “Dalit Processions: Street Politics and Democratization in India.” In Staging Politics: Power and Performance in Asia and Africa, edited by J. C. Strauss and D. B. C. O’Brien, pp. 173–93. London: I. B. Taurus. ———. 2011. “Casting the ‘Sweepers’: Local Politics of Sanskritisation, Caste and Labour.” In Cultural Entrenchment of Hindutva: Local Mediations and Forms of Convergence, edited by D. Berti, N. Jaoul, and P. Kanungo, pp. 273–306. New Delhi: Routledge. ———. 2012. “The Making of a Political Stronghold: A Dalit Neighbourhood’s Exit from the Hindu Nationalist Riot System.” Ethnography 13(1): 102–16. Jha, Vivekanand N. 1975. “Stages in the History of Untouchability.” 2(1): 14–31. ———. 1986. “Candala and the Origin of Untouchability.” The Indian Historical Review 13(1–2): 1–36. Jones, Kenneth W. 1968. “Communalism in the Punjab: The Arya Samaj Contribution.” Journal of Asian Studies 28(1): 39–54. ———. 1976. Arya Dharm: Hindu Consciousness in 19th-Century Punjab. Berkeley: University of California Press. ———. 1981. “Religious Identity and the Indian Census.” In The Indian Census: New Perspectives, edited by N. G. Barrier, pp. 72–101. Delhi: Manohar. Jordens, J. T. F. 1977. “Reconversion to Hinduism, the Shuddhi of the Arya Samaj.” In Religion in South Asia: Religious Conversion and Revival Movements in South Asia in Medieval and Modern Times, edited by G. A. Oddie, pp. 140–59. Columbia, MO: South Asia Books. ———. 1981. Swami Shraddhananda: His Life and Causes. Delhi: Oxford University Press.



Josephson, Jason Ānanda. 2012. The Invention of Religion in Japan. Chicago: University of Chicago Press. Juergensmeyer, Mark. 2009. Religious Rebels in the Punjab: The Ad Dharm Challenge to Caste. New Delhi: Navayana. Kaul, Pandit Harikrishan. 1911. Census of India 1911, Volume Xiv: Punjab, Part I, Report. Lahore: Civil and Military Gazette Press. Kaur, Ravinder. 2008. “Narrative Absence: An ‘Untouchable’ Account of Partition Migration.” Contributions to Indian Sociology 42(2): 281–306. Kaviraj, Sudipta. 1992. “The Imaginary Institution of India.” In Subaltern Studies VI: Writings on South Asian History and Society, edited by P. Chatterjee and G. Pandey, pp. 1–39. Delhi: Oxford University Press. ———. 2010. “On Thick and Thin Religion: Some Critical Reflections on Secularisation Theory.” In Religion and the Political Imagination, edited by I. Katznelson and G. Stedman Jones, pp. 336–55. Cambridge: Cambridge University Press. Keane, Webb. 2007. Christian Moderns: Freedom and Fetish in the Mission Encounter. Berkeley: University of California Press. Kenoyer, Jonathan M. 1998. Ancient Cities of the Indus Valley Civilization. Karachi: Oxford University Press. Khan, Dominique-Sila. 2004. Crossing the Threshold: Understanding Religious Identities in South Asia. London: I. B. Taurus. Khator, Neha. 2017. “The Stink from India’s Past.” The Hindu, June 4, Chennai. Kinsley, David. 1993. Hinduism: A Cultural Perspective. Englewood Cliffs: Prentice Hall. Kipling, Rudyard. 1887. “The Vengeance of Lal Beg.” Civil and Military Gazette, November 3, Lahore. ———. 1901. Kim. New York: Macmillan & Co. Kockelman, Paul. 2007. “Agency: The Relation between Meaning, Power, and Knowledge. ” Current Anthropology 48: 375–401. Kohlberg, Etan. 1975. “Some Imami-Shi’i Views on Taqiyya.” Journal of the American Oriental Society 95(3): 395–402. Kolenda, Pauline. 2003. Caste, Marriage, and Inequality: Essays on North and South India. New Delhi: Rawat Publications. Kolff, D. H. A. 2002. Naukar, Rajput, and Sepoy: The Ethnohistory of the Military Labour Market in Hindustan, 1450–1850. Cambridge: Cambridge University Press. Kolge, Nishikant. 2017. Gandhi against Caste. New Delhi: Oxford University Press.

References 318

Kothari, Rajni. 2009. Politics in India. Hyderabad: Orient Blackswan. Kumar, Dev. 2004. Hāṅ, Hāṅ, Hāṅ, Maiṅ Bhangī Hūṅ. Kanpur: Prabhu Prakashan. Lawrence, Bruce. 1984. “Early Indo-Muslim Saints and Conversion.” In Islam in Asia, Vol. I, edited by Y. Friedman, pp. 109–45. Colorado: Westview Press. Lee, Joel. 2008. “‘It’s the Story of Our Own Village’: A Journey in Indian Street Theater.” Theater 38(1): 73–87. ———. 2014. “Lal Beg Underground: The Passing of an ‘Untouchable’ God.” In Objects of Worship in South Asian Religions: Forms, Practices, and Meanings, edited by K. A. Jacobseen, M. Aktor, and K. Myrvold, pp. 143–62. New York: Routledge. ———. 2015. “Jagdish, Son of Ahmad: Dalit Religion and Nominative Politics in Lucknow.” South Asia Multidisciplinary Academic Journal 11. ———. 2017. “Odor and Order: How Caste Is Inscribed in Space and Sensoria.” Comparative Studies of South Asia, Africa and the Middle East 37(3): 470–90. ———. 2018. “Who Is the True Halalkhor? Genealogy and Ethics in Dalit Muslim Oral Traditions.” Contributions to Indian Sociology 52(1): 1–27. ———. 2019. “All the Valmikis Are One: Bhakti as Majoritarian Project.” In Bhakti and Power: Debating India’s Religion of the Heart, edited by J. S. Hawley, C. L. Novetzke, and S. Sharma, pp. 74–82. Hyderabad: Orient Blackswan. ———. Forthcoming. “Disgust and Untouchability: Notes toward an Affective Theory of Caste.” South Asian History and Culture. Lelyveld, Joseph. 2011. Great Soul: Mahatma Gandhi and His Struggle with India. New York: Alfred A. Knopf. Leslie, Julia. 2003. Authority and Meaning in Indian Religions: Hinduism and the Case of Vālmīki. Hants, UK: Ashgate. Llewellyn, J. E. 1993. The Arya Samaj as a Fundamentalist Movement: A Study in Comparative Fundamentalism. Delhi: Manohar. Lord, Henry. 1630. A Display of Two Forraigne Sects in the East Indies. London: T. and R. Cotes. Lorenzen, David N. 2006. Who Invented Hinduism? Essays on Religion and History. New Delhi: Yoda Press. Lynch, Owen. 1969. The Politics of Untouchability: Social Mobility and Social Change in a City of India. New York: Columbia University Press. ———. 1981. “Rioting as Rational Action: An Interpretation of the April 1978 Riots in Agra.” Economic and Political Weekly 16(48): 1951–56.



