Reflection on "To Be Cared For: The Power of Conversion and Foreignness of Belonging in an Indian Slum" Book by Nathaniel Roberts

N A T H A N I E L R O B E R T S

T o B e C d V e d F o v

T H E A N T H R O P O L O G Y O F C H R I S T I A N I T Y

Edited by Jo e l Robbins

To Be C ared For

T H E P O W E R O F C O N V E R S I O N A N D

F O R E I G N N E S S O F B E L O N G I N G I N

A N I N D I A N S L U M

Nathaniel Roberts

U N I V E R S I T Y O F C A L I F O R N I A P R E S S

U niversi ty o f C a l i forn ia Press, one o f the most distinguished university presses in the U n ited States, enriches lives around the world by advancing scholarship in the humanities, soc ial sciences, and natura l sciences. Its activities are supported by the U C Press Foundat ion and by philanthropic contributions from individuals and insti tutions. For more informat ion , visit www.ucpress.edu .

U niversi ty o f C a l iforn ia Press O ak land , C a l iforn ia

© 2016 by The Regents o f the U niversi ty o f C a l iforn ia

Photographs by N athanie l Roberts

Library o f Congress Cataloging-in-Public at ion D ata

Names: Roberts, N ath anie l, 19 7 0- author. T it le : To be cared for : the power o f conversion and foreignness o f be longing in an Indian slum / N a th anie l Roberts. O ther titles: A n thropology o f C h r is t ia n i ty ; 10 . Description: O akland , C a l iforn ia : U niversi ty o f C a l iforn ia Press, [2016] | “2016 | Series: The A n thropology o f Christ ian ity ; 20 | Includes bibliographica l references and index. Identifiers: L C C N 2015039618I ISB N 9780520288812 (cloth : alk . paper) | ISB N 9780520288829 (p b k .: alk . paper) | ISB N 9780520963634 (ebook) Subjects: L C S H : Pentecostal churches— India— Chenna i. | Pentecostalism— India— Chenna i— H istory. | D a l i t women—Re ligious l i fe— India— Chenna i. | Pentecostal women-—Rel igious l i fe— Ind ia— Chenna i. | Slums— India— Chenna i. | Christ ian ity and other rel igions— H induism . | H induism—Re lations— Christ ianity . Classificat ion : L C C B X 8762 . A 45147 2016 | D D C 289 .9/4082095482— dc23 L C record available at http:/ / lccn .l0c .g0v/2015039618

M anu factured in the U n ited States o f A m erica

25 24 23 22 21 20 19 18 17 16 10 9 8 7 6 5 4 3 2 1

In keeping with a commitment to support environmentally responsible and sustainable prin t ing practices, U C Press has prin ted this book on N atures N atura l, a fiber that conta ins 30% post-consumer waste and meets the minimum requirements o f a n s i / n i s o Z39.48-1992 ( r 1997) {Permanence o f Paper).

This book is dedicated to K . Sagayaraj

D ave Chapelle a nd George Saunders

Everyone is family, every place is home. Neither good nor evil comes from the other. They are not the cause o f our suffering

nor do they bring us solace. Death is nothing new to us. We do not rejoice saying life is sweet. Yet neither do we reject it bitterly. For as the wise know,

a human life is but a raft in the rapids, pummeled by cold rain under a lightning-streaked sky.

And that is why we do not praise the mighty. Still less do we spurn the weak.

T A M IL PO E M , C IR C A 2 0 0 CE

C O N T E N T S

List o f Illustrations ix Acknowledgments x i

T erminological Notes x i i i

In troduction i

i • O utsiders 13

2 • Caste, Care, and the H um an 47

3 • Sharing, C aring, and Supernatural A t tac k 81

4 • Rel igion , Conversion , and the N a t iona l Frame i n

5 • The Logic o f Slum Religion 152

6 • Pastoral Power and the Mirac les o f C hrist 18 5

7 • Salvation , Knowledge, and Su ffering Z17

Conc lusion 24 4

Appendix: T amil N adu Prohibit ion o f Forcible Conversion o f Religion O rdinance, 2002 251

Notes 255 References 265

Index 279

I L L U S T R A T I O N S

1. Typical church interior 14

2. Sewage channel alongside house 3 6

3. Flies on my helmet 38

4. Garbage removal 39

5. H eaped garbage 40

6. W omen at water tank , relaxed 4 4

7. A world apart 4s

8. M ora l isolation: two views 79

9. A ninety-pound laborer, after a long shift 8s

10. A n act o f fa i th 180

11. A l k dancing 203

12. The H oly Spirit washes over women 2 07

13. Spirit-filled praises 208

14 . Inspired prayer 209

15. Umbrella Preacher 229

A C K N O W L E D G M E N T S

I gratefully acknowledge the support o f the Social Science Research Counc i l (SSRC) for the predissertation grant that funded an initial fifteen-month research period from 2001 to 2002, and J . Jeyaran jan o f the Institute for D evelopment A lternatives (ID A) in C henna i for serving as my host; the A merican Insti tute o f Indian Studies (A IIS) for funding my fieldwork from 2003 to 2004; the Economic and Social Research Coun c i l (ESRC) for a one- year postdoc at the London School o f Economics (LSE), where I was fortu � nate to have C hris Fu ller as a mentor; the M ax Planck Insti tute for the Study o f Religious and E thn ic D iversi ty in Gott ingen , my current institutional home, and Peter van der Veer for providing me with the extra time needed to write this book , amid two new research projects over the last four years; the Centre for Modern Indian Studies at the Universi ty o f Gott ingen , my insti � tutional “home away from home”; and D r Mohana Dass and the Department o f Indian Studies at the Universi ty o f M alaya, for hosting me during my final round o f copy edits.

I would like to thank everyone who read or conversed with me about parts o f this book , in no particular order: Sanal Mohan , Gajendran Ayyathurai , U day C handra , L isa B jorkm an , C hristodas G andhi , P. Sivakami, N ic k D irks, Va l D aniel , L ila Abu-Lughod , Gayatri Spivak , O wen Lynch (R.I.P.), K aori H atsumi, A n ia Loomba , Suvir K aul, T anika Sarkar, Lisa Mitchell, Joe l Lee, Joh n ny Parry, O livia H arris (R.I.P.), M a t thew Engelke, M ichae l Scott , Radhika Gupta, D evik a Bordia, Peter van der Veer, D . Ravikumar , Jin-H eon Jung , K ar in Kapadia, Norbert Peabody, E r ik H arms, Lionel Caplan , Penny Vera-Sanso, Preetham Chakravarthy, V. “ C h aks” Chakravarthy, K rit i Kapila, Sahana Udupa, Jyotirm aya Sharma, Jacob Copeman , D anie la D e l Campo, M echthild Nagel, A ndrew N icholson , Gauri Viswanathan , C am Grey, A jay

G andhi , K a thryn T idrick , Susan Billington-H arper , M ichael Bergunder, D avid Mosse, N aomi H aynes, an anonymous reviewer at U C Press, A ngie H eo, Drs. Peter J . Bräunlein , M ichae l D ickhardt , and W olfgang K e m p f o f the Insti tut für Ethnologie at the University o f Gött ingen .

Srirupa Roy and La l i t Vachani have given me warm intellectual friendship for the past few years, and the run o f their Gött ingen residence for my final month o f writing. I am grateful to my series editor, Jo e l Robbins, who believed in this project and wai ted patiently for several years; Reed M alcolm , my commissioning editor; Stacy Eisenstark , also o f U C Press; Elisabeth Magnus, my copy editor, whose pa instaking work greatly improved this text; and my dear friend S. A n and o f N avayana Press, who read all the chapters with great care. I thank my father, M ax Roberts, for also reading the entire manuscript and encouraging me to keep trying to write clearly; my mother, Margaret Lincoln , for believing in me; and my in-laws, Radha and N atarajan Viswanath , for their love and support.

Joe l Lee and Gajendran Ayyathura i have been my brothers in struggle: Gaje for many years, Joe l since recently. M y debt to Joe l is great. The detailed comments he made on my entire manuscript, which he read at the last minute, were transformative. H is unpublished doctoral dissertation , which he made available to me when I was in the final stages o f revising my own manuscript, drew my attention to the importance o f embodiment and my own bodily presence in A nbu N agar and suggested arguments that added to and greatly deepened those I present in chapter 4. K . Sagayaraj was my friend and companion during the most diff icult and persona lly chal lenging periods o f my life, and to him this book is dedicated.

M y greatest debt o f all, however, is to Rupa V iswanath . She has read every page o f this book multiple times, and the personal sacrifices she has made to support me— as best friend , inte l lectual partner, and susta iner— are immeasurable.

X i i • A C K N O W L E D G M E N T S

T E R M I N O L O G I C A L N O T E S

C astes (<o r jd t is) are endogamous groups whose members are thought to share important physical, moral, and inte l lectual traits. Castes are notiona lly ranked, and a person’s caste is hereditary and permanent. In Indian languages

jd t i can also refer to other natural types, such as species o f animal, types o f plants, and so on, as well as to races (e.g., “the European j d t i ” “ the A frican jd t i ” ) and gender (“ the m ale jd t i ” and “ the female jd t i ” ). But it is always clear from context which sense o f the term is intended.

Throughout this book , D a l i t refers to members o f Indian castes outside H indu society’s traditional fourfold social structure known as the varna sys � tem. D a li ts were once known by a variety o f English terms, like untouchables, Pa riahs, and outcastes, as we ll as by native terms such as ava rn a , chanda la , zn&pancham a, and various regional equivalents. Such terms are now seen as insults. D a lit is a relatively recent coinage, originating in the 1920s but gain � ing popularity only in the 1970s. D a lit comes from a M arath i word meaning “ground down” or “ broken” (Rao 2004 , xx-xxi) . Its T amil equivalent, td l ttapattor, means “ those who have been put down ’” (Mosse 2012, 193). Both D a lit and tdlt tapa t tor were accepted by the subjects o f my study, though neither was in common use. M y preference for D a lit follows standard schol � arly practice. Previously untouchable and H a r ija n were also used by scholars, but since the 1990s they have fa llen out o f favor: untouchable, because it is insulting, and H a r ija n , a term promoted by M . K . Gandhi , because it is a spec ifical ly H indu term that excludes C hrist ian and M uslim D al its, and because D a li ts themselves widely reject it. D a lit is the only name originating from with in the community it describes. The bureaucratic term Scheduled Caste (or S C ) is sometimes loosely used as a synonym for D a lit . But that use misleads, because in fact S C refers only to H indu , Sikh, and Buddhist D al its;

the official list o f Scheduled Castes excludes C hrist ian and Muslim D a li ts (V iswanath 2012).

C ast e p eo p l e (Tamil: jd t i m a k k a l) is an idiomatic term referring to mem � bers o f any non-D ali t caste. N o t just H indus, but also non-D ali t Christ ians and Muslims can be described as “caste people.” The term caste people is never used to refer to D ali ts. This is a potential source o f confusion , because D a li ts also belong to part icular castes in the sense o f endogamous descent groups, or já t is. I t might therefore seem that they too should be called “caste people,” but that is not how the phrase is understood. Likewise, “ to have caste” means to have membership in a non-D al it caste.

I refer to the neighborhood where my research was carried out as a sl u m, a word that can have negative connotations in English . But slum is the stand � ard term in India’s scholarly and governmental discourse. I t is also favored by slum dwellers themselves. In the local context it has no strong negative asso� ciations and is preferred over common alternatives, such as céri and colony.

Following native usage, m i d d l e class in this book includes everyone from small business owners, through midlevel bureaucrats and other salaried workers, to those having advanced professional qualifications. W ith Leela Fernandes and Patrick H el ler (2006), I recognize the Indian middle class as encompassing disparate social statuses that share a distinctive modern and pan-Indian culture, a strong sense o f national identity, and the be l ief in the middle class’s own moral leadership. The Indian middle class is properly described as a social and economic el i te, and I therefore use these terms inter � changeably. India’s middle class speak English and their views are reflected in national English-language dailies. To be middle class in India entails being a caste person; those few D a l i ts who have atta ined middle-class status in socioeconomic terms are nevertheless cultural ly marked in India as dist inct from other middle-class subjects by their D a l i t identity.

Sl u m C h r ist i a n i t y and sl u m ch u rch es refer to local slum-based forms o f Pentecostal Christ ianity. These churches are led by D a l i t pastors who in most cases were born in those very slums and continue to reside there. I describe these churches’ members as sl u m C h r ist i a ns. N ot al l Christ ian slum dwell � ers are Pentecostals. Some 10 percent belong to mainstream denominations like the Church o f South India (CSI), or the Roman C atholic (RC) Church. These mainstream denominations do not figure prominently in this study, and when they come up I refer to them by denomination or, more generically, as m a i nst rea m or a ncest r al C h r ist i a n s. This last term stems from the fact that membership in C SI , R C , and other mainstream Christ ian denomina-

x i v • T E R M I N O L O G I C A L N O T E S

tions in India is nearly always hereditary. Since Independence, mainstream Christ ians have ceased active conversion efforts in India. Slum Christians, by contrast, are nearly al l converts. They have converted from either H induism or mainstream C hrist ian denominations. The loca lly significant religious difference among slum dwellers is not between C hrist ianity and H induism , but between Pentecostal slum Christ ianity , on one hand, and both H indus and mainstream Christ ians on the other.

Slum dwellers speak variously o f multiple gods, and o f G o d . Slum H indus view the former as aspects o f the latter, like the T rin i ty only more varied. W hen capitalized in this book , G o d means the supreme godhead o f any tra � dit ion . The singular G od is spoken o f by H indus using masculine, feminine, and gender-neutral terms {dntavar, ammd, and kapavu l respectively), whereas Christ ians ma in ly use the masculine term, rarely the neutral, and never the feminine. W hen discussing the Christ ian god as one de ity among many, I refer to him as a god (without capitalization), but I capitalize in contexts where H e means the singular God . Likewise, when H indus refer to part icu � lar deities I do not capitalize “god,” but when they mean the one godhead I write “ God ,” as appropriate.

A n b u N a g a r (from anpu, a common T amil word for “ love, affection”) is the name o f the slum where this study is set. It is one o f dozens, in a large agglomeration o f slums called K ash t a p p a t t i n a m , in an industrial area north o f C henn a i ’s old c ity center. A nbu Nagar, Kashtappattinam , and all personal names used in this book are pseudonyms to shield my study’s subjects from reprisals.

T E R M I N O L O G I C A L N O T E S • XV

Introduction

Pariah is a cruel word. For most speakers o f English today, only the dimmest memory o f what it once meant survives. But for its victims the memory o f that cruelty has not been forgotten, in part because it is not just a memory.. . . It is a word that causes the descendants o f those it once named to visibly wince. Like nigger it is not just a word. Like nigger it at once references and is itself a part o f one of the most dehumanizing social orders the world has ever known.

R U P A V I S W A N A T H , The Pariah Problem,

They say: “You are a shameful man. Your life is worthless—you earn nothing for your family! What are you even living for?” But Christ will not desert you—he will lift you up. Christ uplifted those who were shamed, those who were pushed away. In the same way . . . he will look upon us lowly creatures. He will lift us up__ He will place you on the highest peak, before your enemies, before people from other castes—hallelujah! In front o f those who spoke ill of you, those who tormented you, who detested you, who pushed you away, who said you were not quali � fied, who said you were without talent or intelligence, who said you were useless—Christ will bind their tongues! Hallelujah!

P A S T O R Y E S U D A S A N

t h e w o r d p a r i a h is u s e d casually by speakers o f English to refer to those who are shunned by others. A socially awkward, foul-smelling, or oth � erwise unpleasant person might be called a “social pariah”; apartheid South A frica , N orth Korea, and Saddam H ussein’s Iraq have all been called “pariah” states. The implication o f our usage is that those we apply it to not only are

ostracized but deserve to be. This is due to the peculiar history by which the term entered our language: we learned it from the original Pariahs’ enemies and oppressors. N o Pariah sees her own child, or other loved ones, as a pari ah. The word comes from colonial South India, a slave society that British colo � nizers wrested from native rulers in the eighteenth century. W here standard nationalist and postcolonial accounts have focused on the con flict ing inter � ests o f the revenue-extracting colonial state and native taxpayers, this tug-of- war is better understood as a fight between two categories o f exploiter over the fruits o f slave labor. W ith respect to those called “ Pariahs,” the colonizer and the colonized were in fac t allies. It is we ll known that Brit ish India depended on the cooperation o f native elites from the village level on up. W ha t is less we ll known is the extent to which these foreign rulers colluded with natives in the perpetuation o f caste-based slavery. The British were at pains to avoid overturning the trough at which both they and native taxpay � ers fed. From the early eighteenth century, historian Rupa V iswanath has shown, Brit ish India hands began to portray agrarian slavery there as “a gen � tle form” o f slavery, based on traditional rights and responsibilities from which slaves themselves benefited, and even as slavery “ in name only” (2014b, 3-6) . This was in order to fend o f f an emergent abolition movement, which threatened to ruin their racket. Though agrarian slavery may not have played an equally significant role in all parts o f British India, the M adras Presidency’s agricultural economy depended on slave labor. M adras was the cash cow o f Brita in ’s eastern empire; it was among the few possessions in which revenues exceeded expenditures, and surpluses generated in M adras were used to fund British operations elsewhere. Thus at a very basic, material level the Brit ish Empire depended on enslaved D a l i t laborers. A history o f the United States that did not take into account the institution o f slavery and its continuing legacy would be ludicrous. But that is exactly the state o f scholarship on colo � nia l and postcolonial India, which either omits slavery entirely or character � izes it in the anodyne and apologetic language o f British administrative dis � course (2014b , 241-46) .

In the colonial period P a r ia h was a name used to refer generically to all untouchable castes, but it derives specifically from the name o f one such caste, the Paraiyar. This book ’s protagonists are from that caste, but Pa ra iya r is rarely used today, and I refer to them mainly as D al its or, following their own self-description, simply as “the poor.” H istorica l ly in T amil country to be D ali t was simultaneously to be a hereditarily unfree agricultural laborer— in native parlance, a slave or a t im a iydl— bought and sold along with the parcels

2, I N T R O D U C T I O N

o f land to which your family was permanently attached, and treated with impunity by masters (V iswanath 2014b). Caste in India is often spoken o f as a complex system comprising innumerable distinctions and subdistinctions, a mosaic o f notiona lly ranked endogamous groups { jdt is) vying with one another for social precedence and bound together by a shared ritual cosmol � ogy. U nlike race, which involves “ ‘th ink ing in blocks’ o f large similar com � munities,” caste, according to this mosaic understanding, “revolves around differences in tiny details” (D ipankar Gupta , quoted in IS H R 2007, 6).

The singular division between D a li ts and non-Dalits, however, is qualita � t ively dist inct from , and irreducible to, the multiple divisions between indi � vidual castes.1 J a t i distinctions among “touchable” castes imply no deep onto � logical or social divide and are minor by comparison. They constitute a graded system o f relative prestige, in which all are accorded some measure o f respect. The division between D ali ts and non-Dalits, by contrast, is premised on the systematic dehumanization o f the former by the latter. It is racelike not just because being a D a l i t or a non-D al it is permanent, based on blood, and seen as determining one’s physical and moral-psychological nature, and but because the characteristics ascribed to D ali ts are (unlike those ascribed to individual castes above the touchability line) o f an overwhelmingly and exclusively nega � tive quality. A l l over India D a li ts are stereotyped as lazy, stupid, licentious, dishonest, suitable only for manual labor, unclean , and repulsive.

Lik e caste people, D a l i ts also belong to numerous local jd t is , each with its own name, but throughout India they are distinguished in native discourse from al l others as chanda las, pancha mas, and ava rn a , and by equivalent generic terms specific to each region (see T erminological Notes). They are also recognized everywhere as being outside the idealized fourfold structure o f traditional H indu social theory. D a l i ts are also commonly referred to by their local jd t i names, which in everyday use often double as terms o f abuse for any outsider. Throughout southern India, D a li ts are forced to live in sepa � rate settlements at a safe distance from the villages where all others reside. To this day they are referred to as outsiders to the village proper. A s late as the 1920s T am il dictionaries defined the collective ethno-national term for T amils, t a m ilan , as explic i t ly exc luding T am il speakers o f D a l i t caste (U niversity o f M adras 1929, 1367), reflecting a popular understanding that remained current in rura l settings much longer (V iramma, Rac ine, and Rac ine 2000 , 282 n. 3).

The scholarly literature on India focuses overwhelmingly on the lives and cultural achievements o f caste people. That has been the priority. To B e C ared

I N T R O D U C T I O N • 3

F o r is about those other Indians, the ones we hear much less about. Specifically, it is about a small community o f D ali ts living in a slum in north � ern Chennai . It is about their struggles both with the world and with them � selves, and how these struggles take on new meaning when women in the slum convert to Pentecostal Christ ianity.

T H E A R G U M E N T

The story this book tells begins and ends in a neighborhood I cal l A nbu Nagar, part o f a large agglomeration o f D a l i t slums, K ashtappattinam , in the industrial sprawl extending northward from C h enn a i ’s old c ity center. The conversion movement it describes began in the early 1990s and continues to the present day. A t the time o f my fieldwork the majority o f converts, some 85 to 90 percent, were women. For that reason, and because o f the unique way slum pastors interpreted C hrist ’s message, C hrist ianity was local ly under � stood not only as the religion o f the poor but also more spec ifical ly as a “women ’s religion .” The book centers on conversion, but in the first three chapters C hrist ianity makes only a fleeting appearance, and conversion none at all. That is by design. Most anthropological studies o f religion focus on ritual contexts and textual practices. Relatively few look seriously at what people do when they’re not “ doing” religion , or what they ta lk about when they’re not talk ing about it. They tell us little i f anything about people’s eve � ryday struggles, at work and at home. Fewer sti ll have attempted to grasp the systematic connections between religion and everyday life, or to show how what goes on in the street and in the home gives meaning to religion— and not just the reverse, as commonly supposed.

Christians everywhere use the same basic text and the same toolbox o f basic concepts—sin, salvation, faith, and so forth. Pentecostal Christ ians, further � more, worship in quite similar ways the world over. I f students o f A frican or Lat in American Pentecostalism were to visit any o f Kashtappatt inam ’s churches, they would feel immediately at home. W ha t they would not grasp, however, is why any o f it mattered to local people. They would not get, for example, why the story o f Adam and Eve’s fal l from Eden was understood by slum dwellers as conveying the message that husbands should always listen to their wives! They would not immediately understand that when slum Christians talked about being “saved from sin,” it was not typically their own sins that they had in mind but the sins o f others; that the prototypical “other”

4 I N T R O D U C T I O N

from whom they hoped to be saved was a caste person; and that “salvation” was conceived not as a spiritual state but as a worldly revolution in which caste itself was abolished. This book shows that to understand why some slum dwellers convert and others do not we must begin with the unique moral problems and cultura l contradictions that structure their existence. W h ile Christ ianity appears throughout the world as the solution to converts’ problems, the prob � lems it solves are local and unique.

The first three chapters are about rejection , its opposite, and the unex � pected ways the two entwine. Chapter i, “ O utsiders,” explores what, for the people o f A nbu N agar, it meant to belong. It is a chapter about the different ways insiders and outsiders are distinguished . It begins with rumors that c irculated intermittently among slum dwellers that one or another local Pentecostal church was receiving “ foreign money.” Foreign money is also a persistent concern in public discourse in India, ma in ly for the threat it is seen to pose to national autonomy. The possibili ty that foreign money was flowing to local pastors was also a matter o f concern for slum dwellers, but not for the same reasons. W here national-level discussions o f foreign money revolve around issues o f autonomy, slum dwellers’ turned on a rather different ideal, that o f care. W hy did no one seem to care for them? That was an abiding concern for the people in A nbu Nagar, and it was linked in complex ways to the distinction between national and foreign and their own uncertain status in relation to both. This turned out to have important implications for me, a foreigner in the flesh. In this chapter I describe my own entry into the field, the nature o f the relationship I formed with the people there, and what this helped me to see about their relationship to the dominant society. The third and last section o f the chapter describes the physical character o f the slum, which residents believed set them radically apart from nonslum neighbor � hoods. It focuses in particular on fi lth and drin k ing water, and the heavy symbolic freight they bore for slum dwellers as reminders o f their own out � cast status.

The second chapter, “ Caste , Care, and the H um an ,” examines how the people o f A nbu N agar and surrounding slums understood their relationship to the dominant society and to humanity as a whole. A ccording to them, they had no caste. They were cal led Paraiyar, they readily admitted, but this was just a name that had been applied to them by others. It had no intrinsic significance; it referred to no underlying essence or truth about them. In reality, they claimed, they were simply human . Indeed, they insisted that although by living together in this slum they had become a community, in

I N T R O D U C T I O N • 5

real ity they had nothing in common with one another: no special traditions, no unique outlook , no common blood. W ha t bound them together, these slum dwellers said, was simply that they were poor— that and what they shared with al l people throughout the world, their humanity.

This opens an exploration o f the moral ethos o f the slum, and the role with in it o f the twin notions o f care and the human . Seeing themselves as lacking in caste— as being merely human— the inhabitants o f A nbu N agar believed they were naturally drawn to others in ways most Indians were not. For to be hum an in the world o f the slum was to be instinctively concerned about those who were in need, whoever they might be, and to feel called upon to care actively for them. To be human was also to be, oneself, worthy o f being cared for by others. Caste, by contrast, was understood by slum dwellers as the denial o f common humanity, a denial they believed was unique to India. O utcasts in their own land, they imagined a special connection between themselves and all who were, in India, labeled foreign—A fricans, East Asians, white Europeans. For foreigners too were envisioned as being merely human and, like their own idealized selves, freely and spontaneously caring for others. The foreignness o f belonging in the book ’s subtitle refers to the dual nature o f their predicament. Belonging was itse lf foreign to these slum dwellers’ experi � ence in the land o f their birth. But they also perceived a world in which they did truly belong, and which awaited them in the form o f a larger humanity that was “ foreign” in the sense o f existing beyond national borders.

The continuity these slum dwellers envisioned between themselves and the foreign depended on a clear-cut division between inside and outside, between themselves and the dominant nonslum society. Chapter 3, “Sharing, C aring, and Supernatural A ttack ,” reveals the instability o f this moral topol � ogy. The stylized contrast described in chapter 2 between freely sharing and humane slum dwellers and their caste oppressors is a moral discourse, not sociological reality. It was true that slum dwellers did not generally discrimi � nate on the basis o f caste and that they were indeed generous with strangers. But they did not always share freely and spontaneously with their more needy peers with in the slum itself, and in two key relationships the gap between moral ideal and actual practice was systematic. O ne was everyday money- lending relations among slum women , where, instead o f sharing freely with one another, they sought profit by charging usurious rates o f interest. The other was between women and the men they were married to: husbands did not always hand over their wages to wives as they should. I refer to these two relations— among women , and between women and the ir husbands— as

6 I N T R O D U C T I O N

moral fault lines because, like geological fault lines, the contradictions they embodied were largely hidden from view.

These fault lines converged on married women , exerting a pressure that was at times unbearable. Chapter 3 accounts for this pressure by trac ing the movement o f money with in and between households and, concurrently, the c irculat ion o f rumor and innuendo about various forms o f supernatural threat. Such ta lk had contradictory effects. W h ile it instigated women to share generously with others and to provide material aid to those in need, the discourse on supernatural threat also atomized slum women and subtly pit � ted them against one another. Supernatural evil, though the opposite o f care, was constitutively linked to care and was, in fact, its dark underside. U nlike the problematic relationship between slum dwellers and the dominant soci � ety, which slum churches were powerless to resolve, the internal contradic � t ions o f slum culture were susceptible to reform . H ow slum C hrist ianity was able part ia l ly to resolve these tensions, and how this resolution was locally understood, are addressed in chapters 6 and 7.

Conversion is important in India, not just to converts themselves, but in a very different way to the dominant national community. O n O ctober 5, 2002, just ten months prior to my arriva l in the field, the state o f T amil N adu issued an ordinance (avacara cat tam, li tera l ly “emergency law”) banning exact ly the sort o f conversions I would go on to study (see Appendix).2 The law remained in effect throughout the period o f my research, and it explicit ly targeted conversion among three demographics: H indu D ali ts, women , and children . This grouping is significant. Throughout India the overwhelming majority o f converts out o f H induism are D a l i t , and in the Indian legal and poli tical context D a l i ts and women alike are frequently m inor it i z ed— in the sense o f being treated as minors, children not fully capable o f m aking their own decisions and therefore requiring special forms o f supervision .

Reports o f Christ ian conversion in India are met with responses ranging from tolerant disapproval among secular liberals to outrage in the H indu national ist camp. Religious conversion is portrayed in national discourse as an attack on Indian culture and on the nat ion itself. It is not even genuinely religious, in fact, because it is based on worldly motives said to be antithetical to authentic spiritua l understanding. It is an inherently dangerous activity that disrupts local communities, leading inevitably to violent conflict. It destroys families. N o t only does conversion lead to violence, it is i tse lf an intrinsical ly violent act. It is an intrusion into the innermost core o f the indi � vidual self.

I N T R O D U C T I O N • 7

W hile this is how national elites see conversion, it is not at al l how it appears to either H indu or Christ ian slum dwellers. Chapters 4 and 5 pro � vide detailed accounts, respectively, o f how differently religion is conceptual � ized in India’s national public sphere and in the slum. Chapter 4, “Religion , Conversion , and the N ational Frame,” examines this discourse in detail. It also shows it to be o f recent historical origins. Though supposedly based on a time-immemorial H indu cultural understanding, the key features o f this discourse in fact date only to the early twentieth century. Except under cer � tain very special circumstances, before the twentieth century H indus were not at all bothered by D a l i t conversion to Christ ianity . D a l i t conversion was not seen as a threat to H induism because D a li ts themselves were not regarded as H indu . It was only in 1909, in the context o f a new sense o f demographic struggle among H indu communalists in opposit ion to Muslims, that D a li ts were first identified as a key strategic resource. C la im ing D ali ts as H indus would put H indu claims to majority status on firm footing. But the novel idea that D ali ts should be embraced as fe llow H indus, and that their conver � sion to C hrist ianity should be opposed, was con fined for the first three dec � ades o f the century to H indu majoritarian activists.

A l l this changed in the 1930s. Under the polit ical and intel lectual leader � ship o f M . K . G andhi , the project o f integrating and retain ing D a li ts w ith in the H indu fold was reenvisioned as not merely a H indu communal interest but a national one. It was during this decade, moreover, that all the key tropes o f present-day national discourse on conversion acquired the commonsense status they en joy today, also as a result o f G an d h i ’s leadership. I t was he who developed and popularized the argument that D a li ts were spiritual minors who required special supervision; that they were easily “ lured” into convert � ing with offers o f worldly benefits; that religion itse lf was a properly spiritual affair in which worldly interests had no legitimate place; that conversion upset local communities, leading inevitably to violent conflict. G andhi thus presented conversion as simultaneously a threat to the nat ion’s autonomy and to that o f converts.

H aving situated the historical ly contingent assumptions about religion and conversion that shape India’s public sphere discussions and anticonver � sion law, the fifth chapter reveals that even today this picture o f religion is not universal. Entit led “ The Logic o f Slum Religion ,” it describes how religion was lived and understood by both Christ ians and H indus in Kashtappattinam , and in so doing it reveals that religious conversion in the slum meant some � th ing very different than it does in the national imaginat ion . The chapter

8 • I N T R O D U C T I O N

directly addresses the question o f why the radical religious division in the slum brought about by conversion did not result in social con flict; why, even when members o f a single household embraced different faiths, family life was not disrupted; why conversion did not pose any o f the problems about free will , autonomy, or cultura l authentic ity for slum dwellers that it did for elites. In short, the chapter shows how anticonversion laws “safeguarded” the religious autonomy o f D ali ts from foreign imposit ion precisely by imposing upon them religious norms that, as this book makes plain , were themselves entirely foreign to their own religious understanding.

A ccording to standard models, the al legedly ever-present potential for conflict between religions in India is mitigated by widespread syncretism, in which adherents o f different fa i ths adopt one another’s practices, thereby blurring the boundaries between one religion and another. The assumption that hard boundaries and mutua lly exclusive identities are productive o f con � flict, whereas fuzzy boundaries and multiple identities mitigate it, is ques � t ionable (cf. van der Veer 1994; G . V iswanathan 1995). In A nbu N agar and surrounding slums, however, religious boundaries were clearly marked and vigorously policed . W h a t is more, the practitioners o f one faith, Pentecostal Christ ianity , forthrightly declared the gods o f the other “mere stone idols” whose worship was “complete useless,” while H indus in turn dismissed Jesus as a weak , overly demanding, and jealous god— a passive-aggressive deity whose alleged miracles were mostly imagined. Yet no one was particularly bothered by such talk , nor did they see any reason they should be.

The reason, I show, is that religion in the slum was not a matter o f identity at a l l—neither national, ethnic, nor personal. Its focus, rather, was on the power-laden and unequal relationships between human beings and gods. Gods existed to help people, and the benefits they provided were o f a funda � mentally worldly and moral, but not spiritual, nature. N e i ther Christ ianity nor H induism , as practiced in the slum, emphasized any sort o f otherworldly telos. D ecisions about which god to worship were based on considerations o f what that god could do for worshippers in this world . These decisions, more � over, were not moralized by slum dwellers. This is because moral ity was seen as a un iversa l property that did not vary by religion . Slum dwellers did not subscribe to the secular modern understanding o f religion as culture and therefore did not see different religions as comprising different “systems o f values.” Christ ians and H indus recognized one another’s gods as entirely moral beings, and they perceived the moral ity they upheld as identical. W h ile gods did not vary morally, they did vary existentially and in their

I N T R O D U C T I O N • 9

powers—some gods were more powerful than others, and some might simply not exist. The truth o f their existence and o f what powers they possessed, however, was never entirely knowable. W hereas in the secular modern under � standing o f religion as culture gods’ “existence” is a funct ion o f the human traditions organized around their worship, slum dwellers subscribed to what I cal l theological realism-, some gods existed and others did not, and nothing humans had to say about it could change this. Accordingly, when H indu and Christ ian slum dwellers debated their various gods’ existential status, they did not do so in the manner o f competing teams whose goal is to w in the argument. Rather, they were engaged in trying to find out which gods were real and therefore worth worshipping.

Chapters 6 and 7 are about the organizational form o f slum Christ ianity and the unique way the Christ ian message was interpreted by slum pastors. Chapter 6, “Pastoral Power and the Miracles o f Christ ,” identifies two sources o f Pentecostalism’s power for slum women . First is the intensity o f pastoral care. Kashtappattinam was home to a large number o f independent pastors in a state o f perpetual competition with one another. Those who were less atten � tive to the everyday needs and concerns o f their mostly female flock lost fol � lowers to their more energetic and innovative rivals. A major arena o f pastoral innovation was doctrinal. O f the same caste as those they served, and in most cases born and raised in the very same neighborhood, pastors shared the worldview o f ordinary slum dwellers. But unlike most slum men, they pos � sessed an intimate understanding o f women’s struggles, built up over years o f dai ly pastoral service. The long hours they spent listening to women shaped these pastors’ understanding o f Christ ianity itself, as detailed in chapter 7.

Slum pastors did not tend only to women . They tried equal ly hard to win male converts, but with less success. U nlike women, whose suffering stemmed both from their caste-based subordination to the dominant nonslum society a n d from the moral fault lines that divided the slum against itself, men’s distress derived mainly from their powerlessness in relation to caste people. A n d this, as already noted, was not a problem slum churches were in a posi � tion to resolve. W ha t they could directly affect were women’s con flicts with one another and with their husbands. This brings us to the second source o f slum C hrist ianity’s appeal to women. For despite appearances that both pas � tors and their flock silently colluded in ma intain ing, the real center o f power in slum churches was not the pastor h im se lf but organized networks o f church women whose relationships to one another were mediated by prayer. Women ’s prayer networks counteracted the harm ful effects o f the slum’s

I O • I N T R O D U C T I O N

moral fault lines by redistributing responsibili ty for women s individual problems across a network o f concerned sisters. W here H indu cosmology subtly atomized women at a moral level, pit t ing them against one another, women ’s prayer networks strengthened the ir sense o f shared interests. W omen who were exploited by other women and by their own husbands drew on this collective basis o f support to assert their right to be cared for. This explains why the majority o f converts were women , and also why men who converted frequently drifted away from Christ ianity but female converts did not.

A key argument o f the book is that C hrist ia n ity ’s success in the slum derives from its abili ty to mediate endogenous moral contradictions and to relieve the existential distress and interpersonal con flict that these contradic � t ions entail. This is why religious conversion did not divide the slum com � munity o f A nbu N agar or lead to violent con flict , as India’s national public discourse predicts it should. By suturing the moral fault lines that covertly pitted slum dwellers against one another, slum C hrist ianity integrated the slum community as a whole a n d irrespective o f religious affiliat ion . The con � version o f some residents to a different religion , instead o f d ividing the slum community, in fac t served to unite it.

In chapter 7, “Salvation, Knowledge, and Suffering,” we turn f inal ly to the content o f the slum C hrist ian message. U n like loca lly available forms o f H induism , slum C hrist ianity is a discursively rich tradition in which wor � ship is congregational and intensely interactive. Converts in A nbu N agar claimed that unlike their former gods Jesus spoke to them—he taught them, empowering them with knowledge. Yet the knowledge they received con � tained irreducible perplexities. The chapter identifies two distinct spatiotem- poral horizons intrinsic to Pentecostal be l ief and practice in the slum. The first was local and concerned everyday relations and the difficulties o f get ting along. The temporal horizon in this case was that o f homogeneous empty time (Ben jamin 1968; Anderson 2006) and involved gr a d u a l transformation through divinely mediated human effort. The target was intraslum conflict, and the moral fault lines along which the slum ethos o f care was routinely violated. H ere pastors offered an alternative vision o f what marriages and other conflict-prone relations could be. They also dispensed advice aimed at teaching women how to deal with conflicts and how to remake their relations w ith husbands and others.

The second spatiotemporal horizon evoked by slum pastors was global and was organized around a binary relationship between good and evil, poor and

I N T R O D U C T I O N • II

rich, casteless and encasted humanity. The temporali ty here was that o f revolutionary-messianic time (Ben jamin 1968; Lowy 2005; cf. Robbins 2007). I t was under this rubric that the slum’s collective subordination to the nonslum society was addressed. The promised transformation was to occur all at once, in some unspecified future, and by means o f divine agency alone. Pastors offered no advice or practical suggestions in this case, because there was noth ing for believers to do but wait . They did, however, provide a steady stream o f evocative images h in ting at what this revolution would look like. This chapter’s second epigraph— an outburst that occurred at the height o f divinely inspired preaching by a slum pastor, Yesudasan— is an example.

The spatiotemporal horizons around which slum preaching swirled were not always kept distinct. O ften pastors employed ambiguous language that could be plausibly understood by believers as referring to either collective or interpersonal suffering, or both simultaneously. In this way slum Pentecostal discourse spoke on several levels at once, and in doing so it made these osten � sibly different forms o f oppression appear as but different facets o f a single underlying problem, “sin.” The fiery and often bewildering content o f slum preaching blurred distinctions and made the messianic appear to erupt con � tinuously with in the everyday (cf. Robbins 2004). The promised total trans � formation was thus at once indefinitely deferred and already under way. The uncerta inty about what was being said was not a weakness o f slum C hrist ian discourse but one o f the sources o f its power. I t incited believers to a practice o f interpretation that, because the rock bottom o f some fina l truth could never be reached, kept them continually on their toes and spellbound. A nd at the limits o f interpretability, meaning itse lf dissolved into what was called anniya pdcai, the T amil term for glossolalia that translates li teral ly as “ the foreign tongue.” H ere we encounter the idea o f a language that is at once totally incomprehensible and yet believed to be the words o f one’s own heart in direct communication to God . C loser than the most intimate relation � ship, yet unbridgeably far, in anniya pdca i the book ’s final chapter thus ends where it began: with the irresolvable predicament, and hope, that I summa � rize as “the foreignness o f belonging.”

A brie f conclusion reviews the ground that was covered in order to reflect more explici tly on care in and beyond the discipline o f anthropology: on the role o f care in ethnographic research and on things that are owed but never given.

I N T R O D U C T I O N

Outsiders

r u m o r s o f c e r t a i n p a s t o r s r e c e i v i n g “ foreign money” surfaced soon after my arriva l in A nbu Nagar. Previously my ideas about conversion in India had been drawn from newspaper reports and the like, and I took claims that foreign money was being used to lure converts at face value. It also seemed obvious to me that conversion must stir social con flict and, in areas where most converts were women , stir trouble between husbands and wives. Thus I assumed an important contribution o f my research would be in eth- nographically charting conflicts, and perhaps in showing how a local com � munity struggled to keep its moral bearings under the assault o f monetized proselytism, borne on a wave o f global capital. I pursued “ foreign money” rumors with the same energy that I sought out signs o f religious con flict between Christ ians and H indus, or with in individual families, over matters o f religion .

W hen I discreetly queried pastors themselves about foreign money and who was receiving it, Pastor Senthil K um ar ’s and Pastor V i jay’s names con � sistently came up. O ther pastors regretted not being contacted by foreign donors w ill ing to back their ministries, but consoled themselves with the idea that G od chooses to bless each o f H is servants in different ways. “ W e cannot presume to know H is plans” was a line I often heard in this context. I did not immediately question Pastors Senthil Kum ar or V i jay about rumors that they had received foreign money. I assumed that direct questioning would put them on the defensive and that they would attempt to evade. So I gathered as much in formation as I could, indirect ly at first, from lay Christ ians and H indus. I planned my approach very carefully, over a period o f weeks. I did not want anyone to know what I was after. I especially did not want to risk tipping o f f any o f the other pastors to the centrality o f foreign money to my

F IG U R E i. Typ ic a l church in terior.

evolving research plans. I knew pastors treated one another prim arily as rivals, but I did not yet fully understand the relations among them, and it seemed prudent to assume that others, too, might be in on the game.

By the time I raised the issue directly with Pastor Senthil Kumar, I had already worked out the basic details o f the case. H e had received a one-time donation that financed a new church building. Compared to most slum churches, his was indeed impressive. W here other churches were simply larger versions o f typical slum dwellings— often either palm thatch or corrugated asbestos sheeting tied to frames o f wooden poles (figure i)—Senthil Kum ar ’s church had “pukka” (i.e., masonry) walls and was bigger than what could have been built on local donations. But I still wanted to question him directly. I needed to gauge his reaction. A fter weeks o f cautious circling, during which time I attended the pastor’s Sunday services and got to know the man h imse lf a bit, informally, I arranged an interview. The interview was casual by design and wandered over a range o f innocuous topics I had preselected to draw the pastor out and establish his trust. I saved my big question for the end.

I am glad I did my homework , but the cloak-and-dagger approach proved unnecessary. The pastor was happy to answer my questions and displayed not

1 4 • O U T S I D E R S

the slightest bit o f embarrassment or defensiveness over having received money from abroad. H is answers confirmed what I had already learned, to which he added several details. The foreign organization that had paid for his new church had later sent someone to take a photo o f the completed build � ing, which they published in a newsletter. H e had had no contact with them since, but he helpfully dug out a copy o f the newsletter carrying a photograph o f his church. H e was proud to have been featured in a foreign publication , which he interpreted as worldwide recognition o f the good works he was doing for the people o f Kashtappat tinam . The newsletter’s purpose, as he understood it, was to “ tell the world” about his m in istry and about the many miracles God was performing among the poor. The newsletter was in English, and reading it one could get the impression that Senthil K um ar ’s success could be credited to their own organization ’s support, and even that he h im � se lf was a creature o f their global program . H e was perplexed when I told him this. “But I built up this congregation myself, with only G od ’s help! Those foreign evangelists had nothing to do with it. A part from the new building, everyth ing was pa id for by the people in my flock who were with me from before, some from the very beginning o f my ministry. The foreign donors heard o f my good works. They came here and they saw that our church was crowded with believers. There was nowhere to sit, and the roo f leaked when it rained. They wanted to help me by build ing this better church for us to worship in. That is all.” In the name o f thoroughness I double-checked his story with other pastors, especially known rivals. Everyth ing he had told me was true.

The other pastor said to have received foreign money, Pastor Vijay, turned out to be just a part icularly charismatic leader, whose superior church build � ing was fully pa id for by donations from his own congregation. Rumors o f foreign money, though un founded , com forted less popular pastors by ascrib � ing V i jay’s success in at trac ting followers to someth ing other than superior abilities or greater effort. W hen questioned more closely, other pastors admitted that they had no posi tive knowledge that he had ever received for � eign funds and that this was merely a matter o f speculation.

W h a t is significant here is the lack , among pastors, o f any sense that for � eign money was moral ly compromised. This is interesting because in India more generally the possibil ity o f “ foreign money” supporting churches is invoked in ominous tones, suggesting mal ign antinational interests. Such ta lk draws on an old trope in Indian politics, the interfering “ foreign hand,” an unspecified threat famously cited by Indira G andhi during the Emergency

O U T S I D E R S 15

to just ify repressive measures and explain away policy failures. Prime M inister G andh i ’s frequent resort to the “ foreign hand” bugaboo has since become a matter o f general ridicule in Indian public life and a way to poke fun at the flimsy grounds on which succeeding generations o f national leaders have invoked the specter o f foreign influence to shore up poli tical support or attack rivals (e.g., B. A l i 1986; A rora 2 0 11; “ D eadly ‘Foreign H a n d ’ ” 1990; “ ‘Foreign H a n d ’ Aga in ” 1999; K aur 2002; Singh 2004 ; Romesh Thapar 1983). But what such ridicule targets is the outlandishness and implausibili ty o f leaders’ claims, not the underlying assumption that foreign influence is mal ign and to be avoided. This assumption is shared by actors across the spectrum o f elite politics in India, though how the label “ foreign” is applied varies according to polit ical ideology. Thus, for India’s communist parties, American influence was foreign but not the dictates emanating from Moscow (Sherlock 1998). For H indu nat ional ists, remittances by overseas Indian Christ ians and Muslims count as foreign influence, but not the substantial flow o f money into H indu nat ionalist organizations from N orth A merica’s powerful Indian business communities.

The rejection o f “ foreign influence” is a feature o f all modem national � isms, which follows from a core precept o f national ideology, that o f national self-determination or autonomy. Self-determinat ion is espec ially dear to postcolonial states, born as they were o f a struggle against foreign rule. But the inte l lectual trajectory o f the Indian independence movement has endowed autonomy with a religio-spiritual dimension that goes beyond the usual nationalist ones and continues to play a profound role in elite common sense. From its origins in nineteenth-century Bengal, the push for self-rule rested on unique arguments (Chatterjee 1986). To most contemporary observers— natives no less than outsiders— India was not a nation at all but a geographic region home to a mult iplic ity o f cultural ly and linguistically diverse peoples, whose historical allegiances had been to mutually antagonis � tic polit ical traditions that actively resisted subsumption under a single state. The closest the subcontinent had come to polit ical un ity was under Brit ish rule, and being forcibly brought together by a foreign power was no argument for national identity. Early nat ional ists in Benga l devised an ingenious response. Ind ia ’s internal d ivisions— l ike its subordination to foreign power—were very real but perta ined only to the vulgar, material domain o f economic and public life. Aga inst this they posi ted an “ inner” spiritual (and implic itly H indu) domain , associated with the domestic sphere, in which the nation was not only already unified but the “undominated , sover �

1 6 O U T S I D E R S

eign master o f its own fate” (Chatterjee 1999, 6 ,121; Sartori 2008). Arguments for national autonomy in British India were thus not merely political. They were premised on claims to autonomy in a uniquely Indian religio-spiritual realm.

This line o f thought was further developed by the “ Father o f the N a t ion ,” M . K . G andhi (1997), whose novel theory o f swaraj (self-rule) sought to link the autonomy o f the nat ion to that o f the individual . N o t only must the nation reject foreign influence in the geopolit ical realm, but, more funda � mentally, every individual must cultivate an ethos o f self-reliance and per � sonal self-control (Terchek 2011). N a t ion a l autonomy depends, in the G andhian worldview, on personal autonomy and not the other way around (Parel 1997, lv). A ccepting a id— monetary or otherwise— from others is anathematized in G andhian ideology because it violates the principle o f self- reliance and opens the se lf to the corrupting effects o f outside influence. A id that comes from Christ ian sources is particularly threatening to the nation G andhi fathered, not just because it is “ foreign ,” but, worse still, according to him , because it i l legitimately mixes worldly and spiritual concerns and undermines persona l autonomy in the most important realm o f all, the religious.

A s we w i l l see, the people o f A nbu N agar did not enshrine autonomy as a sacred moral principle. Their own most deeply held moral intuitions centered on a very different ideal, one that is in some sense its opposite. To be human , for them, was not to be autonomous but to be profoundly and irreducibly connected with others. These connections were at once moral and material, and involved positive obligations to others in relation to which the contriv � ances o f modern nation-states and the regimes o f border control on which they depended were o f no relevance. Slum pastors’ fai lure to stigmatize “ foreign money” was rooted in this alternative moral understanding and not in part ia l i ty toward their own corel igionists abroad. The pastors o f Kashtappattinam were confident that money flowing into India from rich countries in the M iddle East, through Muslim organizations, was also moti � vated by the basic human impulse to care for those in need. A l l over the world, they were quite sure, were people who just wanted to help. The urge to help those in need—poor people such as themselves—was not a specifically Christ ian impulse, according to these pastors, but a basic human one.

The salient distinction for the people o f A nbu N agar was not between conationals and foreigners, but between those who embraced the impulse to care for others and those who, out o f selfishness, refused it. Thus, apart from

O U T S I D E R S 17

not perceiving the world beyond India’s borders as opposed to their interests, they emphatically rejected the idea that conationals were necessarily allies. “ It is only polit ic ians who ta lk badly o f ‘ foreign influence,” ’ one pastor exclaimed in response to my repeated attempts to elicit nationalist distrust. “But that is only because they want us to stay poor!” H e continued, “People from other countries try to send money to help the poor, but the polit ic ians block it. They are afraid. They are afraid that we w il l become educated, and rise up. But slowly, slowly, because o f Jesus, we are rising up anyway. The rich keep trying to stop it but they cannot.” H is name was Mat thew , and like all pastors he perceived Jesus as the poor’s ult imate ally. Yet he equal ly praised the efforts o f Muslims in the Emirates, who he believed were sending money to fund schools and charities in India. Indeed, none o f the pastors I knew saw charity coming from non-Christ ian sources as inferior, let alone threatening, though all maintained that only Jesus could guarantee ultimate salvation . In some ways, even M uslim and H indu charitable organizations were seen as doing Jesus’s work , insofar as the poor benefited. “ W ho is afraid o f ‘ foreign money’ ?” another pastor asked rhetorically when I asked him why money from the M iddle East should be a source o f worry for other Indians. “ O nly the rich! W hy? Because they want the poor to remain their slaves.”

It was not just pastors who saw things this way. O rdinary H indus and Christians in A nbu Nagar, I discovered, also failed to perceive foreign money as threatening or tainted. N or did they share another view G andhi popular � ized, that there was something irreligious about the desire for money, or for the worldly goods that money could buy, and that religion should be directed solely to spiritual ends. Money itse lf was not seen by them as moral ly suspect, and they betrayed no sense that money and religion ought to be kept separate. Money was not a bad thing. It was very useful. It was exactly what slum dwellers lacked and desperately needed. Immora l i ty was a characteristic not o f money but o f people, and the paradigmatic form o f immora li ty in their eyes was the refusal o f the haves to share with the have-nots. Lik e slum-born pastors, ordinary H indus and Christ ians in A nbu N agar envisioned the world outside India’s national borders in curiously positive terms. W hatever money foreigners had sent, they assumed, was sent with love and had been earmarked for benevolent purposes.

A mong the benevolent purposes ordinary slum dwellers attributed to for � eign money was proselytization . U nlike nationalist elites, they saw efforts by religious people o f all fai ths—Pentecostal pastors, the Ramakrishna Mission (a H indu organization), local M uslim preachers, and so on— to propagate

1 8 • O U T S I D E R S

their teachings as a good thing. They were prepared to take at face value the claim by missionaries o f all faiths to be motivated by love and concern for others. L ik e most Indians they had heard rumors that elsewhere in India C hrist ian pastors were offering cash or gifts in k ind to potential converts, and they did not doubt that these rumors were true. But when I asked whether they’ d ever heard o f local pastors giving such gifts they scoffed. “ They say Christ ians are doing that [i.e., offering gifts to those who convert] in other places,” one H indu man laughed in response to my queries about the practice, “ but around here all the pastors do is take!” H e did not see such gift giving as morally suspect, and he added that i f local pastors were as generous as their counterparts elsewhere he would hold them in higher regard. O ther slum dwellers interpreted the absence o f such “programs” in Kashtappattinam— programs they believed were being made available to potential converts in other parts o f India— as yet another symptom o f the area’s neglect. “ H a! Do you th ink anyone cares that much about usT one woman commented. “ D o the politicians care? D o the pastors?” The fact that none o f the local, slum- born pastors made such offers was cited as an example o f how stingy they were. W here national public sphere discussions saw the alleged mixing o f material and religious goods as a self-evidently ill ic i t practice, and as a sign that Christ ian conversion efforts were not sincerely religious but driven by “ulte � rior motives,” many in A nbu N agar took local pastors’ failure to offer gifts as evidence o f their insincerity. I f they really cared about saving people, surely they would put their money where their mouths were!

C ou ld such gifts have “ induced” anyone to convert, as the language in the mainstream public sphere has it? The idea that any sane person would make such a serious decision on this basis was simply inconceivable to the people o f A nbu N agar. D oing so would not have been immora l or irreligious, however, just impossibly foolish. In the slum both Christ ianity and H induism revolved around the hope o f worldly benefits, but the value o f missionary gifts, no matter how generous, paled in comparison to those that gods were capable o f providing— everyth ing from divine help in finding a job to the preservation o f dear life. This did not mean gifts offered to potential converts would have been meaningless or unappreciated. Though strictly irrelevant to evaluating the merits o f different gods, such gifts would be significant for what they said about the character o f the pastor offering them.

To al l outward appearances, most pastors were as poor as anyone else in the slum, although a few had had better jobs than the average slum man prior to receiving their call. So the fact that they fai led to give gifts to potential

O U T S I D E R S • 1 9

converts, as better-off pastors in other parts o f India were assumed to do, was not always held against them. But the idea that foreign organizations might be sending money to Indian pastors was fuel for intermittent , low-level speculation in and around A nbu N agar that pastors were not giving all they could. Thus, i f a local pastor somehow came into possession o f a brand-new bicycle, for example, people might begin to get suspicious. O nce during my stay this was exactly what happened. A pastor was r iding a new bicycle, and immediately rumors began to circulate. A week or two later, this part icular pastor’s granddaughter came to church in a dress no one recalled ever having seen before. Shortly thereafter, I heard that someone else had noticed that the same pastor’s wife was wearing a white sari that looked unusua lly bright. Women in the congregation began to wonder aloud: Was that the same sari she’d been wearing all month? D id anyone recall seeing that one before, or had the pastor gone and bought her a brand-new one? It was not easy to tell with a white nylon sari— one could never be sure— but hadn ’t hers had a small tear at the edge? A n d hadn ’t the pastor h im se lf gotten a new set o f dentures earlier that year? O r had he always had those?

The bicycle, I was eventually able to determine, had not been purchased with foreign funds. It had been given to the pastor by a friend o f his, a D a li t Christ ian police officer who kept his religious identity hidden from fellow officers because, as he told me, he did not want to be passed over for promo � tions. The bicycle he had given was not actual ly brand-new, it turned out, only somewhat new. The pastor’s dentures were not new at all, but quite old, and the pastor wore them only on special occasions. The pastor’s w ife ’s sari, how � ever, was indeed new. The intense scrutiny o f pastors was unusual, but it dif � fered only in degree from what anyone residing in A nbu N agar faced. Slum dwellers kept close watch on one another, too, and even ordinary slum dwell � ers were at pains to downplay any bit o f good fortune so as not to appear to be selfishly accumulating while others were in need. Suspicion fell especially strongly on pastors, in part because they lived on the donations o f their con � gregation— donations that were meant prim arily to cover church expenses and only the basic necessities o f life for the pastor and his family. But the more important reason pastors were subject to such scrutiny was the be l ief that pastors could potentially be in contact with foreigners. A n d foreign money, everyone was quite certain, was money that was meant to help the poor.

The tone o f moral critique that I in itial ly detected in talk o f foreign money had turned out to be based on assumptions quite different from those at work in elite nat ional ist discourse. W hereas national media in India

l O • O U T S I D E R S

stigmatized foreign aid as such, slum dwellers worried that lovingly purposed money was being intercepted and, instead o f reaching its intended beneficiar � ies, diverted to selfish ends. W h a t appeared to go unquestioned in the slum was the assumption that people living outside India wanted to help people like themselves and that foreigners were on the side o f the poor, unlike other Indians, who were against them. W hy?

The imperative to share was a basic axiom in the moral world o f the slum dweller. In A nbu N agar those who had anything extra— food, cash, a kitchen utensil—were expected to share freely with those who lacked. For the most part this was exact ly what they did. I f anything, they were generous to a fault. The demand to share made accumulat ing any k ind o f household savings d if � ficult, though there were a few moral ly acceptable techniques to get around this. W e w i l l examine these techniques in chapter 3. W ha t is important to stress here is that the in junction to care was not just a pious wish and that it was applied to anyone in need, inc luding outsiders. It was central to slum dwellers’ sense o f themselves, and indeed, to their ideas o f what it meant to be human . The failure o f other Indians—politicians, shop owners, employ � ers— to care for people like themselves was, conversely, a def in ing problem for slum dwellers, a perpetual complaint, and a recurrent theme in their own moral narratives.

W hile I do not endorse the stark dichotomy between themselves and the majority o f their countrymen that this narrative implies, let alone the nega � t ive and highly stereotyped picture o f other Indians it rests on, I nevertheless do my best to show why it seemed true to these part icular slum dwellers in light o f their experiences. A s for the myth o f the benevolent foreigner— the idea that foreigners o f al l races were just like themselves, and unlike other Indians, in their spontaneous desire to care for others— I find this much more diff icult to explain . There was no direct empirical basis for the positive image they had o f foreigners; I was the first foreigner most slum dwellers had ever met. M any slum dwellers had never seen a foreigner in the flesh, and I could find only one who had spent significant time with a non-Indian .1 N or are foreigners generally depicted as the good guys in Indian films, television,

or other media— quite the contrary. W hile it is true that D a li ts experienced great gains under foreign rule, and

that subaltern caste leaders in the late colonial period, like the great anticaste leader B. R. A mbedkar and the T amil non-Brahmin agitator E. V. Ramasamy Periyar, tended to see the British as a potential al ly in their struggle against high-caste H indu nationalists, this cannot explain present-day slum dwellers’

O U T S I D E R S • 21

ideas. The slum dwellers I knew had only the vaguest knowledge o f Ambedkarite or Periyarist thought, and I was unable to connect anything slum dwellers said about foreigners to either o f these leaders or to the contem � porary movements they spawned, which in any case do not today emphasize their founders’ positive stance toward the foreign. Finally, although it is con � ceivable that the myth o f the benevolent foreigner originates in some sort o f collective memory among slum dwellers o f their ancestors’ positive interac � tions with foreigner employers and patrons (Roberts 2015a, 240; V iswanath 2010), and al though I did find examples o f such memories persisting in the present day, these memory traces were too few and insufficiently robust for me to posit a causal relation.

In short, how exactly the myth came to be we may never know. But though I cannot offer an empirical account o f its emergence, there is a possible struc � tural explanation . In order to make that argument I must first explore in some depth, in chapter 2, the unique set o f relations the people o f A nbu N agar perceived among caste, care, and the human .

A F O R E I G N E R I N T H E F L E S H

We have seen how, upon closer inspection , rumors o f foreign money in A nbu N agar disclosed something different from , and in some ways opposite to, what I was in it ia l ly led to believe. This was not the first time during my field � work that this happened. The most important reversal occurred a couple o f months into my study, when I fina l ly faced up to the fac t that simply spend � ing my days in the slum was not enough. I would have to live there. A n d what changed once I moved in was the way slum dwellers described the moral ity o f the dominant, nonslum society. It was not that my understanding o f slum dwellers’ words changed, as it had regarding rumors o f “ foreign money,” but that they began to reveal sentiments to me that completely contradicted what they had previously expressed. To account for this reversal, I need to say something about my own posi tion with in the slum and the nature o f my relationship with those who lived there. H ere is where I address what anthro � pologists call questions o f access. H ow did I gain entrée to the field site? W hy were people w il l ing to ta lk with me? C an we trust what they had to say? A nd for this we need to go back to the beginning.

M y arrival in A nbu N agar was completely unannounced and, unusual among anthropologists, I came without any sort o f introduction by a prestig �

1 Z • O U T S I D E R S

ious outsider such as an N G O worker, a local notable, a politician , or an activ � ist. Residents’ response to me was not what I expected. W herever I went in T amil N adu I had always been warmly received and shown a form o f polite � ness known as mariydtai. This is the standard way any visitor, save for a servant or underling, is treated. M a r iyd t a i is usual ly translated as “respect,” and it encodes hierarchy (Scherl 1996, 91-142) , though it does not necessarily mean that the recipient is a superior in all contexts.2 By showing m ar iydta i Tamils demonstrate their own good breeding and manners. A t a minimum this entails using the respectful second-person plural, nihk a , rather than the infor � mal singular, ni. The people o f A nbu N agar were also very warm but showed me no m ar iydta i at all. I was not, for example, immediately offered a seat. More strikingly, I was addressed as ni. W hen I told this to a friend, a middle- class scholar, he laughed with surprise. According to him this just showed how completely uncultured such people were. The only sense that could be made o f their behavior was that they simply did not know even the basics o f how visitors were to be treated, that they were like jungle dwellers. Such people, he concluded, “may speak T amil but in fact they are not even T amil at all.” M y friend was wrong. The people o f A nbu N agar knew as well as he how one was supposed to behave. A s I soon learned, they were perfectly capable o f showing mar iydta i and did so un fa i lingly with other outsiders who visited the slum— N G O workers, police officers, teachers in the local school. W ith such people slum dwellers adopted a scrupulously deferential posture, whether inside the slum or out, whether the outsider was someone they were meeting for the first time or a man or woman they had known for decades.