———. 1990. “The Mastrām: Emotion and Person among Mathura’s Chaubes.” In Divine Passions: The Social Construction of Emotion in India, edited by O. Lynch, pp. 91–115. Berkeley: University of California Press. ———. 2012. “Ambedkar Jayanti, Hierarchy and the Darshan Effect.” In Dalit Art and Visual Imagery, edited by G. Tartakov, ch. 8. New York: Oxford University Press. Malkani, Naraindas Rattanmal. 1965. Clean People and an Unclean Country. Delhi: Harijan Sevak Sangh. Mall, Chaina. 1886. “Musallis – Nau Musallim – Initiation Ceremony.” Panjab Notes and Queries 3(33): 146. Mann, Gurinder Singh. 2004. Sikhism. New Jersey: Prentice Hall. Mariuma, Yarden. 2014. “Taqiyya as Polemic, Law and Knowledge: Following an Islamic Legal Term through the Worlds of Islamic Scholars, Ethnographers, Polemicists and Military Men.” The Muslim World 104: 89–108. Masuzawa, Tomoko. 2005. The Invention of World Religions: Or, How European Universalism Was Preserved in the Language of Pluralism. Chicago: University of Chicago Press. Mazzarella, William. 2010. “Branding the Mahatma: The Untimely Provocation of Gandhian Publicity.” Cultural Anthropology 25(1): 1–39. McGregor, R. S. 1993. Oxford Hindi–English Dictionary. New Delhi: Oxford University Press. Menon, Dilip M. 2010. “The Blindness of Insight: Why Communalism in India Is about Caste.” In Indian Political Thought: A Reader, edited by A. Singh and S. Mohapatra, pp. 123–35. London: Routledge. Messmore, J. H. 1903. The Life of Edwin Wallace Parker, D. D., Missionary Bishop of Southern Asia, Forty-One Years a Missionary in India. New York: Eaton & Mains. Mhaskar, Sumeet. 2018. “Ghettoisation of Economic Choices in a Global City: A Case Study of Mumbai.” Economic and Political Weekly 53(29): 29–37. Mitchell, Lisa. 2014. “The Visual Turn in Political Anthropology and the Mediation of Political Practice in Contemporary India.” South Asia: Journal of South Asian Studies 37(3): 515–40. Mohan, P. Sanal. 2015. Modernity of Slavery: Struggles against Caste Inequality in Colonial Kerala. New Delhi: Oxford University Press. Monier-Williams, Monier. 2004. Sanskrit–English Dictionary: Etymologically and Philologically Arranged with Special Reference to Cognate Indo-European Languages. New Delhi: Munshiram Manoharlal.

References 320

Muhammad, Ghulam. 1973. Hayat-e Bahadur Yar Jang. Karachi: Bahadur Yar Jang Academy. Nagar, Amritlal. 2017. Nāchyau Bahut Gopāl. Delhi: Rajpal and Sons. Nanda, B. R. 1985. Gandhi and His Critics. New Delhi: Oxford University Press. Nārada. 1989a. The Nāradasmṛti: A Critical Edition, Part Two: Translation. Translated by R. W. Lariviere. Philadelphia: University of Pennsylvania Press. ———. 1989b. The Nāradasmṛti: A Critical Edition, Part One: Text. Translated by R. W. Lariviere. Philadelphia: University of Pennsylvania Press. Narain Din, Master. N.d. Hela-Halalkhor. Allahabad: Shri Sahabdin. Narayan, Badri. 2009. Fascinating Hindutva: Saffron Politics and Dalit Mobilisation. New Delhi: Sage. Natrajan, Balmurli. 2012. The Culturalization of Caste in India: Identity and Inequality in a Multicultural Age. New York: Routledge. Nehru, Rameshwari. 1940. The Harijan Movement. New Delhi: Hindustan Times Press. Nevill, H. R. 1904. Bara Banki: A Gazetteer. Being Volume Xlviii of the District Gazetteers of the United Provinces of Agra and Oudh. Allahabad: Government Press. Nietzsche, Friedrich. 1989. On the Genealogy of Morals. Translated by W. Kaufman and R. J. Hollingdale. New York: Vintage Books. Nizami, Khwaja Hasan. 1923. Ḥalālkhor. Dilli: Halqa-e Mashaikh Book Depot. O’Brien, John. 2012. The Unconquered People: The Liberation Journey of an Oppressed Caste. Karachi: Oxford University Press. O’Hanlon, Rosalind. 1985. Caste, Conflict, and Ideology: Mahatma Jotirao Phule and Low Caste Protest in Nineteenth-Century Western India. Cambridge: Cambridge University Press. Oberoi, Harjot. 1997. The Construction of Religious Boundaries: Culture, Identity, and Diversity in the Sikh Tradition. Delhi: Oxford University Press. OED Online. 2020a. “function, n.” June, Oxford University Press. Available at hxNyTP&result=1 (accessed July 14, 2020). ———. 2020b. “identity, n. ” June, Oxford University Press. Available at https:// identity (accessed August 20, 2020). ———. 2020c. “subterfuge, n.” June. Oxford University Press. Available at EcHHzN&result=1 (accessed July 14, 2020).



Oldenburg, Veena Talwar. 2010. The Making of Colonial Lucknow: 1856–1877. In The Lucknow Omnibus. Delhi: Oxford University Press. Paik, Shailaja. 2011. “Mahar–Dalit–Buddhist: The History and Politics of Naming in Maharashtra.” Contributions to Indian Sociology 45(2): 217–41. ———. 2014. Dalit Women’s Education in Modern India: Double Discrimination. New York: Routledge. Pandey, Gyanendra, ed. 1993. Hindus and Others: The Question of Identity in India Today. New Delhi: Penguin. ———. 2002. The Ascendancy of the Congress in Uttar Pradesh: Class, Community and Nation in Northern India, 1920–1940. London: Anthem Press. Pant, Govind Ballabh. 2000a. Selected Works of Govind Ballabh Pant, Vol. 13. Edited by B. R. Nanda. New Delhi: Oxford University Press. ———. 2000b. Selected Works of Govind Ballabh Pant, Vol. 14. Edited by B. R. Nanda. New Delhi: Oxford University Press. Patwardhan, Anand, dir. 2011. Jai Bhim Comrade. Pennington, Brian K. 2005. Was Hinduism Invented? Britons, Indians, and the Colonial Construction of Religion. New York: Oxford University Press. Pernau, Margrit. 2000. The Passing of Patrimonialism: Politics and Political Culture in Hyderabad 1911–1948. Delhi: Manohar. Peirce, Charles Sanders. 1955. Philosophical Writings of Peirce. Edited by Justus Buchler. New York: Dover Publications. Pinkney, Andrea Marion. 2013. “Prasāda, the Gracious Gift, in Contemporary and Classical South Asia.” Journal of the American Academy of Religion 81(3): 734–56. Platts, John T. 2004 [1884]. A Dictionary of Urdu, Classical Hindi and English. Delhi: Munshiram Manoharlal. Plowden, W. Chichele. 1883. Report on the Census of British India, Taken on 17 February 1881, Vol. I. London: Eyre and Spottiswoode. Povinelli, Elizabeth. 2002. The Cunning of Recognition: Indigenous Alterities and the Making of Australian Multiculturalism. Durham: Duke University Press. Powell, Avril A. 1993. Muslims and Missionaries in Pre-Mutiny India. London: Routledge. ———. 2003. “‘Pillar of a New Faith’: Christianity in Late-Nineteenth-Century Punjab from the Perspective of a Convert from Islam.” In Christians and Missionaries in India: Cross-Cultural Communication since 1500, edited by R. E. Frykenberg, pp. 223–55. London: Routledge Curzon. Prakash, Gyan. 1991. “Becoming a Bhuinya: Oral Traditions and Contested Domination in Eastern India.” In Contesting Power: Resistance and Everyday

References 322

Social Relations in South Asia, edited by D. E. Haynes and G. Prakash, pp. 145–70. Berkeley: University of California Press. Prashad, Vijay. 1996. “The Untouchable Question.” Economic and Political Weekly 31(9): 551–59. ———. 2000. Untouchable Freedom: A Social History of a Dalit Community. Delhi: Oxford University Press. Puri, J. R. and T. R. Shangari, eds. 1986. Bulleh Shah: The Love-Intoxicated Iconoclast. Amritsar: Radha Soami Satsang Beas. Pyarelal. 1932. The Epic Fast. Ahmedabad: Mohanlal Maganlal Bhatt. Raheja, Gloria Goodwin. 1988. The Poison in the Gift: Ritual, Prestation, and the Dominant Caste in a North Indian Village. Chicago: University of Chicago Press. Rai, Lajpat. 1915. The Arya Samaj: An Account of Its Origins, Doctrines, and Activities, with a Biographical Sketch of the Founder. London: Longmans, Green and Co. Ramberg, Lucinda. 2014. Given to the Goddess: South Indian Devadasis and the Sexuality of Religion. Durham: Duke University Press. Ramaswamy, Gita. 2005. India Stinking: Manual Scavengers in Andhra Pradesh and Their Work. Delhi: Navayana. Ramchandra. 1933. “Mysore Report for March.” Harijan 1(14): 7–8. Rao, Anupama. 2009. The Caste Question: Dalits and the Politics of Modern India. Berkeley: University of California Press. ———, ed. 2003. Gender and Caste. Delhi: Kali for Women. Ravan, Darshan Ratan. 2010. Ambedkar Se Vimukh Safai Kamgar Samaj. New Delhi: Adhas Samvad. Rawat, Ramnarayan S. 2011. Reconsidering Untouchability: Chamars and Dalit History in North India. Bloomington: University of Indiana Press. ———. 2015. “Genealogies of the Dalit Political: The Transformation of Achhut from ‘Untouched’ to ‘Untouchable’ in Early Twentieth-Century North India.” The Indian Economic and Social History Review 52(3): 335–55. Rege, Sharmila, ed. 2013. Against the Madness of Manu: B. R. Ambedkar’s Writings on Brahmanical Patriarchy. Delhi: Navayana. Richman, Paula, ed. 1991. Many Ramayanas: The Diversity of a Narrative Tradition in South Asia. Berkeley: University of California Press. Ring, Laura. 2006. Zenana: Everyday Peace in a Karachi Apartment Building. Bloomington: Indiana University Press. Risley, Herbert Hope. 1902. Census of India 1901: Report. Vol. I. Calcutta: Superintendent Government Printing.

323 References

Roberts, Nathaniel. 2016. To Be Cared For: The Power of Conversion and Foreignness of Belonging in an Indian Slum. Oakland: University of California Press. Rose, Horace Arthur. 1902. Census of India 1901, Volume XVII: Punjab and North-West Frontier Proovince, Part I, Report. Simla: Government Central Printing Office. Roy, Satcowripati. 1933. “Bengal Report for May.” Harijan 1(23): 1–2. Rupkala, Sitaramsaran Bhagavanprasad. 1962 [1910]. Śrī Bhaktamāl with the Bhaktirasabodhinī Commentary of Priyādās. Lucknow: Tejkumar Press. Russell, R. V. and Rai Bahadur Hira Lal. 1916. The Tribes and Castes of the Central Provinces of India, Vol. I. London: Macmillan and Co. Sahdev, Manjula. 1997. Maharishi Valmiki: Vyaktitva evam Kritatva. Patiala: Punjab University Press. Sastri, R. V. 1933a. “A Day’s Work.” Harijan 1(23): 2–3. ———. 1933b. “Week to Week.” Harijan 1(8). Scott, James C. 1987. Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven: Yale University Press. ———. 2012. “Infrapolitics and Mobilizations: A Response by James C. Scott.” Revue française d’études américaines 131(1): 112–17. Searle-Chatterjee, Mary. 1981. Reversible Sex Roles: The Special Case of Benares Sweepers. Oxford: Pergamon Press. Searle-Chatterjee, Mary. 2008. “Attributing and Rejecting the Label ‘Hindu’ in North India.” In Religion, Language, and Power, edited by N. Green and M. Searle-Chatterjee, pp. 186–201. New York: Routledge. Shah, Waris. 1966. The Adventures of Hir and Ranjha: A Translation into English Prose from the Punjabi Poem. Translated by C. F. Usborne. Karachi: Karachi Lion Art Press. Sharma, Amichand. 1928. Śrī Bālmīki Prakāś. Delhi: Gautam Book Centre. Sharma, C. B. 1934. “Rajputana Report for June 1934.” Harijan 2(26): 201. Sharma, Rama. 1995. Bhangi, Scavenger in Indian Society: Marginality, Identity and Politicization of the Community. New Delhi: MD Publications. Sharma, Satish Kumar. 1985. Social Movements and Social Change: A Study of Arya Samaj and Untouchables in Punjab. Delhi: B.R. Publishing Corporation. Sharma, Shriram. N.d. Patitoddhār. Shraddhanand, Sanyasi. 1926. Hindu Sangathan: Saviour of the Dying Race. Delhi: Arjun Press. ———. 1946. Inside the Congress. New Delhi: Dayanand Sansthan.