I did not mind the lack o f mariydta i. In fact, I found it a bit o f relief. But there was something else, which I did not disclose to my scholarly friend. N ot only did they fa i l to observe the usual forms o f politeness, but from my very first visi t to A nbu N agar I was, in addition , subjected to some rather wicked mockery. This remained a persistent feature o f my interactions with them for some time, in fact , and their banter often involved a sexua lly suggestive dimension , such as asking me in front o f everyone whether I “ l ik ed” a par � ticular teenage girl among the crowd. This was even more embarrassing for the girl than it was for me, and in such cases it was clear enough that she was the main target o f jok ing and not I. This k ind o f sport was not at all to my lik ing, but I grew gradually inured to it. W h a t only gradua lly became clear to me was that al l th is— the lack o f mar iydta i, the ribbing—was to be wel � comed. They were treating me exactly as they treated one another. This is not to say they saw us as identical. Too much separated me from them for that to be

O U T S I D E R S • 2.3

possible: my skin color, the fact that I did not suffer as they did, that at the beginning my command o f T amil was rather mediocre. I was different from them, but it was a horizontal difference, not a vertical one. I was different, but not a social superior. That I should have been accepted to this extent was not because they recognized something special in me. It was entirely accidental. W hether because I happened to be o f foreign birth, because my entry was unburdened by introductions from high-ranked outsiders, or some combina � t ion o f these things, I cannot say. But I am grateful for it because it meant I was halfway there— not to becoming “one o f them” but to being allowed to hear what they had to say about the dominant society and their moral cri � tique o f it.

M y initial plan had been to live in a comfortable flat in another part o f town, with my partner, and to spend only my wak ing hours in the slum. I had no desire—for reasons that w i l l soon be obvious— to actually live there. For my first two months my routine was to arrive in A nbu N agar first th ing in the morning and return home at nightfall . I managed to learn a lot in this period, but at some point I began to sense that I had hit a wall. I had no way o f knowing what exactly I was missing, but I felt sure there was something. So I rented an ordinary, single-room dwell ing for 400 Rs. ($8) per month and moved in, spending just one night per week with my partner in the upscale neighborhood of N ungambakk am . There my rent was 12,000 Rs. per month ($240), which for that neighborhood was a bargain . M y routine was to leave A nbu N agar on Sunday afternoon , after part ic ipating in a gruel ing (for the nonbeliever) two- to three-hour service in one o f K ashtappa t t inam ’s churches, and to return again to the slum on M onday morning. I shared the house I had rented in A nbu N agar with my assistant, Sagayaraj, a young D a l i t man whose parents and brother were landless agricultura l laborers in rural T amil N adu . Sagayaraj had managed to get a college degree, but like so many D ali ts in his position he was never able to land the sort o f job a caste person with his credentials could expect.3

Living in the slum altered my standing among those who lived there. I was no longer just some k ind o f in formation collector, like a census taker, only one who asked many more questions and never seemed to stop. A s a resident I no longer had to just ify my presence. I could just hang out, like everyone else, with no purpose besides passing the time and en joying others’ company. But social personhood created new problems. There is no such thing as a slum dweller in the abstract: all slum dwellers were part o f multiple networks o f kin and friendship, which, though ever shifting in their composition, were

2 4 • O U T S I D E R S

periodical ly at odds with one another. To become an ordinary member o f their social world would have meant be ing part o f one or another clique and tak ing their side in squabbles, or at least expressing sympathy with their posi � tion as against others.

M erging with the flow o f everyday life, becoming an insider: these are experiences many anthropologists prize. I do not know to what extent I could have achieved such a status, had I pursued it. But, for methodological reasons, the ideal o f merging with my informants was not one I sought. I worked hard to preserve my outsider status with respect to intraslum politics, not allowing myse lf to be identified with any group. But outsider status, as many anthro � pologists before me have learned, does not mean having less access to so- called inside information . O n the contrary, by ma in ta in ing strict neutrality with respect to the constant dramas o f slum life and the intrigues slum dwell � ers loved to discuss privately and attempt to decode, I was privy to knowledge I would not otherwise have had access to. A s a social outsider with a reputa � t ion for strict probity and never repeating anyth ing I was told, I was more often than not the one opposing parties in any given con flict would come to in order to discuss their interpretations o f events. Third parties would speculate to me about how much those on either side knew and did not know, what their true purposes were, the nature o f their underlying motives, and so on.

A complete account o f intraslum intrigues could fi ll an entire monograph. In this book I describe only a few such dramas, which shed light on my cen � tral concern with morality, religious conversion , and the structural basis o f con flicts between husbands and wives, and among debtors and creditors. These stories come in chapters 3 and 6. M any o f the dai ly dramas I witnessed had no direct relation to my research concerns but were nevertheless signifi � cant in another way. D issecting social dramas was i tse lf an important part o f slum sociality, and part ic ipating in such discussions was a key component o f my own socialization . A t tending to them, I gained a practical expertise in the informa l logic o f slum life. M y learn ing curve was steep, but I had another advantage fully integrated members o f slum society did not. In addition to rema in ing outside the slum’s various cliques, I was an inveterate information collector. Fu ll participants in social life simply do not have time to keep sys � tematic records o f everyth ing they are told and o f their own firsthand obser � vations. N ote tak ing is a luxury, as Pierre Bourdieu (1990) has observed, available only to those who do not have to worry about the practical necessi � ties o f da i ly life. The fact that I kept da i ly records o f all I was privy to, and

O U T S I D E R S 2-5

spent some two hours or more each day organiz ing and reorganiz ing my notes, meant that I had far more in formation at my fingertips than I could possibly have had by memory alone. Field notes are a powerful technology. A nd the fact that I reviewed my notes regularly allowed me to correlate d if � ferent versions o f events, discern patterns, spot discrepancies, and so on, to an extent far exceeding what my unaided mind would have been capable of. M y apparently wizardlike insights and powers o f recall made me a popular gossip partner, despite the fact that con fidentiali ty requirements meant I could seldom reveal all that I knew. I could tell I was get ting the hang o f slum life when , with increasing frequency, others began to ha i l the uncanny plau � sibility o f my interpretations.

The process I have just described is well understood by anyone who has conducted extended ethnographic research. But for two reasons I feel it is worth mentioning. First, it addresses a perennial question in graduate and undergraduate anthropology seminars: whether an outsider can ever tru ly understand another social world (or “culture”), especially a world that is very different from one’s own. Second, the way “ insides” and “outsides” are con � structed in the first place, and what it means to belong, are themselves central themes o f this book . “ Insider” and “outsider” are not neutral categories that simply describe positions relative to some already existing totality. They are themselves signs whose meanings are variably applied with in different lan � guage games (including that o f the anthropologist), and, as we have already seen, their social salience is not always as ideologies o f national culture and foreignness would have it.

A n ethnographer’s knowledge is always partial , o f course, but that is true o f all knowledge. It is equally so with the knowledge “natives” have o f one another. A l l members o f a social group have secrets, which they share with only a select few or with no one at all. But anthropological knowledge is not we ll characterized as the uncovering o f secrets. Secrets may occasionally be learned, but what is anthropologically relevant is not what is secret but what is public. Before asking how an outsider could understand the insider, one should thus ask how insiders understand each other. W ha t we call “culture” is not a mystical mind-meld join ing insiders but exc luding others. It exists only through language and through other public ly observable signs and actions. O f course people shape what they say to what they th in k you want to hear— or to what they want you to th ink . That is true o f all communica � tion, including among so-called insiders. But what people tel l one another is not always what they intend and is always much more than what they say. The

O U T S I D E R S

only way for ethnographic subjects to keep a diligent anthropologist in the dark about major aspects o f social and cultural life, even assuming that this is something they would want to do, is by refusing to speak to them at all (cf. C lastres 2000 , 96-97) .

Aga in , ethnographic knowledge is always part ia l. O ne early reader o f this ethnography felt it was written in a way that suppressed male voices. I do not quite accept this criticism. Suppression implies deliberately keeping some � th ing from the reader, but the truth is simply that I know much less about men’s worlds than women ’s. I make no claim that the picture o f slum life I offer is comprehensive, only that it is accurate. Choices must be made, and for every path that is taken another is foreclosed. O ne decision I made was to actively avoid being pulled into the world o f male social ity— to becoming “one o f the guys.” A mong other things, this would have involved drinking, going to films, and philosophiz ing about love and life. Various circles o f unmarried men seemed eager to make me part o f their world , but had I allowed this to happen , it would have been difficult for me to do the sort o f research—which was centered on the lives o f married women around my own age— that this book is based on.

That I was so readily accepted by the married women o f A nbu N agar was at first surprising to me, since, except among the cosmopolitan elite, men and women in T amil N adu generally lead very separate lives. N ot so in A nbu N agar. W omen usually spoke frankly with me about what are normal ly con � sidered private or taboo matters, though there were limits on the sort o f information women would share with me directly. They would tell me about their quarrels with their husbands, for example, but not about their sexual relations with them. That was fine with me, as the details o f people’s sex lives were not among the topics I had come to study. In the course o f my time in A nbu N agar I did, however, learn a fa ir amount about this topic, because a few o f the women there happened to feel like ta lk ing about it from time to time—just not directly. O ne o f the most amusing ways one woman found o f gett ing around this unwritten rule was to address her commentary not to me but to a third party— one o f A nbu N agar ’s male-to-female transvestites— who was si tt ing w ith us. W e were all sit ting together in close quarters, and I could hear what she had said as we ll as anyone else in the room . But the important thing, it seemed, was that the woman ’s disclosure was not addressed to me. Formally speaking, I was not the addressee but an overhearer (Goffman 1981). The transvestite, Pushpa, then turned to me and repeated verbatim what the woman had just said. I responded, not to the woman, but

O U T S I D E R S • Z 7

to Pushpa, who then repeated my response to her as i f she had not already heard it!

Pushpa was one o f five transvestites who lived together in A nbu N agar and referred to themselves as a family. Their “mother” lived elsewhere. They moved to the slum several months after me, thereby providing me an oppor � tunity to observe in real time the community’s reaction to a new and very different presence in their midst. Despite much hype about how traditional homosexual transvestites (known as a lis or a ravdn is) are an accepted part o f life in T amil country, and have a recognized place in certain religious rituals, the a lis I knew described a great deal o f discriminat ion and abuse.4 They told me that was why they often lived in slums: no one else was w il l ing to rent to them. Some o f the other alis they knew, they revealed, including one member o f their extended “ family,” lived closeted lives. D uring the day they dressed as men and worked regular jobs, donning female attire and makeup only on weekends and for special evenings out. In the slum, by contrast, they were fully accepted as part o f the women’s world. Young men generally avoided them, for fear that too much interaction might be interpreted as sexual inter � est, and because a lis are stereotyped as sexua lly ravenous. But slum men were usual ly polite to the alis, and I never witnessed a lis subjected to the abuse they experienced outside the slum context.

Though I became fairly close with Pushpa, a lis do not play a major role in the story this book tells. I mention their reception in A nbu N agar for what it says about the rest o f the slum community and their unusual openness to others. A ccording to the moral self-narrative o f the people who lived there, they were unlike other Indians insofar as they did not discriminate against others and accepted all people on the basis o f their shared humanity. H ow seriously to take such talk was not in itial ly clear to me. G iven that no one else ever seemed to want anything to do with them, opportunities to discriminate simply did not arise, and I wondered what their professed openness to al l people amounted to in practice. I was thus lucky to have been presented with this example, which helped persuade me this was not just talk .

Belonging, as I have already stressed, does not imply common identity. Pushpa and her sisters were a distinct presence in A nbu N agar, markedly different from other slum dwellers. They were like me in that way. A nother way they and I were alike was that we hung out mostly with women (though unlike the a lis I spent a fa ir amount o f time with men as well). Despite our visible differences, we were both socially accepted to the extent o f being able to forge intimate relations and build enduring solidarities with others. Social

z 8 • O U T S I D E R S

closeness was not obligatory, and not everyone was interested. Just as unmar � ried men generally shunned the alis, and just as I in turn kept a certain dis � tance from such men, some slum dwellers showed li tt le or no interest in spending significant time with me. There were a few households where, even after almost two years, I never felt free just to “ drop in” any time o f day, without purpose.

Intimacy, as opposed to mere acceptance, implies a substantive connection between persons and includes the possibility o f confrontation . O ne impor � tant way such connection was manifest in A nbu N agar was through a cul � tural practice known as t it tu or t i ttuta l. This can be translated as “scolding,” “ haranguing,” or “rebuke.”5 The people o f A nbu N agar frequently “scolded” one another— that is, criticized one another’s behavior, appearance, and so on—publicly and in private. They had a strong sense o f right and wrong that extended not just to moral matters but also to what seemed merely stylistic or aesthetic. I was scolded, for example, for not shaving one day. “You shouldn’t just let your whiskers grow,” I was told. “ It looks fil thy [acinkam]!” I was also scolded for rema in ing shirtless after my evening bath. Though many slum men walked around shirtless throughout the day, I was told that I must always keep my shirt on. O therwise, my scolder informed me, people would say I was “showing off.” U nlike other men, I was large and we ll fed.

Ce lvi , a married woman just a few days older than myself, whom we w il l meet in chapter 3, once complained to me about how, as a married woman , she was required to wear a sari all the time. H ow much easier and more com � fortable it would be to wear a cutitdr, she exclaimed, referring to the long flowing shirt and pant combinat ion that was commonly worn by N orth Indian woman but in the conservative female dress code o f the slum was permissible only for teenaged girls. But why not just dress as you like? I asked. W h a t would happen i f you just put on a cut itdr one day? “ I couldn ’t,” she replied. “ Everyone would scold me.” I asked how she would react i f she saw another woman in A nbu N agar dressing that way. “ I would give her a good scolding!” C e lv i laughed at the irony o f the situation , but she meant it. Women were especially renowned for scolding in the slum, but everyone took an interest in polic ing one another’s behavior to some extent. Scolding was not merely tolerated in the slum but recognized as an important and neces � sary component o f social existence. Theirs was not a “ live and let live” ethos but a con frontationa l one in which correct and moral behavior was every � one’s concern. N ot a day went by without someone being scolded for some � th ing, and much ta lk revolved around who had been scolded by whom and

O U T S I D E R S • 2 9

for what. Those on the receiving end o f a scolding often simply accepted it, though sometimes they would argue back. They would deny having done what they were accused of, but the principle that those in the wrong should be scolded went unchallenged. There was no saying “Buzz o f f!” or “ M ind your own goddamn business!” in A nbu Nagar.

In the previous section I sketched a distinction between moral paradigms centering on autonomy and on connection. In one the focus is on self-reliance and personal or national self-rule; in the other, vulnerabili ty and the obliga � t ion to materially and morally care for those in need. Scolding was another aspect o f A nbu N agar ’s connection-based moral ethos. It is incompatible with the ideal in which each person is responsible for themselves.

In practice scolding did not fa l l on everyone equally. I never once saw a child scold an adult, for example. Women often rebuked other women , as well as men near to their own age and younger, albeit less frequently. They did not tend to scold men o f their father’s generation, however, at least not to their faces. M en rebuked other men much less readily than women rebuked other women because scolding could lead to fights, and fights between men were, as everyone in A nbu N agar said, serious business. M en rarely, i f ever, scolded women other than their wives or younger sisters—with the exception o f pastors, who scolded both men and women in their care with great alacrity. A s Yesudasan , whom we’ l l meet in chapter 5, explained, “Scolding is a pas � tor’s duty. A pastor is not a llowed to just look the other way.”

A s for me, I never scolded anyone at all. I could not simultaneously study slum society and part icipate in it to that extent. M y purpose was to under � stand slum life as it actua lly was, not push its members to conform to a set o f ideals I did not yet fully understand or endorse. But I did not just refra in from scolding. I also took pains to make clear that my visible abstention from male vices should not be interpreted as moral judgment. Thus I let everyone know that in my country premarital sex was the norm , and that I myse lf had habitually enjoyed a drink or two every evening and continued to do so when not in the slum. This frankness on my part had unexpected and important consequences. M y female informants, in turn , freely admitted to me that virtua l ly all their friends had had sexual relations before marriage and that they thought there was nothing whatsoever wrong with th is— though with � out exception mothers tried to hold their own daughters to a much stricter standard. They also confessed that they really did not mind their husbands drink ing and indeed felt men ought to drink , so long as they did not drink to excess or become violent. In both these cases the views they expressed to me

3 0 O U T S I D E R S

directly contradicted those they expressed in the presence o f high-status Indians, such as N G O workers. To such people slum women professed total opposi tion to male d rin k ing and, in con formity to T amil gender norms, generally presented themselves as chaste beings who had sex only as a duty o f marriage or when forced.6

W hen faced with a discrepancy like this, in which subjects espouse con � tradictory values to different audiences, it is tempting to suppose that one “ l ine” represents their true feelings and the other is just for show. This is the approach taken by James Scott in his inf luential book D om in a t ion a n d the A r ts o f Resistance (1990), where what he terms “ hidden transcripts” or “off � stage” lines are commonly understood by readers as disc losing an inner truth o f subaltern consciousness. W h a t subalterns say to elites or their “public transcript ,” by contrast, is interpreted as a merely instrumenta l discourse designed to secure advantages by appealing to elite preconceptions and values (see also Ba lagopal 1989). This way o f construing the situation commits us to a hermeneutics o f surfaces and depth, inner truth versus outward appearance. Intuit ively this makes a lot o f sense. A disadvantage o f such an approach, however, is its implication that people are always clear on what they really think . It may also be that they th in k different things in different contexts, or that their beliefs and values encompass contradictory tendencies (Crehan 2002). W h ile it is undoubtedly true that at times people knowingly dissimu � late for instrumenta l reasons, to construe a l l discrepancies between what people say in different contexts on this model is to foreclose the possibility that they may be at least part ia l ly hegemonized, as T im othy M itche ll (1990) has argued in a crit ical discussion o f Scott ’s earlier work .

M y account remains neutral between these alternatives, both o f which aim to say something about subaltern consciousness— namely, whether it is hegemonized or not. Rather than posit ing a contrast between surfaces and depths, I simply note the existence o f a discrepancy between what slum women said to different audiences. This leaves open the possibili ty that they really be lieved different things at different times. Consciousness is an inher � ently murky terrain , and we cannot reliably separate truth from fiction in all instances. So instead o f construing the discrepancy between how slum women portrayed themselves to elite advocates and what they told me pri � vate ly in terms o f surfaces and depths o f individua l minds, I map them socially. W ithout suggesting one “ l ine” is mere pretense and the other inner truth, we may nevertheless observe an asymmetry between the two at a purely external, or social, level. Simply put, slum women did not hide from me that

O U T S I D E R S • 31

in some contexts they expressed views that conformed with elite expecta � tions, but— and here is the asymmetry— they did hide their more tolerant attitudes about male drin k ing from N G O workers and other elite moral guardians.

This brings us to a more important discursive contradiction , signaled at the beginning o f this section: the complete reversal I observed after taking up resi � dence in A nbu N agar in what slum dwellers had to say about the dominant, nonslum society. W hen I was sti ll spending only my days in the slum, the people who lived there spoke glowingly o f lnd ian society as a whole and Tamil N adu in particular. Nonslum dwellers were goodhearted people, everyone wanted to help. Slum dwellers respected others, and others respected them. Life in the city was fun , and all men lived as brothers. It was only after I bodily committed myself to A nbu N agar that they began to express ideas and opin � ions that entirely contradicted what they had previously let on. A few o f the men stuck to their original line. These were men with direct ties to one o f Dravidianist polit ical parties ( D M K or A I A D M K ) that had ruled T amil N adu since 1967, and the celebratory picture they presented o f a unified T amil people matched exact ly the one their part ies promoted (Barnett 1976; Subramanian 1999). But these men’s wives told a different tale, as did women in general and the majority o f men. The picture they painted was diametrically opposed to the one I had initial ly been presented with. It was a thoroughgoing and often quite bitter critique o f the selfishness and inhumanity o f those they called “the rich” (panak kdrahka) or “the privileged” (vacatiyutaiyavanka), a category that included virtua lly all nonslum dwellers. W ith time I came to understand, furthermore, that “the rich” referred not just to those who had more money than them. It was away o f ta lk ing about the dist inction between D a l i t and non-Dalit, caste and outcaste, without naming it outright.

As with women’s attitudes toward male drinking, there was an asymmetry in what these slum dwellers said in different contexts, and depending on who was with in earshot. O nce, for example, I was sit ting in a roadside eatery with a slum man, a laborer in George Town , the old commercial center where most slum men worked. H is eyes were locked with mine as he spelled out in fierce tones what he saw as the true state o f affairs in T amil N adu between the rich and the poor, the uncaring attitude o f the former and the historical suffering o f the latter. The “ hotel,” as roadside eateries are called, was empty but for us. W hen a stranger took the table next to us, my friend concluded what he was saying in a whisper and quickly moved onto a different topic. The stranger was not visibly rich, though he was clearly not a manual laborer. W hy the

3 Z • O U T S I D E R S

caution? A nother example w i l l explain . It also nicely illustrates why terms like rich and poor cannot be taken literally.

There was a circle o f male friends in a neighboring slum who on one occa � sion had been part icularly forthright with me about their views about rich and poor. Their critique o f “ the r ich” matched we ll what I was told by the man in the roadside eatery, but I had had to cut the conversation short to make an appointment elsewhere. The men had been very vocal, and I was eager to hear more. I returned about a week later to their in forma l clubhouse, only this time I brought Sagayaraj with me. W hen they saw him they imme � diately offered us tea, which Sagayaraj refused. I caught two o f them exchang� ing what seemed to be a m eaningful glance, but Sagayaraj remained oblivi � ous. Despite my best efforts, I could not get them back on the topic o f rich and poor. Sagayaraj was clearly not rich in the literal sense o f the term; apart from the fact that he wore shoes, and an oversized pa ir o f nylon pants, he looked like any other slum dweller. I could already guess the source o f their reticence, but it was too late to do anyth ing about it. So after chatt ing a while about tota lly banal topics, I said goodbye and made a plan to meet them again the following week. “ Good idea,” one o f them said.

The next week I returned, again accompanied by Sagayaraj. O nce again they offered us tea, which Sagayaraj again refused. This time, however, I was prepared. “ H e doesn’t drin k tea,” I told them, which was true, “ but he’d probably like some warm milk .” They brought him milk , and Sagayaraj drank it. The men smiled. Con f ident now that my intuit ion was correct, I casually mentioned that Sagayaraj was not from the big c ity but a village. H e lived in the “colony,” I added— using a common euphemism for ceri, a spatial ly sepa � rated hamlet housing the D a l i t laborers attached to every caste village. “A h , we thought he was a . . . ” one man said with a smile, not quite finishing his sentence, and all o f them laughed. O ur conversation about “ the rich” then resumed as it had before Sagayaraj first joined us, and they no longer held anything back. Slum dwellers often used class terms to express ideas about the social world beyond the dictionary definitions o f those terms, something we must st i ll account for. For now I pose a more basic question: W hy did the people o f A nbu N agar begin , only after I had taken up residence there, to reveal views about nonslum society quite at odds with what they had initial ly told me?

Ethnographic research is concerned not just with what our subjects say— people say al l sorts o f things—but with what they do, in different contexts. W ha t is often not adequately stressed is the importance, epistemologically,

O U T S I D E R S • 33

o f the anthropologist’s own behavior. Just as we observe them, they also observe us. By taking up residence in the slum I had crossed a line that is rarely i f ever crossed in India. I had given them an opportunity to observe me more closely— bath ing and performing bodily funct ions as they did, eating their food , especially beef—and to decide for themselves whether I was indeed as they had hoped, and unlike those they called “the rich.” Though I had not set out to do so, through my actions I had demonstrated that I did not regard them as fundamenta lly different from myself—that I did not see them as they believed most other Indians did, as less than fully human . Simply being a foreigner, and hence in their view a theoretical ly casteless person, was not enough. I t was as i f I had to perform my castelessness at a bodily level before they could fully accept it.

A W O R L D A P A R T ?

So far as the people o f A nbu N agar were concerned, they and their neighbor � hood were radica lly distinct. That they were people who cared for others was central to slum dwellers’ self-understanding, and it was this more than any � th ing else that differentiated them, in their own eyes, from those they called “the rich,” the term they used to describe al l those who did not live in slums. I explore this dichotomy in their th in k ing and its social effects in the next chapter. For now, it is enough to note that in purely economic terms there is no clear dichotomy between slum dwellers and others, but a continuum . Some nonslum dwellers were in fac t quite poor and often deeply in debt and struggling to make ends meet. A n d some nonslum neighborhoods were only marginally better than those classified by the c ity as slums.

Yet there was a pronounced sense in A nbu N agar that they and others like them were be ing un fa irly treated, that they were not receiving the care that they deserved. They were sure that the city had the resources to improve their living conditions but simply did not care. The fact that more was not being done for them was a source o f simmering resentment. In what follows I describe what life was like in A nbu N agar at a very basic material level, begin � ning with the layout and built environment and its advantages relative to other slums. I then describe the two material features o f slum life that those who lived there complained about more than any other, features they related direct ly to their neglect by those who should be he lping them: trash and water.

3 4 • O U T S I D E R S

The first th ing to make clear is that A nbu N agar was in many ways a much better place to live than most other slums in the c ity and that this was one o f the reasons I chose it. U nlike most C henna i slums, which were illegal squat � ter colonies on margina l or unused lands, A nbu N agar was a “settled slum.” That is to say, the people who lived there (or their predecessors) had been given titles to the land , which was divided into com fortable ninety-five- square-meter plots. The original settlers, many o f whom were sti ll present, had been relocated to A nbu N agar under Congress Party rule in the early 1960s from an illegal squatter colony near the harbor, where many at that time were employed as loaders. The biggest advantage to its being a settled slum was that residents were free to build permanent and flame-proof dwell � ings. The majority o f slums in Chenna i are illegal, and are tolerated only so long as those l iving there make no permanent claim on the land by construct � ing solid dwellings for themselves. Thus in A nbu N agar, unlike other slums where people live under pa lm thatch or plastic tarpaulins, a bit more than h a l f o f the dwellings had roofs made o f tile, corrugated asbestos sheeting, or poured concrete.

A lmost every year one or two slums in C henna i are destroyed by fire. W hen that happens, newspapers reports frequently editorialize in ways that make slum dwellers themselves responsible for the ir own misfortune. Portrayed as feckless simpletons, eager to reap the economic rewards o f c ity life, the rural migrants who fi l l C h enn a i ’s slums are cited as an object lesson in fai lure to plan ahead. H eedless o f the risks, they erect temporary structures rather than saving their money and build ing permanent ones. The reali ty these stories miss is that slum dwellers in illega l settlements face evict ion the moment they at tempt to erect permanent, flame � p roo f structures.7 O ccasiona lly newspapers mention slum dwellers’ own theories about how these fires start— that they are deliberately set by people who want them out and do not want to go through the diff icult process o f having them legally evicted. W h a t I ’ve never seen mentioned, however, is the widespread view among slum dwellers that links slum fires to caste. A mong the most feared weapons at dominant castes’ disposal against rural D a li t slums, known as ceris, is arson.8 The avai lability o f flame-proof roofing in A nbu N agar was a significant amenity and was much appreciated by those who live there.

The streets in A nbu N agar were also considerably wider than those in other slums. A nbu N agar comprised two main streets, each a bit less than one hundred meters in length, and two much shorter cross streets. Illegal slums,

O U T S I D E R S 35

/

F IG U R E 2. Sewage channe l a longside house.

by contrast, are a claustrophobic warren o f narrow, unpaved footpaths with no drainage at all. W aste water accumulates in pools until it either evaporates or soaks into the earth. Roughly h a l f o f A nbu Nagar, on the other hand , was connected to some sort o f underground sewage or septic system. The plot I lived on even had a (nonflushing) toilet, one o f A nbu N agar ’s few. Beginn ing about ten meters up the street from where I lived, however, open sewage filled a small channel known as a cdkka ta i (figure z) running down either side o f the street, just outside the thatched sides o f my neighbors’ homes. D uring heavy rains cakkatais overflowed, f i l l ing the street with sewage that often seeped into adjacent dwellings.

A nbu N agar ’s original plots had been long since subdivided, each now accommodating some half-dozen or more dwell ings. The plot on which

3 6 • O U T S I D E R S

Sagayaraj and I lived accommodated six households and a total o f twenty-one residents. The interior o f my own house, which I rented at market rate for Rs 400 per month, was thirteen feet long and just under eight feet wide. The floor was slightly below ground level and made o f poured concrete; a two- inch lip prevented water from flowing in during the rainy season, which was a good th ing because, like most slum dwellers, we slept on the floor. The walls were masonry, and the roo f was made o f ceramic tiles laid over wooden slats. The dwellings on both sides were slightly smaller than ours. To the right lived A nbu , a casual laborer, with his wife and small daughter. To the left lived a family o f five.

Living in such close quarters with others did not bother people in A nbu N agar. C erta in ly they would have appreciated more space, but they did not regard it as something they were being un fa irly deprived of. N or were they at al l fazed by the things that bothered me most— like the rats that would scamper and leap around my room at night, searching for food and toppling stainless steel containers w ith a great clamor that left me pumped with adrenalin and unable to sleep; or the mice that scurried about in the rafters, mice whose droppings used to fa l l on me and Sagayaraj while we slept, and which later accumulated in the mosquito nets I set up for us; or the flies that, in certain seasons, covered al l surfaces and not just food and shit (see figure 3).

Pests used to freak me out, the rats especially. But they weren’t the k ind o f things other residents there even thought to complain about. W hen I told some young men about the rats knocking over my pots, they just laughed. I said I was th in k ing about buying a trap from the hardware store, and one o f them k indly offered to help me set it up, since he knew how they worked. But he was also honest. H e told me there really wasn ’t much o f a point. It was probably different rats each night, and even i f you caught one o f them, others would just come in its place.

W h a t bothered regular slum dwellers were the things they regarded as unfair, the result o f discriminat ion . W ater access was one o f them, trash was another. Rodents and insects were just symptoms, and they proliferate in any trash-filled setting. Pests are natural. But trash is a human product— and it is social. W herever human beings live they create trash, as any archaeologist can tell you. W e in the “ First W orld” create lots o f it. W e hardly ever think about it, and neither do most people in Chenna i . It just goes away; someone else takes care o f it. But in the slum trash would just stick around. Once in a while a truck might come around to pick up some o f it, but never enough to

O U T S I D E R S 3 7

f i g u r e 3. F l ies o n m y h e lm e t .

make much o f an impact on the visual landscape. A n d any improvement was short lived.

It did not have to be this way, according to residents, and they were right. O ther parts o f the city were serviced regularly by uniformed street cleaners and trash removers—who were, ironically, themselves slum-dwelling D a l i ts— hired to load other neighborhoods’ garbage into smart-looking trucks, not the filthy hauler sent every now and then to Kashtappattinam (figure 4).

“ W hy is our place al lowed to look like this?” a middle-aged woman asked me on one o f my first visits to Kashtappatt inam , in early zooi , before I even began fieldwork. I had told her I was a social scientist and wanted to learn about the problems people were facing. Unequal treatment from the c ity was a constant source o f complaint for the people who lived there, and the fact that garbage was allowed to accumulate was a major bone o f contention . “ We are also citizens [kut imakka[\ o f this c i ty!” one man shouted angrily when I asked him about the trash. “ They don ’t let other people just live w ith so much trash. W hy?” That municipal garbage collectors would fa i l to service illegal slums was one thing. But Kashtappatt inam was not an illegal slum. I t was an

38 O U T S I D E R S

F IG U R E 4 . G arbage removal.

official settlement, theoretically equal in status to any other neighborhood o f the city. A lm ost everyone there took for granted that waste was allowed to accumulate in their neighborhood as the result o f a deliberate policy. A nd they were enthused about my gett ing the word out, tell ing the rest o f the world what was being done to them. Because, recall, they trusted that outside India were people who would want to know this and who would care. Several times I was made to promise that I would not forget to in form people in my country about these things.

Toward the end o f my fieldwork I decided I would need some photos i f I was going to make good on my promise to them. I had hardly started when I was reminded o f something else trash signified for the people o f the slum: shame. I was about to snap a photo when a man I didn ’t know approached me. “ H ey! D on ’t take a picture like that___Show the people in your country something nice about us.” I assured him that I had taken many beautiful photos already but that the reason I had come was to learn the problems people were facing, and many had complained about the trash. “I promised I would tell the truth about how you are living,” I concluded. “ That would be good ,” he replied, looking at me intently, “ I hope you really w i l l show them.” Figure 5 is a photo I took just after this conversation. C erta in that something un justifiable was be ing done to them, the people o f A nbu N agar felt that they themselves were being treated as trash. That was a source sometimes o f out � rage, sometimes o f shame and humiliat ion .

O U T S I D E R S • 39

Because it was a potent symbol o f their condition , because it resonated so powerfully with their feelings about caste, trash was a natural image through which slum-born pastors understood C hrist ’s message. H ere is an example: “ Is the very place we live in a garbage dump? W h a t is o f no use is called trash. W ha t we throw out is trash. W h a t is not needed is trash___[Trash is] use � less,’ ‘unwanted,’ and ‘rotten .’ . . . [Those people,] assuming you were dirt , they threw you away.” H ie man who spoke these words happened to be a Christ ian , but the thoughts he expressed here were not spec ifical ly C hrist ian ones. These thoughts were everyone’s in the slum. For this pastor their wrongful treatment was linked to the promise o f overcoming, the promise o f divine justice. For others, it was just a horrible fact, unlinked to any messianic hope. I have edited the pastor’s words to strip them o f their theological con � text; we w il l revisit this quote in chapter 7 and see what additional meanings trash takes on in slum Christ ian discourse. For now it is only the generic character o f his words that matters.

To summarize, the slum dwellers I knew saw the fi lth o f their surround � ings as deeply un just and not at al l reflective o f who they were as human beings. C henn a i ’s slum dwellers lived surrounded by trash because they h ad to, not because “their tolerance o f garbage is much greater than that o f the middle class,” as the poor o f Kolkata , whom Sudipta Kaviraj writes about,

4 0 • O U T S I D E R S

apparently do (1997,107). N or did they accumulate rubbish and filth around themselves deliberately in order to repel the urban middle class and “symboli � cal ly establish control ove r . . . space” (107). A n d the garbage-filled landscape o f Kashtappat tinam , unlike that surrounding the urban poor whom Dipesh C h akrabarty writes about, did not stem from a rejection o f modernity, let alone a “refusal to become cit izens” (1991, 28). The slum dwellers I knew understood the trash that the c ity al lowed to accumulate in the place they lived as the de n i a l o f their rights as citizens, and indeed o f their humanity.

N ow let’s ta lk about water. “ The sun is so dangerous, Robert. I t can k il l you very easily,” Sagayaraj once told me, as we walked a five-hundred-meter stretch o f Kashtappatt inam ’s main road together one morning. It was only 9:00 a.m. and the sun was already st ingingly hot. This is the sort o f th ing that could have easily gone in one ear and out the other for me. But it stuck because o f something that had happened to me just a few days earlier. I had been sit ting with a group o f women and young men in A nbu Nagar, ta lk ing about the supernatural. They had been te ll ing me about black magic and about the types o f dangerous spirits that prowled the area—pey,picdcu, dvi , ceyvina iydvi. I must not have drunk water in a long time, because I had stopped sweating. Usua lly my body was completely wet throughout the day and my shirt soaked through. But at that moment I was dry. W hen I tried to stand up my vision went dark and I felt as i f I was spinning. M y first thought was that I had been drugged. I slumped to the ground as several pairs o f arms reached out for me, easing me down. Someone brought water, and before long I was back to work .

“ M aybe it was a ceyvinaiydvi,” a teenage wiseacre suggested. “Be careful or you ’ l l end up like her,” he laughed, gesturing to a mental ly retarded woman who had just wandered over. She was reputed to have been normal as a child but to have lost her mental faculties when a ceyvinaiydvi— a k ind o f spirit sent by a magician to do harm—had attacked her. The woman was well loved, but love and rough humor often went hand in hand in the slum, as I mentioned before. Spirit attacks are common in broad daylight, and their symptoms are identical to those o f fa in t ing from heat and lack o f water, which is even more common . Usua lly the only way to tell them apart is that when the cause is heat you recover quickly. W hen it ’s a spirit that did it, you just keep feeling terrible and have nightmares and the like unti l an exorcism is performed. I wasn ’t part icularly worried about spirits, but the feel ing o f losing conscious � ness from the heat was a lesson to me. So when Sagayaraj remarked, a few days later, “ The sun can k il l you very easily,” I heard his words, as I would not have

O U T S I D E R S • 4 1

otherwise. I understood bodily the seriousness o f what he was tell ing me. Sagayaraj was from a family o f landless agricultural laborers, people used to working all day in blazing hot fields, without any shade. The words he spoke to me came from that world, the world o f the rural D a l i t laborer. “ I f no one gives you water,” he continued, “you w il l soon be dead.”

In rural T amil N adu , a village’s best and most conveniently located water supply is reserved for caste people only, though exclusion o f D a li ts from com � mon water resources is technically illegal under the Scheduled Castes and Scheduled Tribes (Prevention o f Atrocities) A c t o f 1989.9 In former times, and to a great extent even in the present day (H uman Rights W atch 1999, 1-10 , 17 9 -104) , i f a D a l i t needed water from this source he or she was required to stand and beg for it. Thirsty D a l i ts cannot just take what they need to live but must find a caste person w il l ing to stand above them and pour water direct ly into their mouth, or into their cupped hands. They can continue drink ing only so long as a member o f the dominant village society is pouring, because when they stop it al l flows away. They are not given an actual cup or direct access to the we ll itself. Their very touch is polluting, and the purity o f the water source must be protected. That is what the caste people claim.

Da lits, it is true, always have their own water sources. But in the hot South Indian sun, a man or woman who needs water often needs it right away. The man or woman who is made to beg for water is not, under normal c ircum � stances, denied outright, though a D a l i t known for insubordination may be toyed with and even turned away. Begging in this manner is o f course also a h ighly stylized gesture. For this reason, in part, the arrangement is typica lly characterized in the ethnographic literature as a “ritual practice,” along with the many other acts o f self-abnegation D a l i ts are required to perform in the presence o f caste people. Some examples include removing their sandals when passing a caste person in the road, dismounting from their bicycle i f they are riding one, and remaining shirtless.

Focusing on examples like these, and on the ritual logic o f purity and pollution cited by those who enforce these practices, anthropologists in the colonial period came to conceptualize caste i tse lf as an essentially ritual rather than political-economic order, and this remains largely the case up to the present day (Roberts 2008, 2015b; G uha 2013; V iswanath 2010, 2014b). Those who collude in preventing D a li ts from accessing water are described in older ethnographic accounts as act ing on “caste scruples,” as i f to render another human being permanently dependent in this way were just an

4 Z • O U T S I D E R S

extreme form o f persona l fastidiousness. But that is not the only way to understand the situation . A noth er way to analyze this ubiquitous water regime is as a strategic practice. To control a persons water access is to put a sword o f D amoc les over their head. You li teral ly have the power o f life and death over them, and you remind them o f this fact every time they must beg “ritually” for water. R i t u a l dom ina t ion is an interesting concept because it can be interpreted in two ways: “r i t u a l dominat ion” and “ritual dom ina t ion .” W e can emphasize the ri tual part , thereby implic itly distinguishing what is going on from actual domination . O r we can put the conceptual stress on dom ina t ion , thereby signaling that a relationship o f domination between two classes o f people is reproduced, in part, through ritualized forms o f bodily tra in ing (Mauss 1973), also known as discipline (Foucault 1991). A s in the elaborate system o f racial etiquette in the American South o f yore, Gerald Berreman reminds us, in India “ deference is dem a nded and not merely inde � pendently given” (1972, 150 , emphasis added, quoting D ollard 1957, 174). H istorica l ly and to the present day, refusal by D a li ts to abase themselves is taken very seriously, and those who do not comply are beaten, subject to a variety o f tortures, or even k il led (Roberts 1015a , 240).

Twice per week a water truck was supposed to come to A nbu N agar to fi ll a communal tank , from which each household was permitted two kutams (large plastic jugs carried on the hip). The water was heavily chlorinated— it smelled like an overchlorinated swimming pool— and contained black sedi � ment that collected at the bottom o f the kutam . This was what the people o f the slum drank and cooked with . Women lined up we ll in advance o f the delivery truck ’s arrival (figure 6), and the scene was frequently tense.

A s soon as the tank was full , squabbles and shouting were the norm. A ny lit tle th ing could start a fight, like someone accidentally spill ing water and wanting to top up their ku ta m again. The fear was always that there wouldn ’t be enough to go around. W hen the water truck fa iled to turn up with in the usual window o f time, tensions ran part icularly high . Far worse was when it missed a scheduled delivery entirely. O n these occasions the mood in the slum was hard to describe. To call it “tense” would be an understatement. People were obviously deeply distressed and on edge, but no one wanted to ta lk about it. It was as i f even acknowledging the situation might make it worse. O nce during my stay in Kashtappat tinam the water truck failed to turn up for a fu ll two weeks. D uring this time the population survived on bore-well water, which was brackish and normal ly used only for bathing and washing clothes. I t tasted terrible. W ith in a few days, residents said, this

O U T S I D E R S • 4 3

F IG U R E 6 . W om en at w a ter tank , relaxed .

water began to give them headaches. By the time water fina l ly arrived they were in a terrible state, their eyes burning and everyone very cross.

“ This doesn’t happen to other neighborhoods! O nly to us!” a woman shouted at me when I asked why everyone was so angry on the occasion o f a missed delivery. I have no evidence that she was right about this, but her assumption was widely shared. I cannot rule out the possibili ty that slum dwellers experienced such problems more often, but I can attest that they were not alone in having to wa it for water in Chennai . A lthough many c ity residents had running water, many did not, and it was not just D a li ts who lacked it. O n the other hand , the cushy apartment I had rented with my partner in another part o f town never lacked for water, not even during the slum’s two-week drought. The most likely explanation for this discrepancy was class, not caste— though, as I w il l show, this distinction may i tse lf rest on theoretical assumptions that are not sound. I suspect that the residents’ asso� ciation o f my partner’s and my building had arranged for a private water supply to supplement what the c ity provided. The problem, o f course, is that when you permit a private water distribution system to compete w ith the public one— and remember, all water comes from the same natural sources—

4 4 • O U T S I D E R S

F I G U R E 7. A w o r l d a p a r t .

inequalities are inevitable. Furthermore, when one section o f the cit izenry is insulated from the hardships o f the other, their sense o f common interests naturally weakens. N o one in our upper-middle-class build ing was even aware at the time o f Kashtappatt inam ’s drought, o f what others in the city were going through.

N o one in the slum explic it ly linked their dependence on the state for water to the rural caste-based water regime. Caste , as we w i l l see, was a topic the people o f A nbu N agar avoided ta lk ing about, i f at all possible. It is tempting to suppose that the issue o f drink ing water, and having to beg for it, would carry associations for them that members o f the dominant society would not share. M any o f them had grown up in villages, and no adult slum dwellers were separated from their rural origins by more than a generation. But this is just speculation . W h a t I can say with certa in ty is that the most immediate concern for them was that they were being neglected, ignored, cut off. The act o f neglecting, ignoring, or cutt ing someone o f f can be translated by a single word in Tamil, purakkanippu. Reca ll the first section o f this chap � ter, on rumors o f foreign money. W hen I asked a woman whether pastors in Kashtappat tinam offered gifts to potential converts, she scoffed, “ H a! Do

O U T S I D E R S • 4 5

you th ink anyone cares that much about us?” Purakkanippu is the act o f not caring, with the connotation o f refusing to care, rejecting. The next chapter w i l l examine rejection and its opposite, caring about and for others.

I began the fina l section o f this chapter with a question: Is the slum “a world apart” ? In the eyes o f those who lived there it was— unquestionably so. Both in their moral sensibilities and in their material circumstances, slum dwellers believed they were different from others. They were also sure that others saw them as different— though not in the same ways they themselves did.

O U T S I D E R S

T W O

Caste, Care, and the Human

M Y i n f o r m a n t s ’ s e n s e t h a t t h e dominant, nonslum society was prejudiced against them was not without basis. But the prejudice was o f a special type, one that went hand in hand with feelings o f social concern and a desire to “uplift” them. This desire was commonly expressed by C henn a i ’s English speakers, those urban elites who are conventionally referred to in India as the “middle class” (Fernandes and H e l ler z oo6). The attitude o f T amil monolinguals, the majority o f the c ity’s residents, toward slum dwell � ers was somewhat different, and I w i l l come to that at the end o f this section. I did not set out to study nonslum dwellers’ ideas about their slum-dwelling counterparts; it was a topic that was quite li teral ly thrust upon me. Prior to beginning intensive fieldwork in 2003, I spent fifteen months in Chenna i between 2000 and 2001 doing pre l im inary research. D u ring this time I vis � ited a wide variety o f slums but seldom spent more than a few days in each. More o f my time was spent meeting with the representatives and fieldworkers o f various N G O s that serve the c ity’s poor, as wel l as with officers and former officers o f the T am il N adu Slum Clearance Board (T N SC B ) , the state bureaucracy most d irectly involved in slum administrat ion . D uring this time, I collected and read whatever government and private studies o f the c ity’s slums I could find, as well as the extensive N G O pamphlet literature. A part from this, I frequently found myse lf in conversation with middle-class citizens whom I happened to meet, and who frequently expressed great inter � est in my research when they discovered I was studying slums.

W ha t struck me about my interactions with members o f the urban middle class was not just the intensi ty o f their interest in slums but also the remark � ably detailed ideas they had about them (cf. Bate 2000 ,222 n. i).1 This was all the more strik ing given that these were people who would never have been

4 7

inside a slum or shared a meal w ith slum dwellers and whose only extended contact with slum dwellers was w ith the women employed in their own households as domestic servants. Yet not only did they have quite a bit to say about the lives o f their slum counterparts, but their accounts o f slum life were h ighly consistent w ith one another, and also with the picture pa in ted by N G O workers and in government reports. I w i l l sketch only the broad out � lines o f this discourse here. The point is not to prove that those who are well o f f harbor negative opinions about the poor, something that is undoubtedly true all over the world. N or is it simply to rebut stereotypes that are false or misleading. The significance o f the stereotypes I encountered is that they correspond to long-standing stereotypes among caste people about D a li ts, or “untouchables,” although they may not be consciously understood as such by those who hold them. Yet the fact that such stereotypes take this form has consequences for relations between the slum and nonslum society, as I w ill show.

A mong the most prominent features o f middle-class, N G O , and govern � ment discourse on slums was its focus on hygiene. Given the lack o f regular garbage collection in slums, and the absence o f sewers, it is understandable that any outsider who ventured into a slum would be struck by the level o f filth. O ne middle-aged M uslim woman from a middle-class background, whom I interviewed , and who was pursuing a Ph D project on C h en n a i ’s slums at the U niversity o f M adras, regaled me with lurid descriptions o f one part icular slum where the garbage was piled as high as the huts and where stray dogs “ loitered” on the roadsides. The undeniable filth o f slum surround � ings was furthermore assimilated in middle-class discourse to the idea that slum dwellers were themselves filthy. In my experience slum dwellers were as scrupulous as any other T amil about personal hygiene. They brushed their teeth and soaped and bathed themselves thoroughly every morning. M en would also usually bathe again in the evening, after work . Yet their alleged lack o f personal hygiene was among the most commonly cited attributes in my conversations with English-speaking c ity dwellers. The same Ph D student mentioned above, for example, segued from describing garbage and flea-bit � ten dogs to a long discourse on personal hygiene: how important it was for slum dwellers to be “ taught” to use soap and bathe daily, and how many prob � lems could be solved thereby. Similar ideas were expressed to me by the heads o f N G O s focusing on the “urban poor.” Personal hygiene loomed larger in most o f my informants’ accounts than even poverty, and it was described in ways that made it seem like the root cause not just o f disease but o f slum

4 8 • C A S T E , C A R E , A N D T H E H U M A N

dwellers’ failure to succeed in life. Together with their d irty and crowded surroundings, poor hygiene was made to seem at once the result o f moral shortcomings and also a cause o f immora lity, and even o f poverty itself.

Examples o f such th in k ing can also be found in government reports and related documents. The chairman o f the T N S C B Board writes in a 1987 publication on urban poverty, “Slum dwellers who were not used to the toilet habits and bath ing in a bathroom do not easily acclamatise \sic\ themselves to the amenities provided___Throwing garbage and waste water from the upper floors merrily without caring for the feelings o f others make the whole environ d irty” (Sampath 1987, i6z).

H ere we see how the ideas o f personal hygiene are linked to environmental pollution , and both to perverted social morality. Com mon middle-class ideas l in k ing fi lth and moral turpitude to the alleged causes o f poverty can be glimpsed in a 1975 Slum Clearance Board survey that explains, “People in these slums live a life that is nasty, brutish, and short.’ The environment is h ighly repellant, sordid and sickening. W ha t lit tle they earn, they spend on drin k ing and tribal festivities, resulting in a life o f eternal poverty stricken conditions” ( T N S C B 1975, 42.).

The idea that slum dwellers’ poverty was due to their love o f drink and desire to celebrate life cycle rituals such as weddings in a fashion inappropri � ate to people o f their station was another common theme in middle-class commentary on life in the slum. The detailed testimony I received during my in it ia l stay in C henna i from ordinary citizens on the economics o f slum households, and on the disastrous effects o f festivals in part icular, was an important influence on my eventual field research. H ere I would like only to distinguish two salient features o f these accounts o f slum life cycle rituals. First, there was the sheer level o f detail, often inc luding exact figures, interest pa id on debts, and so on, that my informants claimed to possess. M y best guess as to how they came to such knowledge, which I cannot believe was simply fabricated, was from ta lk ing to their household servants. Slum women throughout southern C henna i (but not in A nbu Nagar, which is located in the c ity’s northern industrial sprawl) commonly work as domestic help in the middle-class neighborhoods that slums in this part o f the c ity exist largely in order to serve (Ramani 1985).

I do not imagine domestic servants would endorse the strong current o f moral judgment in middle-class accounts o f slum festivals. W here middle-class employers described slum dwellers’ festivals as wasteful and as examples o f status striving, slum dwellers themselves did not see their life

C A S T E , C A R E , A N D T H E H U M A N • 4 9

cycle celebrations as excessive or as aspiring to a status that was inappropriate for them. Moreover, festivals were not in fac t a major source o f slum dwellers’ poverty. A s the next chapter explains, they were among the few loca lly acceptable occasions to purchase desired household items, and they also offered indirect opportunities for accumulation and savings. The true source o f slum dwellers’ poverty— in their own estimation anyway—was that they were pa id so little for the work they did, that they were exploited. That slum dwellers were underpaid did not figure anywhere in middle-class theories o f slum economics.

A second aspect o f middle-class moral iz ing about slum festivals had to do with the fact that slum dwellers served their guests meat. This may seem like an odd detail for concerned citizens to focus on, but it came up with strik ing regularity and was often spoken o f in vehement tones. Ostensibly what was so objectionable about hosts offering meat to their guests was simply that meat was expensive, although in real ity the amount o f meat that each guest received (in biriyani, a rice dish) was very small, and at any rate the cost o f the feast was borne collectively by the sponsoring family’s k in network , not by the individual household. H ow are we to understand my interlocutors’ strange fixation on the fac t that slum dwellers served meat at ritual occa � sions? Q uite possibly they simply overestimated the cost or the quantities served. But this also concerned caste. M y interlocutors were invariably o f castes for whom serving meat at a major li fe cycle ritua l would be h ighly pollut ing and even irreligious, even i f they themselves might also consume meat outside o f ritual contexts. Indeed , the two possibili t ies— “economics” and ritual purity— cannot be entirely separated. H igh-caste status in South India is closely associated with austerity and a “simple” l ife � style, and vegetarian fare in part icular is celebrated not only for its ritual purity but as an emblem o f the notional asceticism—what one might cal l the “conspicuous nonconsumption”— o f South Indian Brahminism and o f high- caste T amil Saivism more generally.

The language o f social reform and o f liberal concern for the poor common to middle-class, official, and N G O discourse envisions slums as sites o f moral and biophysical impurity. W h a t soon becomes clear is that slums are under � stood not only, or even primarily, as a threat to slum dwellers themselves. The close proximity o f slums to “respectable” neighborhoods, and the danger this presents to the latter, constituted a persistent refra in in my conversations with middle-class c ity dwellers. Such worries are evident as well in official reports. A 1975 Slum Clearance Report tells us that “ impure air, inadequate

5 0 • C A S T E , C A R E , A N D T H E H U M A N

water supply, improper sewage disposa l are the dangers prevalent in the slums. They have endangered the heal th o f not only the slum dwellers but also the residents in neighbouring areas” ( T N S C B 1975, 34). O r, as a 1971 Slum Clearance Board publication more bluntly puts it, “A slum is a veritable cancer in the heart o f the C i ty” ( T N S C B 1971 ,14) .

The threat that slums posed to their neighbors was deemed to be not just biophysical but also criminal , and the two, indeed, seemed to be intimately connected. I was warned by people elsewhere in the c ity to exercise extreme caution in Kashtapat tinam , as violent crime was rampant. So many people have been k il led there, I was told. A part from violence, theft was also said to be a major danger. I f you don ’t watch out, I was warned, those people w i l l steal everyth ing you have.

Anthropologists often face danger— in many cases, it is simply the price o f ethnographic knowledge. But I am not one o f them. The picture people sketched o f Kashtappatt inam had li tt le basis in reality. I never once heard o f a violent crime tak ing place during my stay there and never felt the slightest bit unsafe. Some boys once told me about a deadly fight between two slum men, in which one hacked the other to death. But it had happened some sixteen years earlier and was the culmination o f a long-standing enmity. N o one could cite anything o f the k ind more recently than that, and no one had ever heard o f a stranger being attacked. A s for theft, noth ing o f mine was ever stolen. I didn ’t even lock my door most o f the time.

The 1971 report that describes slums as a “cancer” neatly illustrates how biological impurity, immorality, and criminal threat are run indiscriminately together in official writ ings just as they are in middle-class discourse: “A slum is the cradle for disease borne out o f unhygienic and insanitary [sic] condi � tions. . . . W e can see men and women born to live as human beings in civi � lized society lead a life worse than that o f animals___Crim es are cultivated due to the perverted growth o f mental ethics and unbalanced social equilib � rium . The ir life o f want, misery and i l l heal th turns them into cro9ks” ( T N S C B 1971 ,14) .

W omen play a unique role in these moral iz ing portra its o f slum life: “In slum areas women idle away their time and as a result o f it quarrel” ( T N SC B 1971, 53). Slum women are stereotyped as foul-tongued , irascible, and brash, traits that are at odds with T amil cultural ideals for women . Women in slums are also alleged to drin k alcohol, which is associated in public discourse with sexual promiscuity. A ccording to tradit ional T amil gender norms, the ideal woman exhibits an exaggerated sense o f shyness or “shame” {vetkarn), which

C A S T E , C A R E , A N D T H E H U M A N • 5 1

i tse lf is a sign o f her chasti ty (,karpu). A s with poor hygiene, slum women ’s supposed lack o f chasti ty is regarded as both the cause and the result o f immorality. In this case immora lity comes in the form o f crime and vict im- hood, for womens sexuality is implicated as a major cause o f male criminal � ity. Thus among the common causes for murder in the slum is said to be “sexual jea lousies”; apart from murder, “crimes affect ing the human body such as . . . molesting women , rape, etc.,” are likewise said to be “common occurrences in slum areas” ( T N S C B 1975,37).

The T N S C B survey that makes these claims offers no data to substantiate its contention that crime is in fact higher in slums than elsewhere. N or do its authors explain how they came to the conclusion that part icular forms o f criminali ty, such as murder and sexual molestation , predominate there. This lack o f data is strik ing in a report whose pages are mostly dedicated to the presentation o f quantitative data in tabular form . Equa l ly strik ing is the authors’ disarming admission , at the very start o f section on crime, that “no register is maintained, in any Police Station , which gives figures for offenses committed by slum dwellers [as opposed to] non-slum dwellers” ( T BS C B 1975, 37). Given that even such basic data are unavailable, one wonders how anyone could possibly know how often slum murders are driven by sexual jealousy.

The focus in official documents on slum dwellers’ moral shortcomings implies the need for pedagogica l programs to turn them into acceptable members o f society. Thus the T N S C B , in a publication devoted to summa � riz ing its achievements in “slum clearance,” describes how educational films are regularly screened for the purpose o f instructing residents “ how to live in society and adjust with other groups o f people” ( T N S C B 1971,53). A llevia t ing poverty is not enough; slum dwellers’ “adjustment” to modern society and to people requires them to overcome their own moral and psychological limita � t ions: “A part from considerations o f finance, there are psychological factors which have to be overcome i f slums are to be el iminated and the c ity made beautiful. The slum dwellers are in their attitude different from other people” (Census o f India 1961,1, quoted in T N S C B 1971, 98).

W ha t is it about slum dwellers that makes them so different from other people? O ne part o f it may be discerned in a passage appearing in an earlier publication , from which the above quotat ion was taken . Notice that the underlying logic o f this passage is the same as in the previous one; the only difference is that in this case the predominant caste character o f Chenna i slum dwellers is explicit ly stated: “More than 70% o f the Scheduled Caste

52. • C A S T E , C A R E , A N D T H E H U M A N

population o f M adras C i ty live in slu m s . . . . Such segregation w il l not help the proper development o f the Scheduled Castes, who have to undergo a transformation in their outlook” (Census o f India 1961, 39).

In light o f passages such as this, and o f the fact that middle-class stereotypes about slum dwellers closely match caste stereotypes about D ali ts in the preco � lonial period and in villages even today, contemporary discourse is best under � stood as a postcolonial variant o f older caste talk , which has incorporated the language o f scientific hygiene. Indeed, until the 1960s C henna i ’s slums were sti ll openly referred to as certs, the T amil term for an untouchable ghetto.

M any o f the ideas I encountered could also have been taken straight out of V ic tor ian London (Jones 1986), in the moral revulsion they express and in their impulse to uplift— a “c ivili z ing mission” o f sorts. This is not surprising, and not just because o f the legacy o f colonialism . Chenna i middle-class ideas about slums display traffic between V ictorian class prejudices and caste preju � dice because caste prejudice is not a sui generis or unique ly Indian phenomenon .

Sudipta Kaviraj has claimed, in a discussion o f fi lth in the urban public sphere, that “ the Brahminica l concept o f cleanliness and purity was quite different from the emergent W estern ideas about hygiene” (1997, 98). C erta in ly differences in the theoretical underpinnings o f these discourses are not hard to spot, especially i f Brahminica l concepts are sought primarily in textual sources. But caste purity and notions o f biophysical fi lth have never been entirely dist inct—part icularly with respect to untouchability. Since the colonial period defenders o f the caste system have described their “ natural” revulsion at the unclean habits o f D a l i ts, while reformers have promised D a l i ts’ salvation through the acquisition o f cleanliness (Bayly 1998). But the fuzzy distinct ion between the D a l i t ’s ritual impurity and plain old filth is not con fined to the modern period. A s Sascha Ebe l ing has noted, the earliest T amil l i terary account o f a D a l i t ceri, dat ing to the twelfth century, was designed to evoke disgust in high-caste audiences through depictions o f over � crowding, scanty clothing, dark skin , the presence o f carcasses and leather, promiscuous drunken women , and animals and people living in close and unruly proxim ity (Ebel ing zoio , 28-30).

Today, however, the caste-specific character o f C h enn a i ’s slums has been effectively erased from offic ia l and public sphere discourse, al though the associated stereotypes remain f irm ly in place and it is widely understood that slums are inhabited by people be longing to the lower and not higher castes.2 Given that the middle-class urbanites whose ta lk I am describing had no

C A S T E , C A R E , A N D T H E H U M A N • 53

personal experience w ith slums, it is quite possible they were unaware that the slum dwellers they spoke o f were in fac t D ali ts, or that their images o f slum dwellers matched rural stereotypes about the ceri. Most urban elites have no more direct knowledge o f rural India than they do o f urban slums.

A l l the foregoing applies only to the discourse o f English-speaking elites, N G O s , and government bureaucracies. The situation is quite different, in my experience, among monolingua l T amil speakers, who make up the majority o f non-D al it Chenna i residents. A mong such people awareness o f the caste- specific character o f the c ity’s slums is often quite explicit, and slums are sti ll frequently referred to as ceris. Indeed, more than once T amil monolinguals have stunned me by referring to slum dwellers as “paraiyan t a n !” (mere pari � ahs!) and denounc ing them as acinkam (disgusting) or “ the foul/wicked ones” {kettavahka).

“ t h e r e i s n o c a s t e i n t h e c i t y ”

The people o f A nbu N agar were loath to discuss their own caste status, Paraiyar, though they would commonly identify nonslum dwellers by their caste and stereotype them accordingly. They never claimed caste-specific characteristic for themselves. N or did their spontaneous self-descriptions involve any reference to their “tradit ional” occupations, and they were espe � cially reluctant to speak about untouchability. W hen I myself raised the topic o f their caste, they readily acknowledged it but dismissed it as a historical curiosity o f no real concern or relevance to their present-day existence, saying things like “ That means nothing—we all have the same blood, don ’t we?” or “ Isn’t it true that G od created al l people the same?” This was a paradoxical universalism , however, given that they invoked it only with respect to them � selves. Non-D a li ts were often described by slum dwellers in terms o f inher � ited caste characteristics.

In expla in ing caste, slum dwellers commonly resorted to temporal and spatial displacements, as the following examples illustrate. Prior to beginning my fieldwork I spent fifteen months in Chenna i visi ting slums all around the c ity and ta lk ing with residents about their lives. They were on the whole very forthcoming and eager to help, but whenever I expressed an interest in learn � ing about their own experiences o f caste, I was given advice along these lines: “ I f you want to know about caste you should go to a village,” or “ That is a topic for history. In these modern times, caste is no longer a consideration .”

54 C A S T E , C A R E , A N D T H E H U M A N

Such historical and spatial displacements, moreover, funct ioned more or less interchangeably, with the effect that “the vi llage” was not only in a different place but also in a different time. It was not unusual to hear statements such as “ Caste is a th ing o f the past—here in the c ity we have no caste.” Caste , in short, was elsewhere. It was in the historical past, and it was in their own vil lage past. The ir relatives who lived in the village might continue to suffer it, but urban slum dwellers insisted that they did not. “ H ere I can sit next to anyone on a bench in the hotel [canteen],” one slum dweller explained to me. “ I can live anywhere. There is no more caste today.”