References 324

———. 1987a. Swami Shraddhanand Granthawali, Vol. XI. Delhi: Govindram Hasanand. ———. 1987b. Swami Shraddhanand Granthawali, Vol. VI. Edited by B. Bharatiya. Delhi: Govindram Hasanand. Shyamlal. 1984. The Bhangis in Transition. New Delhi: Inter-India Publications. ———. 1997. From Higher Caste to Lower Caste: The Process of Asprashyeekaran and the Myth of Sanskritization. Jaipur: Rawat Publications. ———. 2016. Politics of the Untouchables: Congress and the Bhangis: A SocioPolitical Analysis. Jaipur: Rawat Publications. Sikand, Yoginder. 2003. “Arya Shuddhi and Muslim Tabligh: Muslim Reactions to Arya Samaj Proselytization (1923–30).” In Religious Conversion in India: Modes, Motivations, and Meanings, edited by R. Robinson and S. Clarke, pp. 98–118. New Delhi: Oxford University Press. ———. 2004. “Sikh–Muslim Harmony: Contributions of Khwaja Hasan Nizami.” Economic and Political Weekly 39(11): 1113–16. Simmel, Georg. 1950. The Sociology of Georg Simmel. Translated by K. H. Wolff. New York: Free Press. Simpson, Audra. 2014. Mohawk Interruptus: Political Life across Borders of Settler States. Durham: Duke University Press. Singh, Beohar Rajendra. 1934. “Mahakoshal (C.P. Hindi) Report for June and July 34.” Harijan 2(35): 279. Singh, Bhai Maya. 1982 [1895]. The Panjabi Dictionary. New Delhi: Amar Prakashan. Singh, Bhrigupati. 2015. Poverty and the Quest for Life: Spiritual and Material Striving in Rural India. Chicago: University of Chicago Press. Spivak, Gayatri. 1981. “‘Draupadi’ by Mahasveta Devi.” Critical Inquiry 8(2): 381–402. Srinivas, M. N. 1996. Village, Caste, Gender and Method: Essays in Indian Social Anthropology. Delhi: Oxford University Press. Stewart, Tony and Carl Ernst. 2003. “Syncretism.” In South Asian Folklore: An Encyclopedia, edited by P. J. Claus and M. A. Mills, pp. 586–88. New York: Garland Publishing. Stietencron, Heinrich von. 2007. “Religious Configurations in Medieval India and the Modern Concept of Hinduism.” In The Oxford India Hinduism Reader, edited by V. Dalmia and H. v. Stietencron, pp. 51–69. New Delhi: Oxford University Press. Streefland, Pieter. 1979. The Sweepers of Slaughterhouse: Conflict and Survival in a Karachi Neighbourhood. Assen: Van Gorcum.



Strickler, Herbert Johnson. 1926. “The Religion and Customs of the Chuhra in the Punjab Province, India.” Master’s thesis, Sociology, University of Kansas. Subramanian, Ajantha. 2019. The Caste of Merit: Engineering Education in India. Cambridge: Harvard University Press. Sundar, Chandrasen. 1933. “U.P. Report for April and the First Half of May.” Harijan 1(16): 6–7. Tagore, Rabindranath. 1933a. “Love’s Gold.” Harijan 1(16). ———. 1933b. “Sweet Mercy.” Harijan 1(15). ———. 1933c. “The Cleanser (a Free Rendering from the Bengali of Satyendranth Datta’s ‘The Scavenger’).” Harijan 1(1): 1. ———. 1933d. “The Sacred Touch.” Harijan 1(7). Talbot, Cynthia. 1995. “Inscribing the Other, Inscribing the Self: Hindu–Muslim Identities in Pre-Colonial India.” Comparative Studies in Society and History 37(4): 692–722. Taneja, Anand Vivek. 2018. Jinnealogy: Time, Islam, and Ecological Thought in the Medieval Ruins of Delhi. Stanford: Stanford University Press. Tarde, Gabriel de. 1903. The Laws of Imitation. Translated by E. C. Parson. New York: H. Holt and Company. Taylor, Charles. 2007. A Secular Age. Cambridge: Harvard University Press. Tejani, Shabnum. 2008. Indian Secularism: A Social and Intellectual History, 1890–1950. Bloomington: Indiana University Press. Temple, Richard Carnac. 1884. Legends of the Panjab, Vol. I. Bombay: Education Society’s Press. Thakkar, Amritlal V. 1946. “Harijan Sevak Sangh Report for the Two Years from 1st Oct. ’43 to 30th Sept. ’45.” Delhi: I. M. H. Press. Thapar, Romila. 1989. “Imagined Religious Communities? Ancient History and the Modern Search for a Hindu Identity.” Modern Asian Studies 23(2): 209–31. ———. 1996. “The Tyranny of Labels.” Social Scientist 24(9/10): 3–23. ———. 2005. Somanatha: The Many Voices of a History. London: Verso. Titus, Murray T. 1930. Indian Islam: A Religious History of Islam in India. Oxford: Oxford University Press. Urban, Hugh. 1998. “The Torment of Secrecy: Ethical and Epistemological Problems in the Study of Esoteric Traditions.” History of Religions 37(3): 209–48. Valmiki, Omprakash. 2003. Joothan: A Dalit’s Life. Translated by A. P. Mukherjee. Delhi: Samya. ———. 2008. Safāī Devtā. Delhi: Radhakrishna Prakashan.

References 326

———. 2010. Salaam. Delhi: Radhakrishna Paperbacks. Verma, Mukut Behari. 1971. History of the Harijan Sevak Sangh, 1932–1968. Delhi: Harijan Sevak Sangh. Virani, Shafique N. 2011. “‘Taqiyya’ and Identity in a South Asian Community.” The Journal of Asian Studies 70(1): 99–139. Viswanath, Rupa. 2014. The Pariah Problem: Caste, Religion, and the Social in Modern India. New York: Columbia University Press. Viswanathan, Gauri. 1998. Outside the Fold: Conversion, Modernity, and Belief. Princeton: Princeton University Press. Vološinov, V. N. 1973. Marxism and the Philosophy of Language. Translated by L. Matejka and I. R. Titunik. Cambridge: Harvard University Press. Warner, Michael. 2002. Publics and Counterpublics. New York: Zone Books. Webster, John C. B. 1992. The Dalit Christians: A History. Delhi: Indian Society for Promoting Christian Knowledge. ———. 2007. A Social History of Christianity: North-West India since 1800. New Delhi: Oxford University Press. ———. 2012. Historiography of Christianity in India. New Delhi: Oxford University Press. Weinreich, Max. 1945. “Der Yivo Un Di Problemen Fun Undzer Tsayt.” Yivobleter 25: 3–18. Weitbrecht, H. U. 1886. “Work among the Chuhras.” The Church Missionary Intelligencer and Record XI: 763–66. Wiser, William Henricks. 2000. Behind Mud Walls: Seventy-Five Years in a North Indian Village. Berkeley: University of California Press. Yelle, Robert A. 2012. The Language of Disenchantment: Protestant Literalism and Colonial Discourse in British India. Oxford: Oxford University Press. Youngson, John F. W. 1896. Forty Years of the Panjab Mission of the Church of Scotland, 1855–1895. Edinburgh: R & R Clark. ———. 1906. “The Chuhras.” The Indian Antiquary 35: 82–96, 302–10, 337–56. ———. 1907. “The Chuhras (Continued).” The Indian Antiquary 36: 19–31, 71–83, 106–16, 135–48. Zaidi, A. M., ed. 1987. Inc the Glorious Tradition, Vol. 1: 1885–1920. New Delhi: Indian Institute of Applied Political Research. Zavos, John. 2000. The Emergence of Hindu Nationalism in India. New Delhi: Oxford University Press. Zelliot, Eleanor. 2010. From Untouchable to Dalit: Essays on the Ambedkar Movement. New Delhi: Manohar.


Names marked with an asterisk (*) are pseudonyms. Ab-e Hayat, 71 Arya Samaj, 10–12, 45–46, 50, 78–80, Abhyuday, 103 99–100, 104, 108, 119, 123, 153, Abstinence Department (Madh 156–58, 160, 167, 169, 172, 189, Nishedh Vibhag), 168–69, 177 198–99, 206, 237, 292, 294, 301 Achhutanand, Swami, 33–34, 97 competition with Christian achhutodhhār (untouchable uplift), 10 missionaries, 85–86 Adcock, C. S., 49, 88, 128, 132 shuddhi or “purification” Ad Dharm, 24, 33–34, 134–35, 150 campaigns, 82–99, 114, Adi-Dravida, 24 118–20, 131–33 Adi Hindu Mahasabha, 97 Aurangzeb, Emperor, 284–85 Adi Hindu, 33 Awadh, 21–25, 55, 163–66, 275 Ahmad, Mirza Ghulam, 46, 72, 118 Azad, Muhammad Husain, 71 Ahmadiyya, 45 Al Biruni, 19 Baba, Khak Shah, 167–68, 303 All-India Shuddhi Sabha, 89 Babri Masjid dispute, 25 al-Tusi, Nasir al-Din Muhammad, Bacon, Francis, 218 268 Bahadur, Tegh, 284 Ambedkar, Bhimrao, 14–17, 20, Baleshahi, 160 24, 33–34, 38, 67, 77–78, 98, Bal Mik/Balmik (of Lal Begi oral 100, 124–31, 135–36, 147–51, tradition), 57–60, 109–10, 156–57, 161, 165, 291, 305 112–16, 173, 186–87, 303 Ambedkar Jayanti, 234–36 Balmiki, Achhe Lal, 178, 182, Amritsar Congress session of 1919, 84 197–202, 205–06, 246 Anand, Mulk Raj, 11, 147–48 Balmiki, Kanhaiya Lal, 165–73, Untouchable, 147–48 178–79, 187, 205, 207, 239, 279, Andrews, C. F., 84, 106 285, 289 Appadurai, Arjun, 10, 91 *Balmiki, Munnu, 179, 246–47, Article 17 of the Constitution, 205 279–83, 297–98 Arunthatiyar, 20, 122 Balmiki, Narain Din, 174–75, 177, Arya Patrika, 85 205, 207

Index 328

*Balmiki, Tara, 23, 247–48 Baniya, 122, 231 Bansphor (Basor), 18, 20 Bara Banki, 166, 178, 181, 187–88, 208, 258, 262, 273–74 barbers, 104, 205–06, 208, 262, 296 bartan samiti, 199 Benares (Varanasi), 6, 20, 32, 35, 44–45, 54, 61, 65, 70, 117, 131, 133 bhagaunās, 195–203 Bhaktamāl, 112 bhakti tradition, 80, 144 Bhangi, 17, 104, 121–24, 137–42, 160, 173 Bharatiya Janata Party (BJP), 170, 229–31, 293–94, 296–97 bifurcation paradigm, 41, 52, 73, 195 Birla, Ghanshyamdas, 12, 131–34, 151, 161, 169–70, 174, 294 brahmin, 6, 25–26, 36–37, 59, 83, 96, 102–03, 110–12, 145–46, 242, 245, 297, 304 broom (jhāṛū), 31, 57, 213, 232 “broom-people” (jhāṛū-wāle), 236, 239–40, 296 Butalia, Urvashi, 32–33, 35, 68 carrion, 45, 155–56 caste, 3–9, 13–16, 27–28, 56 Ambedkar’s formulation of, 14–15 and communalism, 9, 185–86, 206–08 nomenclature/titles, 16–18, 140 slavery, 74 sociability, 195–96 Catholicism, 51, 218–21

census of India, 3–9, 25–26, 42–43, 50, 69–70, 110–11, 125, 135–36, 222, 304 chai vendors, 206, 208, 289, 296 Chamar, 5–6, 20, 24, 86, 98, 122, 147, 295 Chandal (chānḍāla), 21, 73, 100, 111–15, 139 *Chaudhury, Amin, 200–02 Cherian, Divya, 73–74 Christianity, 46, 49, 53, 56, 78, 86, 96, 150, 157, 216, 218, 220, 224 Christopher, K. W., 148 Chuhra Christians, 46–47 Chuhra, 18–20, 34, 57, 216, 244, 284 clans (qabīle), 21 mass conversions, 45–47, 49, 52 oral traditions, 21, 64 Punjabi, 21, 42–46, 48, 109n4 treatment of, 68–72 Clark, Lynne, 268 Clean People and an Unclean Country, 159 cohabitation controversy, 179–86 Communal Award, 125–27, 129–31 concealment, 220–22, 255, 269–70, 272, 274, 278, 281–83 Congress Balmikis, 172, 207–08, 250 Conrad, Dieter, 162 Constitution (Scheduled Castes) Order of 1950, 11, 77, 161–62, 194 “Cult of Lal Beg,” 40, 55 Da’i-i Islam (The Missionary of Islam), 115 Dalit politics and religion, 23–25, 27, 30 127, 215–16, 233, 235, 288, 296, 298, 305