Mere poverty was infin itely superior to untouchabili ty (tintamai)-, the fact that they themselves were poor rather than untouchable was a point o f pride, as was the fac t that many slum dwellers had never been persona lly subjected to untouchability. A s one young man explained to me, “ W e live in the city. So we have no caste—we are educated people, modern .” Given that most adult slum dwellers, inc luding the speaker himself, had not studied past the third grade, educated here cannot be taken literally. It functions as a synonym for modern or urbane, untouched by the old world o f caste.

Those residents o f A nbu N agar who had grown up in the village main � tained that life was actual ly much more comfortable there in some respects. The air was fresh and clean, there were lots o f plants, conditions were not so crowded. They earned less, but their expenses were less too. The downside was caste: da i ly humiliations, bullying, rituals o f self-abnegation , the ever-present threat o f violence. They told o f not being al lowed to w a lk on certain streets, being required to get o f f their bicycles and wa lk when passing certain people, being openly derided, being hit by overseers, having funeral processions for � cibly diverted through flooded paddy fields by caste people who would not al low them to use the road. This was true, incidentally, whether the caste people were H indu or Christ ian . Sagayaraj came from a village where the dominant caste, Vanniyars, were al l Roman Catholic , just like the D a li ts they oppressed. N G O s and human rights activists have also reported that D a l i t women are routinely subject to sexual harassment and rape by caste men who regard this as their traditional prerogative (H uman Rights W atch 1999), and my informants in A nbu N agar privately confirmed that this was the case. The mass migration ofPara iyars from rural T amil N adu to Chenna i / M adras has always been seen— along with rura l-urban migration more generally— as driven by economic factors: either the “p u l l” factor o f urban opportunities or the so-called “push factor” o f wanting to escape rural deprivation . O ne does not doubt that both sorts o f economic incentives are relevant for D al its too.

C A S T E , C A R E , A N D T H E H U M A N • 55

But it seems fa ir to surmise that the mass migration o f Paraiyars to the c ity’s slums is not just a flight from rural poverty and landlessness. It is also a flight from untouchabili ty and routine violence: in short, an attempt to become simply “the poor” (Roberts 1015a).

Slum dwellers made up a full 40 percent o f the c ity’s populat ion as o f 1981 (Ramani 1985, zi), and they supply the bulk o f all manual labor. They also have a virtual monopoly on the most physically difficult and dangerous jobs. That slums might therefore be understood as cheap-labor colonies, housing a des � perate subpopulation whose poorly compensated exertions underwrite the relative privilege o f virtually everyone else, remains entirely unnoted in official studies and reports.

C henna i ’s slums are inhabited almost entirely by D al its. Statistics on the caste composition o f Chenna i slums are extremely unreliable in my experi � ence. O ne o f the reasons I chose A nbu N agar was that it was the subject o f a special study by the zooi Census o f India. The published report based on this study suggested that D al its made up anywhere from 70 to 83 percent o f the total population o f the neighborhood, when in truth only 3 out o f a total o f 403 households were non-Dalit. In point o f fact, neither the Government o f Tamil N adu nor the Government o f India collects statistics on D ali ts per se, but only on H indu D al its (known as Scheduled Castes). The report to which I refer lists A nbu N agar as 70 percent Scheduled Caste, implying that D al its constitute between 70 and 83 percent o f A nbu N agar s population (the highest possible figure being i f al l the 13 percent who are listed as Musl ims and Christians are assumed to be D a li ts also). Errors in census reports and other surveys are most l ikely not due to any shortcoming o f the Census, one o f India’s most respected and professional bureaucracies. M any D a l i t slum dwell � ers told me that they misreported their caste category to census workers. My own assessment o f slums’ caste composition is based on extended interaction with residents; in cases where I did not spend extended periods o f time in a neighborhood, my method for determining the percentage o f D ali ts was as follows. Rather than conducting door-to-door surveys, I asked a range o f resi � dents whose trust I had established to identify which households were not D al it . Residents o f face-to-face communities invariably know one another’s caste. W hen more than a dozen respondents independently described the place they lived as being a “ D a l i t neighborhood” (ati tirdvitapaku ti) and con � sistently identified the exact same households as be ing the only non-D ali ts in their neighborhood, I judged that the information I was receiving was accu � rate, and I estimated the percentages o f D ali ts and non-Dalits on this basis.

5 6 • C A S T E , C A R E , A N D T H E H U M A N

H ie residents o f A nbu N agar readily acknowledged that A nbu N agar and surrounding slums, along with every other slum in the city they were person � ally familiar with , were D a l i t neighborhoods. They also readily acknowledged that proportional ly very few D al its lived in nonslum areas.3 They knew that men from slums were predominantly casual day laborers (“coolies” ) and that virtua lly all casual day laborers were slum dwellers. A n d yet they insisted that they lived in such neighborhoods, not because they were prohibited from living elsewhere, but simply because they were poor. They performed difficult physical labor, not because they were forced to— as their ancestors were, even two or three generations ago—but because they were poor.

In proc la iming that “there is no caste in the city,” the poor people o f A nbu N agar appeared to be saying exac tly what India ’s moderniz ing elites and many nationalist intellectuals (e.g., Bete ille 2012; Panini 1996; Srinivas 2003) also say: that caste is a th ing o f the past, that it is a retrograde tendency that survives only in the rural hinterland, or, elites often add, where low-caste poli tical entrepreneurs have given it a new lease on life through the pursuit o f “ identity politics.” A ccording to this line o f th ink ing, caste survives in the modern sphere only because vested interests have kept it on life support. I t is those who complain about caste, its self-identified victims, and those who support affirmative action style policies to overcome entrenched ine � qualities, who are the real casteists. The solution is to just stop ta lk ing about it and for the state to abandon its affirmative action policies. N icholas D irks implies that such a stance, which he dubs the “embarrassment o f caste,” is rooted in bad fa i th because those who take it ignore their own inherited privilege and because they would directly benefit from ending affirmative action policies designed to help those who lack their advantages (2001, 285- 86; cf. Betei lle 2000 , 152-53; H amermesh 1990). W hether he is right about bad fa i th or not, D irks is persuasive in identifying among Indian elites a widespread embarrassment about caste that manifests as a reluctance to talk about it and a tendency to discount evidence o f its continued salience in modern life.

W hile the resemblance between what the people o f A nbu N agar had to say about the absence o f caste in the ci ty and what urban elites say is strik ing, it is also superficial. W ith elites the embarrassment o f caste is based on a sense that, insofar as caste persists, it is a black mark on the reputation o f the nation as a whole. For the people o f A nbu N agar, by contrast, the claim that caste had been entirely eliminated in the c ity was rooted, not in any desire to protect the image o f the Indian nation , but in the fact that memories o f

C A S T E , C A R E , A N D T H E H U M A N • 5 7

untouchabili ty were deeply pa in ful for them and that having been treated in this way was i tse lf a source o f lingering humiliat ion (cf. G uru Z009).

The effect was not just a denial o f caste— and a palpable reluctance to talk about their own caste in part icular— but a double denial, a denial and also a denial o f the very fac t that there is anything there to deny. W h a t are we talk � ing about here? It was not that the people o f A nbu N agar ever made an explicit pact among themselves not to ta lk about the legacy o f untouchabil � ity. I f such a pact had been made, I am reasonably confident that I would have discovered it. W hatever understanding there was among them was a tacit one: an unspoken agreement to look forward and not back, to simply avoid any open acknowledgment o f the very obvious connections between their present state and their origins. The existence o f such an understanding was not merely a matter o f speculation on my part. O n a few rare occasions it came into the open; in the next section I w i l l describe two such events.

RETURN OF THE REPRESSED

Sitt ing one morning with Indira, a mother o f two teenagers and one o f my closest friends in A nbu N agar, she used a word I had never before heard uttered in the slum. That word was ceri. “Around here, everyone is poor,” she said, and then she muttered, “C eri people are always poor.” C eri is the name for a D a l i t ghetto, or, in rural T amil N adu , a separate hamlet housing the D a l i t laborers who once belonged to the landowners o f the main village, known as the ur. I didn ’t say anything at first, but a few minutes later I tried using the word myse lf and Indira froze. “You must never use that word around here, Robert,” she warned me. I pointed out that she herse lf had just used it, an observation she did not attempt direc tly to refute. Instead she responded by saying, “ W e don ’t use that word a n y m o r e It was a word they had used a long time ago in the slum, she admitted, but it was no longer used. She then warned me not to ever use it myse lf and said that i f anyone heard me using it they would get angry— “Espec ial ly the men . . . they w il l get very angry.” But it ’s a perfectly good word, I insisted, and there is no shame in it. I pointed out that Sagayaraj, my assistant who had grown up in a village ceri, used the word freely and that the term was common throughout rural T amil Nadu. Indira listened patiently, then reiterated that the term was no longer used to refer to urban slums. She did not deny that an urban slum was in fact a ceri. Instead she simply extended her right arm toward me, pa lm up, and

5 8 • C A S T E , C A R E , A N D T H E H U M A N

with her other hand pulled at the skin o f her wrist where the veins showed through. “ W e al l have the same blood, don ’t we, Robert?” I t was at once an appeal and a challenge.

Later I mentioned this incident to a friend, Karuppan , a social worker who was active in D a l i t politics. H e h imse lf had grown up in A nbu N agar and was among a hand ful o f young men from the slum to have gone to college, found a job , and moved out. H e shook his head at what he portrayed as Indira’s naive fa i th in words. “ In reality,” he explained , “slum means ceri and ceri means slum. C eri is simply the T amil word for slum— there’s no difference. A ceri is wherever D a l i t people live; wherever D a l i t people live is a slum, a ceril Every village has its own slum, but they use the T amil word. In the c ity peo � ple like to say slum because they th in k it sounds better. They want to say: ‘ We are modern people, we have no caste’ . . . but really, they know the truth.” A ccording to Karuppan , the people o f A nbu N agar— most o f his family, in other words—were perfec t ly we ll aware that they were st i ll “noth ing but Paraiyars” in the eyes o f caste people and that al l claims to have escaped from caste were but a vain hope.

A nother D a l i t friend, who was o f rural origin but now part o f the urban middle class, had a somewhat different response when I told him about Indira’s reaction to the word ceri. Like K aruppan he was involved in D a l i t politics. H e interpreted Indira ’s words as evidence not o f denial on her part but o f genuine ignorance. “ O f course they don ’t know about caste,” he replied, laughing darkly. “ They live in a place where everyone around them is also D a l i t . They can eat their beef and no one cares. There are no houses for rent in the slums that say ‘Brahmins only’ or Vegetarians only.’ A n d the jobs they try to get— coolie work— are only for D a l i ts anyway, so no one w i l l ever tell them no’ [you can’t have this job]. But when they try to rise up and leave the slum, that is when they find out.” In other words, their entire lives are determined by caste in a way they can never see unti l they try to step out o f that life and into another.

Amavasai was a retired harbor worker in his late seventies. H e and his wife had been living in A nbu N agar since having been settled there after the slum where they had previously lived was “cleared” by the T amil N adu H ousing Board in 1964. They were among the most prosperous slum dwellers: they owned the plot on which Sagayaraj’s and my room was located. Amavasai was lucky to have spent his work ing years in a unionized sector—unlike the majority o f slum dwellers who were casual laborers— and he received a hand � some monthly pension o f Rs. 4 ,000 . O n the day he received his pension he

C A S T E , C A R E , A N D T H E H U M A N • 5 9

would buy a bottle o f liquor and a packet o f bidis and then turn the balance over to his wife. H is custom was to drin k the bottle over a two-day period, which he confided to me was the only time he was ever really happy.

A m avasa i’s wife had many friends with whom she would sit in front o f her house in the mornings and evenings, but he always seemed peripheral to that group and didn ’t have any friends o f his own as far as I was aware. Perhaps for that reason he would often drop in on me, and I always made an effort to ta lk to him , even though he seldom had much to say. H is prim ary interest, which I shared, was in doing small household projects like f ixing a light switch, stringing a clothesline, repairing a broken shelf, and so on, and I occasionally even contrived new projects for the two o f us to do together.

O n this occasion, Amavasai, who had just received his monthly pension, wandered into my room with a bottle in his hand. It was early afternoon, the hottest time o f the day, when those who did not have to work remained indoors. I generally used this time to organized my field notes but on this day was in my room reading something about hook-swinging in late nineteenth-century Madras.4 Amavasai, who was illiterate, asked me what I was reading, and by way o f explanation I drew a sketch o f a hook-swinging. “ O h yes, that,” he said, claim � ing to have witnessed the practice as a young boy in rural Tamil N adu before migrating alone to Madras at age nine after both his parents died. This seems to have prompted a string o f memories. H is eyes misting over, he leaned very close, looking at me intensely. “The whole world has changed and all that is finished,” he said and paused, as i f for me to absorb what he was telling me. W hen at last he continued speaking it was in a hushed tone. “ In those days, there was untouchability’ \tïntâmai\ .” H e touched my knee emphatically as he said this.

“ The caste people \ jâti makkaï] held us down, and whatever they said to do, we had to do it. I f not, they would beat us.” H e spoke about the violence, poverty, and hunger they faced, as well as the forced labor {atimai vêlai, lit. “slave labor”) they were made to perform. H is speech was not always clear, because o f drink and emotion, but again and again between each point he would repeat the word tintâmai, lift ing his hand o f f my leg and pressing it down again firmly, as i f to illustrate. A t length he paused and, sitting back, looked at me for what seemed like a long time, his rheumy eyes searching my face for some response or perhaps a sign o f recognition. I managed only to mumble the usual platitudes about how caste is “a social evil” and “we are all the same blood.” I don’t know i f he heard me at this point, or even cared, as I was midsentence when he resumed, bringing his whole speech to conclusion: “But today I have a ‘pension’ !” H is eyes were wide, and his mouth opened in a stylized gesture o f astonishment as he

6 0 • C A S T E , C A R E , A N D T H E H U M A N

pronounced this English word. “Pension!” he repeated, chuckling to himself, and bottle in hand he wandered off.

A few hours later Am avasa i returned look ing distraught. H e took my hands in his and, kneel ing down in front o f my chair, asked me i f I had told anyone about our conversation . Previously we had both been seated at equal levels, and his kneel ing unnerved me. I said that I hadn ’t, but he would not let go o f my hands or get up o f f the floor. Tears were now openly streaming down his face, and he begged me never to tell anyone in A nbu N agar what he had told me. I had to repeat the promise several times before he, satisfied, released my hands and left. “A re you sure you won ’t t e l l? . . . A re you sure?. . . Please don’t t e l l . . . You must not say anything, okay? D o you promise?”

CA STE OR CLA S S ?

Two generations o f husbandly neglect had taken a toll on K a la ivan i ’s family, and her four children were perpetually dirty, each having only one set o f worn clothes save the youngest, a boy o f around two, who wore only under � pants. H er eldest child, a boy named Sudagar, was a regular at the English classes I offered in the evenings. H e was a clever child but often went un fed and for this reason had trouble concentrating. Ascerta in ing this to be the case, I began slipping him small snacks before class, and his abili ty to concen � trate was markedly improved. H e was an avid learner, and I was thus sur � prised when my assistant Sagayaraj informed me that Sudagar was habitua lly absent from the free government school set up for slum children. This seemed odd for a boy who was seldom seen out o f his school uniform , which he wore not only on schooldays but on Sundays and school holidays as well. H is mother explained that Sudagar was a carefree child who simply refused to go to school because he “prefers to run around and play all the time.” I learned from other children , however, that the real reason was that his teachers would verbally abuse him and pull him around the room by the ear because his uniform was d irty and torn. H e wore that old uniform so much, it turned out, because it was the only outfit he owned. W hen I confronted Sudagar about the abuse at school he confessed that that was indeed the real reason he played hooky. H e said that his mother didn ’t know about it and asked me

not to tell her. Though my usual policy was to buy only medicine and the occasional meal

for people in A nbu N agar, I went to his mother and offered to have a new

C A S T E , C A R E , A N D T H E H U M A N

uniform stitched for him . I told her Sudagar had won a prize for perfect attendance in my evening English class. I did not mention that the prize i tse lf had been concocted as an excuse to provide her son with a new uniform . The shop front cloth seller who normal ly supplied the slum was closed for lunch when we arrived, so I took her to another cloth seller, a walk-in establishment in another neighborhood not far away. It was not a fancy place; the shop was quite shabby, and the neighborhood where it was located was a poor working- class one. Yet K a la ivani was for some reason reluctant to go inside. She said she’ d prefer to just wait on the street. But I insisted that she accompany me because I wouldn ’t be able to identify the correct cloth for the uniform . W hen we got inside I was horrified . The merchants openly sneered at Kala ivani , even though we were clearly together and I myse lf spoke to her using the respectful nlhka form. I was so flustered by their behavior I couldn ’t find the words to tell them off. K a la ivani kept her eyes to the floor and spoke in barely audible tones throughout the encounter. I later learned that this sort o f th ing was typical— that a l l but the best-dressed slum dwellers would be treated this way in most commercial establishments and for that reason generally avoided entering them. H aving been sensitized to the hidden forms discrimination could take in Chennai , I soon began to notice it everywhere. Even at roadside tea stands, I often saw barefoot slum-dwelling laborers being deliberately ignored or made to wai t unti l all other customers were served.

Was that caste or class discrimination? Most, inc luding slum dwellers, would emphatically say the discriminat ion was a matter o f class. In Chenna i and other cities, no one knows whether a stranger is untouchable or not, at least not with any high degree o f certainty. But where caste determines to a remarkable degree the divide between slum-dwell ing casual laborers and everyone else, one wonders whether the caste-class distinct ion is theoretically sound. W hen D a l i t slum dwellers say there is no caste in the city, they appar � ently mean that caste in the sense o f ri tual rules governing everyday interac � tions is absent. I f caste is defined, as many anthropologists have sought to define it, on the basis o f its putative ritual essence, then it is indeed absent in most urban milieus. But as the people o f A nbu N agar understood, and at times explicitly stated, caste is not ultimately a matter o f ritual hierarchy. It is a complex social order that is multiply determined, inc luding by endogamy and heritable polit ical-economic roles, and by control over resources— realities a previous generation o f anthropologists characterized as “secular factors” that could be excluded from the definition o f caste as such (Roberts 2.008, zoisb).

6 z • C A S T E , C A R E , A N D T H E H U M A N

Though slum dwellers did not readily refer to themselves as Paraiyars, A d i Dravidas, Scheduled Castes, or D a li ts, they acknowledged the applicability o f these terms. But they preferred to describe themselves simply as “the poor” [elai). Conversely, when speaking o f nonslum dwellers, they generally avoided referring to them by their collective status as “caste people” { jd t i m akkal). To do so would indirec t ly reference their own outcaste status. Instead they described them as “the privileged” {vacatipettavanka, or vacati utaiyavanka-, lit. “those who have com forts”) or “the r ich” {panakkdranka) .

It was not by chance that these were the categories out o f which slum dwellers would fashion a unique language for ta lk ing about caste. In C henna i , al l slum-dwel ling D a l i ts were poor, and apart from one or two exceptional cases al l nonslum dwellers (who were rich, by slum standards) were caste people. This empirical regularity al lowed slum dwellers to use class terminology when speak ing about D a l i t and non-D al it . But why not speak o f caste directly, i f that is what is really meant? For one thing, by referring to D a l i ts collectively simply as “ the poor,” and caste people as “the rich,” slum dwellers asserted that the difference between themselves and others was one o f accident, not essence. W here jd t i refers to someth ing immutable and intrinsic to the person, class is a condition that is extrinsic and mutable, at least in theory. C lass terms, like jd t i , describe a collective identity but one that does not wholly define an individual . In principle one can enter or exit a class without changing who one is. The class idiom is, therefore, not merely a euphemism for caste. It is a theoretical challenge to it. By describing them � selves in class terms, slum dwellers claimed an identity untainted by “pollu � tion ,” one that made no reference to the various inborn and wholly negative physical and moral characteristics that caste people ascribed to them, assert � ing instead that they were simply people whose lack o f access to resources put them at the mercy o f others. Conversely, by consistently referring to caste people, not according to their claimed status as “ having caste,” but simply as “ the rich,” slum dwellers advanced the view that caste people’s superior status was in fa c t nothing more than weal th and entrenched privilege. For after a 11, “ Caste is a mere lie \verum poy],” slum dwellers maintained. “A l l people are the same . . . all have the same blood.”

A t the same time, in practice slum dwellers d id ascribe various inborn traits to “the rich.” A n d they did so using the language o f kunam (Sanskrit:

guna), a term for traits or characteristics that inhere from birth that is part and parcel o f the vocabulary o f caste. This lent what we might call a “caste character” to their concept o f class. Terms like rich and poor in slum usage

C A S T E , C A R E , A N D T H E H U M A N • 6 3

thus took on properties o f both jd t i and class but were reducible to neither. O ne way they differed was that the inherited characteristics ascribed to the rich were, unlike those that caste people used to describe their own jd t is, wholly negative. The characteristics that slum dwellers ascribed to “the poor,” by contrast, were entirely positive. Yet there was a difference: these positive characteristics were not associated with their being D a li ts. That would be to admit that they were a hereditary group whose basic nature was determined by blood after all. They would be just another jdti-, and instead o f chal lenging the ideology o f caste they would have retained it, only with themselves now at the moral pinnacle. W ha t allowed them to avoid this possibili ty was their concept o f humanity. Their own good traits were due, not to any unique inheritance, but to something they shared with all who were merely human— that is, with all who had not selfishly cut themselves o f f from the rest o f humanity by prac tic ing caste. Thus the positive moral traits they claimed to possess were not a special property; the balance o f humanity, l iving beyond India’s borders, who, like them, did not practice caste, were equal ly good by nature.

In anthropological studies o f D al its, much attention has been devoted to collecting and analyz ing so-called untouchable origin myths— legends and just-so stories explain ing how they became untouchable in the first place. W ha t has been found, over and over again, is that some original ancestor was reduced to the status o f untouchable for reasons that were neither legitimate nor rational, usually involving some sort o f cosmic accident, a trick , or a mis � understanding (Deliege 1999; V incentnathan 1993). The unstated assumption in these studies is that myths are an important source for understanding how Da lits conceptualize their relation to the dominant caste society, which can be read o f f their narrative and symbolic elements in a fairly straightforward fashion. W ha t has received little i f any consideration, however, is who actually knows and tells these types o f stories, under what circumstances they are told and to whom, or what either their tellers or their hearers actually make o f them. That such stories are available for anthropologists who specifical ly request them is clear enough, but we know little about when and how such stories are actual ly drawn upon . Presented in isolation from contextualized studies o f everyday reasoning, myth-focused accounts risk giving the impres � sion that D ali ts live prim arily in a world o f mythology that renders them incapable o f coming to the same sorts o f conclusions we ourselves might make.

M y approach was not to seek out myths or the traditional specialists who might be likely to know and tell them but simply to observe how caste

6 4 « C A S T E , C A R E , A N D T H E H U M A N

was addressed in day-to-day life. Insofar as I actively pursued the question o f caste by asking slum dwellers what they made o f it, I never suggested that my in formants should couch their answers in mythological form—nor did they ever spontaneously see fit to do so. W hen I spec ifical ly asked slum dwellers to explain how castes had come to be, they referred not to primordia l events but to ongoing processes. Jd t is, according to slum understandings, had no fundamental rea lity in their own right but were merely the effect o f selfish actions aimed at monopoliz ing resources.

Spec if ica l ly ,^ / / was an effect o f traits such as stinginess and selfish pride, which slum dwellers ascribed to “the rich.” In posit ingja t i as an effect, rather than an independent cause, slum dwellers thus departed from the popular understanding o f jd t i as natural human type. Suresh, a manual laborer and the father o f two, put it this way:

W h a t is jdtP. I t is simply this: Some people w i l l decide they don ’t want to share. For example, suppose the people on that end o f the street decided they didn ’t want to let those o f us on this end use the water pump anymore. “ I f you come near it, we w i l l beat you .” A n d then suppose they also said: “ N o more school for you— school is on ly for us.” In that case we would have no water and would be su ffering horribly. O ur children would not be able to develop [valara m utiyatu]. Those people would get bigger and bigger and we would remain weak . Soon we would have to serve them and work for them. That is jd ti.

The people in this example only become a group {or jd t i) in the act o f pro � c la im ing that they “ don ’t want to share.” Prior to banding together for the purposes o f exc luding others, the excluders were not a corporate group but merely “ the people on that end o f the street” (i.e., not related to one another except by the accident o f l iving on the same end o f the street). In contrast to popular understandings, as we ll as traditional anthropological theories— in which political-economic realities are treated as external “ factors,” rather than as be ing defin it ive o f caste— slum dwellers conceptua lized resource monopolizat ion and domination as in fac t the very essence o f caste (cf. Berreman 1979; M encher 1974, 1980; Lynch 1977; D eliege 1997; Freeman

1986,156). Employing a biological metaphor, slum dwellers described the normal

course o f things as one in which human beings, like al l l iving creatures, would natural ly develop and flourish (valarum /valaruvdnka) i f given the care they required. A ccording to this picture, “the rich” had prevented them from rising, not just by monopoliz ing natural resources, but also by rejecting the

C A S T E , C A R E , A N D T H E H U M A N • 6 5

fellow humanity o f D al its. This rejection was described in terms o f a refusal to care for and help them. A racu was a young man , and father to two excep � tionally bright girls. H e was aware o f their talents, and though he had only made it to the third grade himself, he aspired to much more for his girls. H e followed their progress in school closely and gave them frequent pep talks about how they could achieve anything. But what he did not tell them was that he didn ’t real ly believe it. A s he explained to me privately, “ The way things are today in this world— in India— no amount o f studying and educa � t ion w i l l be enough for someone from the slum to rise up. A child can study and study, but even i f he has talent, it is not enough. Because what you need is for someone to recognize that talent. Someone must notice and say: ‘Lit t le brother, I see you are studying well. Come along with m e!’ It is their duty [katamai\ to reach out and help like that, but they w i l l never do it. Never.” In expressing his and his children ’s poverty in terms o f a need for help, A racu ’s statement was typical o f slum dwellers. It was also typical in that its critical force was limited to identifying a sin o f omission— rich people’s fa i l � ure, on an individual and collective level, to recognize slum dwellers as people with legitimate needs, to treat them as fellow human beings.

N ot included in slum dwellers’ explanations o f caste were any sort o f ritual or religious practices— contra both Brahminica l ideology and much anthro � pological theory. Indeed , not only did d ietary practices and other ritual observations have no causal relation to caste, but many slum dwellers went so far as to deny even an ex post facto connection between dietary and other ritual observances and caste (cf. Vincentnathan 1993). “ Everyone eats beef,” I was assured by slum dwellers on numerous occasions. “ N adar, Iyer, Thevar, Gounder— all eat i t!” one neighbor exc laimed in response to my queries, genuinely incredulous that I didn ’t already know this. “But they are too proud, and w il l never admit it.” I t was not what people actual ly did that determined their caste, in other words, but simply their abili ty to both c laim and enforce precedence over others (Berreman 1979; Srin ivas 1956). A ccording to the slum dwellers I talked with , too, ri tual purity had nothing to do with caste rank in reality. Purity was merely a conceit o f the privileged, a way o f putting on airs. It was an idiom for ta lk ing about social status, not its source. It would be pointless for slum dwellers to give up beef, because eating beef had never been the real cause o f their low status. Similarly, people in A nbu N agar never cited their ancestors’ association with impure tasks as the reason they were regarded as untouchable and deprived o f access to resources and social power. They believed that their ancestors had been

66 • C A S T E , C A R E , A N D T H E H U M A N

forced to perform difficult and degrading tasks because they were poor and powerless, and not the reverse.

The fac t that the people o f A nbu N agar did not see ritual impurity as the true cause o f their low-caste status, so much as a corollary o f it, d id not come as a huge surprise to me. This was entirely consistent with what other anthro � pologists who had worked closely with D a l i ts had reported. W h a t did come as a surprise to me was that they did not see any connection between the inst itut ion o f caste and the H indu religion . This was not what I had expected to find. N o t only have most anthropologists seen caste as an outcome o f H indu ritual proscriptions, but the dominant Dravidianist ideology in the state o f T amil N adu likewise theorizes caste as a matter o f religious prejudice rooted in H induism . Yet when I posed the possibil ity that caste was due to H induism , people in A nbu N agar were un iformly dismissive. “God doesn’t see jd t i ” they claimed, or “Jd t i is a human creation— it was not made by God .” Interestingly, even Christ ian slum dwellers, who had no vested interest in defending the H indu religion, denied any connection between caste and H induism . W hen I pointed out that only Brahmins could serve as priests in the most important H indu temples, and that until recently D a li ts had not been allowed even to enter most H indu temples, slum H indus were not per � suaded. They saw temples as the earth ly homes o f supremely powerful beings, known to bestow important worldly benefits to those who approached them in the right manner (cf. Krishnamacharya [1936?], 8, cited in Galanter 1971, 476), and, as such, temples were yet another important resource—like educa � tion, good jobs, land , and water— that caste people sought to monopolize at others’ expense. The Brahmin monopoly on priesthood and D a l i ts’ previous exclusion from H indu temples were understood as mere human contrivances and not as intrinsic features o f H induism as such (Roberts 2015a).

THE SOULS OF RICH FOLK

Le t us return now to the peculiar fact that, despite preferring to speak o f “r ich” and “poor” rather than “ D a l i ts” and “caste people,” and despite claim � ing that al l people were basically the same and that jd t i was a “mere lie,” the people o f A nbu N agar nevertheless continued to ascribe heritable character � istics to non-D ali ts. In fact , al though they referred to non-D ali ts collectively as “ the rich,” they had no compunctions about referring to individual “rich people” by their jd t i name and stereotyping them accordingly. Was this not

C A S T E , C A R E , A N D T H E H U M A N • 6 7

a contradiction? H ere is how they explained themselves. W h ile it was true that jd t i was not real, slum dwellers maintained , the selfish at titude that caused people to organize themselves this way— the jd t i strategy— had been “soaked into the bodies [of caste people] since ancient times” and was thus, in a sense, indelible. A nother in formant described caste as having penetrated “ into their very bones.”

Furthermore, while each caste was usua lly spoken o f as having its own unique character, or kunam, these could al l be traced back to a common root. N adars and Chet t iars were typically described as having st ingy kunams and caring about money more than people. Gounders, N a idus, M udal iars, and Reddys, on the other hand, were more frequently associated with domineer � ing persona lity traits. These traits were cited as causing people o f these castes in rural areas to be part icularly invested in requiring D ali ts to perform ritual � ized forms o f self-abnegation . Thus, when I asked in formants to describe these part icular castes in more detail, some typica l replies were, “ Gounders say, ‘N o shoes’ ”; “ N a idus say, ‘ N o shirt’ ”; “Reddys are people who say, ‘Low! Ba d !” ’ In the city, people o f these castes could not get away with such behav � ior, one informant added, because D ali ts could not be so easily identified, and also because in the urban milieu these castes did not possess the means to violently enforce their w il l . Brahmins, for their part , were more often described as simply having “no love” {anpu illa i). “ W hen they see you they w i l l not look into your eyes. They w i l l stand at a distance and look at you up and down warily, or else just ignore you.”

A t one level, then, slum dwellers possessed an elaborate typology o f jd t i- specific traits and tendencies. But when I attempted to systematize slum dwellers’ disparate and sometimes con flicting claims about different castes’ characteristic traits, it became clear that al l the various traits they ascribed to different castes were in fact, so far as slum dwellers were concerned, ulti � mately but variations on a single theme. Laughing at my repeated attempts to derive a consistent model by which each caste’s traits could serve to distin � guish it from others, one o f my closer friends in the slum exclaimed, “ K a l lar , Vanniyar, N adar, Iyer, N a idu , Gounder— al l are the same! There is no differ � ence between them!” Behind each o f these castes’ particular traits was a com � mon tendency. M y slum informants summarized this tendency as perum ai. Perum ai, which normally means “pride,” does not commonly figure in popu � lar or social scientific definitions o f caste (but see Gupta quoted in IS H R 2007,7, and D . Gupta 2004 , xiii). Insofar as “pride” is associated with “caste” in mainstream discourse in India, it is as an outgrowth o f caste, an excess,

6 8 • C A S T E , C A R E , A N D T H E H U M A N

not its very essence or cause. But in A nbu N agar that was exactly how it was understood. W h a t is more, in the slumperum a i did not mean pride exactly; it carried the additional connotat ion o f falseness, as in “ false pride.” M any slum dwellers with whom I discussed the term def ined it additionally as meaning “selfishness” {cuyanalam p a r t ia l) and “not caring about others,” definitions not listed in any o f the T amil dictionaries I have consulted.

To summarize, the superficial behaviors and traits slum dwellers ascribed to caste people on a jd t i-b j-jd t i basis boiled down , upon further questioning, to a few basic and highly consistent notions about the character o f caste peo � ple in general. The descriptive plura li ty o f jd t i gave way to a stark dichotomy, w ith non-D ali t castes on one side and D a li ts on the other. The former were characterized by selfishness and pride, the urge to set themselves apart from others. It was this urge, indeed, that slum dwellers saw as the root cause o f division itself.

W hy, according to slum dwellers, did caste people have such kunamsi W hen I asked, I was told it was because they were so privileged , so rich: weal th i tse lf served to isolate them from other people. This causal relation, however, was not straightforward; I was also told that it was only because they were selfish and didn ’t care about other people that they became “rich.” O nly those who lacked fe llow feeling were capable o f doing what it took to become rich. W ealth could be accumulated, according to the people o f A nbu N agar, only by those with hard-driving and selfish personalities. Anyone who cared about others, on the other hand, could never become weal thy as long as there were people in need. Thus the tendencies that caused some to grow rich included not just greed but a cold and heartless attitude toward others— a basic meanness o f character. Even when such people behaved generously, I was told, it was only in order to show off. Thus, when I asked a close friend o f mine in the slum to give an example o f typica l caste behavior, he described the employers o f his aged mother. Though she had served them for almost twenty years, c leaning the ir home and caring for their children , “she has never had a day off, except when someone in our family has died, or when she is sick. But even when she is sick, they won ’t give her even a few rupees to take the bus home.” Yet these same people hosted lavish dinners and gave gener � ously to religious organizations and charities.

The meanness that slum dwellers attributed to caste people was not seen as being directed only at underlings. They were believed to behave this way toward one another also, and even toward their own children and themselves. Thus they were said to priorit ize weal th and social status over the happiness

C A S T E , C A R E , A N D T H E H U M A N • 6 9

o f their own children . A slum woman I was interviewing on the topic o f suicide surprised me by list ing failure in exams as one common cause. I told her that while I was aware that suicide was common in the slum, I had never heard o f this as being a cause. “ O f course not,” she explained. “ That doesn’t happen here”

It is only the rich who drive their children like that. Our children can develop ambition in their studies o f their own accord, and many do. But we do not push them to it. Sometimes they become very studious and work very hard, studying all the time. And yet when they get to the tenth standard they usu� ally fail [because the local government school for slum children does not adequately prepare them for statewide exams—n r ]. When this happens they will be disappointed and become sad for a while, o f course, but they never go to the point o f killing themselves because we do not drive them like that.

Rich people’s alleged disregard for their own children ’s happiness was seen by slum dwellers as a central theme in their marita l practices. Rich people, they claimed, were even w il l ing to marry their beautiful daughters o f f to ugly or very old men, because al l they cared about was the husband ’s weal th and social status. “ They love money more that they love their own daughters— and do you know what happens? Very often the man w il l die and the girl w il l be left a w idow!”

This brings us to endogamy, perhaps the most important strategy for col � lectively controlling resources in India— and one that is virtua l ly definitive o f jd t i (cf. A m bedk ar [1936] 2014). The people o f A nbu N agar rejected endogamy as a matter o f principle and made no effort to support in-group marriages in practice. The overwhelming majority o f marriages in the slum were “ love marriages” (i.e., based on a romance rather than the decision o f the couple’s families), and all parents I spoke with expressed the opinion that it was a child ’s right to choose his or her own partner. In India intercaste mar � riages are sti ll rare, and those that occur are almost invariably love marriages. A rranged marriages are endogamous by defin it ion , and the insti tut ion o f endogamy rests ultimately on the abili ty o f parents and other k in to dictate acceptable marriage partners. To return to Suresh’s account o f the jd t i strat � egy, what is important to understand is that caste endogamy is both a strategy for resource monopolization (insofar as k inship defines heritable property) and an instance o f it (insofar as marriageable young people are important resources, as is k inship itself).

7 0 • C A S T E , C A R E , A N D T H E H U M A N

IN INDIA ONLY THE POOR CARE ABOUT OTHERS

“ D o you notice anything different about this place? About the people here?” A man asked me this a few weeks after I had taken up residence in A nbu Nagar. It was clear from the way he asked that the answer to him was obvious but that he was trying to find out whether I saw it too. I told him that in my experience the people were uncommonly open, they laughed a lot, and it felt like one big family. “ That ’s righ t!” he said. “Around here there is so much affect ion \pdcam], so much love.” The word he used for love was anpu, pro � nounced “anbu.” A nbu was a popular name for children in the neighborhood where I conducted this research. “A nbu N agar” seemed like an appropriate pseudonym for the place where these children lived and played.

U nlike the rich—whom they described as distant, speaking sourly, easily angered, proud, and generally lacking in fellow fee ling— slum dwellers char � acterized themselves as warm and affectionate toward al l people, as freely sharing whatever resources they had, and as going out o f their way to care for those in need. W itness the following conversat ion between my friend Ganapat i, a man from a different slum whom I ’d asked to conduct a series o f interviews for me (for reasons I w i l l get to below), and a man in a slum neigh � borhood right next to A nbu Nagar:

G A N A P A T I : Do rich and poor people differ in character \kunarri\, or are all people basically the same?

S A M P A T H : It is the poor who are good—i f they see someone who is hurt or suffering, they will always help. But the rich will just drive away in his car.

G A N A P A T I : W ho is more generous, the rich or the poor?

S A M P A T H : I just toldyou: it is us, the poor! Even i f we have nothing to give, we will at least offer a glass o f water. But i f you go to the rich, he will not give you a glass o f water—he will chase you away like a dog! But the poor, even i f someone asks us for some water to wash her hands after going to the toilet, the poor will at once bring water for her.

G A N A P A T I : Can y o u give an example?

s a m p a t h : W hat I said just now is an example.

g a n a p a t i : Okay, okay. But are there any exceptions? Have you witnessed any exceptions?

S A M P A T H : Yes, there are exceptions. Once in Central Station I saw a lady, a Hindu [i.e., caste person]. She came in an auto with packets of food, which she distributed among the poor, and then left. So there are good

C A S T E , C A R E , A N D T H E H U M A N 71

people even among the rich. But the poor always help. Suppose someone leaves their village and comes to the city, where no one knows them. Suppose they are lost. I have seen people feeding them, collecting money for them, and helping them to return to their place. Once this happened near where I work [lifting loads in George Town]. A man had left: the place where he lived, a village, an unknown place. He was just looking for a job, but some [rich] people were suspicious of him. They wanted to beat him and hand him over to the police! But some poor people saw this and rallied to help him. They defended him. They inquired about his background. They fed him, gave him money, and sent him safely along.

A n early experience in the slum showed me what it felt like to be cared for in this way.

Most slum dwellers slept on the hard ground with only a mill imeter-thick straw mat beneath them, and when I first moved in , I thought I should at least try to do the same. The first week was dreadful. M y body was not accus � tomed to sleeping this way, and I rarely slept soundly. I might eventually have gotten used to it, but then I developed a neck pa in that got worse and worse. I also began to experience what is known as “cardiac pa in” in my left arm. I lost feeling in three fingers and could scarcely lift that arm. I checked myse lf into a hospital, where an M R I revealed that I ’d reactivated an old sports in jury: a herniated disc in my neck had become inflamed and was put t ing pressure on a nerve. A neurologist assured me that with proper treatment I had an 85 percent chance o f fu ll recovery.

Luck ily for me, my symptoms gradua lly subsided. But the inc ident directed my attention to two things. The first was the seriousness o f bodily incapacity. Most o f the men in A nbu N agar worked as load lifters, and symp � toms like mine were we ll known in the slum. But unlike me, none o f these men would ever see a neurologist or an M R I machine. A lmost everyone got hurt eventually, and few men could do this sort o f work past the age o f forty. The untimely in jury o f a male breadwinner could spell disaster for a family. The second was the sympathy and concern my condition elicited. The people o f A nbu N agar, I learned, were acutely aware o f other people’s suffering, and—with a few exceptions that we w i l l examine in chapter 3— ral lied around those who had fallen .

Even before my neck pa in got bad, women in the slum had noticed that I was not turn ing my head as far in one direction as the other. E lla iyammal and N agamma, my landlady and her sister-in-law, began asking me whether I was okay and i f I was in pain . They said others had also been wondering. I

7 2 • C A S T E , C A R E , A N D T H E H U M A N

was surprised that they had been able to perceive this, as I had not yet men � tioned it to anyone. Subsequently, after the M R I revealed the seriousness o f my in jury, several men offered to massage my neck . A part from my parents and my partner, no one had ever shown me this sort o f concern, and I didn ’t really know how to react. I felt like a complete loser: pale, well fed, weak . I was unable to do even the simplest things. I had barely begun to live in A nbu Nagar, and already I was a burden on everyone around me. The first time I was offered a neck massage, by my next-door neighbor, a laborer named A nbu , I tried to refuse. “ Com e on,” he said, “ it w i l l make you feel better.” Aga in I demurred. “ H ey! This is important,” he said. “You can ’t just say no’ !” H e began to rub my neck , and I let him . “ I f I were hurt , wouldn ’t you rub my neck?” he asked rhetorically, kneading my trapezoids. “ O f course!” I replied, though in real ity it would not have even occurred to me to offer. But that was not the worst o f it. W hen I told people my doctor had instructed me to sleep on a mattress, another man , Suresh, whom I did not even know very well at that point , began collecting donations from each family in the slum to help pay for it! A mattress would have been a big purchase for a slum family, but not for me. The people o f A nbu N agar understood, o f course, that I had much more money than them, but they clearly did not grasp the extent o f our inequality. The sums they were accustomed to dea ling with did not permit them to form an accurate picture o f just how poor they really were relative to others. I was just a graduate student. I f only they could have known how the c ity’s more com fortable residents lived, let alone the truly rich.

I did not know what to do. W as the mattress money a gift, in the Maussian sense so we ll known to anthropologists— an overture that seeks to init iate a relationship o f mutual exchange, and that therefore cannot be refused with � out causing affront (Mauss 1966)? O r were they simply trying to help, con � cerned for me and believing I needed assistance? U ncerta in how to interpret Suresh ’s offer, I hesitated. Look ing into his eyes, but finding no clues, I decided I would not try to second-guess what was soc ial ly proper. I would just do what I thought was right: I could not accept money I did not need from those who needed it. I told Suresh I could buy the mattress myself, that my research stipend was more than adequate. H is reaction con � firmed that I had done the right thing. H e was not offended but relieved. “ That is so good , Robert ,” he said, “ I am real ly happy to hear that.” H e appeared to mean it. I accompanied him as he went house to house, returning the money he had collected so far. The donors smiled at me, clearly glad to have it back.

C A S T E , C A R E , A ND THE H U M A N • 7 3

As I would only later come fully to appreciate, slum dwellers were uncom � monly attuned to other’s discomforts, psychological as we ll as physical. They seemed to possess a truly uncanny abili ty to “see” pa in or discom fort in oth � ers, where my own untrained eyes could detect no external sign o f it. W hen I had not slept well , for example, people would instantly perceive this and remark on it. O f course I too am able to recognize sleeplessness in others’ eyes, but only in extreme cases. I am far more perceptive when it comes to my own children; when my children are sleepy I can see it instantly. Were I as habitually concerned with others’ wel l-being as I am with my children ’s, I suppose I ’d see their sleepiness and other discomforts, too.

The difference between slum dwellers and me, however, was not just the extent o f their concern for others. They could actual ly see discomforts I could not. They were able to perceive, not just pa in , sleeplessness, and worry, for example, but also hunger. I am referring not to starvation or malnourish- ment, which anyone can spot, but everyday hunger. A t least once or twice a week I would get so carried away with my work that I ’d forget about lunch. O ften I would not even realize unti l a man or woman would remark that I “ looked hungry.” H unger shows itself, apparently, in the eyes. People in A nbu N agar described it as a certain k ind o f “ heat,” a heat you can see. A s far as they were concerned, hunger was no different in this respect than sleepless � ness or anger or worry. “Your eyes are burning, Robert ,” someone would say. “ H aven ’t you had lunch yet?”

Could it really be that hunger shows i tse lf in the eyes? I had never heard o f such a th ing before, and try as I might I was never able to see any difference in how my own eyes looked when I ’ d missed a meal. Yet time and again slum dwellers would correctly identify my hunger. I do not th in k this can be explained as just a series o f lucky guesses; while there were times that I was hungry and no one noticed, I was never told my eyes showed hunger when I was not in fact several hours overdue for a meal. Sense perception is learned. I cannot, for example, perceive the scent o f cockroaches. Though I have been exposed to it many times, it is a scent I sti ll do not know how to identify. I was not even aware cockroaches had a scent until, on one o f my first trips to India, a woman we had hired to clean our flat after a two-week holiday stunned me by remark ing that the whole k itchen stank o f them . There were no cockroaches visible at the time. But unbeknownst to our cleaner, I had sprinkled the place with insecticide two weeks earlier and, just prior to her arrival, had swept up and disposed o f the dozens o f dead ones that we found upon returning. W as our maid some sort o f olfac tory savant?

7 4 « C A S T E , C A R E , A N D T H E H U M A N

That is exactly what I thought. But I have since heard from other South Asians, inc luding the extremely elite Indians I went to graduate school with, that cockroach is a familiar scent to them and impossible to miss. It is not that I have a hopelessly bad sense o f smell. W ork ing in restaurants over a four-year period before beginning my Ph D , and sampling customers’ leftover wine on the sly in an effort to educate myse lf in such matters, I eventually learned to discern several grape varietals. W ith sufficient time and motiva � tion one can apparently develop one’s senses quite a bit.

I f there are smells only some people can perceive, might there not be other qual it ies— the look o f hunger, for instance— that we are as yet unaware of, but that are clearly visible to those whose entire lives and social existence revolve around suffering and concern for others? A s anthropologist Joe l Lee argues in a groundbreaking ethnography o f D a l i t sanitat ion workers, our “organs o f perception” are differently “c a l ib ra t e d] . . . according to social loca � tion (2015, 4 6, emphasis added). N ew research in the psychology o f percep � tion suggests that power changes the way the brain responds to others, physi � ca l ly and in measurable ways (Hogeveen , Inz l icht , and O bhi 2014). Specifically, it has been shown that humans and other primates lacking in power, defined as social dominance, are better able to process subtle percep � tual cues and take on “ the visual, cognitive, and emotional perspectives o f others” (2014 , 755). I f we lack the abili ty to see hunger, perhaps it is only because we live in a world where such perception is neither valued nor cultivated.

O nly the poor, slum dwellers often say, care about other people. But why? I t was not, they claimed, because they were born with some unique percep � tual talent or moral sensibility other people lacked. A s I have already stressed, the people o f A nbu N agar emphatically denied that they possessed any inher � ited trai ts— even positive ones— that set them apart from others, going so far as to claim that they had nothing whatsoever in common with one another apart from the fac t that they were poor and rejected by other Indians. Care was conceived by them not as something special but as an impulse all humans naturally felt for others. W h a t required explanation , so far as they were con � cerned, was not why they cared, but why those they called “the rich” did not.

The rich, we now know, were not simply those w ith more money than slum dwellers. The characterist ics and cultura l practices my informants ascribed to “ the rich” were not those o f wea lth per se but o f caste. Those they called rich were often no weal thier than themselves, and they did not apply this label to fellow D a li ts who just happened to have risen out o f the slum and

C A S T E , C A R E , A N D T H E H U M A N 75

into the middle class. W h a t is more, as we have seen, they credited foreigners with the same care and fellow feel ing as the poor, though they understood that foreigners were often extremely rich in the literal sense o f the word. That slum dwellers credited foreigners with an impulse to care for others identical to their own was apparent, for example, in the rumors over foreign money discussed in chapter i. It was intrinsic to the logic o f these rumors and to slum dwellers’ suspicions that money sent for benevolent purposes was being diverted by selfish intermediaries.

The myth o f the benevolent foreigner, as I showed in chapter i, is not easy to explain empirical ly— there are no solid historical grounds for its emer � gence. I alluded previously to the possibil ity that this myth could be explained by way o f a structural argument. It is time now to revisit that claim. First, I should stress that my account o f the myth is not based merely on slum dwellers tell ing me, “ We believe foreigners are good.” This is important to underscore, because no anthropologist worth her salt would simply take flattering statements about foreigners at face value, at least not i f she herself were foreign. To make sure that I was not in some way influenc ing the very phenomenon I was trying to make sense of, I hired a man from another part o f Kashtappatt inam to conduct a series o f open-ended interviews with slum dwellers outside A nbu N agar, slum dwellers who had never met me and were unaware o f my research. The interview questions I prepared for him probed not just their ideas about foreigners but a whole host o f topics, inc luding the differences ( i f any) between the kunams o f rich and poor. In this way I was able to independently confirm what I had already discovered by other means.

The interviewer’s name was Ganapati. I have already quoted from one o f Ganapa t i’s interviews, with a H indu slum dweller named Sampath, above. Later in the same interview, Ganapati shifted the conversation to foreigners.

g a n a p a t i : Is there a difference between rich Indians and foreigners? D o foreigners have the same kunam as the rich, or different?

s a m p a t h : Foreigners are also rich, but there is a difference. I sha ll narrate to you an inc ident that took place, on a very busy road not far from here. A poor man was drunk . H e was com pletely drunk . H e was so drun k he had lost his senses and just lay down right there on the edge o f the road. C ars and trucks were passing very close to the man lying there, uncon � scious. But a ll the people w a l k ing past just looked at h im and continued wa lk ing! But then a fore igner came along. W ithout saying a word , he lifted the drunken m an— a poor man , a very d ir ty m an— in his arms. H e

C A S T E , C A R E , A ND THE H U M A N

carried h im to the side and la id h im down gently, in a safe place, under a tree. Foreigners have a very good kunam. Their kunam is to help people. They treat everyon e equally [emphasis added].

U nlik e the other slum dwellers Ganapa t i interviewed , Sampath was able to substantiate his positive assessment o f foreigners with an empirical example. But it seems un likely that Sampath ’s fa i th in the egali tarianism o f foreigners, and in their spontaneous concern for those in need, emerged fully formed only after he had witnessed this one chance event. Similar ideas about for � eigners were expressed by all o f Ganapa t i s other interviewees, none o f whom were able to cite any actual experience in support.

“ Their kunam is to help people. They treat everyone equally.” Sampath was speaking about foreigners here, but his description matched exactly what slum dwellers said about themselves, the poor. Given that almost no one in the slum had ever known a foreigner before meet ing me, it seems reasonably clear that their ideas about them were in fact a project ion o f their own ideo � logical self-image onto others. They needed foreigners to be like this, and i f no foreigners existed they would likely have had to invent them— indeed, that was exact ly what they did. But why?

The favorable ideas about foreigners that I encountered are unlike the xenophilia anthropologists have observed in Melanesia, where dark-skinned natives often valorize whiteness and “whitemen” at their own expense (Bashkow z oo6; Robbins 1998). It was not white foreigners who slum dwell � ers believed were more caring than India’s “rich” but all the world ’s people— A fricans, East Asians, and those living in the M iddle East. N or did they see themselves as in ferior to the foreigners they celebrated. Theirs was not a hier � archical xenophilia but one in which the goodness o f the foreign was simul � taneously their own . They were moral, the people o f A nbu N agar believed, not because they were in some way unique but because they were merely human. The ir systematic denial o f caste ideology entailed denying also that they themselves possessed any inherited group-based character traits, even positive ones, that would set them apart from other people. Their desire to care for others was not something they had cultivated but a natural part o f what it meant to be a human being.

For slum dwellers humanity was a biological substratum and a shared spe � cies identity. It entailed moral claims about the fundamental equality and value o f al l persons, and a sense o f our potential to connect with one another across polit ical and cultural divisions— in other words, it was quite similar in some respects to European notions o f humanity.5 But in the slum, humanity

C A S T E , C A R E , A N D T H E H U M A N • 7 7

also entailed additional, substantive qualities that are not normal ly treated as central to the W estern version o f the concept. To be human was to be vulnerable and susceptible to harm . It was also to care instinct ively about those who were in need, whoever they might be, and to feel called upon to care for them. A n d finally, to be human was to be onese lf worthy o f being cared for by others.

W ha t I gloss with the English word care corresponded to no single word in slum dwellers’ lexicon .6 It summarizes a constellation o f interrelated words and concepts. O ne such word was par, which li tera l ly means “seeing” or “ looking at,” but which also means “at tending to” someone out o f concern , acknowledging that person ’s presence, treat ing him or her as consequential. This sense o f the word p a r was not unique to slum dwellers. W ha t was unique was the central ity o f this and related concepts to slum dwellers’ moral sensi � bility, a sensibility they perceived as rare or altogether absent among members o f the dominant society. Thus “the rich,” or caste people, were often described as “not looking at” others, especially the poor; “ They don ’t see us,” I was told. “ They w i l l scrutinize you, inspect you, but they don’t ever see you” as a fellow human being. O ther closely related terms in slum dwellers’ vocabulary o f care included kavani (pay attention to), anpu (love), irakkam (sympathy),pdcam (affection), as well as utavi kotu (give aid to) and h i l f (a T amil pronunciation o f “ help”). Just as the active and passive forms o f care were not regarded as separable in actual practice, slum dwellers treated transitive and intransitive terms (i.e., giving a id versus fee ling love) as mutua lly entailing. Thus, when I asked them to tell me what it meant to “ have affection” for someone, they often responded in a manner that made nonsense o f the distinction between active and passive, transitive and intransitive, concepts. For example, as one woman explained to me: “A ffect ion means helping, and helping means feel � ing affection!” {Pdcam nd h i l f h ilfndpdcam ).

The opposite o f care could be summarized with a single word: purakkanippu. Purakkanippu is derived from another word for “ look ” or “see,” kdn, and its everyday meaning is “ ignoring” or “ treating w ith disda in” (lit. “ look ing away” or “refusing to acknowledge”), but it is also the term used for ostra � cism, or “social boycott,” as it is known in India— the organized practice o f treating some individual or group as being beyond the pale o f human society. It is, in other words, a practice o f cruel and w il l fu l disregard. “ To ignore” in this sense was both the most serious violation o f slum moral sentiment and the most quintessential stance o f “ the rich.” The practice o f social boycott was among the most powerful sanctions imposed in traditional caste society and

7 8 » C A S T E , C A R E , A N D T H E H U M A N

f i g u r e 8. M ora l isola t ion : two views (credit B irgi t t Sippe l).

could be temporary or permanent as in the case o f outcasteing. A m ong those who had suffered the legacy o f untouchability, the term purakkanippu could have upset ting connotations and was not used lightly. N ot real iz ing this, I once made the mistake o f jok ing to my assistant that he had been “ ignoring” me. The verb I used was purakkani, from which purakkanippu derives. I w ill never forget the stricken look on his face. “ N o , Robert,” he said, tak ing my hand in his, “ I would never do that. N ever .”

To summarize, “ the poor” were not those who necessarily lacked money. They cared for one another, not because they were poor in the literal sense, but because they did not practice caste. In this way slum dwellers were able simultaneously to account for their rejection by other Indians— a rejection they experienced as personally humiliat ing— and to assert their inclusion w ith in a larger humanity. This, I argue, is the true significance o f their oft- repeated claim to be “merely human .” The care they showed toward one another could not be premised on a k ind o f jd t i patriotism—which would make it exclusive, and therefore immora l—but was instead founded on their lack o f caste, their unadorned humanity. A n d humanity, for them, was a club in which everyone was welcome, though not everyone wanted to be part o f it. Those who refused to care— the majority o f Indians, in their view—were not just rejecting the humanity o f others but negating their own humanity too. By pro jecting themselves as simply human and noth ing more, slum dwellers envisioned themselves as connected to globa l humanity in a way they believed caste people could not be so long as they continued to practice caste. In this way they reversed the dominant picture that isolated people like themselves as uniquely stained, by reframing caste people as isolated from a moral ly superior humanity o f which they themselves were a part (figure 8).

C A S T E , C A R E , A N D T H E H U M A N • 7 9

Structural ly speaking, the foreigner was identified with the se lf and the cona � t ional with the other.

In this chapter we have examined how slum dwellers conceptualized, and attempted to make moral sense of, their relationship to the dominant society and to one another. W e considered how interrelated ideas o f an intrinsical ly caring humanity and o f a benevolent world beyond India’s borders enabled them to reverse the stigmatic identity that excluded them from the fully human . This hopeful vision existed as part o f a rather more tragic everyday reality, by which it was often belied. The dichotomy this chapter describes between the slum as a place o f love and mutual care and the dominant nons- lum society cast in the role o f vi l la in betrays a social and polit ical-economic context in which things were not always so clear. In the next chapter we consider two fundamental fault lines in the idealized picture o f slum moral � i ty presented above: two instances in which the slum was not only similar to the nonslum but functiona lly inseparable from it.

8 0 • C A S T E , C A R E , A N D T H E H U M A N

T H R E E

Sharing, Caring, and Supernatural Attack

I N T H E p r e v i o u s c h a p t e r W E examined a moral discourse in which slum dwellers portrayed themselves as the epitome o f goodness, the very embodiment o f humanity and care, while describing their counterparts in the dominant, nonslum society as unqual if iedly bad. H ere we shift our focus from slum dwellers’ own ideal-typical account o f the slum and its other in order to consider two fraught social relations w ith in the slum that contra � dicted the claims o f a humanity premised on care.

I begin by noting that slum dwellers’ portrayals o f the slum as a moral community were to a great extent true. Slum dwellers were indeed uncom � monly warm and affectionate, and, more importantly, they really did take care o f each other on a da i ly basis and in times o f need. This moral commu � n ity did not, however, arise only from slum dwellers’ unadorned humanity, as they themselves implied, but was also the contingent result o f strategic interactions in the context o f both natural and supernatural threats. A nd two o f the principa l relationships out o f which moral community was con � structed— the relationship between husband and wife and between slum women who lent each other money—were simultaneously the two most pro � found dangers to moral community, and in many ways contradicted the basic tenets on which it was based. These fault lines converged on the individual married woman , exerting a pressure that was at times unbearable. Suicide among women was, by al l accounts, rampant— adult slum dwellers claimed that “ hundreds” o f women had k il led themselves in and around A nbu Nagar over the previous two decades— and frequently in a horrifying manner.1 The most common method o f female suicide was tik kuh tta l (“ bath o f fire”): the woman doused herse lf in “ Krisna o i l” (kerosene) and lighted a match. Men, by contrast, either hanged themselves, consumed poison , or laid themselves

81

down before an oncoming train . W h ile women occasional ly used those methods, I never heard o f a man committing tikkulitta l.

The starkness o f women’s economic dependence was such that it was sometimes envisioned by both men and women in the slum as a k ind o f “class” division . The basic terminological build ing blocks o f slum class theory— “ the privileged” and “the poor”—were thus used in ways that chal � lenge our basic assumptions about what might constitute a class: in the slum it was frequently said that “around here, the women and children are very poor.” Class, we see, did not just separate slum dwellers from everyone else; a “class” division also ran through the very heart o f the family. Similarly, while essential differences between jd t i s were often denied in the slum, gender was sometimes described as a k ind o f caste relationship. “ There are n o ja t is in this world ,” one male slum in formant concluded, after having debunked the popular understanding in some detail. “Jd t i is a mere lie and only exists in the mind .” But then, immediately contradicting h im se lf he added, “ In the entire world there are only two jd t is: the male jd t i and the female jd t i ”

The burden o f this chapter is to delineate the mutual ly constitutive rela � tion between the everyday ideals and moral practices o f slum life and their structural tensions and intermittent tragedies. The data it considers range from the straightforwardly economic to the apparently supernatural, bridged by a series o f ethnographic accounts centering on the strategies women employed in their relationships with one another and with their husbands. We begin with the fact that men were the prim ary wage earners and that women depended on men’s wages to manage their households. The economic relation between husbands and wives cannot be understood apart from its role in the moral structure o f the slum community; a woman ’s management o f the household came with the heavy burden o f her being held ultimately responsible— in a language o f culpabili ty bound up with notions o f wife ly auspiciousness—for her fam ily’s well-being. Because a condition o f familia l well-being was a steady flow o f income from an often unreliable partner, she frequently found herself in an impossible situation, and one that slum dwell � ers themselves cited as the source o f women ’s religiosity. W omen turned to gods, in other words, because their husbands so often fa iled them.

Equa l ly important in married women ’s lives were their relations with one another, the economic and moral bonds among women being a safety net o f sorts in a world where neither husbands nor gods could be fully relied upon. These bonds were enacted both through freely given social and material aid and by means o f interest-bearing loans. Though often forced to repay their

8 2 • S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

debts to one another at usurious rates, slum women nevertheless styled the money that others had lent them as “ he lp” and the lenders themselves as “ friends.” A n d in a sense they were. D ebt was contracted on the basis o f persona l connect ion and was one o f the means by which friendship was articulated . For the lender, debt was also a form o f investment, or even o f savings, that had moral and supernatural benefits as we ll as straightforwardly economic ones. But what this otherwise sanguine picture leaves out is what occurred when these relations went sour, as they often did, and women found themselves unable to make payments. For after they had turned to money � lenders to help make ends meet when their husbands fai led them, their own fai lure to manage their moral and financia l bonds w ith other women could have potentia l ly disastrous consequences. D ebt and marita l strife were together recognized by slum dwellers to be the princ ipa l factors driving women to suicide. D ebt and marriage were thus at once an al l but indispen � sable source o f material care and moral support for slum women— the very basis o f moral community— and at the same time the greatest threat to it.

The chapter concludes by reveal ing how notions o f auspiciousness worked in tandem with vankanavi (the spirits that arose from the envy or malicious desire o f other women) to structure the lived forms o f moral responsibility in the slum by atomizing women and deflecting attention from debt and mar � riage as causes o f women’s suffering. E fforts to avoid provoking vankandvi, moreover, involved strategic misdirection and downplaying o f one’s own weal th. This chapter traces the unstable relationship between slum dwellers’ professed moral principles and the practices that contradicted them. In so doing it identifies one o f the principa l forces behind women’s conversions to slum-based Pentecostal Christ ianity .

MEN W OR K , WOMEN PRA Y

The Rigors o f W aged Work

The majority o f C h en n a i ’s slums are small, physically isolated from one another, and surrounded on al l sides by middle-class dwellings that provide ample opportunities for women to work as domestic servants— cleaning the floor, washing the dishes, and doing the laundry o f nonslum households under the supervision o f the women who lived there. A nbu Nagar, by con � trast, was part o f an agglomeration o f slums in C henn a i ’s industrial north, in which slum settlements vastly outnumbered nonslum neighborhoods, and in

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 83

which opportunities for female employment were correspondingly scarce. Because relatively few married women in A nbu N agar and surrounding slums took part in the cash economy, household income was smaller than in other slums and considerably less stable.

The majority (roughly 80 percent) o f male wage earners in the slum were “coolies.” These were manual laborers who performed physically difficult tasks (mostly l ift ing loads and transport ing them around Chennai) in the so-called in forma l sector and were pa id a dai ly wage known as a k u li (coolie), generally about Rs. 150 ($3). Approximately 10 percent had relatively perma � nent jobs as peons, auto rickshaw drivers, mechanics, and so on, that paid at roughly the same rate but were generally less physically demanding. Two or three slum men earned significantly more than this. A nbu N agar ’s highest earner was employed as a c ity bus driver and earned Rs. 6,000 ($130) per month. I f a man was able to work consistently and did not drin k regularly or in excess, his wife would have little trouble m aking ends meet and could even expect to build up a small savings. But hardship could strike at any time i f for some reason work became unavailable or i f additional costs were accrued, for instance when there was an illness in the family. The worst was when the man h imse lf fell i l l or was in jured, for in this case the lack o f income combined with additional costs, and a prolonged absence could also cost a man his job. Temporary loss o f income did not spell disaster because families could always rely on handouts from neighbors and kin , but families with permanent defi � cits, whether due to an unreliable wage earner, his illness, or his death, showed clear signs o f suffering.

Every man knew, moreover, that i f he left his job or made trouble there were many others worse o f f than h im se lf wa i t ing to take his place. Indeed, the labor market was such that men must frequently pay to work . M y next- door neighbor A nbu , for example, original ly had to take on a Rs. 5,000 debt to buy his way into a job. H is was not a part icularly prized job either; he unloaded containers for an international container and freight service, show� ing up at 10 :00 a.m. each day to see i f he was needed. I f so he worked until the job was finished, usually coming home by 9:00 or 10 :00 p.m. Some days there was no work to be done, and i f this went on for three days or more he began to worry— to the extent that whenever the opportunity presented itse lf A nbu was w il l ing to work two or even three back-to-back shifts all through the night, without rest, and into the next day. O n several occasions I saw him return home after over th ir ty hours o f continuous labor.

W h a t ’s that like, I asked him . “ H ard ,” he said.

8 4 • S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

f i g u r e 9. A n in e ty-pound laborer , after a long shift .

Slum men showed a curious lack o f affect when describing their work . To my surprise they seemed to possess no elaborated language to describe their work ing experiences. They would say that their limbs hurt and some � t imes even declare that they just could not go on, and they would stay home for a day or two or even more. I f this continued, however, their wives would at first complain and eventually instigate a fight, because when a man did not work the wife was the first to suffer. A t the same time, the women were also the first to aff irm that men’s work was hard. The d iff iculty o f men’s work was accepted by everyone. Everyone in the slum agreed, for instance, that men drank because their bodies hurt and that drink was necessary to make the pa in go away so that they could sleep. It was seen as a simple matter o f cause and effect. “You come home, drin k three packets [approx. 150 ml

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K 85

each], and. eat some fish. O nly then w il l you forget the pa in and be able to sleep.”

Though women too accepted that, for the laboring man , d r in k was a necessity, they would not readily admit this to their pastors or to the social workers who sometimes came to the slum to conduct surveys and dispense advice about “upliftment,” as noted in chapter i. I t was in fact only excessive (and therefore costly) drink ing, or drin k ing that led to unruly and erratic behavior, that slum women opposed. O therwise drin k was accepted as being necessary to endure a life o f utalulaippu (bodily toil)— as necessary, indeed, as meat. O nly those who “ do not work” could survive on the vegetarian diet o f high-caste H induism .

Everyone in the slum understood, moreover, the permanent toll that such work could take on a man , for many men in the slum were no longer able to work because o f in juries they had suffered. A nbu h im se lf had several scars on his hands from times when they had been cut open by a sharp object unexpectedly pok ing out o f a load he was lifting. But it could be much worse, he explained, in response to my queries. “ Suppose we are l ift ing someth ing heavy and diff icult to carry . . . someth ing that takes four or more men to lift , like marble slabs. I f it drops, and a man is under it: d l k a li [the guy w i l l be finished].” A nbu had seen such th ings happen , but he betrayed no emotion when speaking o f this. The phrase he used, moreover, is an interesting one: d l simply means “guy” or “ hired worker.” The term k a li normally means “empty,” or “vacant,” as in a house that has been vacated, and is also used to describe a bat tery that has lost a l l its charge and is hence “empty.”

In other cases the source and nature o f the in jury, or pdtippu (lit: effect), was difficult to discern. Ten paces from where I stayed there was a desolate- looking man in his early forties who would just sit all day, outside the house o f the married elder sister with whom he lived. H e had been unable to work for the last ten years, I was told, because o f feebleness (palavm am). I t was never clear to me, however, and I was never able to get anyone to explain , whether the problem was with his body or his spirit. Maybe both, someone replied to my queries, shrugging: “ The body cannot carry on i f the spirit is broken, and i f the body is not well the spirit w i l l sink .” But never was ^ sug� gested that he and others like him were simply lazy or malingering. This was particularly strik ing in an environment where suspicion abounded about others’ true intentions and where crit ic ism was frequently and openly expressed. The very idea that someone like him could be mal ingering was

8 6 • S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

uniformly rejected as ridiculous: “ O f course they want to work . W ho would al low themselves to become dependent like an elderly man?”

Or, for that matter, like a woman . . .

W omens D ependence on M en a n d Gods

The limited avai lability o f waged work for the women o f A nbu N agar and surrounding slums meant that they depended entirely on male breadwinners, usually their husbands. A n d to ensure women and children s well-being, male providers had to not only earn but fa i th fully hand over their earnings and avoid spending too much on such things as drink , cinema, cigarettes, or snacks on the way home from work . M en ’s will ingness to spend any portion o f their da i ly wages on snack food was resented by wives and monitored closely. A lthough not expensive in themselves, snacks purchased at roadside eateries st i ll cost considerably more than home-cooked food . Worse, women knew, the longer men spent snacking with comrades rather than rushing straight home after work , the more l ikely they were to move on to drin k and other wasteful pursuits like cinema and other women . Stay-at-home slum wives pictured the world beyond the slum as a veritable cauldron o f tempta � t ions and vice; snack food was spoken o f in the same disapproving tones as excessive drink , and cinematic heroines shaded into real-world temptresses bent on steal ing their men away. This worry was a bit exaggerated. W hile drin k and cinema were very much with in the reach o f slum men, girlfriends were considerably harder to come by. But just as middle-class discourse envi � sioned slums in general as teeming with loose women , the women o f A nbu N agar saw other, more distant , slums this way. But the real ity was that while unattached women existed in all slums, they were subject to constant surveil � lance by both k in and neighbors. A n d however much these poor women might hope for a man to play the role o f provider in their lives and to make them feel spec ial again, slums’ close quarters and relat ively strict sexual moral ity ensured that such hopes were largely in vain .

The threat posed by other women might be overestimated, but slum wives’ concerns about snack food , drink , and cinema rested on a solid economic logic. A part from water, v irtu a l ly everything necessary to ma in ta in ing a household— food , soap, clothes, fuel to cook with—had to be purchased with the cash that male wage earners supplied. Some outside spending by men was not necessarily a problem , but even without any diversion o f funds men’s meager incomes barely covered household needs. It was thus with good

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 8 7

reason that wives feared any sign o f enthusiasm for the high life. But women’s moral iz ing about male vice was collectively shrugged o f f by men and— outside slum Christ ianity— received no authoritative support. M en were not answerable in any public forum , and women ’s confrontations with their men were contained with in the individual household. Publicly, it was the woman who was held responsible for the appearance o f the children , the feeding o f the man, and the smooth functioning o f the household, and i f these jobs were neglected it was ultimately she who was blamed— even though, as everyone knew, she could not fu lf i l l her wife ly duties i f her husband did not bring in cash.

Thus, al though women appreciated fu ll well the d iff iculty o f men’s lives and labor, and admitted that both drink and the occasional day o f f from work were a necessity, their total dependence on male wages frequently led to contradictory statements on their part. O n one hand, women emphatically denied that slum men were irresponsible louts who simply refused to care for them— as middle-class and governmental discourse would have i t— and placed the blame instead squarely on the fact that their men were forced to labor under extremely punishing conditions for wholly inadequate pay. O n the other, because they lived in constant fear that their own and their chil � dren’s well-being hung in the balance, women were also obliged to play the taskmaster, continuously monitoring the ir husbands’ output, watching obsessively over every rupee their husbands might have spent, and noting how regularly they worked.

O ne day, while sit ting with a group o f slum women , I asked them i f they had ever heard a part icular T amil saying, Kanavane kan kanta teyvam, mean � ing roughly, “ The husband is a l iving god [for the wife].” W h ile it is doubtful that such sayings are ever taken literally, the idea that the wife should obey and serve the husband as i f he were a god is an often stated T amil ideal. These women, however, had never heard this saying and laughed derisively, “ H a! Maybe that’s the way things work when you ’re rich, but i t ’s not like that around here!”

“ W hy not?” I asked.

“Because they don’t work!” several women answered emphatically. “I f only they would work, they would be like gods to us,” another woman added, and they all agreed, laughing. “I f you have a good husband, everything in your life will be good,” Rani explained as the others nodded in agreement. “Some men just refuse . . . or, they will work, but they will waste all the money on movies or drink and pdtta i [sauteed goat intestines]___I f they don’t earn, we su ffer .. . .

88 S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

They run around the city all day and forget about us!” Rani thus defined a good husband, very simply, as one who had a job and worked diligently.

The voices o f complaint subsided and everyone’s attention turned to K a la ivani— a very skinny woman with no jewe lry and four perpetual ly dirty children—who was look ing downward and shaking her head. (We met K a la ivani and one o f her sons, Sudagar, in the previous chapter.) H er hus � band Gopa l was one o f the worst, a fact no one needed to mention. H e drank too much, came home late, and had even pawned her necklace without her permission and hadn ’t made a single payment, ensuring that this, her one personal possession o f any value, would soon be lost forever. Women ’s jewelry was considered to belong solely to them, and for Gopa l to have appropriated his w ife ’s necklace was a serious violation o f local norms. But like so many violations in the marita l relation , it came at no cost because there was no public forum with in which K a la ivani could effectively assert her rights.

“ Your husband is good, Ce lvi ,” K a la ivani said gloomily. “ Yes, he’s good ,” C e lv i admitted , as she and the others rall ied around

K ala ivani . I was surprised to hear C e lvi admit that her husband Raju wasn ’t so bad after all, as her complaints about him slacking o f f work were a con � stant theme in her conversations, and the two would often fight. A fter a bad fight she would not infrequently abandon her own household in protest, relocating to her mother’s home scarcely twenty paces away from her own, where she was surrounded by supporting sisters. “ I ’m never going back,” I heard her announce on more than one occasion. In her absence, Raju would be forced to go to his own mother or to his brother’s wife for food , and, soon t iring o f this additional burden , they would urge h im in increasingly strong tones to patch things up with C e lvi . Eventual ly Raju would be forced some � how to win her back. “ I thought you were never going back ,” I asked her after one such drama. “ W ha t happened?” C e lvi gave me a significant look, smiling with satisfaction . Raju was now working very hard, she informed me, and had sworn o f f cinema entirely.

K a la ivani was not so lucky, and it was not for C e lvi to complain about Raju at a time like this. A lthough K a la ivan i ’s husband Gopa l had a good job, driving an auto rickshaw just like Raju, K a la ivani and her children were very poor. For unlike Raju, Gopa l was a habitual drunk and—just like her father, another drunkard—was frequently absent and rumored to have a girlfriend in another part o f town . Two generations o f husbandly neglect had taken a toll on the family: K a la ivani was bone th in and had four children whose meager c loth ing hung o f f their skinny frames. A l l this because neither

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 8 9

Ka la ivani nor her mother had had good husbands. Indeed, moving to A nbu N agar had been a significant achievement for Kalaivani , who had been born under a temporary shelter on the side o f a street in George Town , C henn a i ’s old business district . She once took me to visi t that street, where she had lived, homeless, for most o f her life; many o f her family members lived there still, and she was strongly attached to it.

The plight o f women like Kalaivani , however, cannot be measured simply in economic terms, nor was their gr ie f confined to their relationship with their husbands. A lthough other women most often treated K a la ivani kindly, she had little real standing in the social world o f the slum and was often dis � paraged behind her back by other women, who spoke o f women like Kala ivani as i f they were ultimately responsible for their own misfortune: “ That woman is totally useless—she can never do anything r igh t!. . . Look at how she keeps her children---- W h a t k ind o f a life has she made for herse lf?. . . N o wonder [her husband] never comes home.” A n d this despite the fact that, as everyone would readily acknowledge, the real problem was almost never with these women themselves but with their husbands.

This paradoxical and— from the perspective o f slum dwellers’ own pro � fessed ethic o f sympathetic care— contradictory tendency to condemn the vict ims o f spousal neglect can be understood in two ways. I t can be under � stood culturally, as arising from the discourse on wife ly auspiciousness. It can also be understood in terms o f the economics o f female homosocial ity. Beginning with the cultural, the moral censure o f less fortunate women drew upon deeply ingrained assumptions, we ll attested in the ethnographic litera � ture on T amil N adu and India more generally, that a woman ’s “auspicious � ness” (pdkkiyam) is the underlying source o f the prosperity and heal th o f her household (Leslie 1991). The fac t that household well-being is credited in India to female spiritual potency is often celebrated as a k ind o f protofemi � nist undercurrent in Indian culture. But when things are not working out or when some tragedy befalls the family, it is the w ife ’s lack o f auspiciousness that is to blame. The discourse o f female auspiciousness is thus a double- edged sword. It may seem odd, even incomprehensible, that a woman would be held ultimately responsible for conditions over which she has no control, when everyone knows that the proximate cause o f her distress is in fac t hus � bandly neglect. But cultures often distinguish between proximate causes and ultimate responsibility. It may be true that a woman ’s suffering is due to her husband, but the question then arises: W hy this woman and not another? A comparative case in which proximate causes and ult imate moral

9 0 • S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

responsibili ty are differently assessed can be found in Evans-Pritchard ’s widely cited discussion o f why collapsing granaries fa ll on one person and not another (1937, 69-70) . W h ile the A zande understand perfec t ly we ll that granaries collapse because termites have eaten away at the supports, they explain the apparently random fact o f who is si t ting under it at the time o f collapse by witchcraft. So too with women whose lives are ruined by bad husbands; only instead o f holding some unknown witch responsible, the discourse o f w ife ly (in) auspiciousness blames the vict im herself.

Female auspiciousness is a common trope in T amil culture, and far from being a prowoman ideology it has the potential to harm women from any w a lk o f life, not just slum dwellers. But the part icular conditions o f A nbu N agar and surrounding slums were such as to exacerbate the negative impact it had on them. U n like women in many other Chenna i slums, or D a l i t and other poor women in cities and villages throughout India, the women o f A nbu N agar and surrounding slums could not participate in the cash econ � omy to any great extent. A s we have seen, they lacked the sort o f employment opportunities available to women in slums further to the south o f the city, where woman could f ind work as household servants in nearby middle-class neighborhoods. Kashtappattinam , by contrast, was almost entirely slum, and the middle-class households nearby were too few to employ more than a small fraction o f slum women . These women therefore had virtua l ly no way o f compensat ing for husbandly fai lure. Moreover, unlike most rich and upper-caste women who were also subject to the burdens o f auspiciousness discourse, even the best-off o f A nbu N agar ’s women lived not far from serious catastrophe because o f their absolute poverty and social marginality.

W e turn now to a second way o f understanding the tendency to blame un fortunate women for their own suffering: the specific economic dimen � sions o f female sociality. The fact that women like Ka la ivani were in perpet � ual straits (unlike most slum women who were only intermittently so) meant they had noth ing to contribute to the various moneylending and debt schemes through which female social ity was articulated. C u t o f f from par � ticipation in these various financ ia l arrangements among women , Ka la ivani and others like her were isolated from an important source o f moral valida � t ion and support. W h ile other women might sometimes commiserate with Kala ivani , they just as often treated her as an outsider and pest—which, from the perspective o f women ’s mutual aid arrangements, she was.

Both the ideology o f auspiciousness and the economic rational ity o f female soc ial ity resulted in women being blamed for their household ’s

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 9 1

poverty. The essential dependence o f women on forces largely beyond their control correlated with another key difference between women and men that is o f considerable importance to the overall argument o f this book . U nlike men, who generally showed little interest in religion on a day-to-day basis, slum women were h ighly punctilious in their religious observances. A lthough slum dwellers occasionally explained this discrepancy in terms o f women ’s inherent nature (kunam)— for example, woman were said to “ have more bhakti” than men—most slum dwellers took for granted that the real reason women were more religious than men was that they had more worries, had less control over their lives, and were so poor. A s one man put it:

For the man , there is only work and no time for religious devotion [paktt\. I f he is able to work , then this is enough for him . But for the w i fe there is always worry: “Le t h im be well, let h im work well, let h im not get in to trouble, let us not fight.” M y wife cannot say these th ings to me direc tly. I ’ l l just say, “Yeah, yeah ,” and then forget about it. O r maybe her pestering w i l l even cause us to fight. You see, the w i fe cannot speak to the man direc tly; she can on ly beg before the god she bows down to. The man , on the other hand , when he prays, prays only for work . So once he’s got work , why would he leave that to run o f f and pray?

THE MORAL ITY OF DEBT

A mong slum dwellers’ many failures, according to the middle-class discourse discussed in chapter z, was their tendency to fa ll into a perpetual state o f debt to usurious moneylenders. They were in debt, so the story went, not because o f their inability simply to make ends meet on the exploitative wages they were paid, but because o f their own profligacy. In addition to slum dwellers’ addiction to vices like alcohol, a major source o f their debt, according to middle-class moralists, was their insistence on hosting marriage and other l ife cycle rituals on a scale far beyond their means and wholly inappropriate to their station . Reca ll an offic ial publication o f the T am il N adu Slum Clearance Board , which complained that what lit tle slum dwellers “earn, they spend on drin k ing and tribal festivities, resulting in a life o f eternal poverty stricken conditions” ( T N S C B 1975, 42).

The authors o f this report neglect to provide evidence in support o f their contention, but unsubstantiated does not mean untrue. Given the gravity o f slum dwellers’ overall plight, I took the problem o f indebtedness seriously

9 2 • S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

and spent a considerable amount o f my time in the slum attempting to meas � ure and understand household debt. W ha t I learned was that while indebted � ness was indeed widespread among slum dwellers, household debt was rarely o f ruinous proportions. Thus, although 86 percent o f 347 households sur � veyed were in debt, it is equal ly significant that the rema in ing 14 percent reported having no debt at all. Further, 94 households reported owing only amounts in the range o f Rs. 1 to 5,000, and a majority o f households (219) reported total debt o f less than Rs. 10 ,000 . W h a t these numbers mean in practical terms is that although interest payments represented a drain on the household, and thereby served to exacerbate the effects o f slum poverty, debt could not in most cases be seen as a cause o f poverty in and o f itself. The great majority o f households I surveyed were either gradua lly paying o f f their debt or holding it steady while m aking interest payments. In the approximately 5 percent o f households whose debt had increased over time, this was due not to compounding but to additional borrowing for purposes other than debt service. It was clearly not the case, then, that slum dwellers’ economic straits were the result o f their own mismanagement. But it is sti ll not unreasonable to ask why—given the toll that even low levels o f debt would take on house � hold finances— so many poor people would risk going into debt at all.

To answer this question I conducted a second survey, this time o f the sources o f slum debt. The picture that emerged from this survey, unlike the previous one, did not entirely contradict the elite stereotype. I discovered that roughly a third o f indebted households had indeed gone into debt to finance li fe cycle ritua ls (prim arily marriages and female puberty rites). Surprisingly, not al l life cycle festivals resulted in debt; in the majority o f cases families hosting the event were able to recoup some or al l o f their expenses from the gift money (moy panam) they received from guests. A nother th ird had taken on debt in order to make some k ind o f investment (32 percent o f investments were for small business ventures or for bribes to ga in access to employment, 29 percent to finance their children ’s education, 39 percent to obtain permanent housing). The remainder o f households’ debts had accumulated gradually from the mismatch between their low and unreliable income and both periodic expenses (usually medical) and miscel � laneous dai ly expenditures (food, cooking fuel, rent).

These three basic categories o f debt— festivals, investments, expenses— may be interpreted in a number o f ways. A ccording to the discourse that sought to hold slum dwellers themselves responsible for their own poverty, for instance, debt arising from expenditure on festivals might be described as

S H A R I NG , C A R I NG , A ND S U P E R N A T U R A L A T T A C K “ 9 3

“culpable” debt (because wasteful and unnecessary), whereas debts resulting from either rational investment or the inability to make ends meet might be seen as relatively innocent. Putt ing a more sympathetic spin on mainstream assumptions, even while ma in ta in ing their general form , we might recharac � terize lavish expenditure on life cycle ri tuals as an expression o f slum “cul � ture.” Behavior that is irrational in economic terms is thus val idated as rational by alternative criteria: slum dwellers celebrate because sensual pleas � ures matter more to them than maximization , for instance, or because these functions give their lives meaning or are in some way essential to their cul � tural identity. Slum dwellers themselves, however, accounted for their spend � ing on festivals in somewhat less sanguine terms: they hosted festivals, not because they loved to celebrate (who doesn’t?), but because this was what was expected o f them.

None o f these interpretations, however, seem to afford any real insight into slum debt; for in each case the alleged cause o f slum dwellers’ spending on festivals remains a black box, whether we choose to label it profligacy, “culture,” or simply “what is expected.” The approach I take in the following pages, by contrast, is to consider debt not simply as the outcome o f some prior cause but, by examining in detai l the specific circumstances and context with in which debts were contracted, as a complex social, economic, and moral strategy in its own right.

Preda tory L e n d in g a n d M o ra l Bonds

D ebt is, to a certain extent, an observer’s category. W hile slum dwellers cer � ta in ly understood the concept o f debt— they could answer survey questions regarding it and use it on a da i ly basis in calculating household budgets— they more commonly spoke o f two types o f debts: those contracted with the pawnshops that N orth Indian businessmen had set up on the edge o f the slum, and those contracted with in the slum. Both types o f debt, moreover, came heavily laden with moral judgment: the former intensely negative and the latter strik ingly favorable.

Pawnshops were known as cettu kapai after their owners, who were referred to as Cettus (“ Seths”), which is the common T amil term for M arwari but is applied to all N orth Indian businessmen regardless o f their actual caste or ethnic background, as well as to pawnshop owners even i f they are Tamil. Cettus were described in the most withering terms and were universally hated in the slum. “A l l the money o f the poor eventually goes to them, little by little.

9 4 S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

W ha t we earn, we can never keep, but they earn without working, just sitting in their shops al l day, taking, taking___W herever poor people live they come and set up shop. First one shop, then another, then another—sons o f whores \tevatippaiyanka\ \ ” Debts contracted between slum dwellers—virtual ly always women— on the other hand, were generally styled as “ help” or “aid,” and slum creditors were described in laudatory tones as “ friends.” It is striking that in � slum creditors would be thus valorized, because the rates they charged were normally much higher than the pawnshops’. Pawnshop owners charged 3 to 5 percent per month, which, though usurious by Western standards, was con � siderably less than the 10 percent monthly interest most commonly charged on intraslum loans. W ha t was also peculiar was that intraslum loans could themselves be subdivided into two discrete categories: while most (71 percent) had to be repaid at 10 percent, a minority (29 percent) o f intraslum loans were between related women (classificatory sisters or mother-daughter relations) and cost only o to 2 percent. But whereas the difference between C e t tu loans and “ help” loans received great emphasis, the more radical distinction within “ help” loans went unmentioned in the slum; while debtors were certainly aware o f the range o f possible rates among intraslum loans, they did not emphasize this fact in describing their debts, and these loans were styled as “ help” regardless o f the rate. Those who had received loans at the much lower rate would admit that they were very lucky or that their creditor was unusually kind , and those who were paying o f f debts at the higher rate would— i f asked— admit that their situation was very difficult. But while the difference in the terms o f the loan itse lf was thereby acknowledged, this did not translate into a corresponding difference in the way the creditor was moral ly evaluated.

It is not surprising that C e t tus were so despised— C e t tu pawnbrokers were outsiders who treated slum dwellers with disda in and would not inter � act with them socially, enter their neighborhood, share food with them, and so on. But why were intraslum moneylenders valorized, despite the economic burdens they imposed being higher than those o f the hated Cet tu? M irroring our observations regarding the social distance between C e t tus and slum dwellers, we might explain the tolerance o f intraslum moneylenders on the basis o f something like “community solidarity”— intraslum moneylenders were members o f the same face-to-face community— or even, given that most loans were contracted between women , as evidence o f “gender solidarity.” However, to explain the positive view o f in-group moneylending in such terms is subtly circular. I t treats solidarity as a given rather than as something

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 95

that i tse lf demands explanation; after all, the ideology o f slum solidarity cov � ers over the actual terms o f intragroup exploitation . W h a t I w i l l show is that intraslum moneylending was in fact a key social practice for creating solidar � ity among women.

The critical difference between C e t tu and intraslum loans is to be found not in the pregiven status o f the lender but in the way intraslum loans were used to create a moral relationship. For the very act o f giving an unsecured loan served simultaneously to affirm the borrower not only as trustworthy but also as someone whose authentic need deserved care and, further, to establish the giver o f the loan as one who cared for the needs o f others. Thus although, when asked how they were able to secure a loan , debtors would often explic it ly mention their own trustworthiness, they placed equal or greater weight on the fact that the lender was “a good person” who, recogniz � ing their own genuine and undeserved hardship, was moved to come to their aid. In this way both creditor and debtor garnered considerable rewards from the relationship above and beyond access to cash or profit—which was why slum women frequently borrowed money even when they did not need to. By engaging in such relations, women established themselves not only as reliable partners for future transactions but, more than that, as quintessential ly moral persons under the protection o f others who were also moral. Both partners worked hard to maintain these arrangements, going to great lengths to portray the relation not as an unequal and often mercenary one between creditor and debtor but as a problem-free exchange between a poor woman and her caring friend. The following events w i l l illustrate.

O ne day, while in the middle o f reviewing some field notes, I was startled by a huge commotion at the other end o f my alley. I ran out to find C e lvi , a normally good-humored woman , and K anakam , also one o f my buddies, screaming at each other to the point o f hoarseness, surrounded by a small crowd o f other women inter jecting their own opinions o f who was at fault. I later went to Ce lvi s house to get her side o f the story. Kanakam , it seems, had refused to repay a Rs. zoo loan to C e lv i ’s elder sister Priya, even though, according to Ce lvi , she now had the money to do so. I asked C e lv i why Kanakam would stubbornly refuse to pay, and C e lvi bitterly complained that some people were just like that: “ They have money in their hands but they just can’t bear to part with it. M y sister w i l l never help her again .”

W hen I later asked K anakam about it, she denied that she had the money to repay Priya and furthermore claimed that C e lvi and Priya were demanding she repay the loan wel l before the date they had agreed on previously, simply

9 6 S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

because they wrongly believed she now had the money to repay it. Yet in recounting these facts Kanakam , to my surprise, refused to speak i l l o f either woman . “ [Celvi] is a good woman , but she is m aking a mistake about me. I don’t have the mon ey .. . . I have pa id her al l I have, and I told her I would give her the balance [at the end o f the week]___ I don ’t know why she would accuse me o f simply keeping the money from her— she knows I w i l l pay— it just doesn’t make sense.” T h ink ing perhaps that she was withholding her true feelings because she knew I was friends with Ce lvi , I later sent my assistant, who was we ll known not to be on good terms w ith Ce lvi , to see what he could find out. Yet he too fa i led to elic it any negative comments from Kanakam , who persisted in portraying C e lvi and her sister as good friends who had inexplicably misunderstood her situation. In fact, as I would only later come to appreciate, it was normal in the slum for debtors having prob � lems with their creditors to persist in c la im ing the creditor as a friend and to locate the source o f the problem in a misunderstanding or in an un fortunate twist o f fate. They had not repaid because o f difficulties in their own lives, which the creditor fa i led sufficiently to understand; debtors begged for patience, tolerance, and trust from their creditors but never seemed to con � demn them or their demands as un just, or to blame the terms o f the loan itse lf for their problems.

Two days later C e lvi and her younger sister D ivya appeared at my door. This was a surprise. Because o f a long-simmering quarrel with my assistant Sagayaraj, Ce lvi , who had once been a regular, hadn ’t come to our hut in many months. She seemed shaken, and her sister was visibly cross. Their visit had nothing to do with their elder sister’s loan to K anakam but with the fact, which they had just gotten wind of, that a week or so back C e lv i ’s daughter M a lar and a friend had come to visi t Sagayaraj when he was ly ing il l with a high fever. This, they were convinced, was a cover for some k ind o f romantic tryst. Rumors o f a romance between Sagayaraj and M a lar were the source o f his and C e lvi s original fal ling-out. I had managed to quell the rumors, which were baseless, but C e lv i ’s and Sagayaraj’s relationship never recovered. N ow she was beside herself with the thought that those rumors had been true after all.

In the midst o f this tense standoff—with D ivya shrieking accusations, Sagayaraj meekly protesting his innocence, C e lvi looking as i f she could faint at any moment, and myse lf desperately trying to calm everybody down— who should appear at the door but Kanakam ! She had heard what was going on and spotted in it a singular opportunity. A t the time I was shocked to see

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 9 7

K anakam show up like this, given that she had been avoiding the sisters because o f her unpaid debt. But no one was more astonished to see her than C e lvi and her sister. M y first thought was that she had walked in on us by accident and that she would quickly make herse lf scarce. But K anakam walked in casually and sat down. The sisters, however, couldn ’t say anyth ing about the loan without detracting from the gravity o f the issue with Sagayaraj. For a moment no one said anything at all. Then K anakam turned to me and asked, as i f it were the most natural thing in the world, “Robert, can I borrow a hundred rupees from you?” Eager to get rid o f her as quickly as possible, I agreed, handed her a Rs. 100 note, and she left.

W hy would K anakam do this? D id she intend to immediately use that Rs. 100 to pay the money Priya had lent her? I f so, why not simply get the money from me quietly and at a more convenient time? In fact, however, K anakam did not use the Rs. 100 to repay any part o f her loan but returned it to me the very next day. In all likelihood , she did not actual ly use the Rs. 100 for any � th ing and had merely kept it until she saw me again. But while she may not have needed Rs. 100 per se, what she did need— and very effectively used— was a public demonstration o f the fact that I supported her, that I was her friend and would enact my friendship according to slum norms, namely by giving her a loan to “ help” her. M y loan to her publicly con firmed her as trustworthy and also as deserving o f care— in short, as a worthy part ic ipant in moral community. W ith this one bold move, in the presence o f those very “ friends” who had accused her o f dissimulating with respect to her own weal th— an accusation that, i f true, would make her undeserving o f the protections and advantages o f partak ing in the slum ’s economic-moral networks— K anakam reversed her fortunes and reclaimed her moral stand � ing. I never heard anyth ing else about troubles between K anakam and Priya and C e lvi , and in the end she pa id o f f her loan on the date she had original ly promised.

N ot al l women were as adept as K anakam , however, and when their debts were large and went unpaid the consequences could be very serious. In no case, however, was repayment ever enforced through threats o f vio � lence. The moral force o f intraslum debt was such that, according to in form � ants, debtors always eventual ly repaid the ir loans, and it was unheard o f for anyone simply to refuse to do so. W hile I encountered some circumstan � tial evidence o f debtor families who had fled or gone into hiding, this was not the same th ing as bucking the moral force o f debt— it was cutt ing oneself o ff entirely from the moral community that generated that force. Some

9 8 • S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

women were unable to make such a radical break. But they also could not face the shame o f public censure by angry creditors day in and day out. A nd though loans were taken on beh a lf o f their entire household and in consulta � t ion with their husbands, it was wives who had to bear the censure and moral harassment—for they were the ones who had contracted the debt in the first place, and who were at home throughout the day to endure creditors’ scold � ing. The shame and public humiliation that accompanied soured debt rela � t ions was cited by slum women as one o f the two most common causes o f female suicide, the other being spousal con flict. Indeed, the two causes were inseparable. A s slum women unanimously attested, the frequency o f fighting between a husband and wife was directly proportional to the degree o f their financia l troubles. The two principa l relations that belied the picture o f an undifferentiated, caring slum “ humanity”— the marita l relation and the rela � t ion between debtor and creditor— could thus at times combine to make the moral pressures upon married women unbearable.

D ebt as Savings

Few slum women were only debtors or creditors. A lthough some women were, on balance, earning more in interest than they were paying out to oth � ers, and were owed far more than they owed to others, it was largely impos � sible to tell for sure who was who. For intraslum debt was organized over complex networks and by means o f financ ia l devices that appeared as i f designed to obscure these very details. A part from creating complex and ambiguous networks o f relations among women , a second key feature o f intraslum moneylending practices was their tendency to prolong relations (temporally) and to require frequent interact ion between borrowers and lenders.

O ne paradigmatic example o f intraslum debt relations was the chit fund (cittu kulu, li teral ly “chit c lub”). This popular savings and loan scheme is found al l over India, and some chit funds have thousands o f participants and professional managers, but in the slum they typica lly had fewer than a dozen members. The rules according to which chit funds operate can vary some � what, but in the slum the basic structure was as follows. A group of, say, ten women would agree to meet each month and make a fixed contribution to a common pool, which they then “auctioned” among themselves. Suppose, for example, that the ten women in the group had agreed to a maximum monthly contribution o f Rs. 300 per person. This meant that a maximum monthly

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 9 9

total o f Rs. 3,000 (= Rs. 300 x 10) was awarded to the highest “ bidder.” In fact, she did not get the fu ll amount because her w inn ing bid was subtracted, or “ discounted,” from that total. So i f the w inn ing bid was Rs. 900, then the total pool o f Rs. 3,000 is “ discounted” by Rs. 900, leaving an actual award o f Rs. 2,100. In this case, the other nine women had to each contribute Rs. 210 that month (= Rs. 2,100/10) rather than the maximum Rs. 300, while the w inn ing bidder would wa lk home Rs. 1,890 richer (= Rs. 2,100 minus her own contribution o f Rs. 210). Each woman was eligible to take the pool only once, so, as the number o f rema ining bidders decreased, the tendency was for the w inn ing bid to get smaller and smaller and for the award to get closer and closer to the maximum . H ie last woman collected the fu ll pot , minus her contribution , for a guaranteed profit equivalent to the average o f the previous winn ing bids. She became in effect a moneylender. The other women became either lenders or borrowers, depending on how we ll they had played the game, their status indeterminate ti l l the penultimate round.

The attractions o f chit clubs were threefold. First, they provided a source o f quick money for women who might otherwise have difficulties saving on their own or obtaining a loan. Second, the element o f chance, strategy, and the potential for profit made involvement in a chit club not only an exciting game to play but a monthly social event requiring trust and cooperation among an extended group over a fairly long time span. Third, chit clubs blurred the dis � t inction between savings and debt; their temporal structure was such as to make it impossible to say in advance who would come out as a net debtor and who a net creditor. These three basic features— economic advantage, the rein � forcement o f socio-moral bonds, and the blurring o f lines between lenders and borrowers—were to some extent found in all intraslum economic arrange � ments, though the precise means through which they were achieved varied.

O ther savings schemes were more straightforward: one woman would make regular contributions into an account held by another woman that was earmarked for some specific purchase. Rani , for example, managed “savings accounts” for five to ten other women , collecting small-change deposits from them on a semiweekly basis. A t some point—which she herself determined— she would then use the money a woman had contributed to purchase the item requested, usually some piece o f jewelry, item o f clothing, or kitchen imple � ment. D epending on the contributor’s creditworthiness, this purchase could come at the point when the full price o f the item had been collected, at the ha lfway point, or even at the very outset. Rani , for her part, deducted a small rate o f interest from the other women ’s contributions, usual ly between 1 and

I O O S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

5 percent, thus making a modest profit for her service. In addition , she was able to use whatever cash she had on hand to make small interest-bearing loans to other women or interest-free loans to herself. W h a t is interesting about R a n i— in this she was typica l— is that in addition to being a very small-scale moneylender, she was herse lf perpetual ly in debt to others. A t the time I first learned o f her small savings schemes, for instance, she was in hock to a local pawnshop for Rs. 6,000.

To reiterate, most slum women were involved in multiple schemes, with the same woman borrowing on one side while lending on another. The com � plexity o f financ ia l interrelations l in k ing multiple women in different ways made it impossible to sort out whether an individual woman was a net credi � tor or debtor. W hile some women were evidently making a business out o f moneylending, and others were clearly in overall debt (to caring “ friends” or the Cet tu), most womens posi tions on the borrower-to-lender continuum were uncerta in even to themselves and were completely opaque to others.

The economic and socio-moral advantages o f women’s financial schemes are clear. But why the pervasive and systematic obscuring o f distinction between net gainers and losers? The aggregate opacity o f these arrangements certainly benefited the minority o f women who were consistently coming out ahead and profiting at the expense o f their less fortunate counterparts. To be seen as taking advantage o f others’ misfortune would contravene the most basic moral principle o f slum community, which might create problems for such women even if, for reasons already noted, their debtors continued to praise them as “good” and “caring” friends. But the ambiguities created by intraslum lending arrangements were extremely useful even to those who did not profit directly or regularly from these arrangements. Such women had their own reasons for keeping the details o f their economic status a secret—not only from other women but, perhaps more importantly, from their own husbands.

Evasion as F in ance

Slum women , though intensely interested in savings, found it almost impos � sible to keep cash in their own houses. First, there was the fact that a woman who was believed to have any significant cash surplus would be besieged by requests for help from neighbors and friends. A n d to refuse such requests would not only be to buck one’s most basic moral responsibility to others but also to court vankandvi, the spirit o f malicious desire or “evil eye.” A more immediate danger in keeping loose cash about the house, however, was that it

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • I O I

was vulnerable to being appropriated by the woman s husband. Except in those relatively rare cases where the management o f household finances was a cooperative enterprise involving both husband and wife equally, the man o f the house was liable to perceive any money that had not been specifically ear � marked for household maintenance as a surplus to which he— as the principal breadwinner—was natura lly entitled. The following example will illustrate.

Meena was sixteen years old when I first met her. She told me that she had recently fai led out o f the ninth grade, but she later admitted that she had been forced to leave school because her mother had just taken a job cleaning middle-class homes many miles away and therefore needed M eena to do household tasks. H er mother, in turn , had sought this work because her hus � band was drink ing away most o f his earnings. Smart and ambitious, M eena soon followed in her mother’s footsteps, tak ing a job o f her own: making cotton swabs in a small slum-based factory that paid teenage girls a piece rate. Meena took home an average o f th ir ty rupees (sixty-six cents) per day, after some ten hours o f work . She turned the majority o f her earnings over to her mother to help with household expenses, carefully saving the remainder with the idea o f eventually buying herself a pa ir o f blue jeans. Though considered too risqué and fancy to be actual ly worn in the slum, jeans were a prestige item that young women like M eena al l dreamed o f one day owning.

O n one occasion her father came home from work late and, quite drunk , woke up his wife to demand money for more booze. The mother, having noth � ing to give, woke up her daughter and ordered her to hand over the sum her father demanded from her personal savings. Meena, not yet familiar with the tactics slum women used to retain surplus cash, had foolishly hidden her small savings in the rafters o f her own house and was eventually forced, after a nasty fight with her mother, to hand over a large portion o f it. I learned all this the next day from Meena, who, though still furious at her mother, recognized that her mother had only acted as she had in order to spare Meena from what would have been a far worse fight with her father. W ithin a matter o f days Meena had prudently lent what remained o f her savings to a friend from the factory, who, borrowing money from several other friends as well, had bought herself a pair o f blue jeans; a month later Meena collected her debt and, after borrowing a bit more from another friend, went out and bought a pa ir o f jeans as well.

W hile Meena’s father indeed behaved brutishly, the idea that a man was entitled to unused cash after a hard day o f work was well accepted in the slum, and even the gentlest o f men would feel this his due every once in a while. But cash kept in another womans house was sacrosanct: whatever cash a women

IOZ • S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

was keeping for her friends was absolutely o f f limits to her own husband, for appropriating such monies would be tantamount to stealing from a neighbor.

O f L ove a n d G o ld

Besides deposit ing one’s savings with neighbors in the form o f loans, another way women put savings beyond the reach o f husbands was by turn ing them into gold. Though a woman ’s gold might have been bought w ith money her husband origina l ly brought home, men had no cultural ly recognized rights to it. O nce any wife accumulated significant savings she could buy a piece o f jewelry. W ith price tags start inga t Rs. 150 for a simple nose ring, gold jewe lry allowed for almost any increment o f wea lth to be conveniently stored. Gold was, moreover, readily liquidated , thanks to the pawnbroker. W h ile it was not a husband ’s to take, he too benefited from this arrangement, for it was understood that i f the household was in need, the woman would pawn her gold for the family: women were usua lly amenable to discretionary spending proposed by their husbands, for purposes such as home improvement, al though it was up to the husband to make a persuasive case.

Investment in gold shared many advantages with small savings schemes, though unlike the latter it did not simultaneously mediate social relations with unrelated women . I t did, however, serve to renew and augment relations among female k in . Gold was often pawned to meet large household expenses w ith in women ’s own immediate families, or to finance the weddings o f their sisters and sisters’ daughters. O ccasionally gold was pawned to pay for the feast, but more often it was given directly to their nieces, sisters, and other female k in as part o f their dr, or “ dowry.” In the slum a dr, which is also a common word for dowry among Tamils in general, differed from most dow � ries in T amil N adu in that it remained exclusively the property o f the bride or became the joint property o f the married couple, to be disposed o f only by mutual consent. Even in cases where the d r was treated as join t property, in practice it remained under the w ife ’s control. For it prim arily consisted o f household implements, which had lit tle financia l value and which the wife would use on a da i ly basis, and jewelry, which invariably belonged to the woman alone— although, as mentioned, she would be expected to pawn it i f the family were in desperate straits.

The practice o f giving d r was on the rise in the slum, and this related directly to both debt and the marita l relation . In the Indianist literature the prevalence o f dowry is often assumed to correlate with patriarchal power, and

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 1 0 3

the shift from brideprice to dowry that has recently taken place among the Paraiyars and some non-Brahmin castes is seen as evidence o f Sanskritization , also closely associated with patriarchy (Gough 1956, 845; 1979, 41; Mencher 1972, 46 ; K apadia 1996, 15, 53, 6iff) . Two facts, however, undermine the applicability o f this interpretation in the slum context. First, as noted, the d r remained—not just in theory but also in pract ice— the property o f the bride. W hile any wife could be counted upon to dispose o f this weal th for the ben � efit o f her husband and children, it was she and she alone who decided how and when to do so. Second, upwards o f 90 percent o f marriages in the slum were “ love marriages”— that is, marriages decided upon by the couple them � selves rather than their parents and other senior family members. The rise o f love marriages meant that the groom was rarely matrilateral ly related to the bride, as was common in arranged marriages in T amil N adu , which followed the Dravidian norm o f cross-cousin marriage (Trautmann 1981). This meant that in marital disputes the young wife could normal ly count on the fu ll support o f her classificatory mothers and sisters without there being any chance o f divided loyalties on their part, and groups o f female matrilateral k in thus remained powerful allies throughout a married woman ’s life, to be called upon for support, inc luding financia l support, in times o f difficulty. Such ties were o f course like any others in that they had to be actively main � tained, and one way they were maintained was through dr. Thus C e lvi had at one point pawned al l her jewelry to help finance her younger sister D ivya ’s wedding. This was not a mat ter o f duty or tradit iona l obligat ion , she explained; “You see, i f I help with her dr, when it comes time for M a lar [Celvi s daughter] to get married my sister w i l l have to help her.” Thus invest � ment in gold jewe lry allowed matrilateral ly related women , by spending on female k in , to make a prudent direct investment in this important base o f mutual support. Purchases o f gold, like the chit clubs and savings schemes considered above, obscured the relations between debtors and creditors and offered distinct economic advantages: al l three promoted and reinforced intense affective as well as economic bonds.

AUSPICIOUS W I V E S AND THE SPIR IT

OF MAL IC IOUS DESIRE

I have described two unequal relations within the slum— those between mon � eylender and debtor and between husband and w ife— that escaped condem �

I O 4 S H A R I NG , C A R I NG , A ND S U P E R N A T U R A L A T T A C K

nation despite violating the core precept o f slum morality, an ethic o f selfless care and love for an undifferentiated humanity. Rather than acknowledging these contradictions, the terms o f moral community were such as to displace ultimate responsibility onto the individual wife. W e have already seen one way this displacement was achieved, through the discourse on wife ly auspicious � ness. A second discourse sought to shift blame from the un fortunate wife herse lf onto her even less fortunate peers, who were said to harm other women both inadvertently, with vankandvi (the “spirit o f harm ful desire,” or “evil eye”), and deliberately, with black magic {ceyvinaiydvi). A lthough such attri � butions for misfortune might garner more sympathy for the individual female victim , in aggregate they served to atomize slum women morally, while also reinforcing the tendency to blame them for misfortunes beyond their personal control. Though men might also unleash vankandvi or ceyvinaiydvi on others, these malicious supernatural forces were assumed in practice to come prima � r i ly from women. A third form o f supernatural harm was likewise female in origin , but it did not direct ly pit (living) women against one another. The source o f harm here was the spirits o f dead slum dwellers (dvi; pi. dvik a l), spirits that attacked primarily fortunate women or their husbands, children, and babies. Significantly, these spirits were also almost always female: women who had died unhappy, with the spirits o f female suicides being the most dangerous o f all. In sum, the root o f most supernatural evil was female envy. Women, it seemed, were prone to dangerous envy and were easily provoked by the sight o f another woman ’s prosperity or happiness.

The symptoms o f these supernatural attacks are worth noting. They might cause delirium and possession in the vict im , but more often they resulted in quotidian misfortune for a woman or her family: sickness or death, the loss o f a job, the inability to get a loan , the failure to get pregnant, and so on— the very same misfortunes that might otherwise be attributed to a lack o f personal auspiciousness. Significantly, one o f the most common indications o f super � natural attack was marital trouble: the husband and wife started to fight incessantly, or he would begin to drin k to excess, not work enough, or take up with another woman . The effect o f a woman ’s recourse to “supernatural at tack” when expla in ing her misfortune was, again, to deny that there was anything inauspicious about herself. H er misfortune was recoded as proof o f her own innate auspiciousness— for only a fortunate woman would provoke the envy o f other women.

In shift ing the blame o f f themselves in this way, women nevertheless con � tinued to avoid implicat ing either o f the most immediate sources o f their

S H A R I NG , C A R I NG , A ND S U P E R N A T U R A L A T T A C K » 1 0 5

suffering: husbands and moneylenders. Thus while the discourse on vankandvi and ceyvinaiydvi was in some sense the opposite o f the discourse on auspicious � ness, the two functioned together in diverting moral condemnation away from intraslum domination and onto forces that were, by their very nature, mysterious and resistant to analysis. Both, moreover, served to atomize women. Where auspiciousness did so by making women individually respon� sible for their family’s welfare, slum women’s constant talk o f vankandvi and ceyvinaiydvi sowed suspicion among women that other woman were secretly their enemies.

Fear o f vankandvi and other, mostly female forms o f supernatural attack had one positive effect. It ensured that women were indeed generous with one another, since one o f the best ways to avoid such attacks was to act always in a selfless manner, or at least to appear to do so. Another strategy for avoiding vankandvi was to downplay or disguise one’s own good fortune, including by constantly nagging one another for loans and other more quotidian forms o f help, even when these were not needed. But because all women employed these strategies, the net effect was to make everyone feel put upon and besieged by requests. A kind o f feedback loop took hold, reinforcing public affirmation o f the slum ethos o f care. Women in Anbu N agar frequently wove references to their own generosity and lack o f interest in wealth and savings into everyday conversation. They also affirmed the same about others. To give but one very typical example, Celvi once averred, “My sister Priya is very generous and shares with everyone, just like my mother. I have that same nature too__ Whatever I have, I soon give it away or share with others___ I could never become rich, because I am always sharing—I don’t even know how to save money.” O n another occasion she explained that she always had neighbors who were lacking a pot or a knife or something and would come and “ borrow” one o f hers. She would never say no, she explained, and she never asked for anything to be returned. A l l her wedding gifts had been lost in this way. She didn’t even remember where any o f the items were, she told me, and didn’t care. The only piece she could locate was a rice scoop (,annavetti), which had been with her sister-in-law for the last twelve years. A fter telling me this, she dramatically summoned her daughter Malar and told her to go next door to fetch the item to show me. Malar returned moments later with the rice scoop, which she handed to her mother, who then handed it to me, pointing out that her name had been etched into the back o f the handle. A fter inspecting it carefully, I returned it to Celvi, who in turn gave it to Malar, telling her to put it back where she had found it. “But why don’t you

1 0 6 S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

keep it?” I asked. “ Or, i f you know exactly where it is, why don’t you simply ask for it back?”

“ I f she wants to return it to me, I will accept it. But that is up to her—I will never ask for it. That is what I am like, Robert. W hat people want, they can have. I don’t care about things.” Needless to say, however, “things” were in fact very important in the moral life o f slum dwellers in general, and even more so when lent—as the case o f C e lvi ’s rice scoop, worth around twenty rupees but remembered all those years, makes plain. Celvi herself, for all her undeniable generosity, knew very well how to count and save money and not only participated in but was a principal organizer o f numerous savings schemes that provided her a small but steady income on the side.

The kinds o f careful misrepresentation o f the sort Celvi offered were described by Indira as ubiquitous in the slum. One day Indira asked me how much my graduate student stipend was. I replied to her honestly, as was my policy with those I knew. The sum was huge by slum standards, however, and I was thus inviting both va.nkana.vi and, i f word were to get out, a barrage o f requests for assistance, loans, et cetera. Indira was shocked, exclaiming, “You never lie, do you, Robert?”

“ How do you know?” I asked her. “Maybe I am lying to you all the time.” “Robert,” she continued, ignoring my question, “around here it is a ll lies!

H a l f o f everything that people say is lies.” I asked her what everyone lied about so much. “ Money, Robert,” she said laughing, “ Money!” “I would even lie to my own father,” she continued. “I f I were carrying ten rupees, I would say I only had five.” Not long afterwards, Indira demonstrated to me how adept she was at the kinds o f misdirection she was speaking about. I came into her house one morning and was surprised to find a large color television dominating their small front room. Indira’s house, not much bigger than mine (i.e., around one hundred square feet), was divided into two rooms that housed Indira, her husband, and a teenage son and daughter. H er husband, she said, had purchased it on installment for an initial payment o f Rs. 5,000, with twelve monthly payments o f Rs. 800 apiece to follow. The total cost o f the television was thus Rs. 14 ,600, Indira lamented, much more than the Rs. i2,,ooo list price. She was sure her husband, whom she regarded as foolish with money and constantly getting cheated, had not negotiated the best pos � sible deal. But despite her misgivings she was clearly immensely pleased with their new possession. I admired the television, which went over well, and we had been sitting and talk ing about the television for about twenty minutes when a neighbor came to the door and saw it for the first time. In the

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 1 0 7

presence o f this neighbor Indira’s tone and affect changed instantly. “ I don’t know how we can possibly pay for it---- We are already in debt and he hasn’t worked all week----Such a ridiculous television! It ’s far too big for a house o f our size; there’s barely any room to move around and I don’t know how we’ ll all sleep now. A television like this isn’t meant for us; it should be in-a mansion like the one Murugan lives in!” Murugan worked for the ruling political party, the A I A D M K , and was known as the richest man in Anbu Nagar. H is two-room house was no mansion, though it was well built and unique among slum dwellings because the ground floor was raised two or three feet above street level to prevent flooding during the rainy season.

The ubiquitous slum habits o f misdirection, outright lying, and downplay� ing with regard to one’s wealth meant that while slum dwellers portrayed themselves collectively as good people who cared about others, they were well aware that not everyone in the slum lived up to this ideal. In fact, accusations, especially among slum women, that others were not adequately sharing what they had was a more or less constant feature o f conversation, as was the charge that others had more money than they admitted to. Yet these practices were deemed necessary for countering the ever-present dangers o f other women’s envy.

A N TA G O N IS TI C PROCESSES IN THE REPR OD U CT IO N OF MORAL C O M M U N I T Y

This chapter has explored the key social practices and financial schemes by which moral community was organized and reproduced in Anbu Nagar. Let us now try to bring them all together into a single picture.

First, consider that insofar as the discourses o f vankandvi and auspicious � ness effectively atomized women by locating blame in individuals, they might seem contrary to the ethic o f shared responsibility that defined moral com� munity in Anbu Nagar. Yet talk o f vankandvi and auspiciousness was also, paradoxically, among the means by which care was reproduced and main � tained. Generosity, for example, was directly and indirectly enforced by fears o f vankandvi. The constant buzz about vankanavi, ceyvinaiydvi, and various desirous ghosts reminded would-be hoarders that the social sanctions on such a strategy could easily combine with more mysterious ones. Even as these supernatural threats atomized women in one way, they also reinforced the sharing and mutual reliance that brought women together. Likewise,

1 0 8 • S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

female auspiciousness, while deflecting blame for household misery from men or creditors onto debtor-wives, was a central theme in the two most important life cycle rituals, marriage and girls’ puberty rites, which were in turn productive o f moral community. These rituals were the concrete acts by which women were invested with auspiciousness, and at the same time they reinforced the social ties that tided families over in periods o f hardship— both generalized social ties with neighbors and specific, intensive ties between matrilaterally related women—by means o f debt-producing expend� iture on the feast and the d r respectively.

The relationships o f debt and marriage that together most seriously under � mined slum dwellers’ vision o f their own undifferentiated humanity were also instrumental in the reproduction o f moral community. Debts, as noted, were o f two types, one involving a relation that was itself considered moral, and the other with the despised Cettu . Despite being condemned, however, the Cet tu was essential to the material basis o f women’s auspiciousness—her wealth in the form o f gold, whether this was bestowed upon her by her family as part o f her d r or accrued from her own gradual savings. W ithout the pawnbroker this gold would remain an inert object, entirely useless to famil � ial survival. The two antithetical forms o f debt, moreover, were in constant interplay. Women would often pawn their gold to get quick cash for an immediate need and then later, when they had managed to negotiate some kind o f intraslum loan from a “ friend,” would retrieve that gold. Conversely, when such an intraslum relationship was starting to go sour, a woman might try to rectify it, either by pawning her own gold or—as was often the case— enlisting other friends or female kin to go together and pawn their gold in order to get her out o f trouble with the original “ friend.” One type o f rela � tionship could thus be used to renegotiate the terms o f another one, both o f which were mediated by debt and transformed in the process. Indeed, many intraslum loans themselves originated when a woman without any gold o f her own persuaded a friend to help her by pawning her own gold and then lend� ing her the cash at a higher rate than what the owner o f the gold herself had to pay to the Cettu .

Finally, as we have seen, the husband’s relationship to the wife was at odds with the ideals o f moral community, inasmuch as neither wealth nor suffering was properly shared. Yet it was only because he labored for long hours under punishing conditions that the cash that was necessary to maintain the family, and thus the reputation o f the wife, that was indirectly shared with neighbors through various quotidian exchanges, and that was also gradually

S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K • 1 0 9

accumulated by women to be invested in gold was brought into the slum in the first place. Here, we might say that the dependence o f wife and family on the husband paralleled and depended upon the husband’s exploitation by his nonslum employers. Men’s work—their subordination outside the slum— made possible the intraslum enactment o f an ethic o f mutual sharing, undifferentiated humanity, and limitless care. It also made possible the forms o f intraslum domination—debt and marriage—that this chapter has examined.

In highlighting the ways that interlocking supernatural and micropoliti � cal features o f slum moral life enabled one another, I emphatically do not mean to suggest that they all necessarily “worked” together in some func � tional sense. These features were too recent in origin, were too contingent, and exerted too heavy a toll in human suffering—particularly on women—to be called functional. W hat I have described was the result o f a complex inter � play o f forces that depended, inter alia, on the lack o f independent economic opportunities for women, on the fact that men were themselves brutally exploited and multiply dominated, and on pawnshops charging an illegal rate o f interest while grossly undervaluing women’s gold. A change in any o f these facts could completely destabilize the total configuration I have described.

The particular configuration o f moral community also depended on the absence o f any competing cosmology to the one that trapped women between auspiciousness and envy. But when more than one cosmological system became available, as happened with the spread o f Christianity within the slum, the possibility arose for slum women and others to begin to reformu� late the terms and practices o f moral community on this alternative authori � tative basis. The nature o f the new social and discursive practices introduced by Pentecostal Christianity, and the political-economic constraints they had to continue to work within, are topics to which we soon turn.

I I O S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

F O U R

Religion, Conversion, and the National Frame

C h r i s t i a n s h a v e b e e n L I V I N G c o n t i n u o u s l y in India for longer than in much o f northern Europe. The widespread embrace o f Christianity by India’s Dalits and its so-called tribals (another “outsider” population) is a more recent development, however, dating only to the nineteenth century. It is also a relatively small-scale phenomenon in absolute terms. Christians, after all, constitute just z.6 percent o f the national population, and while a majority o f India’s Christians are Dalits or tribals, only a small minority o f Dalits and tribals (outside the northeast) are Christian .1 Contrary to popular Western stereotypes, however, India is not a H indu nation. It is a constitu � tionally secular nation-state with no official religion. Yet the conversion o f Dalits to either Christianity or Islam—both o f which are commonly por � trayed as “ foreign” religions—is treated in India as a matter o f vital national concern. W hen 180 Dali t families converted en masse to Islam in 1981, in the remote Tamil village o f Meenakshipuram, it provoked a nationwide uproar, bordering on moral panic. The sense o f threat to the nation itself was such that a steady stream o f politicians and journalists from all over India— including the prime minister herself, Indira Gandhi—held face-to-face meet � ings with the converts to inquire as to why this had occurred and what could be done to prevent it from happening again.

The issue o f conversion flared up dramatically again in early 1999, ostensi � bly in response to reports o f tribals converting to Christianity in the Dangs region o f Gujarat, though conversions had been ongoing there for decades. Aga in the media response was intense. “Are the conversions for real?” blared the headline o f a special report in the Sunday Times, India’s largest circulat � ing newspaper. The lede went on to imply that the entire nation was in an uproar. It is impossible to tell whether this accurately represented the

I I I

reaction o f most Indians or merely the state o f emotions in the editorial board, as the only evidence o f popular obsession we have is this and other newspapers’ headlines. H aving hyped the public outcry over conversion, however, the Sunday Times concluded by claiming to offer a neutral, detached perspective on the issue from above the fray:

The def i l ing o f cricket pitches, k i l ler stampedes at Sabarima la , the A n jan a M ishra rape case— none o f these compel l ing new concerns o f the past week have succeeded in dislodging the conversions controversy from public con � sciousness. The H in du right continues to denounce the b lowing up o f the issue as a “ W estern plot ,” and poin ts to the Sonia [Gandhi] fac tor as the reason for the C hrist ians’ new muscle-flexing___ H ow va lid are the charges against missionaries? W h a t makes a person break that most prim eva l o f bonds and bow to another God? A n d on the eve o f the 21st century, should the nation be torn asunder by bigotry? The Su nday T imes takes a dispassion � ate look at an issue whic h could explode in the face o f the nation . (January

17. 1999)

Coverage o f this nature did not emerge from a vacuum. It followed upon and amplified several months o f drum beating by H indu nationalists over allega � tions that poor tribals were being “ forcibly converted” in the Dangs region o f Gujarat and that “ foreign money” was being used across India to lure con � verts away from Hinduism.

Just five days after the Sunday Times special report, Australian missionary Graham Staines and his two young sons were burned alive in their family’s station wagon by H indu nationalists in Orissa. There was no evidence at the time that Staines, a medical missionary who ran a leprosy home, was actually engaged in conversion activities. Nevertheless, news reports soon began to appear hinting, without evidence, that “conversions may have been behind Staines killing” {H indu, May 21,1999). India’s Central Bureau o f Investigation (CBI) cleared Staines o f this charge (“Staines” 2003), but the allegation con � tinued to be repeated as fact, including by the Supreme Court. Later that year a prominent H indu nationalist leader, Swami Dayananda Saraswati, pub � lished an editorial, which has since been reprinted countless times and expanded into a small book, arguing that religious conversion is itself a form o f violence (Saraswati 1999; see Roberts 2012a). H istorian Sumit Sarkar recalls that even “well-intentioned and progressive” Indians at the time seemed to regard the murder as an understandable, i f not justifiable, result o f Staines’s own alleged conversion activities (quoted in Adcock 2014 , 176 n. 4).