329 Index

Dalit, 10–12, 16–17, 78, 86 caste clusters, 19–20 religious categorization of, 35–39 conversion, 42–49 literature, 28–30 neighborhood or bastī, 3–4, 15, 19, 23, 25–26, 31, 39, 123, 150, 164, 169, 178, 193, 195, 213, 215, 227–30, 232, 234, 240, 243, 246, 255–57, 259, 261–65, 274–76, 279, 282, 287, 289, 291, 294, 297, 299–300 Dalmia, Vasudha, 36, 69, 82 Daniel, Valentine, 266 Das, Bhagwan, 58, 165 Datta, Satyendranath, 144 deception, 281–83. See also concealment, secret/secrecy declamatory religion, 216–27, 247–50, 257, 274, 296, 298 deg, 195–203, 208 Dehlawi, Sayyid Ahmad, 58 Delhi, 20, 24, 44, 54, 61, 94, 97, 99, 107–09, 115, 120, 123, 138, 140, 142, 150, 158–59, 166, 168, 231, 253–54, 284–85, 288, 292, 294, 301, 303, 305 Delhi Rowlatt Satyagraha, 84 Depressed Classes, 33, 79, 84, 91, 95, 99–100, 105–06, 108, 114, 122–32, 134, 136, 140, 142, 145, 149–58, 160–62, 285 Depressed Classes Resolution, 1917, 125, 128 Desai, Morarji, 175 Devi, Mahasweta, 36 Dewa Sharif, 21–22, 163, 280 Dhanuk, 18–21, 40, 230

dharma, 102–103, 143 dharmparivartan, 9 dharmaśāstra tradition, 73–74, 100, 139 Dhed, 122 diksha sanskar, 157–58 disgust. See ghṛṇā Dikshit, Madan, 15 disavowability, 279–83, 288, 299 dīkśā, 103 Dom (Domar), 18–20, 92 Dumna, 92 Dumont, Louis, 15, 37 endogamy, 15, 18, 165 Farhang-e Asafiya, 58 Favret-Saada, Jeanne, 254 food preparation, controversy over, 195–203 Forster, E. M., 148 Gait, E. A., 70 Gandhi, Mohandas, 11, 35, 77, 98, 120–24, 130, 171–72, 177, 208, 245, 280 Ad Dharm, 134–35 approach to untouchable uplift, 148–58 carrion-eating, 155–56 characterization of Arya Samajists, 132 “Harijan tour” of 1933–34, 135 mode of engagement with Depressed Class, 142 mode of majoritarian inclusion, 80 nature of Harijan-ness, 137, 150–51 representational practices, 134–44

Index 330

temple entry, 130–31 views on secrecy and publicity, 222–23 Gandhi, Rahul, 245 Gandhian nationalism, 79 Gazetteer of Poona, 50 Gazetteer of the Gurdaspur District, 48 gesture, 9, 13, 31, 213–16, 218–19, 232–33, 239, 249, 263–68, 278, 283, 293, 299 ghar wapsi or “homecoming” events, 164, 215 Ghazni, 58–61, 110 ghṛṇā (disgust), 80–82, 102, 104–06, 113–15, 295 Gordon, Andrew, 216–18, 250 Gottschalk, Peter, 40 Greeven, Richard, 51n4, 57, 60–61, 62n5, 65, 70 Griswold, Hervey DeWitt, 52 Gupta, Chandra Bhanu, 169, 205

Hind Sweepers’ Sevak Samaj (HSSS), 174 Hinduism, 3–4, 6, 10–11, 20, 23–24, 36, 39, 42, 48, 51 Hinduism: A Cultural Perspective (David Kinsley), 36 Hinduization, 12–13, 50, 78, 117, 151–58, 165, 179, 187, 190, 279–80, 292, 303 Hindu Mahasabha, 79, 156 Hindu majoritarianism, 3–4, 9–11, 24, 90–93, 99–100, 103–04, 114, 132–34, 152–58, 240, 274 Hindu nationalism. See Hindu majoritarianism Hindu Sangathan: Saviour of the Dying Race, 90–91 Home Rule, 96 Husain, Zakir, 174 Husen, Sayyid Altaf, 51n4, 57, 60–61, 65, 70

Hacking, Ian, 12 hadith, 267 Halalkhors, 17–18, 20, 45, 50, 71, 73, 94, 140, 176 Halalkhor (book), 94, 115–20 hard Hinduization, 156, 158, 160 Harijan (weekly), 35, 140–41, 143, 152–53 Harijan politics, 23–24, 165, 172, 207, 235, 250 Harijan Sevak Sangh, 11, 78, 100–01, 123, 127, 131–32, 134, 149–58 161–62, 171–72, 177, 291–93 Heer Ranjha, 57 Hela, 18, 20, 176 Hela-Halalkhor, 176

icon. See sign Ibbetson, Denzil, 57, 116 identity politics, 25–30. See also pahchān Ilaiah, Kancha, 34 Imam al-Din, Mirza, 118–19 index. See sign Indian Councils Act of 1909, 90 Indian National Congress, 24, 96, 100, 124–25, 134, 189, 199, 208, 244 intimate untouchability, 180 An Introduction to Hinduism (Gavin Flood), 36 Islam, 42, 49, 51, 78, 115, 150 izzat (honor, shame), 185–86 Jaffrelot, Christophe, 152 jāgarans, 23


jaisā rājā vaisī prajā, 53–54 Jamadar/jamādār, 17–18, 231–32, 239–40, 296 Jamghaṭ, 190–91, 214–15, 256–59, 261, 263–66, 277–79, 282–84, 289–90, 294, 298–301, 306 Jaoul, Nicolas, 165 Jatav, 24 jhāṅkī, 227–40, 250, 296–97 jhāṛū. See broom Jhaumpra, Khwaja, 59, 63, 66–67, 288 Joshi, Dr. Rita Bahuguna, 244–45 jāti, 14, 27–28, 101–03, 122 Jullundur committee of the Indian National Congress, 84


kāryakrams, 228, 240–50, 296–97 Kaul, Pandit Harikrishan, 42 Kaur, Ravinder, 34 Kaviraj, Sudipta, 90 Keane, Webb, 218 Khalji, Sultan Alauddin, 31, 176 Khatik, 237 Khizr, Khwaja, 62–63, 66 King George’s Medical College (KGMC), 170n3, 177, 227, 229, 262, 264, 294–96, 299 Kipling, Rudyard, 57 *Kumar, Rishi, 191–92, 262, 284–87, 290, 294