I l l • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

The primary accused in the Staines murder, Dara Singh, a member o f the Bajrang Dal, a H indu nationalist group, was eventually found guilty and given the death sentence in May 2005. Gladys Staines, mother o f the murdered boys and widow o f Graham, pleaded for leniency. She expressed her opposition to the death penalty on moral and religious grounds and added she had already forgiven the man who had taken her family from her (Vasudevan 2011). In January 2011 the Supreme Court commuted Singh’s sentence to life imprison � ment, in a judgment endorsing the H indu nationalist position that Staines had been indeed converting people and that conversion itself was the ultimate cause o f his murder. The Court also took the opportunity to insert legally gratuitous editorializing on the illegitimacy o f the religious conversion: “It is undisputed that there is no justification for interfering in someone’s belief by way o f . . . conversion . . . or upon a flawed premise that one religion is better than the other” and to make clear that it regarded religious conversion as contrary to the national interest o f “bringing India’s numerous religions and communities into an integrated prosperous nation” (Supreme Court o f India 2011, j$).2

The issue o f conversion continued to simmer in the media, with stories appearing intermittently in English and Tamil newspapers, until October 5, 2002. O n that day an anticonversion law was announced in the state o f Tamil Nadu, on the heels o f a politically embarrassing news report alleging that a group o f Dalits had converted to Christ ianity in the constituency o f the state’s chief minister, Jaya la l i tha (“ Mutalvar Jeya la l i ta Tokutiyil ‘Matamarram’ ” 2002). The Tamil Nadu anticonversion law, which remained in effect throughout most o f my time in the slum, effectively banned the sort o f conversions taking place all around me.3 Ostensibly these laws ban only conversions based on “ force,” “ fraud,” and “allurement,” but, as we will see, these terms are defined so broadly as to render virtual ly any conversion to Christianity or Islam illegal. This was the fifth such law, all containing nearly identical language, to have been passed by Indian states since 1967. These were the Orissa Freedom o f Religion A c t in 1967; the Madhya Pradesh Freedom o f Religion A c t in 1968; the Arunachal Pradesh Freedom o f Religion A c t in 1978; the Chhattisgarh Freedom o f Religion Act in 1968; and, since then, the Gujarat Freedom o f Religion Act 2003, passed in the wake o f an anti-Muslim pogrom in which the government o f Gujarat under C h ie f Minister Narendra Modi has been implicated (Human Rights Watch 2002; Khetan 2007). The Rajasthan Freedom o f Religion Bill (2006) was passed by the state Legislative Assembly but still awaits presidential approval (SA H R D C 2008).

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • I I 3

A N ELITE CONSENSUS

Scholarly literature and public commentary on these laws explain them as an outcome o f H indu majoritarianism—or “ Hindutva,” as it is known—which holds that India is an essentially H indu nation and that the Constitution and state ought to reflect this fact. H indu majoritarian organizations, known col � lectively as the Sangh Parivar (Basu et al. 1993; Jaffrelot 1996), have indeed been instrumental in promoting anticonversion legislation (Sarkar 2007).4 The circumstances o f their passage and the uses to which they are put, further � more, make plain that they are not designed to prevent conversion as such but target specifically conversion out o f H induism by Dalits and tribals.

To explain these laws in terms o f H indutva, however, rests on a hidden instrumentalism that fails to ask deeper questions about the broader politi � cal, moral, and epistemic context. It fails to account for the contents o f the laws themselves, the governmental rationality they embody, and the logic upon which they were found by the Supreme Court to be entirely consistent with India’s constitution; reducing India’s anticonversion laws to H indu majoritarian ideology short-circuits analysis o f something rather more com� plex and interesting. For these laws are not derived from H indutva but rest on assumptions that are entirely secular in character—about the nature o f religion itself and about its relationship to morality, to family life and com� munity, and to the self. A nd these assumptions, I will show, are passively accepted across the spectrum o f national elite opinion, including by those formally opposed to H indutva ideology and to anticonversion legislation.

Officially, India’s anticonversion laws are presented not as a defense o f H induism but as a necessary measure for maintaining public order, for pro � tecting the “family,” and for safeguarding the interests o f subpopulations the state construes as less capable o f making their own decisions: Dalits, tribals, women, and minors. W hat these laws ban, moreover, is not religious conver � sion per se but only conversions that are not genuinely religious because they were undertaken for the wrong reasons. These are ones in which converts have not undergone a true change o f heart because they have been forced, tricked, or induced to convert by the promise o f material benefits: “ No per � son shall convert or attempt to convert, either directly or otherwise, any person from one religion to another by the use o f force or by allurement or by any fraudulent means.” The explicit banning o f conversions is unusual in modern democratic states, but the proscription on coercive force by nonstate agents and on fraud is entirely consistent with modern legal norms.

1 1 4 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

W hile no one would deny that coercion and fraud are illegitimate in any sphere o f activity, not just conversion, Indian critics o f anticonversion legisla � tion have rightly objected to the particular way force is defined in these— namely, as including “the threat o f divine displeasure”—and to the fact that it is unclear what “ fraud” would mean in the religious domain. To define

forcible conversion as extending to the “threat” o f divine displeasure in effect prohibits members o f those religions from freely propagating the tenets o f their faith, a practice Artic le 15(1) o f the Constitution o f India explicitly protects. For many Christians and Muslims, one such tenet is that their god is displeased when people do not worship him. To ban fraud in religious matters is likewise problematic because it puts police and judges in the posi � tion o f deciding which religious teachings are true and which are not (SA H R D C 2008). Critics have also objected that by passing such laws the state lends a spurious credibility to rumors that Indians really are being forced, in the ordinary sense o f the term, to change religions and that con � verts are being in some way defrauded. States that have passed these laws are thus rightly accused o f participating in the demonization o f India’s vulner � able religious minorities.

These are important objections. But what is more interesting, from the point o f view o f this book, is what critics o f these laws have overlooked. It is here that we begin to discern the outlines o f a shared common sense about the role o f religion in national life that unites secular left liberals with their H indu majoritarian opponents. First, few critical voices have questioned why conversions brought about “by allurement” should be banned.5 W hile one can readily grasp a secular state’s inherent interest in curtailing force and fraud, it is unclear why it should insist on converts’ motives being spiritually pure. In legislating on this basis, the state is in fact adopting a particular theological position, one that says that some forms o f religion are legitimate and others are not. The theological underpinnings o f India’s anticonversion laws may be glimpsed even in the language o f the laws themselves, all o f which passed since 1968 specifically outlaw what they describe as “temptation in the form o f . . . material benefit” (my emphasis).6 Apart from anticonver � sion legislation, Indian laws governing religion do not in any other case man� date that religion must be a purely spiritual affair, nor do they otherwise exclude from constitutional protection religious practices undertaken in hope o f worldly gain. This is a significant asymmetry, insofar as it imposes restrictions on converting religions that are not applied elsewhere. This is not the only such asymmetry. Although conversions brought about by the “threat

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • I I 5

o f divine displeasure” are banned, the use o f such threats to prevent people from converting is not only permitted under law but openly practiced. One o f H induism’s most prominent spiritual leaders, Jayendra Saraswati, the Kanchi Sankaracharya, has warned that conversion to Christianity can lead to droughts and crop failure (“ N atu muluvatum” 1999). A nd as we will see, Gandhi himself declared that conversion was a “ blasphemy against God” and thus presumably displeasing to H im .

A second feature o f India’s anticonversion laws that goes largely unchal � lenged is the idea that religious conversion is in some way inherently socially disruptive. Those who oppose these laws may argue that the social disruption caused by religious conversion does not just ify its banning, but they rarely i f ever question the underlying assumption that conversion is an intrinsically “unsettling event” in the life o f the community and a source o f social conflict. This assumption is not an originally Indian one. It derives from a modern European tradition o f thought, which seeks to understand religion as a socio � logical phenomenon. This functionalist model, which conceives o f religion as the “glue” that holds society together, can be found as early as the sixteenth century in Niccolo Machiavelli’s political writings (Preus 1979, 135). The functionalist thesis was more fully elaborated in Jean Bodin ’s Colloquium Heptaplomeres de rerum sublimium arcanis abditis (Colloquium o f the Seven about secrets o f the sublime), composed in 1593 (Preus 1996). Emile Durkheim’s (1995) more famous version o f this argument goes further, argu � ing that religion itself is ultimately an expression o f the collective being o f the social whole. Because the Durkheimian version is more familiar to most readers, and because it more closely matches the assumptions o f elite dis � course, I will henceforth refer to the complex tradition o f functionalist accounts o f religion as the “Durkheimian picture.” I do not imply that those who reproduce these assumptions in India today are conscious disciples o f Durkheim, merely that they are partaking o f a sedimented modernist com� mon sense o f which he is the most famous proponent.

The idea that religious conversion poses a threat to public order is an important one. It is on this assumption that the Supreme Court upheld India’s anticonversion laws as being compatible with Article 25 o f the Constitution in Rev. Stainislaus v. State o f M adhya Pradesh and Orissa (Supreme Court Reports 1977). Article 25(1) guarantees the right o f “all per � sons . . . freely to profess, practice, and propagate” the religion o f their choos � ing, but makes that freedom subject to the state’s more fundamental respon � sibility to preserve “public order, morality and health” (emphasis added). The

1 1 6 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

High Court o f Orissa had ruled that the anticonversion law o f that state imposed an unconstitutional restriction on religious freedom. It was over � turned by the Supreme Court in 1977 on the grounds that the restrictions the law imposed were necessary because conversion itself is a threat to public order. Religious conversion disturbs the life o f currently existing communi � ties, the Court reasoned, and “ I f a thing disturbs the current life o f the com� munity . . . it would amount to a disturbance o f the public order” and can therefore be legally banned (1977, 617-18).

In the previous section I offered examples o f the consternation that reli � gious conversion causes among national elites and their political representa � tives. But laws banning conversion among Dalits and tribals were not upheld by the Court on the grounds that their conversions were upsetting to national elites. They were upheld on the assumption that conversions were a threat to public order in the very communities in which the conversions were actually taking place. Newspaper and N G O reports on conversion-related violence in India make clear that the most direct threat to public order in areas where conversions take place comes from nonlocal activists affiliated to H indu nationalist organizations. But it is generally assumed that H indutva activists are merely capitalizing on existing tensions, tensions caused by conversion itself (or the perceived threat o f it), and not creating them where none exist. The Court endorses this line o f reasoning. It is very explicit on this point. Its argument is that conversions undertaken for the wrong reasons (i.e., “ in a manner reprehensible to the conscience o f the community”) must be sup� pressed, not because such conversions lead automatically to public disorder, but because they are susceptible to being exploited by persons or organizations bent on raising communal passions.7 Given what the Court itself admits— namely, that the true threat to public order comes from the exploitation o f unpopular conversions by mischief-makers and not from conversion itself— one wonders whether public order might be better served by suppressing the former instead o f the latter.8

Maybe not. For i f the general sentiment against certain types o f conver � sions is indeed so strong as to constitute a veritable powder keg that even the barest effort by provocateurs could spark, the state’s responsibility to avoid violent conflagration may indeed require it to focus on the underlying cause and not just the proximate one. But the empirical question o f whether reli � gious conversion is really so disturbing to the face-to-face communities whose members are converting is not one that occurred to the Court to ask, nor did it think to impose any manner o f legal test on states wishing to ban

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E 1 1 7

religious conversion. Such states were not required by the Court to actually demonstrate that a credible threat to public order existed before enacting such laws.

W hat are we to make o f the Court ’s credulity? Part o f the answer, I argue, is the taken-for-granted nature o f the Durkheimian picture in elite national discourse. In the pages that follow I examine this discourse in some detail, and it is important to pause for a moment to explain what I am referring to. By national discourse I mean, quite simply, writings and other publicly avail � able communications conveyed in India’s lingua franca, English. I describe this as an elite discourse because fluent speakers o f English are, by definition, an elite in the Indian context. They are certainly not the only elite, or even the most powerful. M any o f India’s most powerful politicians at both a regional and national level do not habitually express themselves in English. Rural elites often speak only their regional language, as do certain business elites. Each regional language also has its own intelligentsia, its own intel � lectual elite. English is nevertheless the language o f national discourse because elites speaking regional language cannot communicate with coun � terparts in other regions. (H indi is widely spoken across northern India but is not spoken everywhere and is thus still regional.) It may be possible to find examples o f ideas passing directly from Marathi into Bengali, or H indi into Tamil, without ever appearing in English. But that is not common, and no idea can spread to all regions except through English. A l l o f India’s top-level jurists communicate in English (whatever their mother tongue), and the leaders o f the nationalist movement were all also English speakers. Finally, English is the shared language o f India’s intellectual class, whose members hail from backgrounds more commonly described as “middle class.” But what is called middle class in the Indian context is in fact an elite (Fernandes and Heller zoo6).

Returning to our argument, the idea that religion is the source o f social cohesion and community is passively accepted across the political spectrum o f national elite opinion and is shared as much by those who oppose anticon � version legislation as by those who promote it. I am not the first scholar to note the ambivalence o f secular liberal elite sentiment with respect to reli � gious conversion (cf. Sarkar 2007, 357). In a historical study o f tolerance dis � course in India, C . S. Adcock links this ambivalence to contemporary stere � otypes about “proselytizing religions” as inherently conflict prone, stereotypes she traces to the nineteenth-century European discipline o f comparative theology (2014,39-70). My focus, by contrast, is on speculative sociology and

1 1 8 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

the implicitly Durkheimian understanding o f an intrinsic relationship between religion and community life. Adcock ’s research explains why ulti � mate responsibility for communal conflict is conventionally assigned in elite discourse to the so-called converting religions, Christianity and Islam, and to modern forms o f H induism that are said to have remade themselves in the image o f these “Semitic” faiths. Mine complements hers by tracing the implicitly sociological basis o f the argument that conversion disrupts social cohesion at the local level, a view that is likewise shared by both advocates o f anticonversion laws and secular liberals who oppose them.

To illustrate this assumption, I offer three representative statements by public intellectuals at different points on the spectrum o f elite opinion. The first comes from Swami Dayananda Saraswati a vocal proponent o f anticon � version laws and a leader o f the V H P , a H indu majoritarian organization. It initially appeared in an editorial, styled as an open letter to Pope John Paul II and entitled “Conversion Is Violence,” in one o f India’s leading dailies. It has since been reproduced countless times in print and online.

Conversion implies a conscious intrusion into the religious life of a person, in fact, into the religious person. This is a very deep intrusion, as the religious person is the deepest, the most basic in any individual. When that person is disturbed, a hurt is sustained which is very deep--- People connected to a converted person are deeply hurt. Even the converted person will suffer some hurt underneath. He must necessarily wonder i f he has done the right thing and, further, he has to face an inner alienation from his community, a com� munity he has belonged to for generations, and thus an alienation from his ancestors. I don’t think that can ever be fully healed. Religious conversion destroys centuries-old communities and incites communal violence. It is vio� lence and it breeds violence__ Religion is woven into the fabric of culture. So, destruction of a religion amounts to the destruction of a religious culture. (Saraswati 1999)

Saraswati’s statement is useful because it neatly articulates two central themes o f national elite discourse on religious conversion. The first pictures religion as the authentic source o f a person’s innermost identity and the basis o f psychological well-being. Conversion here is seen as disruptive o f this innermost self. The second links religion constitutively to collective identity. W hen an individual converts, he not only inflicts pain on those around him but becomes alienated from his community and culture. Because communi � ties are based on a shared religion, they cannot survive the introduction o f religious plurality in their midst. Before turning to the next two examples, it

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • I I 9

is important to note that the ideas Saraswati expresses are not unique to India and did not originate there. Both the picture o f religion as an inward, psychological phenomenon that defines authentic selfhood and that o f reli � gious conversion as an attack on collective being depend on secular modern theories whose roots extend back to the sixteenth century and are alive and well even in the contemporary anthropological view o f conversion that pictures it as a “colonization o f consciousness” (Roberts 2oiza).

O ur next example comes from a journalistic essay entitled “ W hen Their Gods Fail Them.” The author, Neena Vyas, is a political analyst who is openly critical o f anticonversion legislation (cf. Vyas 1999). The essay from which the following quote is taken describes conversion as an important strategy for subalterns whose ambitions for social betterment are blocked within the religion o f their birth. Yet even while making this argument, Vyas relies on a sociology o f religion that is not so different from Saraswati s.

There is hard ly any doubt that religious be l ie f is part o f the culture o f a peo � ple. O ften it is their religious fa ith whic h sustains their lives in many ways, deciding not only which god they worship and how, but also helping them construct an internal value system, m ak ing social l iv ing cohesive, and impac t � ing the most important events in ord inary lives by laying down the dos and don ts and the ri tuals that surround birth , death and marriage . W henever a religious conversion takes place, indeed it must be pa in ful , as p a in fu l as it is for a child being weaned away from its mother’s breast. (Vyas 1002)

Once again religion is pictured as formative o f a person’s authentic cultural self, his or her “ internal value system,” and the source o f psychological com� fort and sustenance (like a mother’s breast). Conversion is an inevitably pain � ful event insofar as it disrupts this comforting relation. It is also assumed to disrupt the life cycle rituals through which families are constituted, and thereby implicitly those families themselves. Religion is not just about the inward self, therefore, but also the basis o f social cohesion. Though Vyas does not explicitly say so, it would not be a stretch for readers to conclude that conversion, by introducing a religious division within a social whole, would undermine its cohesion.

A final example o f this implicitly Durkheimian sociology comes from a highly nuanced study o f conversion in India and nineteenth-century England by the renowned public intellectual Gauri Viswanathan. Viswanathan, like Vyas, is critical o f anticonversion laws and sees conversion as a vital political strategy for marginalized people and groups. She was indeed the first scholar

1 2 0 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

to draw critical attention to the political significance o f conversion and anti � conversion discourse in India today, and without her pathbreaking work my own argument in this chapter would not have been possible. For what Viswanathan shows is, first, that national cultures and modern states’ attempts to ground them in monolithic religious identities are artificial and, second, that religious conversion is unsettling in the present day precisely for the powerful challenge it presents to hegemonic nation-state discourse. She does not, in other words, present the anxiety over conversion in Durkheimian fashion as a transhistorical phenomenon rooted in an intrinsic connection between religion and the social whole. Quite the contrary. Yet the preface o f her major study o f conversion in modern nation-states, Outside the Fold, begins thus: “Conversion is arguably one o f the most unsettling political events in the life o f a society. This is irrespective o f whether conversion involves a single individual or an entire community, whether it is forced or voluntary, or whether it is the result o f proselytization or inner spiritual illu � mination__ W ith the departure o f members from its fold, the cohesion o f the community is under threat just as forcefully as i f its beliefs had been turned into heresies” (G. Viswanathan zooi , xxi). Conversion is here pre � sented as a disruption o f “society” or “the community” as such, and not merely the imagined community (Anderson 2.006) o f the modern nation � state. W hile Gauri Viswanathan does not herself endorse the notion that conversion undermines a person’s innermost and authentic self, and indeed has argued explicitly against such a view (2.000, 4), the quoted passage could nevertheless be read as contrasting conversions premised on “ inner spiritual illumination” with those resulting from proselytization (as i f proselytization could not be the cause o f inner illumination).9 It could even be read as imply � ing a parallelism between “proselytization” and “ forced” conversions, on one hand, and “ inner spiritual illumination” and “voluntary” ones, on the other, though the main text o f her study serves to undermine any such contrast. The passage, in other words, is simply a bit o f boilerplate that has crept in where it does not belong. It is significant for what it tells, not about Viswanathan’s own considered views on the topic, but about the power o f a discourse in which such ideas are so widely circulating that they can enter texts whose overall purpose is quite antithetical to them.

The examples o f Saraswati, Vyas, and Viswanathan illustrate the assump� tion that religious conversion threatens social cohesion, and the way conver � sion is pictured as tampering with a person’s innermost self. The former assumption is explicitly endorsed by India’s Supreme Court and is the basis

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E 1 1 1

on which anticonversion laws were upheld as constitutional. The latter is not explicitly cited by the Court, but it pervades national discourse on conversion and features prominently in the recommendations o f the N iyogi Commission Report (M PC M A E C 1957), an official study by the state o f Madhya Pradesh that provides the template for all anticonversion legislation to the present day. We will return to that report later in this chapter.

Hie examples I have offered so far illustrate these two assumptions and substantiate my claim that they are shared by writers on both sides o f the debate. But examples cannot on their own prove that these ideas are charac � teristic o f the discourse as a whole. Examples could be multiplied, but no number o f examples would suffice to show that these assumptions are univer � sal. My argument, however, is not that every single piece o f writing in India’s national discourse on conversion exhibits these ideas. It is rather that these ideas are commonplaces, assumptions that are passively accepted to the extent that no one thinks to question them. Their status as commonplaces is evident in the manner the examples I have cited introduce them. The idea that conversion is intrinsically disruptive is not presented by these authors as an empirical claim that must be demonstrated, or as a position that must be defended, but as a settled fact, a matter o f common knowledge.

Let us return to the third oddity o f anticonversion laws noted above: the ban on conversions brought about by allurement, defined as “temptation in the form o f any gift in cash or k in d . . . [or] any material benefit.” The peculiar thing about this ban, recall, is not simply that the ban implies a theological commitment on the state’s part but that it takes this restrictive position only in the case o f religious conversion.10 H ow are we to explain this anomaly? The answer, I argue, is once again to be found in the discourse o f national elites. W hile neither critics nor defenders o f these laws actually attempt to justify the ban on allurement, this feature o f anticonversion legislation is rarely i f ever explicitly questioned. A nd as Sarkar has noted, even among the “well- intentioned and progressive” Indians (quoted in Adcock 2014 ,176 n. 4) who uphold conversion as a fundamental right, the idea o f people converting for material gain is widely regarded as distasteful. A recurrent theme in the hun � dreds o f journalistic reports on conversion, and in the opinion pieces and letters to the editor I have reviewed, is a preoccupation with the question o f motive: W hat were the convert’s “true motives” ? W hat were those o f the converter? In both cases, were these motives truly “religious” ? Recall, for example, the headline that introduces a special series o f articles in the Sunday Times o f India-. “Are the conversions for real?” The editors never explain what

I2 ,Z R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

they mean by “ for real” because they don’t need to. As the articles that follow make clear, what is really at stake is the converts’ motives. Those whose con � versions have been “ induced” by the hope o f material gain are, by that very fact, not real but, in the prevailing lingo, “ fraudulent.”11

Again, secular liberal voices in national discussions o f conversion are not arguing that people should be actively prevented from converting for nonspir � itual reasons. But such conversions are nevertheless portrayed as undesirable, not just for religious reasons, but because they are seen as somehow contrary to the public good. Take, for example, an editorial penned by the former chief justice o f the H igh Court o f Andhra Pradesh and respected secular liberal intellectual A lladi Kuppuswami. In an editorial devoted primarily to denounc � ing anticonversion legislation, he nevertheless affirms that the question o f “why people allow themselves to be converted” is a matter o f legitimate public con� cern and recommends that “steps be immediately taken to see that [potential converts’] economic and social status is improved to such an extent that they get themselves converted to another religion, not because they can enjoy better status as members o f that religious group, but because they feel convinced of the tenets o f that religion” (Kuppuswami 1999). One assumes the judge would agree that improving the lot o f India’s most impoverished communities is itself a matter o f national concern, quite apart from the effects this might have on their decision to convert. W hat remains unclear, however, is why the basis o f their conversions should also be a matter o f national interest. Why, that is, should members o f the nationalist middle class have any stake whatsoever in promoting one kind o f religiosity (an otherworldly one) over any other?

To summarize, we have identified three oddities, three assumptions that are shared across the spectrum o f opinion, all o f which are directly or indi � rectly reflected in anticonversion law: that conversion is socially disruptive, that it impinges upon converts’ innermost identity and self, and that there is something undesirable and even irreligious in conversions undertaken for the sake o f material gain. Apart from the illegitimacy o f material motives, none o f these assumptions were widely articulated in India before the twentieth century. The view that material motives were illegitimate was current only among Christian missionaries until the early decades o f the twentieth cen� tury (Viswanath 2013). These three assumptions, in other words, were not o f Indie origin. Where did they come from? A nd how did they come to appear a matter o f common sense, at least among national elites, in so short a period? To answer these questions we must begin with a fourth oddity o f India’s national discourse on conversion, its singular focus on Dalits.

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • 1 2 3

CO N V ER TI N G THE OU TCAST: FROM MUKERJI TO S H R A D D H A N A N D 12

Lik e a ll H indus, somewhere deep down inside me I had assumed that H ari jans (D a l its), G a n d h i ’s supposed “children o f G od ,” re legated to the fringes o f society, were part o f the H in du community , part o f “us.”

u r v a s h i b u t a l i a , The Other Side of'Silence

The partition o f British India in 1947 was accompanied by the mass mur � der o f H indus by Muslims, and Muslims by Hindus. It is estimated that up to a million people were killed (B. M etcalf and M etcalf zoo6, zzi), often by hand and at close quarters. W hile collecting oral histories o f this vast human tragedy, historian Urvashi Butalia was astonished by evidence o f Dalits who remained calm amid the violence around them. They knew no one would touch them, one D ali t woman recalls, because the conflict was between H indus and Muslims and they belonged to neither group. H er father never � theless urged her to stay indoors for fear that the H indus might mistake his daughter for a Muslim (Butalia zooo, Z35). Dalits displaced by the violence had no place to go, Butalia later reveals. For “ in a war that was basically cen � tered around H indu and Muslim identities,” she explains, no one set up “camps to help [Dalits] tide over the difficult time. No recourse to govern � ment—all too preoccupied at the moment with looking after the interests o f Muslims and Hindus, no help from political leaders whose priorities were different at the time” (zooo, Z38). Could untouchables not have gained admission to H indu relief camps? A t a time when Dalits were excluded from schools for H indu children—because they were regarded as polluting—and were confined to separate quarters in government prisons for the same rea � son, this most likely would not have been possible. Even in Z 0 04 , in the wake o f a devastating tsunami, Dalits were excluded from relief camps that shel � tered caste H indus (Human Rights Watch Z005 , Z 5-Z 9 ; A nand and Thangarasu zoo6; G ill Z007) .

Butalia is not the only late twentieth-century scholar surprised to discover that “ H indu” and “untouchable” are understood as contrasting categories by ordinary people, even in the present day. M ary Searle-Chatterjee recalls, “I could hardly believe the evidence o f my ears” when she first noticed sweepers in Benares, among whom she was conducting ethnographic research in 1971, refer to “the H indus” as other to themselves (zoo8, 189). These were not politicized Dalits who actively rejected the H indu label in favor o f

1X 4 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

Ambedkarite Buddhism, she explains. “ This was the usage o f ordinary, non � politicized sweepers” (2008,189). They were not using the term in a segmen� tary sense, in which “the H indus” means caste people in contrast to “ H arijans,” and that encompasses both caste people and Dalits in contrast to Muslims. W hile “segmentary [terms] may be more or less inclusive,” Searle- Chatterjee explains, “ in the case o f the Tow’ caste reference to H indus as people other than themselves, something more is involved. Even when Muslims were present, sweepers did not shift to referring to themselves as ‘ H indus’ ” (2008,189).

Butalia’s and Searle-Chatterjee’s sense o f surprise is itself unsurprising. As historical anthropologist Joel Lee explains, it reflects the hegemony o f the contemporary assumption

that “sweepers” and other D a l i ts, insofar as they are not form a l converts to Buddhism , Islam or C hrist ian ity , be longin a taxonomic al sense to the H indu com munity and shou ld therefore see themselves as H indus. This assumption follows logic a l ly from the most basic lessons that students across the globe learn about Ind ian society; to be educated in the world today, whether in Tokyo or C h ic ago or Joh ann esburg or D e lh i , is to know that H in du society has or had a caste system that classed some people “untouchable,” that there � fore “untouchables” belong to H in du society, that therefore “untouchables” are H indus . . . This constitutes commonsense among the educated in urban Ind ia and in the academy as well. (2015, 82)

Today this common sense is backed by the force o f law. Dalits who do not specifically proclaim themselves Christian or Muslim are legally categorized as H indu by default. Exceptions like those discovered by Searle-Chatterjee and Butalia were still common in the late twentieth century (e.g., Lynch 1969 ,162-63).13 But in many urban settings at least—including my own field site—Dalits now accept the government’s new, more inclusive definition o f H indu as including people like themselves (Roberts 2015a).

It was not always so. Before the twentieth century Dalits were not regarded as H indu by others, nor did they regard themselves as such. It is conceiv � able that exceptions may yet be found, but in the absence o f clear documen� tary evidence it is an anachronism to label pre-twentieth-century Dalits H indu (Searle-Chatterjee 2008 ,187; Frykenberg 1997). The only significant context in which “untouchable” castes were treated as H indu before the twentieth century was the decennial census o f the colonial state, which cat � egorized all Indians by caste and religion and which assigned those who did not proclaim themselves Muslim , Christian , or members o f another

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

recognized religion to the category “ H indu” by default. In so doing, the colo � nizer rejected native precedent. Census taking and other enumerative tech � nologies were well established in India’s precolonial states and like later colonial censuses categorized subjects by caste and religion (Guha 2013; Roberts 2015b). But these precolonial censuses did not recognize untouchable castes as H indu. As Norbert Peabody has shown in an important paper on precolonial census taking, even as late as 1835 the H indu kingdom o f Marwar conceived the primary division among its subjects as lying, not between H indus and Muslims, but between the so-called “clean” castes, which included both Muslims and H indus jd t h , and the impure servile castes, namely Dalits, who were understood as distinct from both (2001, 834-36). Unpublished research by Divya Cherian on the same kingdom in the late eighteenth century paints a similar picture. O fficial guidelines on religious duties and prohibitions in that state “categorically divided its subjects into two types: H indus (hinduvan) and untouchables {achhep). The latter category consisted o f leatherworking castes, nomadic pastoral groups, Muslims (,turak), and the sanitation labor castes (halalkhor). Not only were the sweep� ers not H indu, they were the antipode o f the H indu: the order made clear that what actions the state required o f its H indu subjects were precisely those that it forbade its untouchable subjects” (quoted in Lee 2015,120). Cherian’s findings differ from Peabody’s only insofar as the records she unearthed categorize Muslims together with untouchables. Rupa Viswanath’s research in the Madras Presidency similarly finds that the term H indu referred, until the early twentieth century, exclusively to those jd t h eligible to live in the ür and expressly excluded those confined to the cm , the Dali t ghetto (2014c; see also Ebeling 2010).14 Indeed, the association o f H indu with respectable caste status was so well established in nineteenth-century Madras that Christians and Muslims belonging to the so-called clean castes were sometimes referred to in native discourse as “ Hindu Christians” and “ Hindu Muhommedans,” to distinguish them from coreligionists o f untouchable origin, known as “Pariah Christians” and “Pariah Mohammedans” (Rupa Viswanath, per � sonal communication). A nd even as late as 1916, Gyan Pandey records that in Chhattisgarh, “to call a man a H indu convey[ed] primarily that he [was] not a Chamar,” that is, not a D al it (1993, 246).

British census officials departed from existing usage. Colonial observers had long stereotyped Indian subjects as divided into two distinct and antago� nistic religious “communities,” H indu and Muslim. The latter were portrayed as following the religion o f “ foreign” invaders who had ruled much o f the

1 2 . 6 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

Indian subcontinent since 1206, the former as followers o f India’s original religion. By playing up alleged conflict between the two, colonizers justified their own rule as bringing peace to the land and as protecting India’s disen � franchised H indu masses (T. M etca lf 2007, 132-48). Reversing precolonial precedent, the colonial census simply lumped untouchables together with H indus. This policy met with frequent objections by native census takers, typically high-caste H indus, who persistently refused to record Dalits as H indu (Lee 2015, no; Mendelsohn and Vicziany 2000, 27-28; Juergensmeyer 1988, 72, cited in Searle-Chatterjee 2008,191).

Hindu census takers were not alone in rejecting untouchables. Dalits were also banned from H indu temples; access to sacred H indu texts was forbidden them; H indu priests refused them. A distinctly anthropological argument could nevertheless be made for classifying untouchables as objectively H indu, irrespective o f how they classified themselves or were classified by others, on the basis o f three criteria: morphological similarities between their cults and those o f popular (non-Brahminical) H induism, Dali ts’ limited participation in village religious festivals, and the fact that Dalits serve H indus by remov� ing ritually impure substances for them. But it is unclear why forced ritual service to a cult implies membership in it, and by the criterion o f participa � t ion many Indian Muslims and Christians are also H indu, and vice versa (Roberts 2015a, 242-44) . A s for morphological similarities at the level o f practice, these are common also between popular H induism and Islam in India, which is why Peter van der Veer, an anthropologist who has studied these extensively, argues that the only valid criteria for group membership are self-definition and acceptance by others. Morphological comparisons at the level o f doctrine are reviewed by Viswanath (2012a), who argues they do not establish common religious identity. But the classification o f Dalits as H indu for census-taking purposes was never purported to rest on objective criteria. Dalits were recorded as H indu by state fiat. It is thus not surprising that H indu census takers would refuse to comply with this order, only to have their surveys later “corrected” by higher-ups.

Joel Lee’s ethnographic study o f the 2011 Census describes a fascinating historical reversal: the Brahmin census taker he accompanied on rounds recorded untouchables as H indu even when they themselves told him they were not (Lee 2015,3-10). W hat had changed? Since the late nineteenth cen� tury Muslims and a H indu missionary organization known as the Arya Samaj had been competing for converts in the United Provinces and the Punjab. The Aryas focused on converting Christians, Muslims, and wayward

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E 1 2 7

Hindus but at this time still regarded untouchables as beyond the pale, and the one or two attempts by renegade Samajists to convert untouchables were met with a strong backlash within the organization (Jones 1976; Adcock 2007). This began to change when the Morley-Minto reforms were announced in 1909. The franchise was extended, and representation o f dif � ferent communities became tied to demographics. W hat had been a struggle for cultural preeminence became a competition for sheer numbers (Tejani 2008, 141-43). The inclusion o f untouchables within H induism merely for purposes o f census taking suddenly had very real political implications, and Muslims began to argue that H indus’ numbers were artificially inflated by the inclusion o f untouchables (Mendelsohn and Vicziany 2000, 28; Rao 2009, 131; Sartori 2003). A nd all over India untouchables themselves demanded to be recognized as a separate element distinct from both groups (Sartori 2003,272-73; D . Sen 2012; Irschick 1969, 71-72; Viswanath 2014b).

Fairness and accuracy required that the controversial policy o f listing untouchables as H indu be revisited. Census Commissioner E. A . Gait argued in 1910 that H induism should be conceived broadly and not be limited to those holding specific beliefs or practices. “A man may believe in the whole H indu Pantheon, or for that matter, in no god at all—he may sacrifice or abstain from sacrifices—may eat fish and flesh, or abstain from doing so,” and still be legitimately called a H indu, Gai t argued (quoted in Mukerji 1911, v). But he observed that it was “absurd to enter without comment as H indus persons . . . who are not regarded [as such] by others, and do not profess themselves to be H indus,” as previous censuses had done (quoted in Mukerji 1911, v).

The possibility o f losing a large portion o f their official numbers sharp � ened the minds o f high-caste H indu leaders, who began at this point to vociferously claim Dalits as fellow H indus and to accuse the colonial state o f conspiring to undermine H induism by divide-and-rule tactics. It is true that the British would later seek to capitalize on the refusal by Dalits to recognize the leadership o f the high-caste H indu-led Congress Party (Prashad 1996). But “divide and rule” implies a prior unity, and in the case o f Dalits and Hindus the evidence for any such unity is lacking. Interestingly, evidence for precisely the opposite—a lack o f both common identity and regular social relations between caste H indus and D ali ts—can be found in the urtext o f the argument that the British were subjecting H indus to divide-and-rule by separating out untouchables. This was a pamphlet entitled A D ying Race, originally published in serial form by the H indu strategist U . N . Mukerji in

128 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

1909. Even as the author accuses Ga i t ’s memo o f attempting to create a divi � sion between H indus and untouchables, he elsewhere notes,

It w i l l puzzle most H indus i f they are asked as to the inner l ife o f these “ low castes.” Respec table people scarcely trouble themselves about such th ings. There is a sort o f a 11 Ghet to” . . . at tached to nearly every vi llage, far away, o f course, from where the respectable classes live. N obody be longing to the “ high castes” ever th in ks o f visi t ing these quarters. Everyth ing about the . . . people o f that class is pollu t ion— the ir touch is pollu tion , their presence is pollu t ion , water touched by them is pollu ted , their very shadow carries in fec t ion . These people do a certa in sort o f work and, when their services are needed, are tolerated to that extent, but they are the “ it ars”— “ the oth � ers”— quite apart from respectable people. A t other times there is h ard ly any contact . (M uker ji [1909] 1929, 43)

Mukerji’s admission that for H indus untouchables were outsiders, and that H indus wanted nothing to do with them, is not presented as an original observation. That H indus regarded untouchables as beyond the pale was common knowledge; Mukerji’s express purpose was to persuade them to reverse course and enlist untouchables as fellow Hindus. Nationalist scholars have nevertheless treated as an established fact the accusation that Gai t set out deliberately to create a division where none had existed. H istorian Pradip Kumar D at ta calls G a i t ’s memo a “ blatant act o f social engineering. . . . designed to encourage the detachment o f low castes from the ‘H indu ’ cate � gory” and to provoke “ low-caste resentment” (1999, 24, my emphasis). Datta offers no evidence for this startling claim, apart from a quote from Mukerji insisting that it is so, an instance o f circular reasoning on Dat ta’s part that usefully illustrates the common ground between the secular liberal and H indu nationalist on the untouchable question.

Mukerji’s pamphlet was reprinted countless times, and its arguments are repeated by H indu nationalists to the present day (Bhatt 2001, 62-68). Its core message was that H induism was in a demographic struggle with Islam, a struggle in which H indus were literally in danger o f becoming biologically extinct. It also provided the strategic blueprint that would become a central feature o f H indu nationalism from that day onward—namely, that the very survival o f H induism in the face o f a putative Muslim threat (and later a Christ ian one) depended on its ability to incorporate Dalits and tribals within its fold. The necessity o f integrating untouchables within Hinduism would become a key plank in the program o f H indu sañgathan (consolida � tion/organization).

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • 1 2 9

By far the most important proponent o f untouchable integration was Swami Shraddhanand, an Arya Samaj leader whose mission to the untoucha � bles was inspired by a personal meeting with U . N . Mukerji in 1912 (Datta 1999, 22). Mukerji s enduring influence is evident in the title o f Shraddhanand s 19x4 tract, H indu Sangathan: Saviour o f the Dying Race, a text that reproduces and expands upon Mukerji s core argument. A “constant refrain” o f the swa- mi ’s writings and speeches on the untouchable question, as Joel Lee’s study o f Shraddhanand s corpus reveals, was the worrying prospect o f Dalits emerging as an autonomous political force in the Indian landscape (Lee 1015, 141). Perhaps equally alarming to Shraddhanand was the threat o f Dalits convert � ing to Christianity or Islam. Thus the swami openly warns that Dalits who convert to Islam “will become equal to H indus----They will not depend on Hindus, but will be able to stand on their own legs”; those who convert to Christianity will “dream o f entering its halls o f governance” (quoted in Lee 2015,140 ,142). The key to preventing this, according to Shraddhanand, was eliminating the divisive practice o f untouchability from the collective H indu body. As Lee observes, “Shraddhanand insisted that the danger . . . o f [Dalit] autonomy from the Hindus could only be defused i f Hindus radically cur � tailed the regime o f disabilities they imposed on untouchables. Further, he maintained that [checking] this autonomy, [by] bringing the untouchables to accept H indu leadership, equated with the neutralization o f the Muslim and Christian threat, and was an essential, sine qua non . . . for the manufacture o f a H indu nation” (Lee 2015, 143). For Shraddhanand the Hinduization o f untouchables was not merely a H indu communal cause but a national one. In his writings the good o f the H indu community was indistinguishable from India’s struggle for national independence: “ The uplift o f the untouchables and their assimilation in the H indu polity is the very plinth on which alone the edifice o f free India can be constructed” (quoted in Lee 2015,143). As we will see, the mission o f ending untouchability would play an identical role for Gandhi. For Gandhi, too, it was essential to the strength o f both Hinduism and the nation—even to the extent that the good o f the one was often pre � sented as indistinguishable from the good o f the other.

The imperative o f H induizing the untouchable was eventually endorsed, in theory i f not always in practice, by H indu organizations across northern India and from Bengal to Bombay (Prashad 1996). The major bases o f sup� port for this movement were in towns and cities, among modernizing H indu organizations like the Arya Samaj (Jones 1976; Adcock 2014), and among politically minded H indu reformers keen on establishing India as a H indu

X 3 0 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

nation (Bayly 1998). But it was by no means universally accepted. Orthodox H indus remained deeply opposed, wanting nothing to do with those they regarded as untouchable. O pposition was also widespread among rural H indu elites (Jones 1976). A nd in South India, where Muslims were not perceived as a threat, programs for H induizing the D al it found few takers.

The Congress Party passed its first resolution condemning the practice o f ritual untouchability, seen by advocates o f H indu sangathan as the principal barrier to Dalits’ inclusion within the H indu fold, only in 1917. But it took no concrete steps on this matter until the 1930s (Prashad 1996, 553). As early as 1920, M . K . Gandhi, the party’s paramount leader, proclaimed the eradication o f untouchability as essential to swaraj, and in 1921 claimed he had in fact opposed untouchability since childhood (Zelliot 2010,153). Gandhi did not act on this conviction until the 1930s, however, despite urgings by Shraddhanand to join him (Lee 2015,145), and despite multiple opportunities to lend support to autonomous Dalit struggles, including invitations from Dalit activists, which he repeatedly declined (Ambedkar 1946, 251-59). According to J .T .F . Jordens, Shraddhanand’s sole supporter o f any note within Congress until the 1920s was G. D . Birla, one o f India’s leading industrialists, whose vast wealth bankrolled Shraddhanand’s operation (1981,165, cited in Lee 2015,143-44).

As for Dalits themselves, some reacted with enthusiasm to the prospect o f H induization (.shuddhi), which by the second decade o f the century many embraced as an opportunity for social advancement. And just as it was Dalits who first approached Christian missionaries, and not the reverse, with demands to be converted (Viswanath 2014b; Webster 2009), so too did they begin to approach the Arya Samaj (Rawat 2011; Adcock 2014, 48-50; Lee 2015). But finding the promise o f full inclusion as equals illusory, Dalits began to turn away from shuddhi by the 1920s (Lee 2015, esp. 150-53) and, simultaneously, to assert their independence from the Congress Party (Prashad 1996,552).

N A T IO N A L IZ IN G H I N D U I S M : M. K. G A N D H I

A m ong the world ’s religious tradit ions, there are those that con � vert and those th at do not. The non-convert ing religious tradi � tions, l ike the H indu , Jewish and Zoroastr ian , give others the freedom to pract ise their religion whether they agree with the others’ tenets or not. T hey do not wish to convert. I would char � acterise them as non-aggressive. Re ligions that are committed by the ir theologies to convert, on the other hand , are necessarily aggressive___

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • 1 3 1

Re ligious conversion destroys centuries-old com m un it ies .. . . Today, for instance, there is no l iving Greek culture; there are only empty monuments. The M ayan , Rom an and many other rich cultures are all lost forever and hum anity is impoverished for it. Le t us at least a llow hum anity to en joy the riches o f its remain ing mosaic o f cultures. Each one has some beauty, some � th ing to contribute to the enrichment o f hum anity .

In any tradit ion , it is wrong to strik e someone who is u n arm ed . . . . A Buddhist , a H indu , a Jew , are all unarmed , in that they do not convert. You cannot ask them to change the genius o f their traditions and begin to convert in order to com � bat conversion . Because it is the tradit ion o f these religions and cultures not to convert, attempts to convert them is one-sided aggression . It is strik ing the unarmed .

S W A M I D A Y A N A N D A S A R A S W A T I

“ Conversion Is V iolenc e: A n O pen L e t ter to Pope Jo h n Paul I I ” (1999)

The idea that religions divide naturally into converting and nonconverting, and that the latter are inherently tolerant and the former conflict prone, derives from a taxonomic distinction developed in nineteenth-century com� parative theology by the likes o f M ax Müller and C . P. Tiele (Adcock 2014, 61-70). Conversion was said to be the sole purview o f so-called universalist religions, religions whose “totalistic” truth claims were supposed, by students o f comparative theology, to make them intrinsically conflict prone (Asad 2001, 209-12). The opposite was true o f what were termed ethnic or national religions, such as H induism, which were presumed to be inherently tolerant of other faiths (cf. Schwab 1984, 427-34). In Indian national discourse, intol � erance is conventionally associated with the so-called “Semitic” religions, Islam and Christianity (e.g., Radhakrishnan 1940; Balagangadhara and De Roover 2007). Evidence o f H indu aggression is explained within this model as a perversion o f H induism’s authentically tolerant tradition by H indu nationalists, who are said to follow a “semiticized” version o f H induism (Adcock 2014, 61). This is how elites who are critical o f H indu nationalism often seek to reconcile evidence o f H indu intolerance with the postulate that H induism is an intrinsically tolerant faith (e.g., Romila Thapar 1985; G . Deshpande 1985; Nandy 1983, 2004; Madan 2003; cf. Roberts 2012b). Their counterparts in the H indu nationalist camp hold that H indus resort to violence only as a defensive measure, to counter foreign religious aggres � sion—religious conversion, for example.

1 3 2 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

The premise that conversion is anathema to the H indu tradition—that there exists a time-immemorial “ Hindu view o f religion” on the basis o f which H indus have always opposed the practice (e.g., Claerhout and De Roover 2005)—is almost certainly false. Despite the currency it enjoys among the general public and in casual scholarship that treats it as axiomatic, there is remarkably little evidence to support the idea o f a long-standing H indu objection to conversion, and much against it. Examples o f H indus converting non-Hindus in precolonial and modern contexts have been sum� marized by anthropologist M . N . Srinivas (1962, 575-76) and historian Sumit Sarkar (2007, 357-58). Sanskritist A rvind Sharma adduces evidence from a wide range o f authoritative scriptural sources to argue that ancient Hinduism was in fact a missionary religion that actively pursued the conversion o f non- H indus (1992, 177-90) . O ther Sanskritists have described precolonial texts that celebrate attempts by representatives o f H indu and non-Hindu sects to convert one another (Bronkhorst 2007; Granoff 1985). Whether these narra � tives accurately depict ground realities is open to question (Andrew Nicholson, personal communication, 2012). But even i f merely the stuff o f Brahminical fantasy, the celebration o f conversion in these texts proves it was not contrary to precolonial H indu norms.

The claim that H indus have, since time immemorial, viewed conversion as an intrinsically unethical practice must therefore be taken with a grain o f salt. So far as I have been able to determine, this idea took hold only in the 1930s, when Gandhi began vigorously to promote it. W hat about the idea that Hindus have long been specifically troubled by the conversion o f Dalits (and not merely their own laborers) to non-Hindu faiths? This idea is perhaps slightly older. But it appears to have been virtually unheard o f prior to Mukerji’s 1909 tract A Dying Race. Thus, when Srinivasa Raghavaiyangar, a prominent Madras Brahmin and civil servant, advised the colonial state in 1892 that “the best thing that can happen to [the untouchables] is conversion to either the Christian or Muhammadan religion,” this highly publicized remark provoked no hue and cry from H indu religionists. N o one accused Raghavaiyangar o f attempting to “divide” the H indu community. Nor did Hindus call foul when he explained that “there was no hope for them [i.e., Dalits] within the pale o f H induism” because H induism was, by its very nature, the religion o f “a supe � rior race” that had erected “moral barriers” to isolate itself from the inferior one (Raghavaiyangar 1893,154). The sole objection history records came not from a H indu but from a man o f the Depressed Classes: Tamil litterateur Pandit Iyotheedas, leader o f a Buddhist organization that sought to convert Dalits

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • 1 3 3

instead to Buddhism. Iyotheedas pointed out that Dalits who converted to Christianity or Islam did not cease to be regarded as untouchable by the broader society and that even within their new religion they were distinguished from others o f their faith as “Pariah Christians” or “Pariah Mohammedans” (lyotheedas [1892] 1985,17).

Frictions between H indus and Christian missionaries were not absent before the twentieth century. Quite the contrary. But only after the publica � tion o£A D ying Race did H indus begin to object to the conversion o f Dalits as such.15 A nd it was only in the 1930s that the idea that conversion itself was an unethical practice and contrary to the principles o f the H indu faith become widespread. Until that time H indus were still seeking openly to convert Dalits, so the argument that conversion itself was wrong was not available to them.16 It is impossible to say who first made the ethical argu � ment against conversion. W hat we can say, however, is that the first Indian o f any note to promote it on a national stage was M . K . Gandhi, who treated it not as self-evident but as an interpretation whose claim to truth had to be supported by argumentation. Notably, Gandhi criticized conversion not only by Christians and Muslims but equally by fellow H indus like Swami Shraddhanand. Gandhi had expressed a personal distaste for conversion, in both its Christian and H indu forms, since the 1920s and possibly earlier. But his statements o f this position grew considerably more frequent, more impas � sioned, and more elaborate after 1932. This upturn coincided with his own mission to the untouchables—a mission not o f conversion but o f “uplift”— and the founding o f the H arijan Sevak Sangh (HSS).

In this context Gandhi began vigorously to promote three historically novel ideas, ideas that are today accepted as a matter o f common sense in Indian national discourse: that conversion is disruptive o f local communities, that it attacks the converts’ innermost identity and sense o f self, and that there is something especially objectionable about conversions undertaken for hope o f worldly gain. H e also portrayed preventing the loss o f Dalits to H induism as not just a H indu communal interest but simultaneously an Indian nationalist one. In the first section o f this chapter I described the commonsense status these ideas enjoy today in Indian national discourse and posed the question o f where they had come from. I said the answer would require us first to consider a fourth oddity o f India’s national discourse on conversion, its focus on Dalits, which I then traced to Mukerji and the rise o f demographic communalism. The time has now come to bring these two arguments together.

1 3 4 R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

Recall that by the 1920s Dalits had begun to drift away from the Arya Samaj. They had also begun to reject the leadership o f the predominantly high-caste H indu Congress Party— exactly as Shraddhanand had feared they would—because o f the rise o f autonomous D al it political organizations (Prashad 1996). The split between autonomous D ali t politics, under the national leadership o f Ambedkar, and Gandhi’s Congress Party came to a head in the second Round Table Conference (1931-32) in London over the issue o f whether Dalits would be permitted to elect their own leaders. The British had already granted this right to H indus and Muslims. The H indu electorate was labeled the “general” electorate. But with Muslims removed, H indus were a supermajority, making it a H indu electorate in all but name. The sticking point was what would happen to Dalits. Ambedkar argued that they, too, should enjoy the privilege o f choosing their own representatives. Although Dalits did not share a single distinctive religious identity like Muslims and H indus, Ambedkar maintained that their political interests were distinct from those o f caste H indus. Wherever Dalits lived they were set apart from, and dominated by, the majority community—be it H indu, Muslim, or other. A nd because Dalits were a minority in all electoral dis � tricts, including them in the “general” electorate would have meant in prac � tice that Dalits’ representatives would be chosen by the majority community, by H indus (Jaffrelot 2005, 53-58; Tejani 2008, 213-21; Anand 2014, 359-63).

Gandhi insisted that Dalits be categorized with Hindus, because that was what they were, according to him. As for whether a non-Dalit politician, or one Dalits themselves had not chosen, could legitimately claim to represent them, Gandhi was adamant. H e dismissed Ambedkar’s arguments that the representation o f Dalits by H indus was problematic. H e even went so far as to deny Ambedkar ’s own credentials to raise such objections on D al its’ behalf. Ambedkar’s claim to represent Dalits was bogus, according to Gandhi. The Dalits’ only true representative at the Round Table Conference was Gandhi himself. “I say that it is not a proper claim which is registered by Dr. Ambedkar when he seeks to speak for the whole o f the Untouchables o f In d ia . . . . I myself in my own person claim to represent the vast mass o f the Untouchables.”17 This was not the only time Gandhi would insist that he and he alone was qualified to represent Dalits. O n this matter he was entirely consistent. A few years later, for example, Gandhi again asserted his author � ity over an audience o f educated D al it leaders: “I know infinitely more than you do what H arijans are, where they live, what their number is and to what

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E » 1 3 5

condition they have been reduced.. . . I claim to be able to speak o f Harijans from the North to the South, the East and the West o f India, and I know their abject position” (C W M G 57:13, quoted in Lee 2015 , n , my emphasis). Educated untouchables could not represent Dalits, according to Gandhi, because their education had alienated them from their uneducated brethren. O nly he could truly understand the humble and uneducated Dal it masses. It went apparently without saying that uneducated Dalits could not represent themselves. A t least in matters o f religion and religious identity, according to Gandhi, as we will see, they lacked the intelligence and understanding to make such decisions for themselves.

The Committee did not find Gandhi ’s argument persuasive, but they worked out a compromise to allay Gandhi’s worry that i f Dalits did not par � ticipate in the general electorate they would gradually lose any sense o f com� mon political identity. Dalits would be given a dual vote; they would vote once in a separate electorate to choose their own representatives and again in the general electorate. But Gandhi was adamant: D ali ts’ “ having double votes does not protect them or H indu soc iety.. . . I sense the injection o f a poison that is calculated to destroy H induism” {C W M G 5 1 :3 1-3 2 , quoted in Tejani 2008 , 229). The threat to H induism was a source o f anguish for Gandhi, but he claimed the threat was not only to H induism but to the entire Indian nation {C W M G 51:461) . This would not be the last time, as we will see, that Gandhi would identify the interests o f India with those o f Hinduism. O n August 18 ,19 3 2 , he announced he would therefore undertake a “ fast unto death” to prevent the agreement from being implemented. The British responded that what had been negotiated in good faith could not be reversed, unless the Dali ts’ representative, Ambedkar, agreed.

Ambedkar was faced with a choice. H e could renounce what his people had been promised, and Gandhi would live. O r he could refuse and assume personal responsibility for the death o f the Mahatma. Ambedkar was not a powerful or popular man; Gandhi was a living saint and dearly beloved fig� ure. But it was not just Gandhi who was in danger. As leader o f the untoucha � bles, by standing his ground Ambedkar would be putting the lives o f Dalits in villages all over India at risk. Ambedkar pleaded with Gandhi to relent and pointed out that his death would “result in nothing but terrorism by his followers against the Depressed Classes all over the country” (Ambedkar 1946 , 327, quoted in Anand 2 0 14 , 365). Ambedkar was not the only one to fear H indu reprisals against his community. Another D al it leader, M . C . Rajah, urged Ambedkar to give in to Gandhi’s demand, explaining, “ For

1 3 6 » R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

thousands o f years we had been treated as Untouchables, downtrodden, insulted, despised.. . . I f he dies, for the next thousands o f years we shall be where we have been, i f not worse. There will be such a strong feeling against us that we brought about his death, that the mind o f the whole H indu com� munity and the whole civilised community will kick us downstairs further st ill” (Jaffrelot 2005, 66). Ambedkar maintained that no legitimate negotia � tions could be undertaken under the specter o f such a threat, and he called upon Gandhi to postpone his fast until a solution could be reached in con � sultation with all parties. The “Mahatma cannot point a bullet at our head and then coerce us,” Ambedkar pleaded. “Let us have a fortnights time for conversations and negotiations” (Tejani 2008, 231). Gandhi refused. Ambedkar called Gandhi’s suicide threat “a foul and filthy act,” and “the worst form o f coercion against a helpless people” (Ambedkar 1946, 270-71).

W hat Ambedkar characterized as blackmail, Gandhi billed as a wholly religious act undertaken for the sake o f “purification and penance” (CWMG 51:64). No one should “mistake this for a political move,” he warned. “It is a deeply spiritual act” (CW M G 62:132). H is fate was not in Ambedkar s hands, Gandhi insisted, but in God ’s, and “therefore, there need be no undue haste or feverish anxiety to save my life” {CWMG 51:62). Gandhi began his fast on September 20,1932. A l l o f India was in an uproar. Gandhi’s own son appeared before Ambedkar with tears in his eyes, begging for his father’s life (Zelliot 2013,137-38). Unable to withstand the pressure, after just six days Ambedkar agreed to all o f Gandhi’s demands. The Mahatma called o ff his fast; he had been purified.

Critics ever since have sought to impugn Gandhi’s motives, suggesting that his intentions in this and other contests were rarely so pure as he made them out to be. In so doing they pay unwitting tribute to the Gandhian worldview, according to which the ultimate status o f a person’s actions turns on the purity o f his or her motives. Purity o f motive was certainly the crite � rion on which Gandhi himself sought to be judged. It was his trademark. Throughout his political career Gandhi staked strong claims to personal and moral authority on the strength o f his own personal integrity, proclaiming his sincerity and the purity o f his motives whether or not anyone had asked. Equally often, as we will see, he sought to cast doubt on the sincerity o f his opponents, imputing “ulterior motives” to them that he claimed rendered their actions illegitimate. Questions o f Gandhi ’s own sincerity, and the purity o f his motives, are for our purposes strictly irrelevant. For the record, I think it quite possible that Gandhi indeed believed every word he ever

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E 1 3 7

spoke, at the moment he spoke them. But my interest is to understand how the question o f motive became so central to Indian conversion discourse in the first place, and it makes little sense to naturalize such a focus by building it into our analytical framework. In the pages that follow, I therefore focus instead simply on what Gandhi did and said (cf. Skinner 1988, 68-96). We will see how arguments that he was the first to develop and promote would soon come to structure both public discourse and law on conversion.

Recall Shraddhanand’s warnings to Gandhi and the Congress Party on the danger o f Dal it political autonomy, and his advocacy o f untouchability removal as a key strategic imperative in securing untouchables for Hinduism. W ith the important exception o f G . D . Birla, Shraddhanand’s warnings had fallen on deaf ears. But as Joel Lee observes, “ His chiding o f Gandhi [and others] for not prioritizing untouchability removal . . . would sound pro � phetic in 1932” (2015,145). I f Gandhi did not heed Shraddhanand s warnings in the 1910s and early 1920s, that would all change in the wake o f his fast- unto-death. W hat Lee shows is that Shraddhanand’s analysis and his strate � gic program were taken up by Gandhi—with a few key modifications—in a big way and were institutionalized in his Anti-Untouchability League (Lee 2015,145,171-83). The League was conceived and launched on September 30, 1932, just four days after the end o f Gandhi’s fast, and was shortly thereafter renamed the H arijan Sevak Sangh (HSS). Shraddhanand (1857-1926) was long dead by this time. But his erstwhile patron, Birla, was present with Gandhi at the League’s founding and was made the organization’s president. Birla s massive fortune, having bankrolled Shraddhanand’s operation, would henceforth underwrite Gandhi’s (Jaffrelot 2005, 70; Renold 1994).

Though a latecomer to the world o f untouchability removal, Gandhi had expressed personal distaste for untouchability as early as 1920. A nd though we have no evidence that Gandhi ever read U . N . Mukerji’s original pam� phlet, which had so influenced Shraddhanand, he was aware o f the program that that pamphlet outlined and made clear that he approved o f it well before Shraddhanand’s warnings became a reality after the Round Table Conference. In 1927 Gandhi quoted approvingly a speech by the Dewan o f Mysore, who had this to say about the relationship between the H indu and the Dalit: “ These people [i.e., Dalits] ought to be the strength o f our strength. Shall we let them become our weakness? They have a rankling sense o f wrong which only kindness can heal. The aim should be to ‘H induize’ them more and more, for they belong to the H indu community, and to offer them every facility to remain within the fold. [Then they] will be a mighty accession to

1 3 8 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

the strength o f our body politic; i f not, they will be an equally heavy subtrac � tion from it” (C W M G 34:122-13). Gandhi reproduced the dewan’s statement in his English weekly, Young India, on July 7, 1927, in an article that con � cluded by issuing “a warning both to the Christ ian missionary and the Mussalman missionary not to try to wean these suppressed classes from Hinduism” {C W M G 34:123).

W hile Gandhi had long shared Shraddhanand’s and Mukerji’s conviction that H induizing the untouchables was imperative for the strengthening o f Hinduism vis-à-vis other communities, when the time came to put this prin � ciple into action he differed with his predecessors at the level o f method. Where Shraddhanand had focused on converting untouchables to Hinduism by means o f shuddhi, according to Gandhi’s way o f thinking Shraddhanand was grasping the wrong end o f the stick. Rather than attempting to convert untouchables to H induism , G andhi ’s approach centered on persuading them—and perhaps more importantly, the rest o f India—that the untoucha � bles were in fact already H indu. Hence his attempt to rebrand them as “ Harijans,” a name that means “children o f [the H indu god] H ari.” Where Shraddhanand had promoted conversion as the cornerstone o f H indu inte � gration, Gandhi’s “more subtle method. . . would encompass the Untouchables within the H indu political constituency by definition, without the political risk or social provocation o f very substantial changes in the lived relations between castes” that Shraddhanand’s interventionist approach had entailed (Adcock 2014 ,163-64) .

W hat untouchables required was not to be made H indus, according to Gandhi, but to be made better ones. In practice this meant enjoining them to renounce non-Hindu practices—especially the consumption o f alcohol, beef, and pork—and teaching them how to worship Brahminical gods (Lee 2015, 173-79). The consumption o f beef (and not just carrion, as high-caste stereotype would have it [Lee 2015, 176]) was traditional among Dalits and had long marked their distinctive identity vis-à-vis H indus. The cow was sacred for all true H indus, after all, and the absolute taboo on consuming its flesh had been a key tenet o f H induism for millennia (though not in the Vedic period [Jha 2002]). The sacrality o f the cow took on additional impor � tance in the nineteenth century, when it became a mainstay in the assertion o f a pan-Indian H indu identity—and the simultaneous othering o f Muslims (Pandey 1983). I f H indus were to recognize untouchables as coreligionists, beef eating would have to stop. For Gandhi, the important thing was con� verting untouchables to H indu practices, rather than attempting to force

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E 1 3 9

Hindu beliefs upon them. Thus in response to a Christian who asked whether, on the basis o f their beliefs, untouchables should perhaps be classi � fied not as H indus but animists, Gandhi replied, “ We have never bothered ourselves with their beliefs, animistic or otherwise. Superstitions and unde � sirable things go as soon as we begin to live the correct life. I concern myself not with their belief but with asking them to do the right thing. As soon as they do it, their belief rights i tself” {C W M G 32:262). Gandhi here acknowl � edges that untouchables might not be H indu at the level o f belief. But for Gandhi the untouchables were nevertheless H indu simply by virtue o f the fact that they were Indian. W hat they believed, including what they believed themselves to be, was irrelevant. H induism was defined by Gandhi as the “ancestral religion” o f all Indians, and the untouchables were also therefore H indu unless and until they professed adherence to a non-Indic faith {C W M G 35:167; cf. Bauman 2011). It was not that instructing untouchables in properly H indu beliefs was in fact irrelevant to Gandhi. As Joel Lee’s research reveals, much o f the H SS ’s practical operations revolved around instructing untouchables in normative H induism, a fact the organization kept largely hidden from sight (2015, 172-74). But the most important way to change their beliefs, according to Gandhi, was to eliminate non-Hindu behavior and initiate them into the practice o f worshipping H indu gods. “Kneel down, move your lips in prayer . . . and you will believe,” as Blaise Pascal is reported to have said (Althusser 1972,167).

Gandhi also hoped that eliminating practices like beef eating and alcohol consumption, and instructing untouchables in “ hygiene,” would make them more acceptable to caste H indus (Prashad 2000 ,121-22; Lee 2015,175-78; cf. Bayly 1998,120). This brings us to the second o f Gandhi’s significant depar � tures from Shraddhanand: the extent to which his reform efforts were focused on caste H indus themselves, and on persuading them to eliminate untouchability from their hearts and to accept Dalits as fellow H indus. Dalits’ alleged filthiness and love o f beef were among the most significant barriers to this. Shraddhanand too had sought to persuade H indus to give up the practice o f untouchability and to embrace the untouchables as coreligion � ists. But this aspect o f his program enjoyed only very limited success, largely because Shraddhanand had required H indus actually to share meals with untouchables and to permit bodily contact. For most H indus this proved too much. Where Shraddhanand had insisted on eliminating the practice o f untouchability, Gandhi proclaimed the elimination o f untouchability as a bodily practice irrelevant. W hat he sought to change, rather, were H indu

1 4 0 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

attitudes. Shraddhanand’s more uncompromising approach limited his reach to, at most, a few dozen H indus at a time. Gandhi’s goal o f attitudinal adjust � ment, by contrast, could be pursued on a far greater scale. Through the medium o f his flagship English-language publication, Harijan, Gandhi preached on a weekly basis to an audience that included, potentially, India’s entire English-speaking elite.

It is worth reiterating this point, because the truth o f it is so surprising: Gandhi’s message to his fellow caste H indus was not that they must actually touch the untouchables but simply that they must cease to regard them as polluting “others.” H is position on this matter was consistent and honestly expressed:

W h a t I am a im ing at is not every H in du touching an untouchable, but driv � ing un touc h abil i ty / r tw his he a r t . . . . In ter-d in ing or in term arrying is not the poin t . I may not dine w i th you, but I ought not to ha rbour the fe e l i n g that i f I d ined with you I should be pollu ted . . . In the programme o f the H ari jan Sevak Sangh we don ’t ask orthodox H indus to inter-dine or in term arry with “untouchables.” (C W M G 64:34 , my emphasis)

In terdrink ing, in terdin ing, in term arrying, I hold , are not essential for the promotion o f the spirit o f dem ocracy . . . and I decl ine to consider it a sin for a man not to dr in k or eat w i th any and everybody. ( C W M G 18:85)

Insistence upon in terd in ing as part o f the programme o f promotion o f fel � lowship, in my opin ion , retards the growth o f goodwill by raising false issues and even false hope. W h a t I am trying to remove is the idea o f po l lu t ion and super ior i ty . . . . Self-imposed restrictions [on interdin ing] have a sanitary as also a spiritua l value. (C W M G 26:569, my emphasis)

Residents o f Gandhi’s ashram and members o f his inner circle were famously held to a higher standard. Unlike the mass o f H indus, they were required actually to expose themselves to the untouchable’s touch and were made to perform ritually polluting labor, like cleaning their own toilets. But such activities were presented as a form o f spiritual exercise, which only a miniscule Gandhian elite was ever expected to take up. “ In my Ashram . . . one o f the ‘untouchable’ inmates dines with the rest without any distinction. But I do not recommend anybody outside the Ashram to follow this example” (CWMG 25:512). For the mass o f H indus, what mattered to Gandhi was sim� ply that they give up the attitude o f untouchability. Gandhi’s paramount goal, he never tired o f repeating, was to persuade them to embrace untouchables as Hindus. So long as they continued to recoil inwardly from the untouchables

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • I 4 1

in their hearts, they could never be persuaded to regard them as being, as Urvashi Butalia so aptly put it, “part o f the H indu community, part o f us’ ” (2000, 235).

Maintaining that the untouchables were, and always had been, an integral part o f the H indu community, Gandhi overcame another key weakness o f Shraddhanand’s program. For i f the division between untouchables and Hindus was “only seeming”—that is to say, i f it existed only in caste H indus’ minds and not in reality—Shraddhanand’s focus on conversion was not just superfluous but gave entirely the wrong message. In saying that untouchables required shuddhi (purification) to be counted as H indus, Shraddhanand gave credence to the view that they were not fully H indu already (Lee 2015,175). This conceptual break with Shraddhanand over conversion afforded Gandhi a powerful new weapon with which to attack Christian and Muslim rivals for untouchable converts. It allowed him to characterize conversion itself as a one-sided act o f aggression by proselytizing faiths against H induism. A nd by distinguishing his own mission to the untouchables as being a mission not o f conversion but merely o f “upliftment,” Gandhi could portray conversion itself as politically motivated and spiritually bankrupt, as disruptive o f the social order. Even as antinational.

Consider the following examples, in which Gandhi presents the threat o f Dalit attrition from H induism as a threat to the whole o f India.

Transfer o f H arijans from the H indu fold to some other, no mat ter by wha t name it is ca lled [i.e., conversion or separate electorates] . . . must mean frat � ric ide---- [It is] an attempt to cut the H in du body in tw o____A n d it w i l l be an evil day for unhappy India i f such a ca lamity descends on her. (CW MG 63:234, my emphasis)

[Missionaries work ing among H arijans] do harm to us. They do harm to those amongst whom they work , and they do harm amongst whom they do not work , i.e. the harm is done to the whole of India. (CW MG 64:98, my emphasis)

I f the leaders o f [the C hrist ian and M usl im] religions in Ind ia ceased to com � pete with one another for enticing the H ari jans in to their fold , it would be we l l for this unfortunate country. (CW MG 63:235, my emphasis)

In these and many other instances Gandhi wrote in such a way as to persuade readers that by luring untouchables—untouchables who, he consistently implied, belonged to the H indus— Christians and Muslims did not just weaken Hinduism vis-à-vis other communities but in so doing harmed India.

1 4 2 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

India’s vulnerability to this manner o f attack by non-Indic religions was due, once again, to untouchability. H indus were therefore called upon to expunge untouchability from their hearts not just as a spiritual duty but as a patriotic one too. Here is one last example from the pages o f Harijan, in which Gandhi identifies the interests o f the nation with those o f the H indu community, making plain for readers that untouchability was a matter o f life and death o f the nation: “ The untouchability question . . . is one o f life and death for H induism . As I have said repeatedly, i f untouchability lives H induism perishes, and even India perishes” (CW M G 64:33).

Gandhi often characterized untouchability as “a poison.” Eradicating that poison was essential not merely to the internal well-being o f H induism but to its power to defend itself in a world filled with enemies. “So long as the poison o f untouchability remains in the H indu body it will be liable to attacks from outside”; the division between H indu and H arijan was a weak � ness in the H indu fortress, which “will be proof against . . . attacks only when a solid and impregnable wall o f purification is erected in the shape o f com� plete removal o f untouchability” (C W M G 63:210).

W ho were the enemies that threatened Hinduism? A t the Round Table Conference Gandhi presented the threat to H induism and the nation as coming from two quarters. O n one side were the British, India’s colonial rulers. O n the other stood Ambedkar, who, by questioning the moral and social basis o f the H indus’ claim over Dalits, was portrayed by Gandhi as aggressively threatening to “vivisect” H induism (C W M G 49:191) and to “cut the H indu body into two” (C W M G 63:234).18 But what the conversion issue teaches us is that questions o f caste and o f communalism are deeply entwined. H ad Ambedkar succeeded in separating the untouchables from Hindus, this would have rendered “the H indu body liable to attacks from outside.” A situation Gandhi most commonly presented as involving just two enemies, the British and Ambedkar, was occasionally acknowledged to involve a third: the Muslim . In a private conversation with his ally Vallabhbhai Patel, recorded for posterity by G andhi ’s personal secretary Mahadev Desai, Gandhi confided his fear that “the separate electorate will create division among H indus so much that it will lead to blood-shed. ‘Untouchable’ hooligans will make common cause with Muslim hooligans and kill caste-Hindus” (Desai 1953, 301). Gandhi is best remembered for his efforts to cultivate Muslims as a key ally in the struggles against foreign rule from 1916 to 1922. But his interest in H indu-M usl im unity waned as Muslims became a less important ally with the breakdown o f the Khilafat

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • 1 4 3

movement in the early 1920s (Adcock 2014, 147-48 ; Gould 2004; M etcalf and M etcalf 2006 ,180-81).

This much is well known. That Gandhi should actually have sought to strengthen one o f India’s religious communities at the expense o f others, however, will come as a surprise to many readers. And it is more surprising still that he should identify the interests o f his favored community, the H indus, with that o f the nation as a whole. That is because most o f us, whether Indian nationals or foreigners, have imbibed the image o f Gandhi presented by Richard Attenborough’s 1982 biopic Gandhi, in school text � books, and by Gandhi’s countless hagiographers. In these sources Gandhi is held forth as a symbol o f a religiously inclusive nationalism and as a bulwark against the H indu chauvinism o f the Sangh Parivar. This is the Gandhi who in a much-quoted passage o f H ind Swaraj proclaimed, “India has ever been such a country [in which] there are as many religions as there are individu � als. . . . I f the H indus believe that India should be peopled only by Hindus, they are living in dreamland. The Hindus, the Mahomedans, the Parsees, and the Christians who have made India their country are fellow country � men. . . . In no part o f the world are one nationality and one religion synony� mous terms: nor has it ever been so in India” ([1909] 1997: 52-53). Gandhi composed these words in 1909, as he traveled by steam from London to South A frica. A rather different message was conveyed on a weekly basis to readers o f H arijan and in other official pronouncements throughout the critical final decade o f India’s independence struggle. A t this time the integration and retention o f Dalits within H induism were presented repeatedly as a matter o f vital national interest, as a precondition for Indian independence. Thus the otherwise inexplicable tempest over conversion that periodically erupts in India’s national media, with which this chapter began. Thus the steady stream o f prominent regional and national leaders—up to and including Prime Minister Indira Gandhi herself—who in 1981 arranged personal meet � ings with a group o f Dalits in the remote Tamil village o f Meenakshipuram, Dalits whose sole claim to fame was their en masse conversion to Islam.

Rather than presenting the untouchables’ H induness as a proposition, the truth o f which requires argumentation and evidence to establish, Gandhi’s writings throughout the 1930s typically insinuate it as a self-evident premise in an article whose overt purpose is to make some entirely different point. O ften this involves vil ifying those who refused to accept that only Hindus should be permitted to bring Dalits into their fold—that is, Muslim or Christian proselytizers. Here are a few examples. “The present [Christian]

1 4 4 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

effort,” Gandhi informs readers o f H arijan, “ is to uproot Hinduism from the very foundat ion .. . . It is like an attempt to destroy a house. . . that has served [the Harijan] and his ancestors for ages” (C W M G 64:441). This passage is typical for the multiple messages it encodes. Most obviously it represents proselytism not as a legitimate expression o f the Christian faith, as missionar � ies themselves claimed, but as a strategy motivated ultimately by a desire to destroy H induism . More subtly, by describing Christian overtures to untouchables as an attack on “ Hinduism,” Gandhi reinforces his underlying contention that they are in fact H indu. In another such article Gandhi announces he has in his possession a pamphlet that Muslim proselytizers have been caught distributing among untouchables. W ithin the pamphlet, “everything held sacred by H indus is caricatured,” Gandhi tells readers, “so as to excite [in Harijans] disgust towards H induism” (C W M G 63:209). This is not surprising. Propaganda aimed at tearing down religious rivals was distributed not only by Muslims but also by H indus in great quantities throughout the entire decade during which Gandhi wrote the above words (Gould 1004 , 35-130). But Gandhi does not present the Muslim pamphlet as a routine salvo in an ongoing propaganda war between two proselytizing religions. O n the premise that only Muslims engage in the spiritually vulgar practice o f proselytizing, Gandhi here portrays the pamphlet as a one-sided act o f Muslim aggression against a tolerant Hinduism. H e then informs readers o f H arijan that the pamphlet also “ holds out material hopes to H ari jans. . . i f they are ever tempted to forsake their ancestral faith” (C W M G 63:210). Once again, Gandhi writes in such a manner as to imply that untouchables’ “ances � tral faith” and “Hinduism” are one and the same. This message is conveyed merely by-the-by. More overtly, the article conveys another standard refrain in Gandhi’s public statements on conversion: the accusation that H indus’ rivals for untouchables’ attention use worldly goodies to lure simpleminded untouchables. In one fell swoop Gandhi thus impugns the motives o f untouch� able converts and their converters, who by employing irreligious means reveal their own true motives to be political and not, as they claim, religious. In this instance Muslims stand so accused. But Christian missionaries, too, are repeatedly reported to “ dangle earthly paradises in front o f [untouchables] and make promises to them which they cannot keep” (C W M G 64:18).

Gandhi did not deny that Christian missionaries sought also to appeal to untouchables’ minds and hearts. The problem, according to him, was simply that the missionaries’ attempt to convey spiritual teachings to such people was a farce because it aimed altogether too high. Scoffing at Christians’ claim that

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • 1 4 5

untouchables too experienced spiritual hunger and not just bodily, Gandhi recorded a dialogue in which he informed one well-meaning but naive Christian, “The poor Harijans have no mind, no intelligence, no sense o f dif � ference between God and no-God” (C W M G 64:18). This was no mere slip o f the tongue. The inability o f the “poor Harijan” to grasp higher truths was a theme Gandhi often returned to. Thus in another exchange a visiting American Christian is presented as approaching the Mahatma in a most obse � quious and praising manner, saying, “You have put your life-blood into [Harijan uplift], you have suffered and triumphed, and I want you to help me to a profound understanding o f what the issues are and tell me how I may help, for I do not want to hinder” (C W M G 64:33) J 9 Christians had in reality been working for untouchable uplift for nearly ha lf a century by the time Gandhi got involved, and his methods were borrowed largely from them. But Gandhi disclosed no hint o f this history to his readers; he consistently presented uplifting the untouchable as his own original idea and spiritual property.

Gandhi replies to the visiting American evangelist by telling him to stay well away from the H arijan and leave their uplift to the H indus. For the entire Christian missionary project, he explains, is built on dubious founda � tions: “You made an appeal to those who have not even the mind and intel � ligence to understand what you talked; they have certainly not the intelli � gence to distinguish between Jesus and Muhammad and N anak and so on” (C W M G 64:35). Christians should therefore give up any thought o f conver � sion. But should they not nevertheless at least try to convey Christ ’s teach � ings, the American visitor asks. “ Would you . . . preach the Gospel to a cow?” Gandhi replies, “ W ell some o f the untouchables are worse than cows in understanding----They can no more distinguish between the relative merits o f Islam and Hinduism and Christ ianity than can a cow” (C W M G 64:37). Apparently this last comment struck even some o f Gandhi’s own followers as unfair, and he was forced twice to defend himself in print. In one instance he assured readers that “there could be no offense meant to Harijans, because the cow is a sacred animal” (C W M G 64:218); in a second he explained that his own sixty-eight-year-old wife and five-year-old grandson were also cow� like in their incapacity to understand spiritual truths, though both, like the H arijan , were “objects o f tender care and affection” for him (C W M G 64:440). H aving clarified that no insult had been intended, Gandhi reas � serted his position that “ it is a travesty o f religion to seek to uproot from the H arijan’s simple mind such faith as they have in their ancestral religion and to transfer their allegiance to another” (C W M G 64:440).

1 4 6 � R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

Let us take stock. W hen combined with charity, proselytism illegitimately lures untouchables with worldly temptations; appeals to the H arijan’s spir � itual understanding, on the other hand, are illegitimate because he has none. But that is not all. The very attempt by Christians and Muslims to transmit their faith to others is portrayed as intolerant and harmful, in itself and irre � spective o f method. This is one o f Gandhi’s subtlest and most difficult argu � ments to grasp, as we will see. O ften he conveyed such lessons by way o f dia � logues in which he presented himse lf as successfully winning over some critic, typically a Christian missionary. In one such dialogue Gandhi leads a missionary nurse gradually to understand that conversion itself is a form o f intolerance. H e prepares her for this conclusion with one o f his more quotid � ian arguments. “ O nly the other day,” he informs her, “a missionary descended on a famine area with money in his hip pocket, distributed it among the famine-stricken, converted them to his fold, took charge o f their temple, and demolished it” (CW M G 61:47). The nurse does not doubt the veracity o f Gandhi’s story or ask why anyone would distribute cash to starving people rather than food. Gandhi s editor Mahadev Desai simply notes that “the lady seemed to be touched” by Gandhi’s story, and he quotes her in such a way as to suggest she accepts that Gandhi’s description o f missionary methods is accurate: “ We [i.e., Christians] should not take an undue advantage o f people in distress” (Desai 1935: 99).20 The nurse then asks, “But, Mr. Gandhi, why do you object to proselytization as such [i.e., even when no undue advantage is taken]?” The dialogue ends with Gandhi offering the earnest young woman a lesson in the proper understanding o f Christianity, which, he informs her, would forbid such practices as proselytization: “I f Jesus came to earth again, he would disown many things that are being done in the name o f Christ ianity” (CW M G 61:47; see also CW M G 64:98). Why? Because proselytism implies the belief that you are right and others are wrong. “You condemn a large part o f humanity,” Gandhi explains, “unless it believes as you do” (CWMG 61:47).

Here we arrive at Gandhi’s most challenging teaching: proselytism is in and o f itself a form o f “ intolerance” (CWMG 34:261). It may seem strange that Christians and Muslims, merely by believing they are right and others wrong, and wanting to convey the truth as they see it to others, should be accused o f intolerance. But this is a perfectly logical inference to make, once we accept two basic propositions o f Gandhian thought. First, all religions are equally true. Second, whatever religion a person is born into provides the permanent and immutable source o f his or her authentic identity. In the first

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E • 1 4 7

Gandhi adopts a version o f the familiar modern idea o f cultural relativism. “A ll the great religions o f the world are true,” he tells us, and “A ll the great religions are equal” (C W M G 64:20 and C W M G 61:457). A l l are true, but each one is true in its own way. A l l are equal, but they are not equal in all respects. Each faith has its own unique value system, and each is superior by the standards o f its own sui generis moral vision. The unique superiority o f Hinduism, according to Gandhi, lies in its unparalleled tolerance and peace � ful nature, and in its capacity to encompass the virtues o f all the other faiths o f the world within itself: “ I have found [Hinduism] to be the most tolerant o f all rel igions.. . . Its freedom from dogma makes a forcible appeal to me inasmuch as it gives the votary the largest scope for self-expression. Not being an exclusive religion, it enables the followers o f that faith not merely to respect all the other religions . . . [but also] to . . . assimilate whatever may be good in other faiths” (C W M G 35:166-67). True conversion, according to Gandhi, means moral and spiritual development within the religion o f one’s birth: that is, adopting the good characteristics o f other religions, but with � out ever forsaking one’s own ancestral faith.

Attempts to convert people from one religion to another are a travesty o f true spiritual understanding, according to Gandhi, an understanding that H indus possess and adherents o f proselytizing religions lack. Proselytism is therefore not genuinely religious at all. It arises from what Gandhi character � izes as an irreligious spirit o f “competition” that betrays the ultimately politi � cal character o f Christian and Muslim motives (C W M G 27:205; C W M G 63:235; C W M G 64:73). Gandhi incessantly accuses other missions o f attempting to “compete” with him, as i f to imply that he is not also compet � ing with them, and o f elbowing in upon tur f that belongs legitimately to him alone.21

Attempting to convert others is not only irreligious, according to Gandhi, but dangerous: “[Conversion] is an error which is perhaps the greatest imped � iment to the world’s progress towards peace . . . [and] a potent source o f vio � lent quarrels leading to bloodshed” (C W M G 64:347). This brings us to the second o f Gandhi’s propositions. Religion, Gandhi emphasizes to readers, is the divinely ordained and immutable source o f both personal and collective identity. Conversion is therefore dangerous because it alienates people both from themselves and from their own authentic culture. We have already seen hints o f this in Gandhi’s portrayals o f conversion as an attack on not just H induism but Indian culture. But here he goes a step further. Conversion harms not only the cultural whole into which the convert was born but also

1 4 8 R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

converts themselves and, indeed, God. For in addition to all being true, all religions “are all God-given, and . . . necessary for the people to whom [they] were revealed” (C W M G 57:147, my emphasis). According to Gandhi, God himself has assigned each o f the earth’s peoples a religion uniquely suited its national character and situation. To reject the religion one has been assigned to is thus not only an affront to one’s ancestors but “a blasphemy against God and the se lf” (C W M G 17:205). A nd it is a dangerous thing to tamper with divine providence. As we have already seen, threats o f divine displeasure are illegitimate when used to persuade people to change religions, but not so when invoked in order to keep them in place.

“Religion is not like a . . . cloak that can be changed at will. It is a more integral part o f one’s self than one’s body” (C W M G 62:37). Just as “a thick woolen coat would be the thing for one living in the cold regions o f earth, [and] apiece o f loin-cloth for another living near the equatorial regions . . . it is a dangerous thing to present [Christian beliefs] to those who have been brought up [in] a different fa ith” (C W M G 61:456), and “ Though all soils have the same predominant characteristics . . . the same seeds do not fare equally well in all soils” (C W M G 64 :440-41) . W hat Gandhi is getting at here is simply the late twentieth-century notion o f cultural authenticity and alienation (Taylor 1992; Mamdani 2000; Kuper 1999). In cutting themselves o ff from their true cultural identity, Indian converts to non-Indic faiths are said by Gandhi to suffer psychologically, do violence to their country, and undermine relations with loved ones. “I see many Christian Indians almost ashamed o f their birth, certainly o f their ancestral religion, and o f their ancestral dress. The aping o f Europeans. . . is a violence to their country . . . [and it is a] crime to part with [a custom] when . . . that giving up would deeply hurt relatives and friends” (C W M G 28:92). Conversion will lead to “ blood feuds between Harijans and themselves.. . . [It will] set father against son and son against father” (C W M G 64:99). “It is the cause o f much avoid� able conflict between classes,” according to Gandhi, and “has meant the disruption o f the family” (C W M G 61:46). By dispensing humanitarian serv� ices with the “ulterior motive” o f converting, missionaries are misleading India’s “unsophisticated villagers . . . and destroying their social superstruc � ture” (C W M G 61:457-58). Proselytization is “the deadliest poison ever to sap the fountain o f truth” (C W M G 64:203, quoted in K im 2003, 33) and “will mean no peace in the world” (C W M G 64:20). Last, in holding out hope of social equality to Harijans, the missionary in fact destroys their only chance o f ever being free, because “ for Harijans there is no social equality, no real

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E 1 4 9

freedom anywhere except when it is first obtained in H induism” (C W M G 6 4:46).

Gandhi’s influence was profound. H is core contention that Dalits were and always had been H indu would become a matter o f law in secular India. As legal scholar J. Duncan Derrett explains, in Indian law “the test o f whether a person is a H indu . . . starts with ethnic and geographical tests, which raise a presumption that can be rebutted not by proof o f absence o f belief or pres � ence o f disbelief but only by proof o f exclusive adherence (or conversion) to a foreign (i.e. a non-Hindu) fai th” (1968, $z). Marc Galanter concurs, adding that “ heterodox practice, lack o f belief, active support o f non-Hindu religious groups, expulsion from a group within H induism—none o f these removed one from the H indu category. The individual could venture as far as he wished over any doctrinal or behavioral borders; the gates would not shut behind him so long as he did not explicitly adhere to another religion” (1971, 471 n.). The definition o f H indu in Indian law is for all intents and purposes identical to the definition famously proposed by the founder o f modern H indu nationalism, V. D. Savarkar, in his 1926 tract H indutva (Savarkar 1969). That the Supreme Court, with its reputation as a bastion o f secular liberalism, should have endorsed the H indu nationalist understanding o f Hinduness is presented as an anomaly by Ronojoy Sen (2007).22 But it is less surprising in light o f what this chapter has shown: that this was not just Savarkar’s understanding but also Gandhi’s.

I trust that by this point the reader will also have noticed in Gandhi’s arguments all the elements o f the India’s present-day national discourse on conversion, as outlined in the first section o f this chapter: namely, that con � version strikes at the heart o f both the individual soul and collective being, that it leads to social conflict, and that it is a threat to the Indian nation; conversion brought about by means o f material temptations is especially dangerous, because it deceives those judged not fully able to make such deci � sions for themselves, Dalits. W hether all o f these ideas are in fact original to Gandhi one cannot say, but they were not always a matter of common sense in India, and he worked assiduously to promote and establish them as such. A l l o f these arguments were picked up by the so-called Niyogi Commission Report o f 1956 (M PC M A E C 1957), which is recognized as the authority on which all India’s postcolonial anticonversion laws are based.

1 5 0 • R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E

Would Gandhi himself have endorsed a legal ban on religious conversion? The man died two decades before the first o f today’s anticonversion laws was enacted, and speculating on historical “what ifs” is an uncertain business. H e did, however, in a 1935 interview that was reproduced in the pages o f H arijan, say: “I f I had the power to legislate, I should certainly stop all proselytizing” (C W M G 61:46). This is admittedly at odds with Gandhi’s reputation as a trenchant critic o f centralized state power (Mantena 2012) and as a subscriber to the ideology “that government is the best which governs the least” (C W M G 62:92). Whether or not Gandhi really meant what he said in 1935, I believe this chapter has established that postcolonial India’s anticonversion laws depend on the arguments that Gandhi himself developed and tirelessly promoted throughout the critical final decade o f India’s independence movement.

I

R E L I G I O N , C O N V E R S I O N , T H E N A T I O N A L F R A M E » 1 5 1

F I V E

The Logic o f Slum Religion

H i n d u M E N i n a n b u n a g a r often poked fun at their slum Christian counterparts’ abstemious, unmasculine ways. “Look at him. No drinking, no smoking, no dirty words. H e thinks he’s Jesus!” Christian women were gibed for praying all the time, for their “ugliness”—that is, for not wearing flowers, makeup, or jewelry—and, most cuttingly, were said to dress like widows. “Look at her! She looks so sad, so p la in . . . . It’s like her husband’s croaked \kanavan cettu ponan pola \ .” H indus acknowledged the divinity o f Christ but suspected that he was a relatively ineffectual deity, whose demands on devo � tees, in any case, far exceeded the blessings he provided in return. Christians, on the other hand, maintained that their god was in fact all-powerful, whereas H indu gods were “mere stones,” idols made by the hands o f men, whose “powers” were figments o f the imagination. Christians accepted that H induism, like all religions, was an inherently moral enterprise but criticized the general laxity o f H indus’ ritual observances and the half-heartedness o f their bhakti (pakti), shortcomings they attributed to the fact that H indu deities, being mere stones, neither rewarded the good nor punished the wicked. Christian pastors, meanwhile, were often described as hypocrites by slum Hindus and were accused o f exploiting their poor congregants while they themselves grew rich on the “ foreign money” they were rumored to receive. Christians too made these complaints, though never about their own pastor, and pointed out that, in any case, the shortcomings o f individual pastors had nothing to do with Christ himself. They said H indu temples are the real scam. “A ll those gods care about is money, gold! . . . [But] Jesus has no need for that. H e is a real god—he cares for the poor.”

Despite this sort o f talk, collective hostilities between religious groups— called “communal conflict” in India—did not exist in Anbu Nagar and the

1 5 2

surrounding slums. They could not, because Christians and H indus were not distinct communities there. Slum dwellers identified individually as follow� ing one or the other o f these mutually exclusive faiths, but they lived within households in which, more often than not, someone else followed a different religion than they did. A nd all slum dwellers had close kin whose religious identity was different from theirs. Even parents and children could not be assumed to follow the same religion. O ften one or more children worshipped different gods than at least one o f their parents. This was not simply an ad hoc adjustment to the fact that neither “side” could compel conformity. Men and women both insisted religion was the individual’s choice, and neither tried to pressure the other to conform, even as everyone continued to insist that their own path was best. Suresh, a thirty-four-year-old married H indu man with one living son and a daughter who had died, gave his sister’s family as an example: “ My sister, like me, is a H indu. But she married a Christian and is living in her mother-in-law’s house . . . yet in that house she worships the H indu God. H er husband is a Christian, her children worship the Christian God. But she continues to practice the H indu rel igion .. . . It all depends on a person’s own wish.” “A nd yet,” he concluded, “each person claims their religion is best—I too will talk big about my religion!”

Religious identity in the slum, in other words, did not map onto any other social division around which collective interests could mobilize. This was true even at the level o f religious practice. Although all Anbu Nagar Christians worshipped in similar ways, they did not worship collectively except at the congregational level, and each congregation was an autonomous entity with no formal ties to other congregations and no supervening institutional struc � ture. Individual Christians, moreover, attended only one o f more than twenty mutually exclusive congregations, and it was not at all unusual for two Christians in a single household to attend different churches. W hile slum pastors sometimes informally met with one another and occasionally even cooperated on a limited basis, they mostly treated one another as rivals. As for H indu forms o f worship, these provided even less basis for collectivity than Christian ones. Most slum Hindus worshipped individually and in their own homes. When they visited temples or participated in other rites, they did so either individually or in informal and transient groups o f one or two friends or family members. There were no rites involving all or even most Hindus in Anbu Nagar. Certain events, like the consecration o f a refurbished temple in a neighboring slum that happened a year into my stay, drew large crowds. But worship here was aggregative, not collective—a clamoring mass of individuals,

T H E L O G I C O F S L U M R E L I G I O N • 1 5 3

each jostling with the next to see and be seen by the god, a personalistic exchange known as darsan (Eck 1998). Such crowds had no lasting social basis and involved no coordination or communication among participants.

The lack o f connection between religious affiliation and collective identity explains why collective hostilities could not arise along religious lines. But what explains this noncorrespondence between religious affiliation and col � lective identity in the first place? In Indian national discourse religion is assumed to be the very nexus o f collective being. W hat religion a person follows is also assumed to determine who one is at a deeply personal level. The people o f Anbu Nagar did not see it this way. Religion was supremely impor � tant to them, but it did not define who they were, their very selfhood. They did not, in other words, share Gandhi’s view that a person’s religion is “a more integral part o f one’s self than one’s body” (CW M G 62:37).

This helps explain why they were not ruffled by the sort o f interreligious barbs described above. Rough humor and mockery were a pervasive feature o f slum life and a source o f much enjoyment, as we have already seen. But when Christians referred to H indu gods as mere stone idols, they were not joking. They really meant it. So did H indus when they poured scorn on Jesus’s uniqueness and cast doubt on his alleged miracles. That the faithful in India are uncommonly sensitive, and that any real or imagined insult o f their religious symbols, let alone gods, will result in “wounded religious senti � ments,” has been a key premise o f Indian law and governance since the colo � nial period. The basic idea is that religion is a matter o f sentiment and primi � tive group pride, and that when Indians’ religious sentiments have been sufficiently wounded they cannot help but lash out violently against offend � ers (Viswanath 2012b).1 Yet slum Christians were little bothered by the pre � vailing H indu sentiment that their own supposedly almighty god was in fact a bit o f weakling, that praying to him for hours on end every day was simply ridiculous, and that the enthusiasm o f their Sunday services had more to do with mass hysteria than with the presence o f the H oly Spirit. H indus, for their part, simply smirked at Christian attempts to persuade them their own gods were in fact “mere stones” and at the ridicule Christians heaped upon them for wasting time and effort on them. For two reasons slum dwellers did not take any o f this personally. We know the first reason already: because religious affiliation was separated from both individual and collective self � hood, criticisms o f their religion really weren’t personal.

The second reason is a bit more subtle and has to do with how the people o f Anbu Nagar understood religious truth and the relationship between

1 5 4 • T H E L O G I C O F S L U M R E L I G I O N

truth and faith. Given their professed lack o f interest in controlling one another’s religious choices, why did everyone nevertheless “talk big” about his or her own religion and proclaim it “the best” ? I f the point was not to recruit new members for one’s team—because, so far as these slum dwellers were concerned, there were no “teams”—why talk like that? Accounting for their apparently contradictory behavior requires us to look in more detail at how they conceptualized religious truth. A nd this will mean unpacking some o f our own received relativism, a relativism that can be difficult to see fully because it is so deeply embedded not only in anthropological discourse but also in widely circulating discourses according to which “all religions are true” in some way.

This chapter’s task is to explicate the basic logic and normative structure o f religion as such, as it was lived and understood by the people o f Anbu Nagar. This account will answer the questions raised so far, while also provid � ing what readers need to make sense o f the following chapters, which focus specifically on Christianity and conversion. For want o f a better term I refer to the generic understanding o f religion that this chapter will detail—the shared conceptual and practical background against which H induism and Christianity appeared as alternative forms o f one and the same thing—as slum religion. Slum religion comprised the taken-for-granted assumptions about gods, humans, and the relationship between them that both slum H indus and Christians shared. Though I call this “slum religion,” I do not assume that such an understanding o f religion is confined to the slum. My aim is to describe what religion meant to people in Anbu Nagar; as far as they were aware, this was religion’s only meaning. Quite possibly, Dalits and the urban poor across India would agree, save those who have fallen under the influence o f H indu nationalist organizations or the H arijan Sevak Sangh. But whether that is so is an empirical matter, for others to investigate. W hat matters for my own argument is that religion, as it was lived and understood by the people o f Anbu Nagar, differed in several significant ways from reli � gion as constructed in national discourse and law. It also differed in a few noteworthy respects from standard scholarly accounts o f popular Hinduism.

Before laying out a detailed account o f slum religion, let us briefly go over the ground to be covered. The first substantive section o f this chapter describes how the slum’s residents conceived the relationship between reli � gious tradition and truth. For them there was but one truth. It was the same for all people and had no necessary link to any existing religious tradition. It might turn out that one tradition had gotten it right, but it was also possible

T H E L O G I C O F S L U M R E L I G I O N • 1 5 5

that none had. It was in everyone’s interest, however, to figure out what the truth was, and the people o f Anbu Nagar treated this as a matter o f ongoing and ultimately collective inquiry. In making a principled distinction between tradition and truth, they implicitly rejected secular anthropology’s cultural relativist position and neo-Hinduism’s pluralist ontology, which say all reli � gions are in some sense true. Their views also differed from standard inter � pretations o f Christianity as a revealed religion whose truths are guaranteed by the tradition that hands them down.

The next section focuses on the reason slum dwellers were so interested in “getting it right” vis-à-vis gods. They worshipped gods, not for their own sake or for some distinct spiritual goals, but for the worldly benefits they provided. Worldly interests are widely recognized as legitimate in popular religion across India, but in the scholarly literature on popular religion such interests are said to rank lower than “pure” (i.e., otherworldly or impersonal) ones. This was not the case in the slum. Residents not only felt no need to apologize for putting worldly interests first but even denied that otherworldly goals had any special value. Slum religion’s worldliness did not, however, mean worship was an amoral free-for-all. I f slum dwellers had a clear interest in “getting it right” vis-à-vis gods, gods too had interests. W hile the existence and potency o f different gods were a matter o f doubt in the slum, its residents agreed among themselves that one quality that all gods— even nonexistent ones— shared was that they were supremely moral and that they wanted human beings to be moral, too. The basic unit o f slum religion, a dyadic relationship between individual human beings and gods, was understood as an intrinsi � cally disciplinary one in which gods dispensed worldly benefits to humans, who were, in turn , required to participate actively in their own reformation.

W hat being moral meant, however, did not vary across religious tradi � tions. The residents o f Anbu Nagar regarded morality as universal and a mat � ter o f common sense, rather than the property o f any religious tradition. Thus their basic assumptions were at odds with the common anthropological understanding, according to which morality varies culturally and across dif � ferent religious traditions. W hat this meant in practice is the topic o f this chapter’s third substantive section. Here we see another way that slum reli � gion differed from standard accounts o f both popular H induism and Pentecostal Christianity. Existing literature says H indus regard gods as morally ambiguous beings or as following a morality that either differs from that o f humans or is simply incomprehensible by human standards. This was

1 5 6 • T H E L O G I C O F S L U M R E L I G I O N

not how the H indus o f Anbu N agar and surrounding slums saw their gods. For them gods were supremely moral beings, and the substance o f this moral � ity was identical for both gods and humans. Likewise, while scholarly litera � ture on Pentecostal Christianity says Christians everywhere regard their own god as supremely moral and see rival gods as diabolic, this was not the case with the Pentecostals o f Anbu Nagar and surrounding slums. According to them, H indu gods were not devils in disguise but supremely moral beings who, in principle, promoted one and the same morality as their god. Their deficiency was not moral but ontological; unlike Christ, the One True God, H indu gods did not actually exist.

The final section considers the relationship between faith and faithless � ness. A l l gods demanded faith from worshippers. To enter into a relationship with a god was not simply a matter o f mechanically following their ritual and moral precepts. Devotees were also expected to cultivate in themselves an attitude o f personal faith in their god, both by rendering themselves willingly dependent and vulnerable before them and by expressing total confidence that that god was indeed a reliable provider o f blessings and protection. And yet, as everyone knew, gods did not always do what worshippers hoped they would do, and for this very reason people routinely gave up on them. W hat was called religious conversion—a shift: from a H indu god to the Christian one (or vice versa)—must be understood against the backdrop o f slum dwell � ers continuously taking up and dropping gods. Despite claims o f unshakable faith, the reality o f slum religion was one in which persons moved continu � ally across and within religious traditions; H indu gods were just as likely to be dumped in favor o f other H indu gods as for the Christian one. W ith two exceptions, people in Anbu Nagar moved from god to god in what appeared to be a random walk. This is exactly what we would expect i f divine interven � tions in the world (“miracles”)— the raison d ’etre o f worship—were them� selves in fact random. The two exceptions were that women tended to stick with Christ more than any other god, whereas men who embraced Christ tended to abandon him at higher rates than they did other gods. Most likely this was because the prohibitions Christ imposed on his devotees were espe � cially onerous for men, in that they required renouncing traditionally male prerogatives such as drinking and tobacco. The unusual “stickiness” o f the Christ cult for women is addressed in chapters 6 and 7. The point I wish to make here is simply that despite professing an unshakable faith in whichever god or gods they were currently worshipping, Anbu N agar’s residents in fact routinely abandoned them. This was not a matter o f people saying one

T H E L O G I C O F S L U M R E L I G I O N 1 57

thing and doing another; at the level o f norms this was no contradiction, so far as slum religion was concerned. Although all gods demanded total faith within the relationship o f worship, none—including the god o f the Christians—ever insisted on being worshipped in the first place, and none were believed to punish apostates.

THEOLOGICAL REALISM

Once per year a certain pastor, from another Kashtappattinam slum, would come to Anbu Nagar, set up a screen in the middle o f the main street, and, as soon as the sun had set, show an old reel-to-reel film depicting the life o f Jesus. The film was a Western production dubbed into Tamil, and although everyone in Anbu Nagar had seen it many times, they would nevertheless fill the street, year after year, to watch it again. I had missed a screening early in my fieldwork period and was inclined to skip this year’s too. The film was quite long and seemed unlikely to contain any surprises. But as I stood watching people gathering in the street, Simbu, the leader o f a gang o f H indu youths, approached me with his friends. They wanted to know whether I was planning to watch the film and, i f so, whether I would sit with them. I didn’t agree right away but was interested to know what they made o f the event and why they thought this pastor came to Anbu Nagar every year to show a film about Jesus.

The purpose, they patiently explained, was simply to provide information. “It is so that we can know about the life ofjesus.” But why? W hy should the pastor want the people o f Anbu Nagar to know this? “You know,” Simbu shrugged, “ it’s like when social workers come and tell us about something. . . like savings schemes or family planning. O r when the city builds a school. It is all for education. That is their duty \katamai\, that is what they must do.” Likewise, according to Simbu, it was this pastor’s responsibility to show the film once per year—“Anyway, the people who gave it to him, most likely foreigners, would not want him to just keep it for himself, right?”

Simbu and his friends had answered all my questions and now wanted to know: Would I watch the film with them or not? They assured me it would be fun—the film itself wasn’t bad and there were one or two really great scenes. I admitted I hadn’t seen it before and said I would be happy to watch it with them i f they didn’t mind my questions. “ Don ’t worry,” Simbu said, “we’ ll explain everything.” W hat this turned out to mean was that they’d

1 5 8 • T H E L O G I C O F S L U M R E L I G I O N

provide me with a raucous commentary on what was happening in each scene and what was going to happen next. Cinematically, according to these young men, the movie’s best scene was undoubtedly the one in which Jesus was made to wear a crown o f thorns, savagely tormented, then nailed to a cross and killed; and they urged me to keep watching at least until this point. They remained undecided, however, as to which was the most important scene in the movie, and this, it turned out, was the real reason they had been so keen for me to watch it with them.

W hen the movie finally ended, the young men insisted that I linger for some time to discuss it with them. They were eager to know my reaction and, most o f all, whether I thought it was true. I interpreted this as their way o f asking me whether I was a Christian, and I patiently reiterated that I was not. They already knew this. W hat they wanted to know, they repeated, was whether I thought the information in the movie was true or not. The charac � ters in the movie were o f course played by actors, they explained, but was the story itself accurate? Though I am sure they would not have taken my word as final, they were quite interested in any added information or insight I might have: Was this the same story told in America? H ad I heard any alter � nate accounts? D id I think it was plausible? They asked me, for instance, whether I thought Jesus would have really let himself be treated this way. Would he really have let himself be humiliated and killed like that? In the course o f talking it became clear that their doubts were not about whether Christ was a god or even about his greatness. They were likewise entirely prepared to accept that someone had been crucified on the cross and then, three days later, had risen from the dead. W hat they doubted, rather, was whether that person—the man portrayed in the film—was actually Christ.

For the anthropologist, gods do not exist apart from the traditions that represent and worship them. Their personal characteristics and activities are whatever believers take them to be. It would make no more sense to ask whether the crucifixion story is really about Christ than to wonder whether the hero o f Arthur Conan Doyle’s famous novel series is really Sherlock Holmes. Yet for Simbu and his friends it was entirely possible not only that Christ was a god, even the One True God, but that the information their Christian neighbors had about him was all wrong. Indeed, this was precisely what they suspected. I f the stories they told about him were wrong, this would call into question their whole mode o f worship. Was it really necessary to give up smoking, to spend countless hours in prayer, and so on? Did Christ really demand that his devotees give up all other gods? Whether or not their

T H E L O G I C O F S L U M R E L I G I O N • X59

Christian neighbors had “got it right” was not the stuff o f idle speculation for these young men but a matter upon which securing a safer future might hinge.

We have here two fundamentally different ways o f thinking about reli � gious truth. O n one side is the basic assumption o f slum religion, best char � acterized as theological realism, insofar as they regarded gods as real beings whose characteristics—including their existence or nonexistence—were completely independent o f anything humans might happen to think about them. Human beings might have gotten it right, but they might also have it all wrong. The opposite, theological nominalism, conceives o f particular gods as intrinsically linked to the human traditions that worship them.2 This is anthropology’s default position, and it provides anthropological relativism’s underlying rationale. Unlike the people o f Anbu Nagar, anthropologists do not ask which religion (if any) is right, because anthropologically speaking all are equally true in their own terms. Gods are a cultural reality for anthro � pologists; they are a social fact, and as such are as real as anything else, so long as people worship and believe in them.

Philosophically speaking, these two positions leave no middle ground. But most people are not metaphysicians, and their attitude toward gods may comprise a complex mix o f ideas whose meaning and social force owe nothing to the canons o f professional philosophy. I would argue, however, that profes � sional anthropologists are theological nominalists by default, whether or not they have given the matter any thought. This is due not to personal predilec � tion but to the fact that the discourse o f anthropology is itself based on a naturalistic picture o f reality that excludes by definition the possibility that gods are a causal force acting upon the world independently o f human agency. And while few religious people would be likely to agree that what they call “God” subsists entirely in human representations and practices, elements o f the nominalist position can be clearly discerned in much modern religious discourse. It is evident, for example, in M .K . Gandhi’s neo-Hindu doctrine that “all religions are true” within their own sphere and for those who believe them. The anticonversion activist Swami Dayananda Saraswati similarly conceptualizes religious truths in cultural relativist terms when he argues that any attempt to dispute such truths is necessarily arbitrary (and therefore violent) because “on the basis o f reason, no non-verifiable belief is going to fare any better than any other non-verifiable belief” (1999).

This was not how people in Anbu Nagar saw it, and although the term theological realism was not in their vocabulary, I argue that it correctly

T H E L O G I C O F S L U M R E L I G I O N

describes their attitude. For them truth was not multiple; there was only one truth, even i f no one was entirely sure what it was. Maybe those who believed Christ was the only god were right and maybe those who believed he was just one o f many were. But they could not both be right. On this Anbu Nagar’s residents agreed. They took for granted that there was a fact o f the matter and that it was independent o f anything humans might have to say about it. For this reason, they took discussion and debate about religious doctrines both more and less seriously than do implicit nominalists, who link religion to tradition. They took such discussion seriously, insofar as they saw a point to it: they were deeply invested in figuring out which claims were true and which were not, and they were hungry for any form o f evidence or argument pointing one way or the other. A t the same time, because they did not take religious truth as consubstantial with the human teachings and traditions on which collective identities were (for others) premised, they did not treat dis � cussions o f religion on the model o f a debate team competition, whose aim was simply to defend one’s own position at all costs.

O n the contrary, because gods’ existence was treated as separable from the traditions and institutions gathered around them, all human ideas about gods could be subject to fundamental doubt and treated as contingent. Thus, even i f Christ should turn out to be the One True God, as Simbu and his friends made clear, this would not translate automatically into any worldly allegiance to self-proclaimed purveyors o f his Word. A nd for this reason as well, the people o f Anbu Nagar tended to take with a grain o f salt efforts by the state and politicians to interfere in such matters. O n August 27,2003, the government ofT amil Nadu announced that it would begin to enforce a mori � bund 1950 law banning animal sacrifice (S. Viswanathan 2003). The H indus o f Anbu Nagar rarely sacrificed animals to their gods, though they would have liked to. They considered it something only “very high caste” people did, because animals were expensive. “ W ho around here has that kind o f money?” one man asked rhetorically when I inquired about the practice. I asked whether he would sacrifice animals i f he had the means. “ W ho wouldn’t?” he replied. It was, after all, a highly effective way o f getting God ’s attention, and the payback was often significant. O n another occasion I discussed the ban with a H indu woman, Lakshmi. She scoffed at the idea that a politician could actually ban animal sacrifice. It was not that she doubted the state’s ability to enforce such a law. But by encroaching on the prerogatives o f a being infinitely more powerful than herself, the chief minister was playing with fire: “ In a fight between [Chief Minister] Jayalalitha and God, who do

T H E L O G I C O F S L U M R E L I G I O N • l 6 l

you think will win?” I wanted to know what she thought could actually hap � pen to Jayalalitha. “You can’t deprive the Mother o f food!” my friend replied, laughing at the thought o f it. “ IfJaya keeps up with it, [the Goddess] may just decide to gobble [her] up instead!” In other words, the chief minister would be brought to a horrifying end. Lakshmi was utterly confident, however, that the law had been passed “ just for show” and that the chief minister didn’t really mean it. “She’s not a fool, after a ll!” A nd sure enough, to the surprise o f no one, a few months later it was announced that the law banning animal sacrifice was no longer in effect. Pastors and lay Christians were simi � larly dismissive o f the anticonversion law. Though they resented it in princi � ple, and felt vulnerable on a personal level to the harassment it exposed them to, they were certain no power on earth could withstand the will o f God. “ They will all be washed away like ants,” Amudha, a Christian woman, declared, referring to the politicians who thought they could actually ban conversion.

Slum dwellers’ attitude toward divine power was, needless to say, very dif � ferent from the one evinced by H indu nationalists and Gandhians, who rely instead on the power o f the state to shield their “ancestral tradition” from competing notions o f truth.

WORLDLY IN TE RE ST S A N D D IV IN E DISCIPLINE

R O B E R T S : Please tel l me about Periyappalayat tamman .

K A L A I V A N I : She is a very good god and gives many blessings. I f I need something, I go to her.

R O B E R T S : W h a t d o y o u a s k f o r ?

K A L A I V A N I : I can ask for anything. M oney [for example].

R O B E R T S : A n d then what do y o u do?

K a l a I v a n I : I t depends. M ost ly I w i l l simply make an offering and go worship her \ k umpit ap o v e n } . . . . But i f I need some very big blessing, I must also pledge someth ing big, right?

R O B E R T S : For e x a m p l e ?

k a l a i v a n I : For example . . . Look here [showing a pla in string around her neck], I have no t a li [wedding meda ll ion],3 see? That is because after marriage for many years I was unable to get pregnant. I prayed and prayed, but st il l nothing. F ina l ly a neighbor told me: “ They say that i f you give your t a l i to Periyappalayat tamman , you w i l l get pregnant im m edi � ately.” So that is what I did.

I Ó2 • T H E L O G I C O F S L U M R E L I G I O N

R O B E R T S : A n d i t worked?

k a l a i v a n i : W ha t? You know I have four children! I became pregnant the very month I gave her my ta li.

Information, anecdotes, speculation, and rumors about gods’ relative potency and responsiveness circulated continuously in the slum and were a matter o f vital interest to those who lived there. This was despite the fact—or per � haps because o f it—that when it came to gods, everyone accepted that what was true and what was false were ultimately, intrinsically, and irreducibly uncertain. O r at least so they said when speaking in the abstract; when speaking about whatever god or gods they themselves were currently wor � shipping, these same people shifted immediately to the language o f total certainty.

Everyday talk o f gods revolved around specific things different gods were supposed to have done (or failed to do) for particular friends, neighbors, rela � tives, and so on, as well as the outstanding requests it was hoped they would soon fulfi ll . This was equally true with both H indu and Christian deities. More importantly, the blessings that Christians and H indus sought were the same: recovery from illness, success on an exam, a baby, a marriage for one’s daughter, a job, or a loan, as well as relief from marital strife, rel ief from heartache, or even the removal o f some personal flaw, such as a drinking habit or bad temper. In addition, Christians and H indus relied on their respective gods to protect them from misfortune and from supernatural attack (demons, black magic, ghosts, and so on). Hearsay and firsthand accounts about gods’ actions, or failure to act, were carefully weighed and discussed. Was some alleged miracle really due to divine intervention, or was it just a coincidence? Or, when a prayer went unanswered, was it due to some fault o f the worship � per’s, or was it a problem with that particular god? Some gods were more reliable than others. Some gods, furthermore, responded to some kinds of requests but not others, so the problem was neither with the god nor the worshipper but was a mismatch o f interests. These were the sorts o f questions potential worshippers focused on.

Asked how the two religions differed, both H indus and Christian resi � dents answered by explaining, often in exhaustive detail, the various forms o f worship required by H indu and Christian gods and the different sorts o f rules H indus and Christians were expected to follow in their daily lives. Here I juxtapose some typical examples o f how H indus and Christians in Anbu Nagar characterized these differences:

T H E L O G I C O F S L U M R E L I G I O N » 1 6 3

W e worship by praying and singing songs, H indus worship by pu t t ing f low � ers and sandalwood paste on id o ls . . . . C hrist ians must pray several times a day and are constantly studying the B ib le . . . . W e worship by doing pu jas, they worship by shouting “ H a l le lu jah .” . . . H indus abstain from eat ing meat on Tuesdays and Fridays, but Christ ians fast on Sa tu rd ays . . . . H indu gods love bright colors and flowers, but C hrist ians are not allowed to put flowers in their hair. They refuse to go to temples and avoid using foul language or chewingpd f ipd k k u [betel nut and acacia leaf].

They distinguished Christianity and Hinduism in terms o f their methods, in other words, not in terms o f any distinct spiritual vision or moral ideal. The purpose in either case was to obtain divine blessings and protection, o f an identical sort.

The basic unit o f slum religion was the hierarchical relationship between an individual human being and the divine.4 Human beings submitted to codes ofbehavior gods demanded and undertook whatever forms o f worship were thought pleasing to them. Gods, in turn, blessed and protected those who correctly worshipped and obeyed them. Gods took humans under their tutelage, not because they needed human worshippers, but because humans needed them. And because gods were intrinsically benevolent beings whose nature was to help those in need.

Gods helped humans mainly in two ways: by providing blessings and pro � tection and by leading them away from sin. In Anbu Nagar, people defined sin not primarily as rebellion against divine will but as harmful action. Though sins came in a great variety o f forms, their defining feature was the harm they caused to others, to oneself, or to the relationship between se lf and other. Stealing and murder were commonly cited examples o f the first; drug taking and other harmful habits, o f the second; promise breaking and cheat � ing, o f the third. Although to sin was also, necessarily, to rebel against God, what made a sin a sin in the first place was not that God opposed it but that it was immoral. The only partial exception was the sin o f breaking a specific ritual rule that a particular god had arbitrarily imposed on his or her own devotees, such as not eating meat on Tuesdays or Fridays. W hat made break � ing ritual rules a sin, even when the forbidden act was not in itself immoral, was that it meant breaking a promise to one’s chosen god. Thus eating meat on Tuesdays or Fridays was a sin only for those who had pledged not to do so.

W hile ritual requirements varied from god to god, the expectation o f morality was invariant. Although gods were liable to punish sinners whether or not the sinner was among their personal devotees, by pledging themselves

1 6 4 • T H E L O G I C O F S L U M R E L I G I O N

to a particular god and beseeching his or her support, people rendered them� selves especially vulnerable to that god’s punitive potential. By worshipping a particular god, a person in effect invited that god to monitor his or her behavior especially closely. In addition to the threat o f punishment, to sin was antithetical to worship at a more pragmatic level; while not all sins were serious enough to provoke instant punishment, sinful behavior was likely to undo whatever divine favor a worshipper had accrued through acts o f devo� tion. The relationship between humans and gods was thus an intrinsically disciplinary and reformative one, in which worship and obedience were offered and worldly benefits were received in turn.

That residents o f Anbu Nagar described worship as a frankly interested act is not unusual. O n the contrary, it is a commonplace in the scholarly literature on popular religion in India that most people, most o f the time, approach the act o f worship hoping for some worldly blessing. Where slum religion differed profoundly from popular religion as existing literature portrays it was in the normative priority o f worldly interests over spiritual ones. For a second main � stay o f this literature is that even those who accept the legitimacy o f worldly motives nevertheless acknowledge the intrinsic superiority o f disinterested worship (meaning worship undertaken with purely spiritual interests, or for the god’s sake alone) over the more common, self-interested variety. As anthro� pologist Christopher Fuller has observed in his authoritative synthesis o f this literature, while most H indus accept the legitimacy o f worldly interests in worship, and only an elite minority deny them outright, such interests are commonly seen as inferior to purely spiritual ones. Ideally, one should worship God for God ’s sake, as it were, and not for one’s own; in the spirit o f disinter � ested service, rather than for worldly benefits (Fuller 2004, 71-72). Thus a common thread in the literature Fuller summarizes is the idea that only “pure” acts o f devotion for God ’s sake alone are beyond reproach, whereas interested worship, though widely accepted in practice, requires justification.

Slum dwellers, by contrast, did not valorize purely spiritual worship. Nor did they betray any sense that worship undertaken for worldly reasons was inferior, much less that it required apology. Whenever I suggested it might be better to worship God “ for God ’s sake” rather than for one’s own, they assured me I was wrong. Gods want to help people, I was told. A t least one woman, I recall, was positively outraged by the suggestion that there was anything superior about disinterested worship. The very idea that people ought to put their own needs aside seemed to strike her as monstrous. It was as i f I had suggested that the needs o f the poor were themselves somehow

T H E L O G I C O F S L U M R E L I G I O N • 1 6 5

illegitimate. W hat kind o f god would demand that? Such selfishness was typical o f demons, but certainly not a god. As for the claim that some people worshipped purely “ for God ’s sake,” this struck people in Anbu Nagar as just another way for rich people to show o ff—a way o f bragging about the fact that they had no problems in their lives. People like that, they maintained, knew nothing about either God or what it meant to be human.

To say that slum residents considered worldly interests as valid as other � worldly ones, however, risks implying that otherworldly goals had more o f a place in slum religion than they did. In reality neither version o f slum religion (Christian or Hindu) posited any state o f existence—an afterlife, ecstatic union with the divine, perception o f a higher truth beyond the illusion o f phenomenal existence, and so on— in relation to which the concerns and cares o f this world might be weighed. Talk o f ecstatic union or piercing the veil o f illusion was entirely absent. The Brahminical theory o f an endless cycle o f rebirths, and the goal o f escaping from such a cycle, were utterly foreign to slum dwellers’ worldview (cf. Kolenda 1964). As for the afterlife, while resi � dents o f Anbu Nagar conceived o f it as potentially better than this one, nei � ther H indus nor Christians regarded this as the goal o f religious practice, or treated it as a significant reference point around which believers’ worldly and religious activities ought to center.

This is admittedly odd, especially because Anbu Nagar’s residents gener � ally regarded worldly existence rather darkly as filled with suffering. As a close neighbor o f mine, a casual day laborer, once told me: you should never wake someone unnecessarily, because sleep is the only time in this life any o f us are truly happy. The aim o f slum religion, however, was not escape from worldly existence but the world’s material and moral transformation.

P R IM IT IV E MORAL U N IV E R SA LI S M

S A G A Y A R A J : W h a t is the difference between magic and religion?

P A S T O R : M agic means destroying human l ife . . . [and] religion is about re leasing us from magic.

S A G A Y A R A J : But do a l l religions teach good things?

P A S T O R : Yes, [religions teach] only good things.

The people o f Anbu Nagar and surrounding slums valued gods mainly for the worldly blessings and protection they provided, and secondarily for the fact

1 6 6 • T H E L O G I C O F S L U M R E L I G I O N 1

f 'P

m

that gods helped them to become better people. I emphasize the former, because that is what they themselves emphasized when ta lk ing about their own religious choices and their reasons for worship. These slum dwellers’ lack o f interest in controlling one another’s religious choices was inseparable from the fact that neither slum C hrist ianity nor slum H induism sought to subor � dinate worshippers’ existing concerns—which were broadly the same for everyone and revolved around their ongoing effort to cope with extreme pov � erty and social deprivation— to some other, dist inct ly religious agenda. But what about moral transformation? H ere there was, for both religions, a clear effort to transform subjects’ existing impulses and habits. I f C hrist ianity and H induism promote different moral visions, might this not emerge as a poten � tial source o f con flict between the adherents o f each?

It is commonly supposed that each religion encodes or expresses a unique set o f values. This is often assumed to follow from the more basic anthropological intuit ion that religion is fundamenta lly connected to morality. For most anthropologists, as for most moderns more generally, morality is not objectively present in the world itself but is a human creation that varies by culture. Each culture, in this view, has its own morality, and religions contribute to the ritual and theoretical framework with in which persons seek to realize these cultur � ally specific moral visions. This relativist understanding o f religious morality is not limited to anthropologists but is a commonsense starting point in many secular discussions o f religious difference today. It is taken for granted by both secularist opponents o f anticonversion laws and their supporters. Recall the journal ist quoted in the previous chapter, who even in defending religious conversion acknowledges that “religious fai th sustains [believers’] lives in many ways, deciding not only which god they worship and how, but also helping them construct an internal value system, making social living cohesive, and impact ing the most important events in ordinary lives by laying down the dos and don’ts and the rituals that surround birth, death and marriage.” Religious conversion’s opponents, from G andhi to Saraswati, likewise take for granted that religions offer distinct interpretations o f moral truth. However, they argue that whichever religion a person is born into is best for him and that ripping a person from this context creates moral confusion . M ainstream Christ ians agree that morality varies by religion, but unlike Gandhians believe the suppos � edly unique moral vision o f C hrist— typically seen as centering on such values as love, forgiveness, and charity— is in fact best for all people.

It is possible to find support in slum dwellers’ own words for the view that the debate between riva l religions was indeed a debate over rival moral

T H E L O G I C O F S L U M R E L I G I O N • 1 6 7

systems. In the slum it was usual, especially among women, to seek and share informat ion about gods. A nd when these discussions turned, as they so often did, to the respective merits o f C hrist ian and H indu deities, C hrist ian women adhered to the standard line that theirs was the O ne True God , that H e was both generous and all-powerful, and above all, that H e was a loving

god who cared for the poor. H indus, for their part, acknowledged that C hrist did many good things for those who worshipped him , and they would readily pass on tales o f the blessings H e had bestowed upon their Christ ian friends and neighbors. But they treated Christ ians’ claim that theirs was the O ne True God as misguided and even risible. In contrast to the unrivaled potency that C hrist ’s followers ascribed to him , the general sense among H indus was that C hrist was a weak , though undoubtedly good-hearted, deity. H indus claimed that their own god was by contrast “a k ingly god” who would always punish those who had sinned, unlike C hrist , who “ just keeps on saying, ‘Forgive, forgive.’ ”

W hen slum Christ ians described their god as a caring god who “ loves the poor,” or when H indus responded by raising doubts about his potency, they might seem to be counterposing two dist inct value systems, the one based on love, the other on power. But appearances are deceiving. Christ ians did not deny the value o f divine potency or seek to give it a secondary status com � pared to divine love. O n the contrary, they insisted that it was because their god was the only god with any real power to intervene in human lives that they worshipped him . Power and love, moreover, were not opposed as two contrary values in slum C hrist ian discourse, or, for that matter, by slum H indus. The idea o f a trade-off between love and power might seem natural in a culture that regards compassion as a sign o f weakness, or in which emo� tional detachment is a manly virtue, but such associations were absent in the world o f the slum. There was no intrinsic reason that a god couldn ’t be both loving and powerful , and what I have described as H indu “ doubts” about C hrist ’s potency were precisely that; while they remained unconvinced by Christ ian claims, H indu slum dwellers were nevertheless prepared to con � sider the possibility that C hrist was everything he was made out to be (i.e., all-powerful).