Lal Begi–Balmiki muqadma, 200–02, 215 Lal Begi, 8–10, 12, 16–17, 21, 26, 40–41, 48, 50, 52, 55–68, 77, 150, 194, 202, 205–06, 208, 254, 285, 301 oral traditions, 56, 82, 109–10, 156, 284, 286, 301 path of salvation, 65 religious practices, 163–64, 263–67 The Legends of the Panjab, 51 liquor, 155–56, 169, 171, 178–79, 196, 200, 208, 214, 216, 246, 257 Lucknow, 20, 31, 40, 79, 123, 163, 165, 194 bastīs of, 15, 19, 193, 195, 215, 257, 265, 279, 287, 294 cantonment, 21, 165–70, 177, 227 Chowk, 206, 227–29, 232–33, 235, 246, 248, 250–51, 263, 296 Daliganj, 231–32, 296 Hazratganj, 234 Lal Bagh, 228, 241, 243, 248–51, 296 Parivartan Chowk, 227–29, 232–33, 235, 246, 248, 250–51, 296 Rakabganj, 183–84 sanitation labor castes of, 20, 163–66 Valmiki movement in, 165–208

*Lal, Bhairav, 289–90 *Lal, Gauhar, 53–55, 194 Lal, Rai Bahadur Hira, 69 Lal Beg, 55–68, 186–91, 223, 253–54 rites of, 189–91, 258–63, 266 shrines, 60–61, 71, 117, 163, 188, 256–58, 263, 287, 299–300

MacDonald, Ramsay, 84, 126 Madiga, 20, 122 Mahar, 20, 24, 98, 147 majoritarian inclusion, 11–12, 15, 78–80, 82, 99, 103–04, 114, 149, 158–62, 165, 170, 215, 237, 254, 273, 284–85, 293, 295, 301

Index 332

majoritarianism, 3–4, 9–11, 24, 91, 240, 274, 290, 295, 298 Mala, 20, 122 Malaviya, Pandit Madan Mohan, 84, 127, 131 Malkana controversy, 93–94 Malkani, K. R., 159 Malkani, N. R., 151, 158–61 Mall, Chaina, 45 Mallah, 237 Mang, 20 manual scavenging, 48, 100, 117, 138, 151n10, 159, 245, 253, 301 marginal Hinduism, 72 Mass Movement Christians, 47 Masud, Sayyid Salar, 57 Mazhabi, 18, 20, 44, 46, 48, 254 meat, 22, 81, 156, 164, 179, 208, 263, 266 meat controversy, 195–203 Megh, 89, 92, 216 Mehtarani, Jumma, 22 Mehtar, 15–17, 108, 112, 140, 173 Modi, Narendra, 293 Mohan, Sanal, 30 Montagu-Chelmsford Reforms of 1918-19, 90 Moplah Rebellion of 1921, 93 Mukherji, Colonel U. N., 90 multiaccentuality (also see sign) 16–17, 240 musalmānā names and traditions, 53–54, 71, 163–64, 178, 186, 192, 194, 201, 203–04, 208, 254, 277, 282, 287 Muslim dafālī community, 263, 300 Muslims, 10–11, 24–25, 32, 34, 40, 50n3, 54, 61–63, 65–68, 73, 96, 99–100, 169

Namashudra, 24, 98, 122, 147 Nasir, Shah, 71, 287 Nehru, Rameshwari, 151, 154, 157, 161 Nizami, Khwaja Hasan, 79, 94 Halalkhor, 115–20 Od, 89, 92 oral traditions, 56, 81–82, 109–10, 117, 156, 284, 286, 301 otherness, 72–74 pahchān, 25–30, 33, 46, 54, 94, 165, 186, 200, 202, 215, 235, 239, 281, 295, 302 paighambar, 10 Paik, Shailaja, 204 Paine, Thomas, 38 panchayat or caste council, 21 Panch Sau Tirasi (583), 8, 21–23, 53, 163, 165–66, 192, 200 Pandey, Gyanendra, 34 Panjab Notes and Queries, 57, 116 Pant, Govind Ballabh, 169–170, 280 Paraiyar/Pariar, 20, 98, 122, 147 paternalism, 124 Patitoddhār (Upliftment of the Fallen), 100–08, 112, 114–15, 119, 176 Patit Prabhākar (The Fallen LightGiver), 100 patriarchy, 15, 181 Peirce, Charles Sanders, 16 Phule, Jotirao, 38 *Phulmati, 188, 260, 277 Poona Pact, 124–34, 147, 162 Prasad, Govind, 177, 199–200, 205, 208–09, 280, 283 Prasad, Lalta, 170–71, 178, 187, 195, 199–200, 204


Prashad, Vijay, 122, 165, 253 Protestants, 51, 56, 83, 85, 87, 198, 209, 217–22, 224–226, 233, 250, 257, 264, 266–67, 270–72, 275–76, 296 Pundri, 58–59 Punjab, 20, 23, 42–48, 52, 54–55, 64–65, 68, 70, 72, 83–86, 89, 91–92, 95, 97–98 Punjab Report of the Census of 1911, 42–43 Purusha Sukta of Rig Veda, 14 qaum, 10, 57, 65–66, 72–73, 77, 96, 117, 122, 165, 302 qawwāl (musician), 23, 242–243, 263 Rahtia, 86–89, 92, 95, 107 Rai, Lala Lajpat, 83, 88, 103, 131 Raidas (Ravidas), 19, 144–48 Rajah, M. C., 154 Rajput, 122 rakṣāmaṇḍal, 182–83, 185–86 *Ram, Daulat, 256–57, 277–78, 283, 299 Ram, Lala Munshi. See Shraddhanand, Swami Ram, Mangoo, 33, 98 Ram, Pandit Lekh, 85 Rangreta, 44 Rao, Shiva, 207 Rashtriya Swayamsevak Sangh (RSS), 159, 164, 215, 273–74 Rawat, Ramnarayan, 19 religious alterity, 10, 33–35, 38, 73–74, 110, 298 religious conversion, 42–50, 85, 133–34