In A nbu Nagar, the defin ing characteristic o f gods and the standard by which they were assessed was what they could do for their worshippers, as we have seen. This made the choice o f which god to worship an overwhe lmingly practical one and not an expression o f residents’ “ highest intuit ions” or “most sacred values.” Christ ians were every bit as concerned as H indus that the god

1 6 8 T H E L O G I C O F S L U M R E L I G I O N

they worshipped be a potent one. A n d H indus, no less than Christ ians, sub � scribed to the slum ’s care-based moral ethos. Even C hrist ian pastors, who were perhaps the only slum Christ ians w ith any vested interest in the goal o f spreading the faith, and who frequently praised their god for his loving and tender heart, never suggested that these qualities were the reason people ought to worship him . O n the contrary, they recommended C hrist precisely for the many benefits he bestowed upon those who praised and obeyed him.

W hen people in A nbu N agar said that C hrist loved the poor, or when they suggested that he might be a relatively “weak” god, therefore, they were not m aking moral claims. To say “ C hrist cares about ‘the poor ’ ” in the slum was to say something quite different from what these words convey in Europe or America, or, for that matter, in many churches in India. In such contexts these words would most natura l ly be understood as an expression o f Christ ian ethics, and as implying “ We too should love and care for the poor, the poor are worthy o f our respect, they are no less important in the grand scheme o f things,” and so on. But when one slum dweller said these words to another, what they were saying was “ H ere is a god who is especially likely to help people like us.” Conversely, when a friend or neighbor responded by rais � ing the possibili ty that C hrist was, nevertheless, a relatively weak god, her statement referenced the commonplace that one might be better o f f aligned with a more powerful deity, even i f that deity did not have a special commit � ment to the poor as such. Thus, however we ll liked and admired Jesus was in the slum, this admiration did not translate into an equally widespread wor � ship; many who professed to greatly admire Christ , and even to feel affection for him , did not worship him .

A nbu N agar s residents shared the anthropologist’s assumption that reli � gion and moral ity were deeply entwined . W h a t they lacked was the idea that each religion had its own unique moral system; so far as they were concerned there was only one moral ity for all people. Though they knew that different people had different customs, and that tastes varied, they did not extend this insight to morality. The idea that there could be distinct moral systems pre � supposes not just a basic sense o f cultural relativism but the additional idea that moral ity i tse lf is merely cultural . A nbu N agar ’s moral universalism can be described as prim itive, not in the derogatory developmentalist sense, but in the sense that mora l ity i tse lf was understood as a prim ary reality, not derivative o f something else, such as divine w i l l or some complex interpretive exercise. Gods might differ in how they brought about moral behavior, and in the rigor with which they enforced part icular moral norms. The Christ ian

T H E L O G I C O F S L U M R E L I G I O N • 1 6 9

god, for example, was seen as being especially strict, even going so far as to forbid minor sins such as cigarette smoking and the use o f unclean words, even as he was also more inclined to forgive. But these were differences in method, not ends, and what counted as moral was the same for al l gods. C igarette smoking was equal ly immoral for all people, and a sin was a sin whether it was punished , forgiven , or simply ignored. Likewise , though C hrist was reputed to be especially attentive to the poor— thus exemplifying the prevai ling moral ideal o f care— this did not mean caring for the poor was a uniquely C hrist ian virtue or that H indu gods did not also support it in principle.

Religions were thus not just moral ly equal in slum dwellers’ estimation but identical. For this reason, they did not conceptualize the choice between Christ ianity and H induism as a choice between compet ing moral systems, nor was religious conversion i tse lf the subject o f moral iz ing discourse in the slum, as it is for both elite H indus and a great number o f C hrist ians worldwide.

This is surprising for at least two distinct reasons. First, moral ity is a cen � tral feature o f much C hrist ian missionary discourse, and the claim that Christ ianity offers a unique and superior moral vision is shared by both lib � eral and conservative Christ ians. For al l but the most theologically liberal Christ ians there is, furthermore, the additional idea that worshipping rival gods is a sin and that those gods are themselves evil imposters. This idea is generally seen as following from the O ld Testament prohibit ion on idol wor � ship. Pentecostal Christ ians typica lly go even further, holding not only that idol worship is a sin but that other religions’ gods are l i tera l ly devils or demons (Caplan 1989; Meyer 1999). W hy slum Christ ians differed from their coreligionists worldwide in not seeing C hrist ianity as moral ly unique, and from their fe llow Pentecostals in not diabolizing rival gods, requires some explanation .

The second surprising th ing about slum dwellers’ view that C hrist ianity and H induism were moral ly identical is that it appears to contradict most literature on popular H induism . A ccording to this literature, the majority o f popular deities are moral ly ambiguous beings whose worship has nothing to do with morality. W h ile H induism is rightly hailed by most observers as conta in ing a h igh ly evolved eth ical discourse, deities are typica l ly not regarded as moral exemplars. A dded to this is the common claim that for H indus the demands o f moral ity are not universal but vary according to caste and station . This does not accord at al l w ith my own finding that both

1 7 0 • T H E L O G I C O F S L U M R E L I G I O N

C hrist ians and H indus in the slum regarded H indu gods as entirely moral beings and moral ity i tse lf as universal and not varying by caste; I therefore owe the reader an account o f exactly how slum religion differed from what has been previously reported, and my best guess as to why this might be.

W e have, then, two distinct sets o f problems, the first having to do with the discrepancy between slum C hrist ianity and Christ ians elsewhere, and the second between slum H indus and what has been generally reported about the relationship between popular H induism and morality. I start with the lat ter and then turn to Christ ianity . In discussing H induism I overlook the d ist inc t ion between so-called folk deities and those o f the high Brahmin ica l tradit ion because this is not a distinction my field subjects themselves recognized.

I begin with the idea that popular H induism is i tse lf a fundamenta lly amora l enterprise. The received view is we ll summarized by L . S. S. O ’Malley. Though written in 1935, his reference work on popular H induism expresses an enduring scholarly consensus on the amoral character o f H indu gods:

The idea o f religion enterta ined by the majority is not that o f the W estern theologian , to whom it connotes a code o f mora ls resting on theology. In the minds o f the ord inary vil lager there is no direc t connexion between religion and the mora l code. The gods do not come w i th in the mora l category. The func t ion o f the gods is not the direc tion o f morals but the distribution o f blessings and, i f not duly propit iated , o f curses___F ew enterta in the idea o f a G od who, loving righteousness and ha t ing in iquity, is grieved by breaches o f a mora l law whic h has been ordained for man ’s spiritua l we lfare---- The gods, according to them , are offended , not by sin, but by neglect. They are pleased by offerings and ceremonies rather than by repentance and a new life. Stress is la id on the v irtue o f such th ings as the continued repet it ion o f the name o f a god more than on the pract ic a l pie ty o f one who does good to his neighbour. Sin does not carry with it the idea o f wrong done to God , whose divine love w i l l be wounded , or whose anger w i l l be provoked , by wrongdoing. (1935, 69, 72-73 ; cf. Fuller 2004)

O ’M a l ley relies prim arily on missionary and colonial sources, but dismissing his remarks as merely the product o f missionary prejudice or W estern ethno- centrism would be wrong. H e has no doubts at al l about the moral ity o f even the simplest unlettered H indu . N or does he claim that elite H induism as such is bereft o f moral teachings, only that such teachings are not part o f the religion o f the masses. I f al l he were saying was that the moral common sense o f ordinary people was not derived from religious sources, the H indus and

T H E L O G I C O F S L U M R E L I G I O N • 1 7 1

Christ ians o f A nbu N agar would certa inly agree. N either group, however, would accept his idea that H indu gods themselves are unconcerned with moral order.

As discussed above, A nbu N agar ’s resident considered the promotion o f moral ity to be one o f the two main roles o f gods. The fact that not only O ’M a l ley but the majority o f scholars to the present day envision the gods o f popular H induism as moral ly ambiguous beings may originate in missionary theories, though it should be added that elite H indu commentators have also described the religion o f the unlettered masses in such terms. But i f scholars have been systematically wrong about this, their error cannot be explained as just elite prejudice. M y own argument is not that scholarly opinion is neces � sarily wrong to see the gods o f popular religion as amoral beings in general, only that this does not accurately capture slum religion (Roberts 2009; cf. V iswanath 2012a). There may, however, be more systematic reasons for why my own account differs from current scholarly consensus. The first involves a theoretical ambiguity in the commonly accepted model in which popular deities exist on a continuum with other supernatural beings, such as demons, as well as humans. The second, also theoretical, stems from the way moral ity itse lf is conceptualized. Both privilege the etic over the emic.

O ne reason H indu gods have been described as morally ambiguous is their alleged continuity with other beings, inc luding malevolent spirits and humans. This continuity is theorized in both psychological and ontogenetic terms. Psychologically, “Puranic” deities— those whose adventures are nar � rated in a medieval Sanskrit literary-mythological genre known aspuranas— have been described as morally ambiguous insofar as they exhibit al l the same psychological tendencies as imperfect humans. They are jealous and lustful, and they commit acts that are straightforwardly immora l by human stand � ards. This may be the case, but it is o f questionable relevance to slum religion because the stories in which gods display these traits did not in form slum dwellers’ basic ideas about divine morality. The inferences scholars make from puranic tales to theories about the amoral character o f H indu gods should be understood as an etic account. W hether such inferences are made also by natives (H indu or Christ ian) should be treated as an empirical matter, and in A nbu N agar and surrounding slums they were not. O ne reason may be that, al though puranic tales were certainly known to slum dwellers, they played no major role in their religious life.

W hereas scholars describe puranic deities as amoral because they exhibit all the foibles o f human psychology, they often see popular deities as continu �

1 7 2 • T H E L O G I C O F S L U M R E L I G I O N

ous with other beings at an ontogenetic level. These are said to have begun life as power-mad demons or the vengeful spirits o f dead humans (typically virgin suicides, sorcerers, wayward yogis, murder victims) who were subse � quently “ domesticated” through worship and thereby deified. Worship, in this theory, is motivated at least part ly by fear. W he ther this accurately describes how popular deities in fact come into being is not my concern. W hatever the historical realities o f their emergence, people in A nbu N agar did not understand them this way. H indu gods were not seen as “ deified malevolent spirits” by those who worshipped them, nor were they regarded as such by slum Christ ians. To slum dwellers, gods were eternal and wholly good. C hrist ians and H indus differed as to which part icular gods were real, but both took as axiomatic that gods were completely and absolutely moral. There was a clear dichotomy, moreover, between gods and all other beings. Though ancestors and even polit ic ians might sometimes be venerated in a manner similar to gods, when I suggested that this veneration amounted to worship I was promptly corrected. They honored the spirits o f their departed family members using ritual forms similar to those used in worshipping gods, they told me, but there could be no con fusion between ancestors or other beings and gods. In addition to being, at least in principle, all-powerful, gods were fundamenta lly unlike all other beings, according to slum dwellers: they were wholly moral, and all others were not.

A second reason H induism has been seen as disconnected from moral ity has to do with the way moral ity i tse lf is conceptualized. In the literature on H induism , moral ity is often described as context specific and the require � ments o f moral behavior as varying by caste and status. W h ile eating meat is said to be immora l for a Brahmin , it is perfectly permissible for other castes. Even beef is permitted for untouchables, though to consume beef is a grave sin for al l others. Likewise, gods may very well act in ways that are immora l for humans, but human standards are inapplicable to gods. N one o f this, however, is relevant to slum religion, because for the slum dwellers I studied moral ity did not in fact vary by caste or social status, and the moral ity they ascribed to H induism was universalist, not part icularist , in form .

This raises a key problem . For caste i tse lf is widely assumed—by scholars, by anticaste critics o f H induism such as B . R . A mbedkar , and by most H indus— to receive its theoretical rationale from H induism . Yet for those in A nbu N agar caste was virtu a l ly the paradigm o f immorality. H ow could slum dwellers, who regarded caste as evil nevertheless see H induism as moral? The answer lies in how they defined H induism , and religion in general—

T H E L O G I C O F S L U M R E L I G I O N • 1 7 3

namely, as a relationship between humans and gods. I f H induism were just whatever H indus did when they said they were ac ting as H indus, it would not be moral at all, because people acted immora lly all the time. But what H induism really was, was what H indu gods commanded, and they could not possibly endorse something so odious and patently immora l as caste. “God doesn’t see j a t i ” slum dwellers said, or, “Jd t i is a human creation— it was not made by God .” Significantly, not just H indus but also Christ ians in the slum, who had no vested interest in defending the H indu religion , emphatically denied any connection between caste and H induism .

This makes perfect sense given the common understanding among the people o f A nbu N agar that caste was ultimately a strategy for monopoliz ing resources and not a religious phenomenon at all. The fact that among the most important resources caste people sought to control was access to gods, whose temples were understood as divine powerhouses and the key to worldly prosperity (cf. Krishnamacharya 1936O], 8, quoted in Galanter 1971,476) , did not make caste i tse lf religious. That D a li ts were traditional ly excluded from H indu temples was not taken as evidence that the H indu religion supported caste. It was not God who barred their entry, after all, but selfish caste people! M y slum-dwelling D a l i t informants were equal ly unimpressed when I pre � sented them with the argument that caste hierarchies received support from H indu scriptures, and that these same scriptures gave pride o f place to Brahmins. “Scriptures?” one woman replied contemptuously. “ W ha t are the so-called scriptures?” Another woman I tried this argument on just laughed: “ W ell, maybe that’s what Brahmins say!” (cf. Berreman 1971, 23).

Let us turn now to slum Christ ians, who likewise affirmed the moral ity o f H induism and H indu gods. U nlike their Pentecostal counterparts else � where in the world, who diabolize their own former gods, slum Christ ians maintained a very clear distinction between the gods o f H induism and demons, devils, and other evil beings. H indu deities were not demons falsely cla iming to be gods but lifeless stone idols mistakenly viewed as such by igno � rant humans. In this respect, slum Christ ians’ views were more in line with those o f mainstream Protestants. But unlike mainstream Protestants, Pentecostals regard demons, devils, black magic and all manner o f evil spirits as very real presences and serious threats. Slum Christ ians believed in and feared the very same evil spirits as their H indu relatives and neighbors. H indus and Christ ians differed only in the methods they used to protect themselves from these harm ful forces. W here H indus would visit temples and mosques, seek specialists to employ countermagic, and pray to both

1 7 4 • T H E L O G I C O F S L U M R E L I G I O N

H indu and Christ ian gods for relief, slum Christ ians relied exclusively on C hrist and on the prayers o f their pastors and fellow worshippers.

Christ ians saw H indu gods as useless in this regard and classified idol worship not as sin but as error. H induism was not devil worship or magic, moreover, but exactly what it claimed to be: a religion . It might involve wor � shipping nonexistent gods, but they were gods al l the same, not devils. This was hardly what I expected, but slum C hrist ians were entirely consistent on this point . To question a god ’s powers made perfect sense in the logic o f slum religion, and doing so was in fact quite routine; the Christ ian posi tion that some gods had no powers at all was simply an extreme example o f this. To ask whether a god was really a devil, however, was completely o f f the map. “ Gods are gods, and devils are devils!” one Christ ian exclaimed, when I asked her whether she thought H indu gods might be devils in disguise. “A god can’t be a devil, and a devil can ’t be a god!”

The distinction between devils and gods was not an empirical truth in the world o f the slum but an analytica l one. A god might turn out not to exist. The problem with H indu gods, as slum Christ ians never tired o f repeating, was that they were noth ing at all. But to be a god, even a nonexistent one, was to be intrinsical ly moral and good by definition , whereas to be a devil was to be wicked .5

Significantly, not just ordinary believers expressed these views. They were equa lly upheld by pastors, many o f whom had some formal theological tra in � ing. So surprising was this to me that I sent my assistant Sagayaraj to speak privately with a range o f pastors, to see whether he could ferret out some moral criticism o f H induism that I could not. I prepared him with a series o f leading questions that sought, for example, to see whether pastors could be induced to describe H induism as ak in to magic or demonolatry. W ha t ensued was a series o f often comic back-and-forths, in which Sagayaraj’s fa i l � ure to find evidence o f moral antagonism helps flesh out what I have outlined so far about the logic o f slum religion . I quoted part o f one such an exchange, with a pastor named Yesudasan , at the start o f this section . H ere is more o f it:

s a g a y a r a j : W h a t is the difference between magic and religion?

P A S T O R : M agic means destroying hum an life . . . [and] religion is about releasing us from magic.

S A G A Y A R A J : But do a l l re ligions teach good things?

P A S T O R : Yes, [religions teach] only good th ings.

T H E L O G I C O F S L U M R E L I G I O N « 1 7 5

S A G A Y A R A J : Then why do some people say that certain religions do not tell us good things?

p a s t o r : A religion—okay, the Hindu religion. It is a different religion [i.e., but still a religion]. And in that particular religion what they say is “ To get this, give a hen,” “To get that, sacrifice a goat,” or “Apply an oil” . . . whereas in [Christianity] what we do is we pray. We pray Jesus’s name—there is only prayer for us, [because] prayer alone is successful. The Hindu is running around wasting his time.

S A G a y a r a j : Okay, then. But what about the basic moral principle? How is the Christian’s different from the H indu’s?

p a s t o r : [They are] not different! But he offers sacrifices—old-style sacrifices, Old Testament period sacrifices. He sacrifices, we pray.

S A G A Y A R A J : But R C —

p a s t o r : The Catholics too [like the Hindus] are still Old Testament religion. They still have rituals, sacrifices . . . bright decorative sacrificial places. Just recently, a few Catholics have made some progress, little by li tt le.. . . But only the New Testament reveals the truth. That is why the Old Testament was pushed aside!

Yesudasan’s comments on Catholic ism here echo ones pastors often stressed. In a Sunday sermon delivered just a few weeks prior to this conversation , for example, Yesudasan had mocked St. Anthony, the patron saint o f a nearby Catholic church, in terms identical to those he and other Pentecostals used for H induism : “ In that St. A n thony’s church . . . that A n thony is just a clay statue! Yet they c ling to that statue. They hold on to that statue and cry, ‘ O h, Anthony! O h, Anthony, save m e!’ . . . [Yet] it is just a doll. They polish it, repaint it, do alterations on it—but there’s noth ing more to it! H e doesn’t see your tears, he doesn’t hear your voice. C an A nthony hear? N o! H e w i l l not hear. Even i f you shout! Even i f you cry!” The pastor’s point about A n thony worship was that it was, like al l idol worship, useless. It was never described as sinful .

But the significance o f Sagayaraj’s in terview is the distinction it illustrates between magic and religion . For this distinct ion , unlike that between reli � gions, was a moral one: religion was good and magic evil. The difference between the two was thus not at all, as some sociologies o f religion have sup � posed, between instrumental and spiritua l goals. In the slum, magic and religion were equally instrumental. But where the former was regarded as inherently harm ful, the latter could be used only for good. Though both sought to channel supernatural forces for human ends, religion could be used only for ends that were themselves in keeping with the moral order. “ W hen

1 7 6 » T H E L O G I C O F S L U M R E L I G I O N

magic is involved,” as another slum resident once explained to me, “someone always gets hurt.” The pastor, like others in the slum, did not shrink from proc la iming his own religion’s superiority, but he did so via a logic wholly different from what my assistant had proposed. W here Sagayaraj had spoken o f morality, the pastor replied in the idiom o f efficacious action: “Prayer alone is successful,” “ The H indu i s . . . wasting his time.” In assimilating H induism to O ld Testament religion , however, the pastor shifted the terms o f distinc � t ion from a dichotomous identitarian logic built around H in du-C hrist ia n difference to one in which other Christ ians and H indus were alike assimi � lated into a single narrative, even a single family tree. Slum C hrist ianity ’s “others,” both H indus and C atholic Christ ians, were pictured as legitimate forebears, along with the ancient Israelites, in the story o f humanity’s divinely orchestrated movement from error to truth.

E V E R Y D A Y A P O S T A S Y A N D T H E D E M A N D S O F F A I T H

A ny given person in A nbu N agar worshipped, at most, a small hand ful o f deities. But when asked which gods they prayed to, H indus typically pro � claimed themselves devoted to “al l the gods.” They often proceeded to reel o f f a list o f examples— typica lly inc luding not just a wide variety o f H indu gods but also the likes of jesus, Buddha, St. Anthony, A llah , and so on. These were understood by slum H indus as be ing at some level just different names for one single God , and all residents spoke in different contexts o f part icular deities and also o f a single “ God .” W hen H indus prayed or performed other acts o f worship, however, they always directed their efforts to some part icu � lar, named god associated with some unique shrine or temple. This was because even i f al l gods were ultimately just different forms o f one and the same a l l-powerful God , in pract ice it mattered a lot which god one wor � shipped. Worship was always oriented to some part icular end, and for this purpose each god was a unique person. In their personal aspect gods varied in the part icular forms o f worship they required , in their propensity to respond to different sorts o f requests, and, indeed, in their powers.6 A l l gods had their specialties. In addition to being the O ne True God , for example, Jesus was also known as the god who spec ialized in “women ’s issues” (or “ family problems”).

A nyone ’s personal roster o f deities could shift and evolve over time. N ew gods were added; others were dropped. H indus, unlike Christ ians, typically

T H E L O G I C O F S L U M R E L I G I O N 1 7 7

had little to say about the gods they had abandoned. But i f asked directly they were quite open about which gods they had given up over the years, and it was common knowledge in the slum that gods were sloughed o f f just as often as new ones were taken up (cf. Srinivas 2012, 330). Because a person’s time and resources were limited, devotion was necessarily a zero-sum game. Choices had to be made. The main (and probably only) reason a god was dropped was that he or she had fai led repeatedly to respond to prayer requests or had unac � countably fa iled to protect a devotee from misfortune. The perpetual search for aid was also why new gods were adopted. Worship was serious business, and which god one invested one’s resources in might very wel l be a matter o f life and death, so far as slum dwellers were concerned. Gods were discarded, but never casually or without cause; new gods, likewise, were not embraced all at once or without their having first provided some tangible evidence o f both their capacity and their will ingness to meet the worshipper’s needs.

Earl ier I described the relationship between the individual devotee and his or her god as a d isc iplinary one, in which obedience and worship were exchanged for blessings and protection . I now flesh this picture out with an account o f an existential stance that defines that relationship’s subjective core: fa ith . In addition to “ keeping fa i th” by not breaking the specific ritual rules the deity had set forth for his or her followers, or the moral ones that all gods upheld, the relationship between human and god also required a certain “ leap o f fai th .” Slum religion placed great stress on rules, but beyond merely following rules, human worshippers were expected to demonstrate existen � tial commitment to their god. W orship was not just a periodic act but an enduring relationship that pervaded the devotee’s everyday life. I t was a dec � laration o f dependence (Ferguson 2013), in which human beings put them � selves at the mercy o f their god by rendering themselves w il l ingly vulnerable to harm and pledging to rely exclusively on that god for protection.

“I trust in God alone”; “ I fear noth ing in my life, because I know God wil l always protect me”; “A m m a protects me, because I trust in her alone.” These were typical, even formulaic, statements o f fai th among slum H indus, uttered routinely in the face o f pervasive uncerta in ty and whenever specific dangers loomed. Equivalent claims were central as we ll to slum Christ ianity . W ho was the intended audience for these declarations? Though they were spoken to fellow residents, the point was not to portray onese lf as either especially pious or worry-free. People knew each other too we ll for this. They knew one another’s personal fai lings and weaknesses. They also knew that, whatever anyone might say, life was indeed precarious and gods could not in fact be

1 7 8 • T H E L O G I C O F S L U M R E L I G I O N

relied upon . But declarations such as these were meant, not for human ears, primarily, but their gods’.

To simply declare one’s dependence on a god was not, however, enough. T a lk is cheap. Believers had to also perform their faith, both through specially designed ritual acts and through the observance o f ri tual protocols that gov � erned everyday life and sought to lim it risk through deliberate and selective exposure to it. Typical in this regard were the many potential ly dangerous ritual tests H indus periodically subjected themselves to, like fire walk ing, having one’s tongue, muscles, or other flesh parts pierced with large needles, or wa lk ing long distances balanc ing heavy pots on one’s head in the noonday heat. Possession rituals can also be seen in this light, insofar as they entailed rel inquishing control o f one’s body and mind and putt ing oneself fully in G o d ’s hands. A critical feature o f these acts, which may seem almost too obvious to state, was that the devotees performing them had to rely on their god alone to protect them. They could in no way hedge their bets. For exam � ple, when having their tongues pierced H indus had to not use analgesics, antibiotic ointments, and so on. Long walks in the sun were performed not only without shoes but also without shade or drin k ing water. The possessed person, likewise, could not resist possession or try to stay conscious to guard against crashing into obstacles. In al l cases, performers had to hurl them � selves fearlessly into the act.

W here slum Christ ians differed from H indus was that their leaps o f fai th were rarely dramatic or ritual ized— though diff icult fasts might be seen in this light, as might the fact that some C hrist ians eschewed pharmaceutical medicine in favor o f prayer alone. Primarily, however, Christ ians’ acts o f faith were o f a day-to-day variety; what Christ ians lacked in dramatic bodily feats, they made up for through an intensification o f the everyday fa i th that was common to slum H induism as well. For C hrist ians were required to forego al l the supernatural, astrological, and magical remedies that H indus nor � m a l ly relied on, in addition to their gods, to protect them from invisible threats. It is hard to convey to readers how frightening and all-pervasive these threats were for people in A nbu N agar, and how great a step will ingly forego � ing al l supplemental protections was.

Slum dwellers were keen observers o f omens and signs, and by their own account, the majority were obsessed with avoiding inauspicious and threaten � ing influences. The ordinary person’s world teemed with potential supernatu � ral danger. The number o f actual dangers was wide ly held to far exceed the known ones. Professional astrologers and folk knowledge were seen as

T H E L O G I C O F S L U M R E L I G I O N � 1 7 9

possessing only an incomplete picture o f the dangers and potential signs that existed, and residents constantly looked out for new ones. O ne man , for instance, discovered that i f he heard a certain word uttered before noon something bad would happen later that day. A nother became convinced that the same would happen i f his eyes fell upon a part icular person (his neigh � bors wife!) when he first left his house in the morning; he then took great care to avoid let t ing this happen , and when it did he was known to refuse to leave his home until dusk. These are admittedly extreme examples, and were described to me in mocking tones to by other slum dwellers. But they accu � rately characterize the pervasive sense in the slum o f world filled with danger � ous invisible influences. Such fear may seem irrational, but the rea lity o f slum life was that very bad things happened to people al l the time, and these

FIGURE IO. A n ac t o f fa i th .

1 80 T H E L O G I C O F S L U M R E L I G I O N

entirely visible misfortunes constantly reminded A nbu N agar ’s residents o f unseen threats assumed to lie in wait.

Thus simply to face these threats w ith no effort to avoid them except through C hrist was no mean feat. O ccasiona lly Christ ians denied such threats were real, but what most (including pastors) said was “ W e are not supposed to even th in k about those things,” and “ They cannot harm us so long as we place our fa i th exclusively in C hrist and do not fear them .” W orshipping the Christ ian god thus required not merely abstaining from alternative sources o f protection but strict m ental discipline. For as long as one’s mind remained tota lly focused on C hrist , Pentecostals said, neither worldly nor supernatural threats could harm you. The latter could do great damage— but only i f you feared them. To al low even a moment’s doubt that one was safe in G od ’s hands could become a se lf-fulfil ling prophecy.

For this reason, perhaps, most slum Christ ians were reluctant even to discuss astrological and other such threats. “ I see only Christ ,” they typica lly replied when asked about their fears. O ften they illustrated with a character � istic gesture— hands first cupped on either side o f their eyes, then extended slowly into the space before them— suggesting a k ind o f w i l l fu l tunnel vision, a refusal to look either to the left or to the right. Ravi , a thirty-year-old con � vert who had done well for h im se lf sell ing scrap metal by the roadside, explained to me what it meant to rely tota lly on C hrist: “You drive a motor � cycle, right? W ell, i t ’s like when you ’re weaving through traffic and you see a gap between two cars. Everyth ing is happening very fast, and the gap is clos � ing. You don ’t know i f you can make it, but you can’t just stop and th ink , ‘ W ill I be able to make it or not?’ N o! You just go straight through. I f you falter, you w il l never make it through . That is what l iving with C hrist is l ik e— al l the t ime!” Readers w ith firsthand knowledge o f C henn a i ’s roads conditions, which are surely among the most hair-raising in India, i f not the world , w i l l appreciate R av i ’s metaphor. H is point was not that life with C hrist was full o f needless chance taking. L ife i tse lf was perilous in A nbu N agar, but by re lying on C hrist alone one could move boldly from success to success, free o f the debili tating anxiety central to slum residents’ lived experi � ence. For Christ ians to face their fears o f these threats was no abstract medi � tation, any more than was H indus’ self-subjection to physical pa in and risk. It was a bodily practice where believers put themselves in harm ’s way to dem� onstrate unswerving faith.

Through both words and action, then, slum dwellers declared themselves personally dependent on gods. In this way devotees sought to ensure that their

T H E L O G I C O F S L U M R E L I G I O N • l 8 l

gods would indeed provide for and protect them. Given such widespread dec � larations, it would be easy for a casual visi tor to overestimate worshippers’ actual commitment to their gods. But a correct understanding o f slum reli � gion requires going beyond people’s statements at any one time to consider the history o f their shifting relationships with deities and what they said about those left behind. The truth was that gods could not always be relied upon, as the people o f A nbu N agar well knew. W hile the protocols o f worship ensured that no one would ever express doubts about a god they were currently wor � shipping, once that relationship had ended they retrospectively admitted that doubts and dissatisfaction had been building for some time.

That worshippers could at one moment declare total confidence and satis � faction in a god and at the next moment unceremoniously drop that god from their roster may appear contradictory. But it is not. There is no contra � diction because their declarations o f dependence carried no implication that that relationship would endure no matter what. T erminating a relationship o f worship violated no norm in slum religion, and no one felt the need to apologize for or just ify such decisions. W hen A nbu N agar ’s residents pro � fessed total confidence in a god, it was not a promise o f perpetual loyalty. Such professions were simply part o f what it meant to worship a god in the first place. The requirement o f devotees to maintain fa ith with in the relation � ship o f worship was central to that relationship’s logic and entailed, among other things, speaking o f that god in a certain way. But the expectation that believers had to keep the fa i th within that relationship implied no require � ment to maintain that relationship in perpetuity.

Gods did not condemn or in any way punish devotees who stopped wor � shipping them. H indu gods did, however, punish those who voluntarily made, and then neglected to fulf i l l , part icular vows to them, and conse � quences for vow breakers could be quite serious (cf. Raj and H arm an 200 6; Fuller 2004). In extreme cases vow breaking could mean a child ’s death, the complete destruction o f one’s household, and the like. D isaster was normal ly averted by resuming the original vow and addit ional ly undertaking some act o f atonement. But what happened when a god was dropped entirely? For H indus, worship more often than not included pledging to observe a regular fast in some god ’s name, or tak ing an annual pilgrimage to their shrine. A n d one could not drop such a god from one’s personal pantheon without simul � taneously breaking one’s vows to him or her. Even i f there were no conse � quences for exiting the relationship as such, would not the vow breaking that exit entailed mean that apostasy was punished after all?

l 8 z • T H E L O G I C O F S L U M R E L I G I O N

For various reasons, the people o f A nbu N agar did not see it this way. First, they dropped a god only when that god had repeatedly proven unreli � able, and gods with no demonstrable power to help were not seen as a source o f harm either. Second, when H indus dropped one god it was only in order to embrace another, more reliable one. H aving thrown themselves at the mercy o f a new deity, they were henceforth under that god ’s protection and therefore had nothing to fear. Finally, because al l gods were in real ity just different aspects o f one and the same God , whatever harm might otherwise come for neglecting one aspect was not seen as applicable in the context o f a renewed relationship to the Supreme. This was no less the case i f that “aspect” happened to be the Christ ian one. T ak ing the ultimate un ity o f God seri � ously meant— for slum H indus, though not for al l H indus who claim to believe G od is O ne— that conversion could never entai l “ blasphemy against G od and se l f” (CWMG 17:205).

The Christ ian god was likewise not believed to smite apostates, or “ back � sliders,” as pastors called those who reverted to the worship o f H indu deities. The most serious consequence backsliders faced, according to pastors, was being deprived o f G o d ’s blessings and protection . Pastors also stressed that, without G od ’s active guidance, former believers usual ly reverted to bad hab � its like card playing and cigarette smoking and became mired again in quar � rels with neighbors and spouses. The prohibit ion on idol worship laid out in the book o f Exodus (20:3-5) was understood by slum residents not as a blan � ket moral condemnation o f the practice but as just one o f many ritual pro � scriptions— like absta in ing from astrological and magical remedies— that applied, not to al l humanity, but only to those who worshipped the biblical god. For those who wished to remain w ith in that relationship, idol worship was indeed a sin; for anyone else it was, as noted above, merely an error.7 Like any other such relationship, the relationship between a slum dweller and the C hrist ian god could be terminated at w il l . A n d when that relationship ended, the ritual proscriptions that governed it no longer held.

In sum , though worship was seen as an intrinsical ly moral practice in A nbu Nagar, human beings were under no moral obligation to engage in the inherently unequal relationship it entailed. They did so because it was very stra ightforwardly in their interest to. Gods in turn protected and blessed those who asked for their help, not because they had any need for worship, but because that was their nature. For in addition to their vastly superior power, what distinguished gods from al l other beings, according to the fa i th ful in A nbu N agar, was their inherent benevolence. Though gods might

T H E L O G I C O F S L U M R E L I G I O N • 1 8 3

punish wrongdoing and broken promises, the slum’s residents unanimously rejected the idea that they ever harmed people out o f anger or jealousy or on a whim . Such behaviors were typical o f demons, tyrants, and even the spirits o f one’s own deceased family members, but never o f gods, who were under � stood by residents o f A nbu N agar as entirely moral and in this respect fundamentally different from al l other beings, natural or supernatural.

1 8 4 • T H E L O G I C O F S L U M R E L I G I O N

S I X

Pastoral Power a n d the M iracles o f Christ

" W I T H T H E r i s e o f S L U M Christ ianity, a new form o f power took hold in the world o f the slum, insinuat ing itse lf into the fabric o f everyday life and subtly altering the terms o f moral community. Most notably this new power targeted relationships among women and those between man and wife. For some this power was none other than that o f Christ himself, a power that they themselves could tap into direct ly and that they credited for the various trans � formations taking place before their very eyes. For others, however, this power had a more recognizable face, that o f the pastor. O ccasiona lly praised, more often mistrusted , and in some corners condemned— by Christ ians no less than H indus— the pastor cut an ambiguous figure in the world o f the slum.

T ak ing slum dwellers’ own theories o f pastoral power as a guide, this chap � ter begins by examining the inst itut ional and spiritua l role o f the pastor and the sources o f slum C hrist ian ity ’s overt pastor-centrism. For the religion was organized and exclusively practiced with in churches that were the personal property o f the pastor who led them, and that were widely perceived as an extension o f his personality. Pastors themselves emphasized their own role in channeling divine agency as central to C hrist ’s work on earth, even to the extent o f giving the impression that it was their own spiritual powers— cul � tivated through ceaseless prayer and fasting— that shielded their congrega � t ions from supernatural attack and ensured collective well-being. W h ile slum men overwhelmingly endorsed this pastor-centric understanding, their fema le counterparts pa in ted a rather different picture. A c cording to Christ ian women , who made up 85 to 90 percent o f al l believers, it was their own direct relationship with C hrist that was central. Pastors played at best a marginal role in the miracles C hrist had worked in their lives and in drawing his love upon them.

1 8 5

These two ways o f conceptualizing spiritual power in slum C hrist ianity corresponded, in turn , to two distinct inst itut ional tendencies in the slum church and two bases o f social power. The first was overt and hierarchical and centered on the person o f the pastor. O f equal or greater significance, how � ever, were quasi-autonomous organizational networks among church women. These women’s networks drew little attention to themselves despite being far more active and interventionist. The interests o f the pastor and those o f his female followers did not automatically coincide. Yet the success o f each ulti � mately depended on that o f the other; the result was a strategy o f mutual accommodation and delicate operational harmony.

P A S T O R - C E N T R I S M A N D

T H E Q U E S T I O N O F C H A R I S M A

Each church in Kashtappattinam had not one name but two. The first was its official name, the name that was used in form al announcements and was painted on a signboard in front o f the church. Some examples were “Apostolic Liberation D ivine Assembly,” “Loving Prayer H ouse,” or “ O pen-Door Fu lly Evangelical Church .” Names such as these were not in common use, however, and even those who walked past a church every day might not be able to recall its official name unless they themselves were a member. Instead , churches were referred to as “ Samuel’s church,” “ C hinn a iyan ’s church ,” “ Yesudasan ’s c hurch”— or, more commonly , simply as “ Samue l,” “ Chinna iyan ,” “Yesudasan .” Thus “Yesudasan” was at once the name o f a part icular pastor and also o f his church, a church that he not only presided over but also owned, having purchased its physical structure from another pastor who had fallen on hard times. Yesudasan ’s congregation, however, was not purchased, nor were congregations ever transferred between pastors. The sole exception was when a pastor retired or died and handed over his flock to a designated successor, his son or a nephew whenever possible. Congregations were built up by pastors over years o f effort, but they were composed o f mem � bers who might come and go as they liked. Pastors struggled endlessly to keep their f lock intact, and a pastor who fai led to do so was no pastor at all.

That “Samuel,” “ Chinna iyan ,” and “Yesudasan” referred to both a man and his church was fitting. A slum church was not only the pastor’s personal property but his l ife ’s work , the worldly manifestation o f his God-given spir � i tual powers, and in many ways an extension o f himself. W hatever contribu �

1 8 6 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

tions others made to his church— build ing and m a in ta in ing its physical structure, paying electricity bills, feeding the pastor and his family, and fi ll � ing its air with a tumult o f praise, clapping, crying, and much more when required—were commonly portrayed by the pastor as his own doing, since it was his prayer and inspirat ional preaching that prope lled others to make these contributions.

Even my own deeds were not immune from incorporation . For several months Pastor Yesudasan prayed publicly for a new cordless microphone. More specifically, he prayed that I would purchase it for him . The microphone became an obsession, with Yesudasan bringing it up in our every personal con � versation. A t times it seemed as i f the microphone was the only th ing he ever wanted to talk about. Formerly our conversations had unfolded over the Bible, with Yesudasan flipping forward and backward between passages to illustrate his observations about life in the slum and beyond. N ow they centered on a tattered old electronics catalog and the reading aloud o f technical specifica � tions. The pastor behaved as i f he and I were simply working out the final details o f the purchase, going so far as to advise me on which o f the city’s electronic shops were Christ ian owned (and therefore the most trustworthy) and on how I ought to go about barga ining with their proprietors, while I struggled to persuade him that I had no intention o f complying. But at length, it seems, Christ touched the anthropologist’s heart. Sick o f the pastor’s pestering, I gave in and purchased the pastor a microphone. It cost me about h a l f the price o f the one he had been hankering for, in a H indu-owned shop in Chenna i ’s electron � ics bazaar. That Sunday the pastor announced to his congregation that, by his steadfast prayers, a miracle had occurred. A brand-new cordless microphone had been bestowed on the church . . . by Jesus Christ!

The tendency o f pastors to treat others’ actions as the fruit o f their own spiritual discipline and devotion was no mere delusion or idiosyncrasy. In so doing, pastors recreated and upheld a picture o f pastorship and pastoral power shared by H indus and Christ ians alike in the slum. A n d it was a picture the pastor could not escape. It confronted him from all sides and was constantly affirmed by the public at large— among whom a pastor’s reputation waxed and waned with the size and enthusiasm o f his congregation. For whatever the actual circumstances under which any given individuals might have joined his church, the assumption was that they were attracted not simply by C hrist ’s miracles but by the miracles that C hrist had performed for that particular congregation and none other, in response to thatparticular pastor’s prayers and the prayers o f those under his direct supervision and guidance.

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 1 8 7

Although pastors did not, officially, have any powers o f their own— osten � sibly they were mere conduits for one and the same G od— slum dwellers could not but notice that some pastors’ prayers were answered more fre � quently, more reliably, and more forcefully than others’. Thus, however much pastors might attempt to renounce personal glory— and no pastor neglected to credit C hrist as the ultimate source o f his achievements— they continued to be perceived as possessing wide-ranging spiritual powers, and some as hav � ing far greater powers than others, whether because o f the intensi ty o f their prayers, fasting, and other spiritual disciplines, because o f the inherent purity o f their souls, or simply because God loved them best.

The most powerful pastors o f all seemed to exist only in legends and rumors, which, however fanc iful, served both to illustrate and to reinforce this widespread conception o f pastoral power. Such legends often told o f the dramatic services that slum pastors provided by casting out evil spirits, for example. These activities were a part icularly fertile source for tales o f pastoral power when the evil spirits turned out to have been deliberately sent by a human foe. In such cases the pastor might find h imse lf locked in a dangerous and often protracted struggle with the sorcerer who had sent the evil spirits in the first place. A typical example o f such a tale, relayed to me by Paul, a C SI Christ ian , began with a local pastor who, after repeatedly fa i l ing to expel an evil spirit that had been afflicting one o f his flock, deduced that it must have been sent by a powerful sorcerer. Real iz ing that he could not by h imse lfhope to match his adversary’s powers, the pastor contacted a certain evangelist, renowned throughout Chenna i for the strength o f his faith, and begged for his assistance. H aving agreed to help, the more powerful pastor soon became locked in an escalating series o f attacks and counterattacks, his own spiritual powers stretched to the breaking point. In the final battle, which was said to have lasted nearly two weeks, the evangelist fasted and prayed continuously, while the congregation o f the pastor who had retained his services prayed, beat drums, and sang songs o f praise in twenty-four-hour rotation . The strain on the evangelist, who was plunged repeatedly into and out o f fever, was said to have been immense. In the end the sorcerer was destroyed and his vict im freed. The evangelist himself, however, was finished. A l kali. Immediately after achieving victory he collapsed and, after lingering in a semiconscious state for a week , “returned to C hrist .”

However fantastical, accounts such as this found a para llel in the routine services slum pastors provided. A l l such men fasted and prayed, cast out evil spirits, engaged in spiritual battles, and suffered physically for their flocks.

1 8 8 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

The idea that a pastor could be k il led in the process— that he might, in effect, sacrifice his own life to save a single worshipper from evil just as C hrist had given his to save all people from evil— underscored the intensely personal and even bodily forms pastoral power took in the slum.

Consider next a bit o f gossip concerning Pastor Yesudasan , relayed to me by a H indu man known as Pa lani . Pu lling me aside one day, Palani seemed eager to discover what I might know about Yesudasan: how close I really was to him , what sorts o f things I might have observed or heard in the time I had been hanging around his church, and, especially, what I might know about a woman named Ce lvi , who had recently left Yesudasan’s church in order to join Pastor Peter’s. I happened to be good friends with C e lvi and to have already heard the story o f her defection from a number o f different perspec � tives, inc luding that o f Ce lvi herself. But not wanting to betray anyone’s trust I professed complete ignorance o f the matter. O nce I had persuaded him that I was indeed quite oblivious to what had occurred, Palani shared his own take on these events.

Several women had already abandoned Yesudasan for other pastors in recent years, Pa lani in formed me, and as time went by more and more were sure to follow them. The reason? It was all because Yesudasan was get ting old, he claimed, and was f inding it more and more difficult to pray steadily from dawn until dusk . A n d because his diabetes had worsened, it was nearly impossible for him to fast for more than a single day. “ H e is a good man ,” Palani said, shaking his head in the appearance o f sympathy, “so o f course this is very sad. But what can he do? A lready he has to rely on others’ help when heal ing people!” Palani reiterated that those who had already quit his flock were just the beginning, and he claimed to know o f many more women who were merely biding their time before jumping ship, lingering out o f sheer pity. Soon only the old and the blind— those who have been with Yesudasan since the very beginning—would remain . O n this dramat ic note Palani concluded.

Pa lani ’s account, though concerning a real pastor and real events, was no less fanc iful than the tale o f the legendary pastor who had sacrificed his own life fighting evil. N e i ther Ce lvi nor anyone else had left Yesudasan because he could no longer fast as he once had. But apart from its empirical shortcom � ings, Pa lan i ’s account raises an important question: W hy should it have mat � tered to those being healed whether Yesudasan required his wife and one or two o f his friends to assist him while praying? This apparent gap in Palani s account, however, made perfect sense given the pastor-centric perspective

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 1 8 9

that it, along with the previous example, illustrated— namely that the pas � tor’s own spiritua l exertions were the decisive factor in the miracles that occurred in his congregation.

The pastor-centric perspective was compatible with a range o f ideas about the ultimate sources o f pastoral power and whether it was a good or bad thing. Its basic premise was shared both by those who envisioned the pastor’s power as a personal capacity, comparable to what a yogi or wizard might have attained, and by those who regarded it, in accordance w ith standard C hrist ian doctrine, as having a source neither with in the pastor h im se lf nor at his command. Indeed, as the next example w i l l illustrate, this premise was accepted even by those who denounced pastors as charlatans, and their feats as illusion.

“ The problem with Christ ians,” Ravi explained as we sat in his house one Sunday eating lunch, “ is that whenever anyth ing good happens— anything at al l— they proclaim it a ‘mirac le!’ ” Ravi, a H indu laborer, was known in the slum for his outspoken criticisms o f slum Christ ianity. Like many other slum dwellers, inc luding Christ ians, Ravi criticized pastors for living o f f the dona � t ions o f those who were often poorer than themselves and accused them o f secretly amassing wealth at their congregations’ expense. Lack ing any evident basis in fact, rumors o f financ ia l impropriety expressed slum dwellers’ sense o f disquiet over the unprecedented influence that pastors appeared to wield, in violation o f the slum’s anarchic and egali tarian ethos. But where other slum dwellers’ suspicions o f pastors were subdued and ironic in tone, Ravi tended to get rather worked up, sometimes to humorous effect. This was why I often sought him out. “Sometimes in life things go the way we want them to,” Ravi continued, “and sometimes they don ’t. W e all pray for good things, o f course, but does that mean that every time something good happens it is because o f God? M aybe the same th ing would have happened anyway. We can never really know. But for Christ ians i t ’s always, ‘Mirac le, miracle, mira � cle! This is a miracle . . . that is a miracle.’ I f this cup were to fa l l o f f the chair, they’ d call it a mirac le!” H is wife Vennila, who happened to be a Christ ian , smiled lovingly. “ Com e on, Ravi ,” she replied, gently chiding him for his recalcitrance, “ how can you deny what Jesus has done for your sisters?” A l l four o f Ravi ’s sisters had converted to C hrist ianity several years earlier, for reasons we w il l come to soon. “Look at how nicely [your sister] M a z h il in i ’s children have grown up.” A t this Ravi laughed triumphantly. “See!” he exclaimed, turning from his wife to me. “Just listen to how that woman talks! ‘The children have grown up.’ The children have grown up? That’s what chil �

1 9 0 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

dren do, right? They grow! But noooo, it must have been a miracle. The pas � tor’s got her convinced that i f it weren’t for Jesus those children would have remained small forever!” I laughed too that day, having witnessed a number o f such miracles myself.

W here some saw a spiritual adept, Ravi saw an adept con man— a slick Rasputin apparently able to convince otherwise sensible, hardheaded slum women to accept the absurd, over and over. Compared with the slum’s Pauls and Palanis, Ravi seemed to me remarkably clear-sighted, his commonsensi- cal outlook be ing one I myse lf shared. I t seemed as obvious to me as it was to Ravi that the source o f the pastor’s powers was the persuasiveness o f his per � sonality, what M ax W eber terms charisma (1978, 241) , rather than the chan � neling o f divine forces. That I should have seen slum pastors as “charismatic” is hardly surprising. W eber’s term is adapted from the same theological sources as Pentecostalism itself, in which charismata (sing., charism or cha- risma)—powers o f glossalalia, heal ing, prophecy, and instruction (1 Cor . 11 :4 )— are divine gifts bestowed by the H oly Spirit on exceptional persons. Scholarship on Pentecostalism, moreover, generally classifies it as part o f a broader phenomenon called “charismatic C hr ist ia n ity” (Robbins 2004), whose name refers both to W eber’s concept and to the prior theological one. So to describe Pentecostal pastors as ‘ charismatic” seemed almost redundant, and for this reason I hardly gave it much thought.

W h a t I completely fai led to see, therefore, was that for all his skepticism R av i ’s vision o f pastoral power was o f the same basic form as Paul ’s or Pa lani ’s— or my own . For while Ravi and I questioned the supernatural basis o f pastoral power, we never questioned the real ity o f pastoral power as such. Paul, Pa lani , Ravi, and I each had our own interpretations o f pastoral power: for Paul it was a bodily sacrifice that drew upon and reproduced the power o f C hrist ’s original gift, for Palani it was a more esoteric power akin to that o f a yogi, for Ravi it was a confidence game, and for me it was “charisma.” We could have argued endlessly among ourselves and in the process only deep � ened our conviction in the assumption we al l shared.

W h a t I only later came to see was that my account o f pastoral power as having its basis in “charisma,” no less than Pa lani ’s esoterism, was explain ing obscurum per obscurius. By using “charisma” as an interpretive crutch, I had unwitt ingly accepted at face value the most basic postulate o f pastor-centric discourse— namely the rea lity o f pastoral power— and had therefore failed to notice two very important things. First, the pastor was h imse lf the prod � uct o f forces that were wholly beyond his own control and that continued to

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 1 9 1

dictate how he must comport and present himself. Second, his formidable capacities were made manifest only through the testimony and affective per � formances o f those very women who were ostensibly under his sway, and whose ongoing cooperation was by no means assured.

T H E M A K I N G O F A P A S T O R

A pastor’s tendency to see his church as an extension o f his own spiritual powers— or, more theologically correct, as a measure o f how G od chose to bless and use him as a vessel—was no private fantasy. It was an effect o f how he was seen and portrayed by others. This was a picture from which the slum pastor could not escape, even i f he wanted to. It was one he was compelled, whatever his personal feelings on the matter, to relentlessly promote so long as he wished to remain a pastor. To understand why this was the case, it is necessary to consider how he became a pastor in the first place, and the nature o f the world into which he thereby stepped.

Before their ordination , which in most cases had occurred less than ten years before I began my research, the majority o f Kashtappatt inam ’s pastors were ordinary slum residents (and all were Dalits). Their only distinguishing characteristics were that they were male and at least part ia l ly literate, though by the time they were ordained all had learned to read fluently. Some were raised in C hrist ian families, though most were not; a few had been ordinary laborers, although the majority, because at least partial ly literate, held slightly better positions. Yesudasan , for instance, had worked in a government prin t shop; another pastor worked in a factory assembling electrical transformers; a third had been an auto rickshaw driver. A l l had feared giving up the secu � rity and steady income their former jobs provided, and in no case had they initial ly set out to become pastors. But gradual ly they had became more and more attracted to G o d ’s work , usually beginning as an ordinary member in another pastor’s church, listening and slowly learn ing what the religion was all about. For those who had not been raised as Christ ians this meant that they had simultaneously to learn the rudiments o f C hrist ianity as well; those who were raised as Christ ians (in al l but one case in mainstream denomina � tions) had only somewhat less to learn .

Somewhere along the line these men experienced some sense o f a “calling”— a vision, a dream, a voice from the sky, or, less commonly, simply a feel ing or vague notion— that they described as unwanted and in all cases unexpected.

192 - P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

Their narratives, at this point, were highly conventionalized. In most instances their experience o f being called provoked in them an internal struggle. Save for Yesudasan , the son o f a pastor himself, all reported being afraid and seek � ing further “evidence” that this was indeed what God wanted them to do. This entailed deepening their search for knowledge by undertaking an intense program o f prayer and fast ing and by seeking guidance from pastors and lay preachers al l over the city. Going from church to church and joining multiple Bible study or prayer groups, each future pastor eventually settled into an informal apprenticeship o f some kind. A t some point, usually at the urging o f their mentor, these men tried their hand at leading a Bible study group o f their own or hosting a weekly prayer meeting in their home. These groups were usually composed o f a few family members, neighbors, or friends, although sometimes the pastor they were apprenticing under helped them gather a small group from among his own congregation. This stage typically lasted between two and four years, and there is reason to believe that for every man who became a pastor there were many more who plateaued as prayer group leaders and progressed no further, either because they did not wish to or because they could not. There were a large number o f informal prayer groups o f this sort in Kashtappatt inam— four or five times the number o f official churches— many o f which had been in existence for years.

W h a t those who went on to become pastors had in common was the dis � covery that they possessed some k ind o f talent, or, what amounted to the same th ing, G od demonstrated his desire for their ordination by blessing those for whom they prayed with miracles, healings, and so on. Typically in such cases word soon spread and more and more people began to join their group. These joiners normal ly included members o f some local church—who generally continued to attend their usual church on Sunday— as we ll as H indus who had only recently become interested in C hrist and had not yet begun at tending a regular church. In cases where the future pastor’s prayer group had included many new converts, these would eventually provide him with his first congregation—which was perhaps why, at this point, a number o f such men began the process o f seeking ordination from one o f the many pastors they had gotten to know in Kashtappattinam or elsewhere in the city. A ny pastor could perform ordinations, and I know o f one case in which a man was ordained on the spot by a complete stranger he had met at one o f the mass open-air Bible meetings that were held each year on M adras’ M arina Beach, which often featured renowned evangelists from South Korea, North America, Europe, and, less commonly, A frica .

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T 1 9 3

Frequently would-be pastors felt the need for some more form al theologi � cal train ing before being ordained and enrolled themselves in one o f many training centers located throughout Chenna i , other parts o f T amil N adu , or elsewhere in South India (e.g., Bangalore).1 There were a half-dozen or so major centers, and a great many smaller Bible schools known to the pastors o f Kashtappatt inam , inc luding one located on the slum’s fringe, and the qual i ty and rigor o f the curriculum varied widely. The leading mainstream Protestant seminaries (the U nited Theological College in Bangalore and G uru k ku l Lutheran Seminary in Chenna i) also accepted a number o f Pentecostal students each year, though prestigious institut ions such as these seem to have been beyond the reach o f ordinary slum dwellers. Courses might be part time or full and could run for as little as a single week to as long as a year or more. Partia l scholarships were often available, though full-time programs generally required students to pay for at least their own room and board. Though it was by no means an easy undertaking, i f a man was suffi � ciently determined he could eventually— through a combination o f loans, fund-raising among friends, and his own personal savings— find some way to get h imse lf through a Bible school.

In some cases graduates might have the option o f being orda ined at the completion o f their course. But even when ordination was part o f the deal, it entailed no enduring connection between the newly minted pastor and either the inst itut ion i tse lf or the denomination or missionary organization that ran it. N or did ordination normal ly create last ing ties between a pastor and the man who had ordained him . W hether a pastor chose to form ally or infor � mal ly associate h im se lf with others was entirely up to him , and al l the slum pastors I knew were independent operators. A lthough some might, at various points in their careers, enter into either formal or informa l arrangements with others, slum pastors remained autonomous for all practical purposes. This followed from the fact that, whether or not a pastor chose to affiliate his church to any other church or local denomination , he was responsible for building his own church and at trac ting a congregation. These were personal assets that remained with him ; whatever affiliations he entered into or left, they entered or left with him . This was equally the case with those relatively few pastors who began their professional life with in the fold o f one or another local denomination . In such cases pastors were never “assigned” to a church (as was the practice in mainstream denominations), but had to—just like any other slum pastor—build their own church from the ground up, raising their own funds and gathering their own flock.

1 9 4 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

W ith this brief overview in hand, let us now take stock. The first significant feature o f the pastoral field is what economists call low barriers to entry. The path o f the pastor required little initial investment and was open to virtually any adult male, so long as he was at least semiliterate and o f reasonable intel � ligence. Despite its low initial investment, the path to the pastorate was a diff icult one, requiring much time and effort in the face o f considerable risk and uncertainty, as the would-be pastor had to continually and publicly prove h imse lf—or, as he would see it, receive repeated divine aff irmation— i f he was to reach his destination. Those who were fina lly chosen, therefore, had made a considerable psychic investment in the certainty o f their call ing and had emerged successfully from a lengthy tria l at once supernatural and social: supernatural because it required evidence o f divine favor; social because it meant not only attract ing followers but doing so in a field that was crowded with other pastors and would-be pastors who were compelled to do likewise.

Compet it ion among pastors was heightened by the lack o f a supervening ecclesiastic authority to regulate or coordinate church-building efforts or to provide slum pastors with a measure o f financ ia l support and therefore o f security in the face o f the inevitable fluctuations in the congregational dona � t ions on which their livelihoods depended. A n even more significant source o f interpastor competition , however, was a demographic bottleneck in which new converts were in short supply. This meant in practice that when a pastor built his church he often did so by either directly or indirec tly poaching sheep from professional colleagues.

The most important and difficult aspect o f church building, therefore, was the struggle not merely to attract but to retain a congregation— the social and spiritual body o f a church, without which the pastor h imse lf was nothing. A n d in order to do this he was obliged at all times to project an air o f con fi � dence and to present, whenever possible, his every success, not as produced prim arily by mundane human effort, but as flowing from the spiritual gifts his god had bestowed upon him . The regnant pastor-centric discourse was thus a double-edged sword. Success bred success, but there was no easy way to explain away failure as other than as fa ilure o f the pastor to w in G od ’s favor. For this reason a pastor was compelled assiduously to promote h im se lf and his church in pastor-centric terms, and therefore to invite being judged a fa i lure— and having his flock gradua lly stripped from h im — i f he was unable to deliver G o d ’s promised blessing.

Reca ll , for example, the case o f C e lvi . A ccording to Pa lani she had left Yesudasan for Pastor Peter because Yesudasan ’s prayers were no longer

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 1 9 5

effective. Specifically, Palani alleged that Peter had successfully freed C e lvi from a stomach pain that Yesudasan had been unable to cure. The fact that Pa lani should have narrated C e lv i ’s move in precisely these terms, however, was not accidental. For Pastor Peter had recently hit upon the idea o f printing a fortnightly newsletter for distribution to the local community, in which he announced C e lv i ’s “cure” and subsequent decision to join his flock as being all about her stomach pain . The newsletter had tac tfully omitted to mention that C e lvi had until this point been part o f Yesudasan ’s church, a fact that was in any case a matter o f common knowledge in A nbu Nagar . N or had the newsletter intimated anything about Yesudasan ’s alleged fa ilure to cure Celvi, thereby causing her to seek out the services o fh is younger rival, Pastor Peter. There was no need to mention these additional details, which readers could be counted on to infer for themselves.

For when C e lvi had origina l ly converted to C hrist ian ity and joined Yesudasan’s church he too had publicly claimed to have cured her o f stomach pain , and also to have cast out an evil spirit that was at tack ing her. Though that part icular spirit had never returned, the stomach pa in had continued to plague her. By c la iming not only to have cured C e lvi , but spec ifical ly to have cured her o f stomach pa in , Pastor Peter had in one deft move added a feather to his own cap and plucked one from Yesudasan ’s. A n d what o f Pa lan i ’s theory that Yesudasan’s failure to have cured C e lvi was a result o f his own inability to properly fast on account o fh is diabetes? There is reason to believe that this detail too had originated with Peter, al though the good pastor had never suggested such a th ing himself, at least not in prin t and not publicly. For just around this time a number o f teenage boys whom Peter had taken under his wing— and who, prior to this, had not expressed the slightest con � cern for Yesudasan’s health (or anyone else’s, so far as I was aware)— appeared to have developed a remarkable interest in that pastor’s blood sugar levels, even going so far as to inquire among members o fh is congregation as to how he was managing to maintain his weekly fasts.

Yet the reali ty o f C e lv i ’s decision to join Peter’s church had nothing what � soever to do with him curing her stomach pain . She in fact remained uncured. A lthough C e lvi never public ly denied Peter’s claims, she continued to seek treatment in various local clinics and to present me with the prescriptions she had been given, which I would then f i ll for her. Nor, for that matter, had her initial decision to convert to C hrist ianity in the first place had anyth ing to do with Yesudasan’s casting out the evil spirit that had been plaguing her. A s she explained to me privately, al though Yesudasan had indeed cast the spirit

196 P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

out, she had been interested in C hrist ianity for some time before that and had already accepted C hrist in her heart and “ thrown all [her] H indu swamis away.” That is, she had dismantled her household shrine and consigned her former gods to the trash heap.

But there is more. The fu ll story o f C e lv i ’s leaving Yesudasan for Peter was more complex than what I have thus far told. O ne o f the teenage boys who had been recently been converted to C hrist ianity by Pastor Peter was C e lv i ’s son, H ari . The fact that mother and son should attend different churches was by no means unusual, and never in the ten months since H a r i ’s conversion had any o f the concerned part ies—Peter, Yesudasan , or C e lv i— ever sug� gested that things should be otherwise. This would all change, however, on the last day o f Pastor Yesudasan s annual open-tent Bible meeting. A s in pre � vious years, Yesudasan hosted a meal for al l those who had contributed labor to making the meeting happen—guest preachers, lay organizers, the electri � cian who had wired the loudspeaker system— and had, according to estab � lished precedent, invited other local pastors as well. A mong the invitees was Peter. Peter, however, chose not to attend in person and sent his group o f teenage boys to participate in his stead.

N ot knowing where these youths had come from , and not having enough food on hand to feed them, pastor Yesudasan had sent them away. Piqued by the pastor’s rejection , the gang o f youths began shouting and created a big commotion . C e lv i ’s son H ar i was among their number and, st ill angry days later, threatened his mother that i f she did not quit Yesudasan ’s church he would leave home forever. The threat was not a serious one; H ar i would have had no place to go and would have probably just moved into his grandpar � ents’ place across the street, or into Peter’s church and personal residence some f ifty meters away. This was the very tactic C e lvi herself had used to bring her own husband to heel, as we saw in chapter 3, but still she was shaken by her son’s threat. She sought the guidance ofYesudasan , who, she said, told her, “ Go to Peter; do not leave your son.” This was later confirmed for me in a private conversation with Yesudasan .

Between the official pastor-centric narrative and what had actual ly tran � spired between pastors like Peter and Yesudasan and women like C e lvi there was an obvious gap. The ethnographic significance o f this gap, is not simply that it existed but that even those who knew full we ll the falsity o f more widely c irculating versions— as did al l three o f the principals in the case o f C e lv i ’s defect ion— did not publicly challenge those narratives or the pastor- centric assumptions they rested on.