as “ritual-political assertion,” 48–50 to Buddhism, 6, 33 to Christianity, 33, 46–48 to Hinduism (see shuddhi) to Islam, 33, 45 to Sikhism, 44–45 Rose, Horace Arthur, 42 Roy, Dr. Bidhan Chandra, 151 Russell, H. V., 69 sacrifice, 189–91, 266 Sadhu Yodharam (or Yodhnath), 107 sahdharmī, 102–04 sangathan, 131, 135 sanitation labor castes, 3, 7, 10, 12, 15–24, 138, 140–41, 160, 293 accessing education, 192–95, 204 as pivotal to Gandhian framing of untouchability 121–23 tolerance of intercaste cohabitation, 15, 180–81 collective designations, 15, 17–18 conversion among, 42–50 tradition of religious autonomy, 10, 31–35, 40, 55–68, 73–74, 117, 255–63, 279–80, 299–300, 303–05 participation in majoritarian violence, 24–25, 301 naming practices, 191–95, 204 shaharwāle and kanṭwāle, 165–66, 169–70, 176–78, 180, 187, 200, 207, 227, 279 Sanskritization, 165 sanātan dharmī or sanātanist, 83, 92 Saraswati, Swami Dayanand, 83 Sarvadeshik Arya Pratinidhi Sabha, 84

Index 334

Satnami , 70 Satnami Mahasabha, 97 Satyashodhak Samaj, 97 Scheduled Castes, 5, 17, 20, 26, 35, 141, 162, 193 Scheduled Castes Order of 1950, 11, 77, 161–62, 194 Scott, James, 13 secret/secrecy, 10, 44, 47, 57, 85, 180, 217, 220–25, 250, 254–55, 257, 267–75, 278, 289, 299 Sekra, 45 semiotic ideology, 198, 218–24, 226, 233, 249, 257, 266–68, 270, 275, 296, 299, 302 Shah, Bullhe, 27 Shah, Waris, 57 Shahi, 59, 62, 62n6, 63, 65. See also Lal Begi Sharma, Amichand, 101, 107–11, 119–20, 160, 167, 173, 187 Shastri, Lal Bahadur, 174 Shi’ism, 53, 223, 226, 267–72, 274–275, 285 in Awadh, 275 Shraddhanand, Swami, 79, 83–84, 86–90, 92–99, 107, 115, 120, 124–25, 131, 135, 149, 156, 158, 160, 172, 293 shrines concealment of, 255, 280–82 destruction of, 47, 188-189, 294–95 Lal Beg, 47, 60–62, 71, 117, 163, 188–89, 256–58, 263, 287, 299–300 Lal Mahadev, 287–88 misrecognition of, 71, 284, 287–90

of Nizamuddin Auliya, 94 of Dewa Sharif, 21 shuddhi, 45, 82–94, 98, 118–20, 124, 129, 132, 157, 164, 215 procedure, 84–85, 88 of Rahtias, 87–88 Shudra, 13, 96, 102, 107, 111, 122, 236–38, 245 sign icon, 214, 218–21, 226, 229, 231, 238–39, 265–66, 287, 295, 299, 302 index, 29, 38, 62–63, 88, 122, 140, 152, 168, 195, 222, 243, 264–65, 277, 295, 304 multiaccentual, 16–17, 240 symbol, 219, 229, 259, 265–66, 302 Sikh Chuhras, 43 Sikhism, 42–44, 49, 51 Sikhs, 35, 40, 44–46, 51, 55, 65, 86–87, 94,109, 118, 125–26, 129, 135, 302 Simmel, Georg, 224, 255, 269, 277–78 *Singh, Gulab 303–05 Singh, Guru Gobind, 284 soft Hinduization, 156, 158 Spivak, Gayatri, 36 Śrī Bālmīki Prakāś, 107–15, 167–68, 173, 244 Srinivas, M. N., 165 strikes, 48n2, 108, 119, 173–74, 289 Bombay sanitation workers’, 139 Calcutta sweepers’, 176 Gandhi’s opposition to, 139–40, 176, 293 Strickler, Herbert, 44 Sufi pirs, 31 syncretism, 50–55

335 Index

tabligh, 82, 94–95, 115, 118–19 Tagore, Rabindranath, 11, 35, 144, 146–47 bhakti cycle, 145 Talbot, Cynthia, 36 Taneja, Anand Vivek, 288 taqiyya, 53, 223, 267–75, 286–87 Tarde, Gabriel, 53 Taylor, Charles, 248 Temple, Richard, 51, 57, 65 temple desecration, 295 Thakkar, Amritlal, 151, 154 Thapar, Romila, 41 traditional occupations, 19 Tārikh al-Hind, 19 tyauhārs (holidays, festivals), 214 Uka story, 121–22, 147, 161 ummat, 10, 55–68, 72–73, 302 untouchability, 4, 12, 49, 66, 81, 86, 95–96, 98–102, 104–05, 107, 112, 121–25, 127–29, 131–32, 135, 143, 147–49, 152, 158, 161, 171–72, 174, 177, 180, 198, 204–07, 237, 273–74, 278, 296 myth of Muslim invention of, 99–100, 158–60 collective responsibility for, 128 curtailment of in Lucknow, 205–06 Untouchable (Mulk Raj Anand), 35 untouchables, representations of, 35–39 “untouchable uplift” programs, 79, 151–58

Uttar Pradesh, 3, 24, 194, 201n7, 207, 227, 230n2, 234–36, 238, 244–45, 256, 292, 305 Uttar Pradesh Balmiki Seva Samaj, 241 Valmiki, Omprakash, 28 Valmiki/Balmiki (caste title), 7–8, 16–17, 21, 24, 30, 60, 112, 301 Valmiki Jayanti, 13, 215–16, 227, 282 jhāṅkī, 227–40, 250, 296–97 kāryakrams, 228, 240–50, 296–97 Valmiki (Rishi), 11–12, 60, 109, 123, 165, 168, 214, 238, 243, 245 Valmiki Sabhas, 98 Valmikism (Vālmīkiyat), 177–79, 203–04, 207–08 Vankar, 122 varṇa, 13–14, 36, 122, 245, 304 violence of untouchability, 3, 13, 24–25, 34, 87, 92–93, 135, 147, 220, 231, 234–35, 236n3, 238 Vishwa Hindu Parishad (VHP), 164, 215 Viswanath, Rupa, 49 Vološinov, Valentin, 16 Washington, Booker T., 106 Weitbrecht, H. U., 45 Yelle, Robert, 218 Young India, 127, 132, 147, 222 Youngson, John, 62–64 zāt, 28 Zelliot, Eleanor, 122, 130, 152