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 1 9 7

THE BURD ENS OF CARE: Y E S U D A S A N ’S CONFESSION

In the previous section we sought to account for pastor-centric discourse by si tuating it in the strategic context o f interpastoral competition, and in par � ticular how the exigencies o f that context compel a pastor to project his every success— and indeed, his church itse lf—as an extension o f his own personal powers. A pastor’s often fraught relationship with his professional colleagues was not, however, the only or most difficult relationship he had to negotiate. Even more important was a pastor’s relationship to his own , mostly female flock, and it was in negotiating this relationship that the slum pastor was most continuously challenged. For a slum church was not, in reality, simply an extension o f its pastor; it was a complex social arrangement comprising numerous people whose own visions o f how things worked did not automati � cally coincide with his.

“ The pastor must accept al l evils,” an old and weary-looking Yesudasan told me one Sunday afternoon some time after C e lv i ’s defection. “ H is role is not to command.” W e had been si tt ing together silently in his now-empty church— the service had ended and the members o f his congregation had all gone home to their families— and the pastor’s remark had come quite out o f the blue. “ H e can only give warnings,” Yesudasan continued. “ H e can say, ‘Danger, don’t go there! I t ’s not a good place. Liquor, gambling, adultery: they w il l only lead to harm !’ But a pastor cannot command. W hen the peo � ple go astray, he must help them come back; when their bones are broken, he must give them first aid. H e must make them strong again.” That the pastor could not command his flock was literal ly true, for he had no authority to do so. Yet Yesudasan , like other pastors, did in fac t command. This did not mean his commands would be obeyed, o f course, and the only sanction a pastor could impose on the disobedient was to expel them from his congregation.

“ D o you remember Subramanian Moses, ‘ M a n i ’ ?” the pastor then asked me, after a pause. M ani used to compose heartbreakingly beautiful songs, which he would often sing, unaccompanied, during the Sunday service. “Yes. O f course you would. W ha t a voice that man had— like an angel. O n the outside he was so pure. A n d sincere too— there were even tears in his eyes. But he was a secret sinner: he had quit drink ing, but then he would start again, quietly. This happened so many times, and he was causing so much suffering in his family. So final ly I had to tell him : ‘G o !’ ” A pastor could not al low this sort o f behavior with in his congregation, even i f he h im se lf would

1 9 8 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

have preferred to look the other way. This was not just a matter o f preserving his own reputation in the eyes o f the larger slum public, or o f st icking up for M a n i ’s wife Rosie. For i f even a few among his flock had been allowed to stray— or even, for that matter, i f they were insufficiently enthusiastic in their prayers or in singing songs o f praise— it could affect the entire congregation. In practice, lack o f enthusiasm was never a real danger because the pastor h im se lf would be there to continuously urge them on: “Clap louder! . . . Sing! Sing w ith more fee l ing!” But M a n i ’s secret drink ing, even i f it had been known to none but the pastor himself, could have caused God to withdraw his favors, no longer answering Yesudasan ’s prayers or those o f his congregation.

“Backslid ing” was what the pastor called it, using an English term that normally referred to reversion to H induism but that, like its T amil equiva � lent pinmdrram, could refer to any fa l l from grace. “ ‘Backsliding’ and ‘pastor- hopping,’ ” Yesudasan continued, “ these are the two great worries in every pastor’s l ife”— the loss o f one’s flock , in other words, to other gods or to other pastors. For a pastor to admit to worry was something new for me; the pas � tor’s total confidence in both h im se lf and his god were, after all, his modus vivendi. “Backsliding,” he then in formed me,

is very, very common in this place. It is most common among men, but it can occur among women as well, Robert, you would be amazed. . . [and] you would never be able to see it coming. They will be there, every day, praying, praying . . . so when you don’t see them for a week, you think nothing of it. But then another week will pass, and then—next thing you know—bang! They have gone back! [Mimics the Hindu practice o f holding a lit oil lamp before the deity in worship, and the daubing kuhkumam, a decorative powder that marks one as Hindu, on his forehead.] I f you go to their house, it will be full o f swami pictures. And do you know what happens next? It is very sad, Robert, very sad. Drinking, smoking. Gambling, adultery, fighting, tears! How will it end?

A nswering his own question, the pastor pantomimed a woman opening a bottle o f kerosene, pouring it over her head, and strik ing a match: tikkulittal.

The pastor was at a loss to explain why backsliding should ever occur, but his distress about the phenomenon (but not about “pastor-hopping”) made perfect sense on both theological and ethical grounds— ethically, because backsl iding led to harm ; theologically, because the pastor’s job was to keep his flock from danger (Ezek. 34 :1-16) . The threat o f “ backsliding,” moreover, has been a standard trope o f missionary discourse in India since the early

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T » 1 9 9

nineteenth century at least, and the idea that the poor and unlettered are especially prone— not just in India, but in Europe as we l l— to “revert” to polytheism (this being understood as the natural and primordia l fa i th o f all but an enlightened elite) has been a mainstay o f Eurocentric thought since the time o f H ume ([1757] 1957, 48-51; cf. Brown 1981,14-15) . W hether back � sliding actual ly occurred with any frequency among slum Christ ians, how � ever, was far from clear. For although it was by no means unheard o f for male converts to lose interest in the church, and to remain C hrist ian in name only, I had never known this to occur among slum C hrist ianity ’s most significant, female demographic. W hen I pressed the pastor for examples o f female back � sliders, he could come up with only three names. The first was someone unknown to me, who had left the slum long ago. The other alleged backslid � ers were M azh il in i and V iolet, a mother and daughter who lived nearby and who, according to the pastor, had reverted to idol worship only quite recently.

Intrigued, I visited the home o f M azh il in i and V iolet unannounced the very next day and was surprised to find them dressed pla in ly with no flowers in their hair and with no trace o f makeup or jewe lry on their bodies (i.e., in the manner o f devout slum Christ ians). The interior o f their one-room dwell � ing, which the pastor had assured me was now th ick with incense and swami pictures, was as unadorned as the women themselves, save for a portra i t o f their Lord , Jesus Christ , hanging on an otherwise bare wall. The ir alleged backsliding, I discovered, was con fined to a personal fa l l ing out they had with their pastor. They hoped this break would prove temporary and stoutly denied any intention o f abandoning Christ , to whom they sti ll prayed thrice daily, with or without Yesudasan ’s help. They added that they would prefer to be part o f some congregation but planned to wai t a while to see i f things could be patched up with Yesudasan before they looked for another church to join .

I never found out the nature o f their dispute with the pastor, but it seemed clear that, unlike with M an i and men o f his ilk , neither they nor any other Christ ian women had ever been actual ly kicked out o f a church for fa i l ing to comply with C hrist ianity ’s moral codes. Slum women , C hrist ian or other � wise, simply did not engage in acts— like drin k ing or smoking, for exam � ple— that would require such extreme measures. The pastor’s struggle was therefore not to keep women in line but simply to keep them. A n d yet, com � pared to the relations between a pastor and his few male congregants, which were normally quite close and largely uncomplicated on a personal level, the

2 0 0 * P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

relations between pastors and the women o f their congregations were far from straightforward and a good deal trickier to manage.

H ence Yesudasan’s other great worry: “pastor-hopping.” U nlike backslid � ing, pastor-hopping presented no obvious ethical or theological problems. Yesudasan at any rate made no attempt to claim that the pastor-hopper her � se lf was in any k ind o f moral jeopardy, nor were there any significant doctri � nal differences to distinguish one pastor from another. The only apparent connection between the pastor’s “ two great worries,” indeed, was the identi � cal setback they represented to h im se lf and his reputation as a pastor. But where Yesudasan had had little to say about the causes o f backsliding, pastor- hopping seemed to demand an explanation . “ ‘Pastor doesn’t care about me anymore,’ they [i.e., the women o f the congregation] w i l l say. ‘Pastor only asks about so-and-so’s hardships, he only pays attention to this sister or that sis � ter— but what about me? W ha t about my family’s problems?” ’ The pastor evoked female congregants complaints in a whiny tone. “Seriously, Robert , that is what they w il l say. ‘Pastor doesn’t love me any more.’ M e, me, me. Do you remember Celvi? For ten days straight I prayed for her. Praying and fast � ing, praying and fasting— al l to remove the evil spirit that was at tacking her. I removed that spirit, but now she has gone to [another pastor]. W ha t can I do, Robert? This is my burden— the burden o f the pastor— the burden o f Christ . I bear everything for you, oh Christ! M y Lord! For you I have . . . [breaks into incoherent prayer].” W h a t is significant here is that, even while revealing aspects o f his call ing that were usually kept well hidden from public view, Yesudasan ’s discourse remained as pastor-centric as Paul’s, Pa lani ’s, or R av i ’s. A lthough he was prepared to admit the vulnerabili ty o f his position , it seemed that whatever happened was still al l about him and his own alleged central ity in the lives o f his congregation .

W e know, however, that at least in the case o f Ce lvi , pastor-directed emo� tional neediness had had noth ing to do with her decision to leave his church for Peter’s. This is not to say that the sorts o f emotional demands Yesudasan described were not real; I had seen enough already to know that the pastor’s relationship with his f lock was often quite emotiona lly intense, and I did not doubt that complex interpersonal negotiations played a role in pastor- hopping from time to time. The point is simply that the pastor would continue to see C e lv i ’s defect ion in such terms, despite what he knew to have in fact taken place. The pastor had already conf irmed all the details o f C e lv i ’s account to me in a previous conversation. It is certainly possible that, in addition to everything else, C e lv i had been feeling ignored and

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 2 0 1

neglected by the pastor. But even i f such were the case, the prim ary factor in her defection clearly had nothing to do with any k ind o f neediness or per � sonal dissatisfaction with Yesudasan . Indeed , even after having left Yesudasan’s church, she continued to express posi tive sentiments about him in my private conversations w ith her. Just a week after her split with Yesudasan , for example, C e lvi told me she hoped I would buy him the microphone he had been so ardently praying for. Conversely, when C e lv i ’s relations with Peter began to sour some weeks later there was never any ques � tion o f her leaving his congregation on that account. A lthough she might very well have done so under other circumstances, what was abundantly clear to me from all my time in the slum was that the emotional dramas between a pastor and the members o fh is flock and the phenomenon o f “pastor-hop � ping” were not directly connected. That such personal dramas should weigh so heavily on the pastor seemed to have less to do with their actual role in determining his fate than with their being one o f the few things he could latch onto and hope to control. In real ity his fate was determined by a whole host o f factors largely beyond his power, part icularly with regard to the women o f the slum.

“A l l pastors are bad,” C e lvi declared with disgust, as we sat together in her house one morning. “ There are no good pastors— only Jesus is good!” C e lvi here invoked a commonplace among slum women; while pastors saw them � selves as vita l conduits o f divine power and as G od ’s chosen representatives on earth, they figured only margina lly in women ’s own accounts o f slum Christ ianity, which revolved around their own direct relationship with God . N ot only had C e lvi grown increasingly dissatisfied with Peter (for reasons I was never to learn), but her assessment o f Yesudasan had simultaneously taken a turn for the worse— even though only recently C e lvi had spoken fondly o f her former pastor and had had no interactions with him in the intervening period.

“Poor th ing!” Pushpa, an a li (figure n) who had been si tt ing with us, responded teasingly, “you ’ve had so many disappointments with those two men.”2 Pushpa, a Pentecostal Christ ian herself, conceded that many pastors were indeed “ bad” and that the important thing was to trust in Jesus. But she chastised C e lvi for assuming that all pastors were alike. “M y pastor isn’t like that, Celvi. H e ’s good, very good. Just like Jesus.” Pushpa perceived a natural affinity between alts and Pentecostal Christ ianity , because, as she explained, “ Christ loves allpeople. H e accepts everyone, irrespective o f personal tenden � cies [>kundticayankal]. H e is the god o f love” Though admitt ing that most a lh

2 0 Z • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

f i g u r e I I . Alh d a n c i n g .

were in fact H indu , Pushpa claimed that the percentage o f Christ ians in the a li community was considerably higher than in the populat ion as a whole and was growing steadily. She also insisted that, in fact, a great many Pentecostal pastors were themselves a lis “who dressed as men.” I was puzzled by this and asked Pushpa what she meant, because I had always assumed that an a li was, by defin it ion , a man who dressed as a woman . “ O h, come on, Robert ,” she replied, “ don ’t pretend you don ’t know. They are like the a lis in your country, gay.’ ”

A C T IO I N DISTANS: T H E R E A L I T Y O F M I R A C L E S

R O B E R T S : W hy did you become a Christian? V E E R A M M A L : Because o f hardship—why else?

R O B E R T S : W h a t t y p e o f h a r d s h i p ?

V E E R A M M A L : Money troubles, fighting with my husband [kutumpa kastam\.

R O B E R T S : And since you’ve become a Christian, have your problems been solved?

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T � X 0 3

v e e r a m m a l : Yes, o f course. No other god would have solved them— and none [of my so-called friends] was willing to help!

R O B E R T S : So does the church give you money? Or food?

V E E R A M M A L : No! Not the church—that pastor doesn’t do anything for us. H e’s never given us anything! [Makes a gesture o f disgust.] It’s only Ecucami [“Jesus Swami”/ “Lord Jesus”] who helps. H e’s the only one who was able to solve my problems.

W ha t distinguished C hrist from other gods was not simply that he answered slum dwellers’ prayers, or even that he was said to love the poor and the weak . H e was all that, o f course, but even more he was known as a god who special � ized in helping women with their marita l and household problems. A mong slum dwellers, C hrist was famous as the god women turned to when fights with their husbands had become too severe or too incessant to bear, when their every aspiration had been frustrated, when household woes had become unmanageable, and— most significantly—when al l other remedies, both human and divine, had been exhausted.

“ I f you become a Christ ian , your husband w il l stop drink ing,” a H indu woman informed me early in my stay. “ That is what everyone says.” I asked her whether what people said was actual ly true, and she affirmed that it was. Then, qua l ifying her statement somewhat, she explained that converting to Christ ianity had worked “ in many cases” but that the method was not fool � proof. This proved to be the consensus among both H indu and C hrist ian women alike, the only difference being that the latter maintained that their god was able to stop any man ’s drink ing, even i f he did not choose to do so in all cases, whereas H indu women remained open to the possibili ty that some men might be so set in their ways that even C hrist might not be able to change them.

My init ial assumption when hearing o f this and other o f C hrist ’s family- related miracles was that, i f they were indeed occurring with anything close to the regularity women claimed, it must be because the husbands o f the women whose prayers had been thus answered had been persuaded to convert as well. O nce the man had become a Christ ian he would be required to stop drink ing (or, at any rate, would risk divine sanction should he continue to do so) and, more generally, would be subject to the pedagogical and disciplinary structures that church life provided. Pastors did indeed, as the next chapter w ill show, incorporate what was essentially a program o f marital reform into their preaching. They also actively intervened and counseled congregants in their marriages when opportunities arose.

Z O 4 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

Recal l, for example, the case o f Ravi , the antipastoral skeptic whose four sisters had converted to C hrist ianity some years earlier. The in itial “miracle” that had persuaded them to do so was that C hrist h im se lf had, through the agency o f a local pastor, reformed the husband o f Ravi ’s eldest sister Lakshmi. L aksh m i ’s husband Mohan had been an abusive lout who had abandoned Lakshmi years earlier. Mohan was no C hrist ian at the time, but later, for unknown reasons, he became one. W hen his new pastor came to know that he had once had a wife and had abandoned her, he counseled Mohan to pray to G od to forgive him . Mohan agreed and, in the process o f praying, found h im se lf overcome with remorse. Though he knew G od had already forgiven him , M ohan— or James, as he was now known—was said to have become inconsolable. The pastor advised him to return to Lakshmi on bended knee and beg her forgiveness as well. Escorting James to A nbu Nagar, the pastor explained to her that her husband was a genuinely changed man. In it ia lly skeptical, Lakshmi agreed to give the man who had deserted her another chance and— over the strenuous objections o f her brother Ravi— took him back into her home on a probationary basis. To Laksh m i ’s delight, and the astonishment o f her entire family, what the pastor said proved true. By all accounts James never spoke i l l to her again and remained a model husband. Lakshm i herse lf was so overwhe lmed by this miraculous turn o f events, events she attributed to C hrist rather than to Jam es’s pastor, that she herself converted to Christ ianity , as did al l three o f her sisters, one o f her younger brothers, and several nieces. A l l became members o f the church whose pastor had brought her errant husband home. But Ravi drew rather different con � clusions. By his own admission this event had sparked his lifelong animosi ty toward pastors. H e was furious that his sister had been reunited with Mohan /

James. Ravi detested “Jam es” and could never forgive him for what he had done. H e insisted that the man ’s new found piety was a sham and that Lakshmi would have been better o f f without him .

This was not an isolated incident. Pastors spent a great deal o f time coun � seling aggrieved women and at tempting to persuade their mostly H indu husbands to reform. O ccasionally they succeeded. But direct pastoral influ � ence did not explain the majority o f C hr ist ’s miracles, for the simple reason that very few men in A nbu N agar had in fact converted to Christ ianity, and non-Christ ian men were rarely w il l ing to be counseled by a slum pastor. N or did slum women ’s own accounts o f how C hrist s miracles occurred lend sup � port to the idea o f pastoral influence. According to both H indu and Christ ian women I questioned on the matter, whether or not the husband himself

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 2 0 5

became a C hrist ian was tota lly irrelevant to whether C hrist would , at his w ife ’s behest, soften his heart or cure him o f the drin k ing habit. I f it was the wife who was asking C hrist to remove a man ’s desire for drink , one woman explained, “then how can C hrist ask the man to worship him?” “ I f I am ask � ing God for something,” another woman chimed in, “then I am the one who must do something for it; i f Saranya is asking God for something, then it is she who must do something for G od—she can’t tell you that you must also praise God and beg H im to do it! . . . The woman ’s husband has no interest in giving up drink , right? So why should he have to worship C hrist?” The only time a man would need to pray to and worship Christ , according to slum women , was i f he h im se lf wanted to stop drink ing, I was told, and wanted C hrist ’s help in doing so.

I f what both Christ ian and H indu women claimed was true— that mirac � ulous transformations o f husbandly behavior were indeed occurring with some regularity— it could not be explained as the result o f any direct pastoral action, as my own pastor-centric assumptions had led me to suspect. For such men were, both before and after their miraculous reforms, entirely outside the inst itut ional controls o f the slum church. A n d yet, I w i l l argue, the organizational and discursive form o f slum C hrist ianity was indeed produc � ing a k ind o f “action at a distance”— enlisting in forma l networks to exert an indirect disciplinary and moral control well beyond its own membership.

C H R I S T , W O M E N , A N D T H E S O C I A L T E C H N O L O G Y

O F P R A Y E R

The Sunday morning service o f the slum church was a weekly spectacle in which the miraculous powers o f C hrist were dramat ical ly displayed— not just for the congregation , but by it and for an audience that included all with in earshot. For all but the smallest o f churches possessed loudspeaker systems, the speakers o f which were pointed , not inward toward the congre � gation itself, but outward at the surrounding slum. V iewed through the standard pastor-centric lens, the appearance was o f a man channel ing divine power and inspiration , which then washed over, and produced a number o f impressive effects upon , those seated before him (figure 12). From another perspective, however, we might perceive the women themselves as produc ing dramatic evidence o f pastoral and C hrist ly power. They did not merely reg� ister the effects o f the divine, in other words. They performed it.

2 0 6 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

f i g u r e iz. The Holy Spirit washes over women.

Swaying, clapping, singing, creating a constant drone o f incomprehensible chanting interspersed with ecstatic and often equal ly incomprehensible out � bursts o f glossolalia, women’s performance o f divine power had both audi � tory and kinesthet ic components (figure 13). A n d though confined to non � speaking and supportive roles during much o f the service, at a number o f critical junc tures women ’s voices took center stage.3 A t one point in every service, for instance, women passed the microphone among themselves, so that any woman who wanted to could offer what were called “voluntary prayers”— a torrent o f inspired and often furious invocations o f divine power similar to, but far exceeding, the ecstatic prayers sometimes offered by pastors at other points in the service (figure 14). A lthough this was technical ly open to both male and female congregants, it was generally understood that the only ones capable o f such feats were women .

“Voluntary prayers,” however, though providing women an opportunity to display their own direct relation to divine power, al lowed little scope for them to actual ly say anything meaningful to their fellow Christ ians or any � one else. The second point in the service when women took center stage, on the other hand , provided exactly that. D ur ing what was known as catci nemm (“the time o f testimony”) any member o f the congregation might

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 2 0 7

f i g u r e 13. Spiri t-f i lled pra ises.

again take the microphone and tell the congregation— and, significantly, the surrounding slum— about her life, her problems, about how she had been “spattered with Jesus’s blood” (i.e., blessed). Though occasiona lly ecstatic, women’s testimonials were more often tearful tales o f the wrongs and abuses they had suffered— at the hands o f shopkeepers, other women , their husbands— or about being ill, having no rice in their house, or strug � gling to obtain , or repay, a loan: “ I have paid! I have already paid, and yet I must pay Rs. 1,500 m ore . . . . [But] my husband drinks a lot and creates so many troubles.. . . O nce Rs. 250 was lost. ‘You must have lost it when you were drunk ,’ I said. ‘ N o! You are the one [who lost i t]!’ he said. H e came home only to fight. H e wanted money, but I could only give h im Rs. 50.” The tone was not one o f defeat but o f defiance. For C hrist , they reminded the

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

F I G U R E 14. Inspired prayer.

world, was on their side: “Because o f my tears I was not even able to pray aloud. So I prayed in my h e a r t . . . [and] these prayers—he always hears these prayers.. . . Even in my sleep C hrist hears my prayers, he sees my tears.” It was not enough, however, for these women simply to claim that their god loved them and that he would “raise [them up] and destroy [their] enemies.” They had to also produce some “evidence” (cdtci) that this was the case, usu � al ly some example or instance in which they had in fact surmounted some d iff iculty— “ bathed in C h r ist ’s b lood”— over the course o f the preceding week .

Complete ly unique in the world o f the slum, cdtci neram provided slum women with a platform from which to condemn the injustices in their lives and to proc laim publicly that— despite appearances to the contrary— they were indeed loved and deserving o f love. A part from using the occasion o f cdtci neram to assert their al liance with C hrist and to in form others o f their personal struggles, the women who took the microphone normal ly at some point also included a direct appeal for prayers: “So you must al l pray for me,” “ D o not forget me,” or “ Please, everyone, I need you to pray.” Equa l ly important , however, was the way cdtci neram combined with other

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 2 0 9

practices—practices that occurred outside the Sunday service and that we w il l come to shortly— to constitute among the women o f the church a new k ind o f social organization and a new understanding o f moral community. But before turning our attention to these extra li turgical activities, let us con � sider a third point in the Sunday service at which womens voices assumed center stage, albeit indirectly.

This was in what were known as “prayer requests.” Far less dramatic than the previous two examples, in this case it was the pastor who spoke, reading aloud from a list o f writ ten requests, previously submitted to him by women o f the church, o f things they would like him and the rest o f the congregation to pray for. (Examples o f prayer requests, a window into women’s dai ly strug � gles, are provided in an addendum at the end o f this chapter.) These appeals, though lacking the intensi ty o f those that women expressed direct ly during cdtci neram, funct ioned similarly in enlisting all members o f the church to address themselves to one another’s needs. The relatively calm presentation o f this segment o f the service, however, was deceiving. For its true impor � tance lay not in the way the pastor read from the list o f requests but in how that list was i tse lf produced and in the weekly cycle it completed.

From Monday to Friday, days that the pastor h imse lf spent in prayer and Bible study—recuperating and preparing h imse lf physically for his Friday fast and Sunday morning service— the women o f the church were hard at work . The precise details o f these activities varied from one slum congregation to the next, but the basic idea was the same. In Yesudasan’s church nine experienced female congregants were organized into three “prayer teams” o f three mem � bers each. A l l the slums o f Kashtappattinam in which congregation members resided were divided into three “zones,” one for each o f the prayer teams. Every morning the teams met together, prayed, and charted an it inerary such that by the end o f every week they would have visited each and every household in their zone in which a member o f Yesudasan’s congregation resided, as we ll as a number o f households (both H indu and Christ ian) that did not have mem � bers in Yesudasan’s congregation but that had requested the prayer team’s services through a friend or neighbor who was a member.

In each household they visited, the women o f the prayer team might spend anywhere from fifteen minutes to an hour or more. They began by praying over and blessing the woman (or man) o f the house, as well as anyone else who was present and also wished to be prayed for. N ext they discussed the ongoing struggles in that person’s life or in the lives o f their loved ones, offering wise counsel and heartfelt sympathy. They also noted any positive developments

2 1 0 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

that might have occurred since the previous week ’s visit and discussed any new problems that might have arisen. They then prayed once again, intensely, over all these problems. Finally, before leaving, they solicited and recorded any specific prayer requests (usually these were identical to the ones they had been praying over) on a special form that was then submitted to the pastor. The pastor then prayed on their beha lf during Friday fast ing prayers, and again during the “prayer requests” segment o f the Sunday service.

A part from following this preset program , the prayer teams also served in forma lly to update members o f the congregation about the most pressing problems o f those visited earlier in the day or earlier in the week , so that each household ’s problems were discussed over and over again, with great sympa � thy, in each successive household that the prayer team visited. O ften these previous households’ problems were prayed for as well. The itinerary that the prayer teams followed varied from week to week , so that those who were visi ted first one week might be visited last in the subsequent week. The three prayer teams also met together midway through the week to pray over and discuss the problems o f al l the households each had visited up until that point, and in this way ensured that knowledge and prayers were circulated among al l households o f the congregation irrespective o f “zone.”

W omen ’s prayer teams, in short, created and continuously maintained a dense web o f knowledge, sympathy, and prayer that joined each and every one o f the congregation’s households— as we ll as a dozen or so nonmember households— to every other. A n d in each o f these households there was someone who stopped what she was doing three times a day to recall and concentratedly pray upon the problems o f (at least some of) her counterparts, k nowing that al l over Kashtappa t t inam others would be doing the same exact th ing for her. The organization o f mutual concern over dispersed social networks was the practical expression o f what the sermons o f the slum church exhorted: the Christ-l ike adoption o f others’ suffering as one’s own spiritual responsibility.

The Sunday services o f the slum church provided a moral ly sanctioned plat � form from which slum women could—with the fu ll force o f both their con � gregation and their god behind them—publicly proclaim their most private struggles and declare before all their need for care. Yet this declaration did not take place just once a week: it was reenacted on a da i ly basis via the novel technique o f wom ens extra l i turgica l social organiz ing (through prayer

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • Z I I

teams), which continually reinforced dense webs o f knowledge, sympathy, and shared spiritual responsibility. W here the woman had once been indi � vidual ly responsible for the state o f her household, and for bearing alone the problems between herself and her husband or with her creditors, these now become matters o f public concern. H er successes or failures in the home now depended, not, as before, on her individual auspiciousness, but on the prayers and moral force o f each and every women in her congregation, just as their successes and failures depended on her own dai ly prayers and concern. W hen misfortune struck, therefore, there was less need to shift blame onto super � natural at tacks— vankandvi, ceyvinaiydvi, and so on— brought about by other women’s envy.

The social technology o f Christ ian prayer functioned in the slum, not just to bring women together in relations o f mutual aid, just as the discourses o f auspiciousness and vankandvi had done, but also to induce them to actively embrace their own ult imate responsibili ty for one another’s fate. Slum C hrist ian prayer brought women together without simultaneously atomiz ing them by hold ing women individua l ly accountable for their own success or failure (as did the discourse o f auspiciousness) or by engender � ing in them a perpetual wariness o f one another’s envy (as did ta lk o f vankandvi).

Critically, the effects o f these novel social techniques were not confined to the community o f C hristian women . For the prayer teams, as we ll as the Sunday services, served as channels through which a woman ’s suffering was not only shared with others like herse lf but made public on a much wider scale. The Sunday speeches, recall, were not only heard by other Christ ians but broadcast on loudspeakers to all w ith in earshot. W hether or not a wom � an’s husband (or anyone else who was causing her grief) happened to hear her, one could be sure that others who knew him would . Furthermore, as the prayer teams made their dai ly rounds through the slum, not only every single congregant, but al l the other members o f each congregant’s household, were made privy to the hitherto largely private travails o f other women and their families. These factors produced what I referred to above as “action at a dis � tance,” informal networks that exerted an indirect but potentially powerful moral influence. The harm ful effects o f men’s drink ing or truancy or woman � izing— and the fact that these violated the slum’s basic ethic o f care—were no longer matters o f simmering discontent for isolated wives. They were rehearsed before an extended network o f friends and neighbors, over and over again. The burden o f each woman ’s situation , moreover, was construed

2 1 2 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

as be ing the ultimate responsibility o f her congregation as a whole, rather than as an individual matter rooted in some personal deficiency.

A D D E N D U M : P R A Y E R R E Q U E S T S

The following is a partial l ist o f prayer requests collected by the three prayer teams o f Yesudasan’s church over a two-month period. D uplicate and repeti � tive requests have been removed.

Area I: Rubavati (team leader); Ester Matilda; Taralammal

The sister called Kirubavati suffers from “sent spirits” [mischievous magic]; let Christ destroy them. Also pray that God will help her to start an idli [steamed dumpling] business.

The sister called Tulasi’s elder brother Murugan has been sick for ten days; let the Lord command a complete cure for him.

The sister called Jennifer Ammal is suffering from headache, fever, and sick� ness. Her husband Suresh has foot pain. Let us pray for her, her husband, and her children.

Rukmani Rebecca prays that Ramesh will be liberated from his drinking habit and that Christ himself will send Guhan to work.

Violet prays that her son Bosco will be freed from his drinking habit and will be married to a nice girl.

Christina prays that Ramu will be freed from drinking; that she herself will not be hindered from coming to church; that no barriers will prevent Bavani from getting married and that it will be a nice boy; that God will help her elder brother go to Pondicherry; and that Jayakumar and Paul will get jobs.

Sagaya Mary prays that Christ himself will help her return the money she has borrowed and that he will do great things \_periya. kariyahkaf\ for her.

Priya prays that no barriers will be placed before Dasaprakasan’s marriage and that it will be a peaceful marriage; that her sons Prabalan, Logu, and Paramurugan will obey/submit to \kUpati\ their mother and give her money; that they will meet the need o f money for Indira’s marriage; that Amul will get a good job and find a place to live; that Saminathan will get a good job and will be able to do cement work on their house; that Lakshmi Ammal will get money; that Mazhilini’s husband John Peter will get freedom from drink; that Sagayaraj will leave his job for a better job, that he will take care o f his

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • Z I 3

family, that he will give respect to his mother, and that more will be added to the family income; that Shambli will pass the Plus Two Exam [an exam taken at the end o f secondary school].

Area II: LidiyalM ary (team leader); Rosie; Savari A m m al

The brother Murugan’s hand was injured; Kirubavati prays that Christ will cure it and also that Murugan will be released from drink.

Violet prays that brother Manimaran will be released from debt problems and that Christ will give money to Velu and Narayanan quickly.

May God break the devil that is in the family o f Moses Murti and Deva Sundari.

Mariyal Muniyammal has had an X-ray at the government hospital because she was not well. Pray that they say nothing is wrong and that an operation is not necessary. Pray also that her husband s plan goes well and that he is given a good income and is not prevented from coming to church.

Sagayam prays that her son Amul Raj will keep away from his [bad] friends \nanpankalai vittuvilakavum\. Also, he is not going to work often; we pray that he will go to work and that his father will stop getting angry.

Kasturi prays that she will get a loan.

Margaret’s child is often crying because o f leg pain. Pray that he is released from leg pain and that brother Terry will be commanded [i.e., by God] to go to work.

Matilda Jegatha wants her son to be called for work and prays that Ramesh will be freed from drinking and that Alice’s daughter will pray and that her prayers will be answered.

Kirttikka prays that her loose motions will be removed, and that her bodily weakness [palavinam] will be cured; that Ramu will become close to his wife and that their family will find peace.

Beula prays that she will be relieved from fainting and bodily weakness and that the Lord will command that her son Jeyakumar gets a good job.

That the Lord gives Celvi relief from the swelling in her throat. Her body is tired and she is suffering from exhaustion and weakness. Pray that her hus- bands auto-driving salary will be increased and that there will be peace in their marriage.

The brother called David has a boil in the middle of his urinary tract that is causing great pain. Let the Lord give him a full recovery.

2 . 1 4 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

The sister called Jeyanthi is being betrayed by Nelavati and Usha; pray that God makes them come together. They are speaking very cruelly; pray that God will touch their tongues.

The one called Saroja: her mother’s arms and legs won’t move. Pray that she is fully cured.

Area III: Christinammal (team leader); Nagomi; Margaret

Lakshmi Amma. Pray that her son John will be liberated from the drinking habit, that he will obey his mother, and that he will be kind to his wife. Pray that John and his wife will have peace.

Saraswati. Pray that Christ will put electricity into her house and will deliver her from debt problems.

Sekar and Rekha pray that they will be able to come to church without hindrances.

Suriyammal prays that Latha’s husband will stay away from his bad friends [tura mdrkkamdna nanpanka], that he will go to work daily, and that he will have no problems; that Loyal and her husband will get a loan; that Usha will have a safe delivery; that Job will get relief from leg pain.

Vaila’s son Kumar is going to a new house next month [i.e., he has just gotten married]. Pray that all their needs will be met and that there will be peace between them. Her son Nagaraj has left his wife and has joined with another woman. He is speaking to her [Vaila] with great cruelty. Pray that this son will happily join his wife again and that he will break with that other woman.

Let Christ give Rita Vennila relief from debt problems. She needs to give two months’ rent, and they have asked for the money all at once. She must give it on Wednesday. Let the Father ensure that she may meet these needs.

The brother called Kanniyappan is suffering from mischievous magic that is being sent by a woman named Aravalli. Pray that the Father breaks all of these bonds.

Ellammal says that the brother called Karan has forgotten his mother and his father. Let the Lord smash all obstacles and help Karan to give love to his mother and father. Also, someone is sending spirits to create divisions and fighting between herself and her husband. We pray that these spirits will leave them and that the problems in their family will come to an end.

Suriya’s mother’s heart is blocked and she is suffering from weakness and pain. Pray that Christ gives her health.

P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T • 2 1 5

Hie man in the opposite house is speaking against Sathiya, and she is also having troubles with her daughter Mohana. Pray that this stop.

Anita prays that all worries will leave her, that her husband will join her again [i.e., return home after having left for another woman], and that God himself will change his heart.

Savari Ammal owes money to the chit fund she has joined. Her husband is not working. Hema Latha is speaking against her.

Sagayam is getting boils that frequently grow to become huge and cause great pain. She is suffering from fever and hip pain. Pray that these problems are relieved. Pray also that her son Ganga and his wife are relieved from debt problems and that they find peace in their marriage.

Gandhi M athis husband is having doubts about his wife and falsely accusing her. Let us pray that these doubts will be removed, that the family will have peace, and that he will be released from the drinking habit.

Saminathan asks that he be called for work.

Saral needs a Rs. z,ooo loan. Pray that the Lord will help her to find someone to lend this amount to her.

Z l 6 • P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

S E V E N

Salva tion, K nowledge, a n d Suffering

t h e s e r v i c e s o f s l u m c h u r c h e s were frequently lampooned by the H indu and mainstream Christ ian residents o f A nbu N agar as being just a pandemonium o f shouting, misdirected enthusiasm, and, when all was said and done, “ N othing but ‘ H allelu jah! For participants, however, these weekly productions were endlessly novel, witty, and replete with important and often startl ing insights. As many churchgoers described it, for the first time in their lives they were ga in ing genuine “ knowledge” (,arivu). “Before we had nothing, we knew nothing . . . only now have we begun to learn . W ha t is Christ? Knowledge! [ecu nd arivu tan!] ” M any church women had had little or no schooling, but all save the totally illiterate owned Bibles and read them, often pa instakingly scrutiniz ing every passage. A lways with in reach, these Bibles were gradua lly transformed, inscribed with the traces o f each w eeks revela � tions. Slum women never came to church without that Bible in hand, and the moment a verse was cited they scrambled to find the page with help from those around them, underlining significant words as they were read aloud, with cross-references scribbled into margins. A n d on separate sheets o f paper folded into the pages o f these Bibles were lists o f chapters and verses read on each date, and notes on their interrelations, kept for later reference and exploration. “ Those H indu cdmis never speak—how can they? H a! They are nothing but stone! O nly our god is a true god. O nly our god gives knowledge.”

W h a t was the C hrist ian god ac tual ly tell ing these people? A n d why did converts find it so unique? It was not entirely true that the C hrist ian god was the only one ever to speak . H indu gods also spoke to human beings, during what was known as cdmi dttam (divine possession). Christ ians o f course mainta ined that possession i tse lf was feigned because H indu gods were imaginary. But that was not the most significant difference between the two.

2 1 7

The divine speech o f H indu and C hrist ian gods differed in form , in content, in the social situations in which it was produced, and in the ways the distinc � tion between the human and divine was organized and maintained. Yet they were in other respects similar. Both were oriented toward the solution o f worldly problems; both revolved around sin and its removal; and both in fact provided knowledge, albeit o f very different kinds.

H indu gods spoke in a terse, straightforward manner. They spoke, not as part o f a weekly ritual cycle, but only when a particular person or family approached the medium in search o f an answer to some question. The ques � tions people approached gods with typically concerned why something bad had happened and what needed to be done. A medium (cdmiyati) in A nbu Nagar, for example, was consulted by a couple wanting to find out why their daughter-in-law had not been able to get pregnant and what to do about it. The god revealed that both she and her mother-in-law had sinned in some way and that to atone, a series o f ri tual offerings were required. This was a typica l case. Ritua l and moral infractions were among the most common reasons things went wrong in people’s lives, according to H indu gods. The others were malevolent spirits, wicked people attacking them magically, and astrological evils. These consultations were often private and only rarely attracted more than a smattering o f onlookers. Even when others were present, the god ’s words rarely addressed anyone other than those who had approached them. A n d though the H indu tradition as such is replete with philosophical complexity and sophisticated moral reasoning, gods did not reveal matters o f general theoretical importance through mediums.1

W here H indu divine speech was terse, the Christ ian god was endlessly prolix; where H indu gods spoke concretely, the Christ ian god spoke meta � phorically and through stories that required (and sometime defied) interpre � tation; where H indu gods addressed specific situations and questions, the C hrist ian god conveyed abstract or general lessons; where H indu gods spoke sporadically to a limited audience, the Christ ian god spoke publicly and on a weekly basis. Finally, al though both Christ ian and H indu gods might be said to speak through human lips, they did so in markedly different ways. The words o f the Christ ian god were spoken, not through an individual in trance, but through the lips o f a fully conscious medium , typica lly the pastor, but also ordinary believers. Sometimes he even spoke through many people simultaneously, to cacophonous effect.

W hen a H indu god spoke, there was no ambiguity about who was speak � ing and when. In cami attain, this distinct ion between human and divine

218 • S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

voices was ensured by the fact that the social person, whose physical form the god took possession o f in order to speak , remained entirely absent from the scene so long as the deity was present. She was neither conscious of, nor in any way responsible for, what the god might say or do while in possession o f her body; her physical form— as much as that o f a temple idol— li teral ly became the god ’s for the duration o f the possession and had to be respected as such. The C hrist ian god ’s voice, by contrast, mingled in uncerta in proportions with those o f his human mediums. Both human and divine voices were co � present (cf. Volosinov 1986) in a single utterance, each contributing in intrin � sically uncerta in proportions to what was said. Rather than tak ing fu ll con � trol over the believer’s vocal apparatus, as H indu gods did, the Christ ian god entered into and in some unknown fashion refracted human speech such that the end result was neither entirely that person’s nor directly G od ’s own. Thus al though the human and divine authors remained, in principle, dis � t inct, there were no clear-cut means to determine where one left o f f and the other began. In this indeterminate m ix o f human and divine voices, absolute truth and merely human additions were not easily disentangled. A n d because the content o f such speech was, moreover, often highly abstract, metaphoric, and sometimes downright cryptic, no utterance could ever simply stand on its own . The onus o f interpretation , moreover, fell not only on listeners but also on speakers themselves, who had to work hard to make sense o f what they themselves were saying. Speakers thus constantly referred back to and sought to interpret their own words.

This recursive aspect o f slum C hrist ian discourse combined with addi � tiona l features to ensure that no interpretation was ever final. In contrast to the self-contained and monogeneric speech o f cdmi âttam, the weekly serv � ices o f the slum church comprised multiple speech genres, inc luding the uninterpretable anniya pâcai (glossolalia). L itera l ly meaning “the foreign/ other language,” anniya pâca i was never tota lly distinct from other forms o f church talk , into which it was interwoven , and was perhaps best understood as just one extreme in a range o f spoken forms that varied from everyday speech to incoherent bursts o f poetic metaphor. The speech genres o f the Sunday service spanned lucid textual exegesis, the singing o f songs, homey personal anecdotes, and free-ranging verbal performances in which the expo � sition o f present-day events was seamlessly interwoven with archaic biblical verse, or in which a pastor’s stream-of-consciousness commentary on G od ’s words might dissolve, imperceptibly, into incoherent flurries o f otherwise recognizable words or ecstatic invocations o f divine glory or glossolalia, only

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2 1 9

to coalesce once again into a relatively straightforward discussion o f some entirely new topic.

Sunday services were also highly interactive. Pastors might ask questions o f their audience, and they o f him. Bible verses used in preaching were read aloud (or recited) by audience members and were then incorporated into the sermon by the pastor in an impromptu fashion. The Bible frequently quotes God directly, as well as other people, and was i tse lf regarded as G od ’s word. Pastors would often call out a certain chapter and verse, and everyone in the room would rifle through their Bibles to find the passage as quickly as possible, several people at once beginning to read. O ne voice would emerge more clearly and authoritatively above the others, who would then fa l l silent. F inding verses quickly and reading them aloud in an authoritative tone was a skill, but it was understood that God himse lf determined which human voice would rise above all others and it was God who in fact spoke through them. The pastor deferred to that speaker, often asking him or her to repeat what had been said or to correct him i f he had understood incorrectly. The multiplicity o f speakers (sometimes speaking simultaneously) combined with the multi � plici ty o f voices (human and divine), the multiplicity o f genres, multiple layers o f meaning (literal, metaphorical, allegorical, allusive), and the recursive qual � ity o f the speech performance (the fact that people referred to and interpreted their own and others’ spoken words). This produced interpretive puzzles. Thus, although the Christ ians’ god spoke to them, he did so in a way that required their active and ongoing interpretation. Theirs was the god who “gave knowledge,” but the knowledge he gave was never complete. There were always new revelations to be had, new discoveries to be made. This was no doubt part o f what kept people coming back week after week, and what allowed them to remain in a heightened state o f attention for hours on end.

Going to church in the slum was exciting, but it was not just the way the message was preached that really mattered to people. It was the message itself, the knowledge about their predicament and how to escape it. This was where the most significant differences between the speech o f H indu and Christ ian gods emerged. Both , as I said, provided knowledge; both were oriented toward the solution o f worldly problems; both revolved around sin and its removal. The Christ ian god, recall, addressed all the same sorts o f worldly problems that H indu gods did. But the C hrist ian god ad � dressed other problems besides, problems that loomed large in slum dwellers’ lives but were not among the ones H indu gods appeared concerned with . These were, first, the oppression slum dwellers collectively experienced at the

2. 2.0 • S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

hands o f the dominant society and, second, the il l treatment slum women suffered at the hand o f other slum women and their own husbands.

The critique and transformation o f existing social relations were central to the discourse o f the slum church. The problematic social relations Christ ianity addressed— those between slum dwellers and others, and among slum dwell � ers themselves—were not clearly distinguished from one another. Both were subsumed with in what slum Christ ians called “sin.” By using the same word, and the same set o f metaphors, to describe mistreatment in the interpersonal realm and systematic class or caste prejudice, the two were made to appear, in the tumult o f divine speech, as aspects o f a single problem.

S A L V A T I O N F R O M S U F F E R I N G

In fa ith all w i l l be done— H a l le lu jah! So be fa i th ful to C hrist the Lord in order to be rid o f all th is in just ice. In Zach ar iah 3:9 we read: “ In a single day I w i l l take away the sin o f th is land; I w i l l one day wipe away all atrocities \_akkiramankalaiyelldm n ik k i \ , all in just ice.” . . . H e knows about your strug � gles . . . [and] he w i l l bring about equa l ity by means o f a revolution [oruperiya cam a tt uva t ta ippura t ciyu ta n erpa t u t t uvdr \ !

To encounter the Bible in A nbu N agar was to discover a book that reverber � ated with the drums o f revolut ionary just ice and the promise o f human equality. Consider the following biblical verses, common among those quoted from memory and strung together by slum Christ ians during church services as a k ind o f ecstatic battle cry.

H e who raises the poor and weak from the dust [Ps. 113:7], we praise y o u ! . . . He who rescues the oppressed, and brings down those with haughty glares [Ps. 18:27], we praise you!. . . He who has saved us from our enemies, and has humiliated those who hated us [Ps. 44 :7] , we praise you!. . . He who punishes caste folk [Ps. 94:10], we praise you! Oh Refuge of those who are disgraced [Ps. 9:9], we praise you! . . . He who has broken our chains [Lev. 26:13], we praise you! He who has made us walk with our heads held high [Lev. 26:13], we praise you! . . . He who stands against the proud [1 Pet. 5:5], we praise you !. . . He who raises the strength o f those who have none [Isa. 40:29], we praise you! . . . He who rescues the weak from the hands o f the strong [Ps. 35:10], we praise you!. . . He who hears the pleas of those who suffer [Ps. 10:17], we praise you!. . . He who hears the appeals o f the poor [Ps. 69:33], we praise you! . . . He who pursues justice for the poor [Ps. 140 :12] , we praise you! . . . “I will deliver you from the hands o f the wicked, and I will wrest you from the grip o f the terrible” [Jer. 15:21]—for this promise, we praise you!2

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2 2 1

It was no accident that many slum Christ ians believed, as one pastor affirmed, that “ the aims o f K arl M arx and the aims o f Jesus C hrist are identica l” The only problem with Marxism , according to slum Christians, was that it mistak � enly believed humans could transform the world without G od ’s help. But while no credible communist party existed in T amil N adu , many slum Christians indicated that i f one did they would gladly vote for it—just as D a li t Pentecostals in the neighboring state o f Kerala often do (Thomas 2008).

This vision o f revolution and the centra li ty o f their own suffering to the unique form o f Pentecostal C hrist ianity that existed in the slum was evident even with respect to the most basic C hrist ian tenets, such as the concept o f sin. Lik e Christ ians worldwide, members o f slum churches were intensely concerned with be ing “rescued from sin.” But unlike Christ ians elsewhere, the sins that occupied them most were not necessarily those they themselves had committed. More often, being “saved from sin” in the slum meant being saved, not from one’s own sins, but from the sins o f others. “ H e who rescues us from sins—he is with m e . . . . The Lord C hrist has promised, ‘I w i l l rescue you.’ Yes. God wil l rescue us \nammai\from the wicked a nd ev il people.” W ho were the wicked people? A n d who were the “we” who w il l be rescued from them? O ne th ing is clear: “we” did not refer to the members o f the part icular congregation, or, interestingly, even to Christ ians as a whole. Slum congrega � tions did not pray for their own benefit or for the benefit only o f other Christians. They prayed for themselves, for each other, and for the well-being o f their neighbors and relatives irrespective o f religious affiliation . O ne did not need to be C hrist ian to benefit from their prayers. The open-ended “we” who needed to be saved from (other people’s) sins in fact extended to all those slum dwellers typically referred to as “ the poor.”

The sinners from whom the poor had to be rescued were never explic it ly named. But various hints and verbal clues made clear that the paradigmatic “wicked ones” were caste people. In chapter 2 I explained that slum dwellers were in most circumstances extremely reluctant to speak in general terms about caste people as such and for various reasons preferred to describe them simply as “the rich” or “ the privileged .” A l l humans beings were created the same, slum dwellers said, and caste was a “mere lie.” Bet ter to describe their collective oppression as what it really was, an effect o f worldly power and privilege. O n this point, like so many others, slum pastors were personally very much in sync with other slum dwellers. O n top o f this they believed that it was against the “rules o f C hrist ianity” to talk about caste. The argument was that “ G od does not see caste” and therefore we should not either.

I l l • S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

N ot seeing meant not acknowledging, or at least not openly acknowledging. Yet slum pastors in fact talked about caste al l the time, without explicitly naming it as such; caste was the conceptual template through which all social suffering, or sin, was understood. N ot explic it ly labe ling it as “caste” had a distinct cognitive advantage: it al lowed pastors to conceptualize sin in a way that tapped into slum dwellers’ deeply felt resentment at their collective caste-based subordination , while extending its significance to other, more intimate experiences o f rejection and abuse.

Consider the following extract, which is typica l o f the way messianic jus � t ice was tied to highly persona lized and, for slum dwellers, evocative exam � ples. A lso typica l here is the way the pastor slips seamlessly from the personal to the collective (“ H e w i l l lift you u p . . . he w i l l l ift us up”). The only unusual th ing about the following is that the pastor seems to have become so lost in the moment that he forgets entirely the rule that Christ ians are not allowed to ta lk about caste. This is the only time, to my knowledge, that such a slip occurred during the entire period o f my research.

In our families there is no blessing. . . there is dire poverty. Please transform these tears! . . . Christ sees the tears of everyone who cries! Hallelujah! . . . They say: “You are a shameful man. Your life is worthless—you earn nothing for your family! What are you even living for?” But Christ will not desert you—he will lift you up. Christ uplifted those who were shamed, those who were pushed away. In the same way . . . he will look upon us lowly creatures. He will lift us up . . . He will place you on the highest peak, before your ene � mies, before people from other castes [pirajdti makkalakku]—Hallelujah! In front o f those who spoke ill o f you, those who tormented you, who detested you, who pushed you away, who said you were not qualified, who said you were without talent or intelligence, who said you were useless—Christ will bind their tongues! Hallelujah!

In this way slum dwellers’ sense o f persona l and collective disgrace was imbued with cosmological significance and tied to the promise o f messianic justice. Their oppressors were not usually identified explic it ly as caste people; the previous extract is exceptional in this regard only. Yet when the pastor said this in the heat o f the moment, no one seemed to notice. The real ity o f caste was no secret, and in the discourse o f the church it was never far from the surface.

More often the caste character o f the division between slum dwellers and their non-D al it oppressors was acknowledged only obliquely, metaphorically, or in other ways that permitted multiple interpretations. Take, for example,

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2 2 3

the next extract, from a different sermon, which centers on the locally potent symbols o f impurity and rejection— the garbage and the waste-strewn land � scape o f the slum itself.

What is it that we call trash? Is the very place we live in a garbage dump? \Namma urkuppai tottiydf] What is o f no use is called trash. What we throw out is trash. What is not needed is trash . . . Only now—did you see it on tel � evision?—from trash they have produced electricity! From the trash [that is spurned as] “useless,” “unwanted,” and “rotten,” today, in the name o f Christ, they are making electricity!. . . .

And the villagers [kirdmattdr], everybody, will come to know of it! Assuming you were dirt they threw you away. [But] you are like an unseen garden. . .Jesus says: Give me your suffering. . . and I will clean it. I will make you invaluable. I will make you respected!

In this extract, the promise is again made that those who have been brought low w ill be raised up. A nother typica l feature o f slum church ta lk that this extract nicely illustrates is the likeness between the knowledge C hrist brings and scientific truth (“ W ha t is Christ? Knowledge!”). The prejudices o f others are displayed here as not only cruel but ignorant and behind the times. Though caste is not spec ifically mentioned here, it is strongly implied in the contrast that is drawn here between the unwanted ones and the kirdmattdr. The literal meaning o f kirdm attdr is “villagers,” a strange way to speak o f other people in an urban setting. The term refers more spec ifical ly to inhabit � ants o f the village proper, the ur, as distinct from the D a l i t hamlet or ceri. A n d it is diff icult not to see the residents o f a garbage dump (kuppai totti) as a reference to the slum itself, or ceri. A t the same time, “what is o f no use,” “what is not needed,” and so on could refer to anyone who has been under � valued in their relations with others. Women frequently described themselves this way, and the slum women who felt misunderstood and underappreciated by their husbands or other women might understand the words “You are like an unseen garden” in a more personal way, in addition to whatever collective message they took from this sermon.

The caste division between slum dwellers and the dominant society was thus a recurring, i f encrypted, theme in the message o f the slum church. Yet the fact that it was not explic it ly named meant the discourse was flexible enough to encompass other forms o f domination and indignity. N o t all forms o f suffering are reducible to caste, and in addition to the collective caste/class suffering o f the slum as a whole, preachers also referred more explicitly to the suffering o f women . A t other times strife was described in

Z Z 4 S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

terms too general to pin down. “ The O mnipotent is with me, so those who torment me shall not destroy me___Either their actions w i l l go to waste and they w il l be shamed, or they w i l l be subjected to unforgettable [and] . . . extreme d isgrac e .. . . God knows what is right: he w i l l bring suffering on those who make you su ffer!” This extract, like the previous two, refers to human cruelty and to divinely orchestrated justice. But the identity o f the tormenters in this case is not at all clear. N or does the preacher specify the nature o f the torment, and therefore the identity o f the victim . Is it a woman whose neighbors have turned on her? Is it a man or woman who is habitually mistreated at work , or by those he or she owes money to? Is it all slum dwell � ers, or even everyone who has ever been humiliated? By not spec ifying the vict im ’s identity, such passages are capable o f ha i ling a wide range o f listeners, seeming to speak direc tly to their personal struggles.

Given the often open-ended and mult ifarious way victims o f in justice were envisioned in the discourse o f the slum church, it is strik ing that one collectivity in particular was not identified in the victim role: Christ ians. It is even more strik ing given that my fieldwork was conducted at a time when H indu majoritarians in Ind ia were increasingly target ing C hrist ians and when a law had recently been enacted in the state o f T amil N adu banning conversion. There were at least two distinct reasons Christ ians were never cast in the victim role. First, although slum Christ ians, like others in A nbu N agar, identified themselves as victims o f the dominant society, they did not see themselves as vict imized for being Christ ians. A fter all, the ways slum Christ ians suffered were mostly identical to the ways non-Christians suf � fered, and Pentecostal Christ ians in the slum saw themselves not as vict ims but as liberators. The targeting o f Christ ians by H indu majoritarians and others identified by slum dwellers as “ the r ich” was characterized by slum Christ ians, not as a sign o f their own vulnerabili ty to India’s most powerful forces, but as a rearguard and ultimately hopeless attempt to prevent the poor from rising up.

A s I have noted, when slum Christ ians prayed for the cessation o f suffering, they were not praying for just themselves and for other Christians. They prayed for their husbands, their wives, their sisters, their friends and neighbors—who were H indus as often as not. Likewise, when a slum pastor spoke about those who were demeaned, mistreated, insulted, and “cast out like garbage,” he was reaching out not just to his fellow Christ ians but to all who had been treated this way. A l l who were poor and whose humanity was rejected by others would be uplifted. W hen Christ ian women exulted in the promise o f triumph over

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2.2.5

collective enemies, they were not exulting in a triumph that would exclude their H indu husbands, neighbors, sons and daughters. Their devotion to Christ heralded a revolutionary justice that would transform the earth for everyone. That was how the universality o f the Christ ian message was understood in A nbu N agar and the surrounding slums.

Second, identifying the collective victim role with a victimized “ Christ ian community” would have contradicted the primary message o f slum churches, in which C hrist ianity was identified not with vict imhood but with victory. C hrist ’s own suffering was often invoked by slum Christ ians, and his suffer � ing presented as a victory. W hen slum dwellers became C hrist ian their suf � fering was similarly refashioned, in ways we w i l l soon come to, as a tool o f liberation. For this reason, when a biblical reference for collective suffering was required, it was normally an O ld Testament example that slum preachers reach for. The collective suffering o f A nbu N agar ’s residents was thus likened to that o f the tribes o f Israel. But the tribalism o f the O ld Testament was also explicitly rejected as a model for Christ ianity , which was conceived not as an exclusive identity but as coextensive with the human: “Jesus was born into the nation o f Israel, in the caste \jdti\ o f D avid . But he was made human, so what did he do? H e did not remain just with that caste. H e came out o f it, and blessed each and everyone!”

S U F F E R I N G A S S A L V A T I O N

When they spat on Jesus Christ’s face, what did he do? Did he spit back? When they insulted him, did he insult in return? The Bible says he did not open his mouth__ And yet we keep on asking: “ W hy must I endure so much pain?” “ W hy are there so many troubles for us?” Why? Why? Why? The answer is: it is for salvation! . . . It is for salvation that you must endure! Endure, endure, endure! Keep enduring! How long must you endure? Until the end! Yes. We must endure until the very end. Endure or die! That is the choice. For i f you do not endure, you cannot live in this world. So you have to endure, dear children of God, you must learn to endure.

In te ll ing believers they “must learn to endure,” slum C hrist ianity could be accused o f making a virtue out o f necessity. W h a t sense did it make to tell the powerless that they must endure, when the truth was that they have no choice? Enduring was what these people did; it was what they had always done. In the pastor’s own words, “ I f you do not endure you cannot live in this world.” The only escape was death: “ Endure or die! That is the choice.”

22 . 6 • S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

H ie re was more to it than that, however. First, what slum Christ ianity told believers was that they were not just l ife ’s losers but the “ dear children o f God .” Slum dwellers’ degradation was no longer an inexplicable injustice but a harbinger o f their own inevitable triumph (cf. N ietzsche 1997; Geuss 1999). Second, believers were led to understand that i f they embraced their suffer � ing, and did not fight back against husbands and neighbors who mistreated them, their suffering i tse lf would become the means o f their own liberation .3 Suffering was in this way transformed from a passive experience to an active one (Asad 2003), a “task” that had to be embraced as the very means o f their salvation: “ W hen you are smarting [at the apparent success o f your tormen � tors], what that means is that the Lord C hrist is giving you a great big task, a great big challenge. A n d when the Lord gives you a big task like that, what does it mean? It means that that there definitely will be equality. So i f you look now, and all you see is sin, do not worry. Because that sin w il l be removed and there w i l l be equa l i ty!” U nlike Christ , slum dwellers had no choice but to suffer. But the more they were able to embrace their suffering and to suffer will ingly, the more Christ-l ike and powerful they would become. Just as C hrist ’s suffering served a purpose, so did theirs.

In the previous chapter we examined women ’s prayer networks, through which the personal struggles o f other slum dwellers, H indu as well as Christ ian , were repeatedly made present to believers’ minds, and an ethic o f shared suffering was cultivated . Sunday sermons offered the theoretical rationale for these prayer practices, as believers were reminded again and again o f how C hrist had suffered for the poor. To suffer for others was to be like C hrist . A t the same time, the suffering o f the poor was described as be ing li tera l ly one and the same as the suffering o f C hrist himself. Poor believers were told that when they suffered for others they were in fact suffering for their god. By deliberately suffering for others, the poor thus acted simultaneously as C hrist and upon him : “ W hatever you do for G o d ’s children , you do for G od himself, and whatever help you give to others is giving to the Lord .” W h a t slum Christ ians did for the poor, further � more, they did not only for their god but also for themselves. A s another pastor explained, “ G iving to the poor is giving to the Lord . A nd who is the poor? A ll o f us here are the poor!” By su ffering actively for others, slum Christ ians occupied the role o f C hrist himself. Just as C hrist was wrongfully tormented by those who fai led to recognize him , so too were they made to suffer by those who did not see them as they really were—human , but also in some way divine.

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2 2 7

The sense o f identity that slum Christ ians envisioned between themselves and the suffering C hrist was powerfully revealed to me by a ioi-year-old man known as Umbrella Preacher. Most o f his life had been spent as an itinerant, a wandering preacher whose only shelter from sun and ra in was a black umbrella with Bible verses pa inted on it. Born a Roman Catholic , at a time when D ali ts were st i ll known as “Pariahs,” he had had little religious educa � tion as a child and his family seldom went to church. H is father had been a poor laborer who made extra money as a practitioner o f black magic, a profes � sion he had passed on to his son. In his early twenties Umbre lla Preacher experienced a series o f unexpected and terrifying visions that left him run � n ing half-naked in the streets. A f ter wandering in a confused state for many days, he eventually recovered his senses and began seeking answers. H e found them, eventually, in Christ . H e then took up the umbrella and began the life o f an independent preacher.

W hen I met Umbre lla Preacher, he was housed with a family just outside Kashtappatt inam . N o longer able to preach, he spent his days in solitary prayer. A lmost everyone had forgotten about him , and I learned o f his exist � ence only accidentally, after more than a year in the field, from my good friend Pastor Yesudasan , who mentioned him in passing. Yesudasan was Umbrella Preacher’s son. Umbrella Preacher was a diminutive, soft-voiced man, and I found it difficult to imagine h im capturing the attention o f a crowd. But even at his advanced age he remained completely lucid. A mong the many things I asked him about were his recollections o f caste in the pre- Independence period. H is answer surprised me because he replied with a story about preaching the gospel. Umbre lla Preacher had been preaching on M arina Beach, a popular gathering place in M adras, when he found h imse lf suddenly surrounded by a gang o f caste H indu men. This was in the 1930s, at the height o f G an d h i ’s H ari jan uplift: cum anticonversion campaign , a period when anti-Christian sentiment, which had always been rare in T amil coun � try, seemed for a brie f time to have surged among caste people. “You talk about Jesus, but we can’t see h im !” the men taunted, l ift ing the little Umbrella Preacher o f f his feet and carrying him to the edge o f the ocean. They were going to toss him into the surf. U mbrel la Preacher (figure 15) did not know how to swim , but he neither begged for his life nor made any attempt to resist. Instead he began to silently pray that his attackers would be forgiven for what they were about to do, that they would be saved. “ W here is Jesus now?” the men demanded. “ W hy doesn’t he save you?” A crowd o f spectators had gathered, and out o f that crowd a M alayal i appeared. “You want to see

Z l 8 • S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

f i g u r é i s - U m bre l la Preacher.

Jesus?” the M alayal i said, speaking to the men. “ That is him , the one you are carrying.” Stunned by the M a laya l i s words, the men released Umbre lla Preacher and silently dispersed.

I had asked the old man about caste, and he had answered with a story about h im se lf and Jesus. H ad he not understood my question? In fact he understood it perfectly. In the minds o f his attackers, Umbre lla Preacher was a not only a C hrist ian but a rebellious Pariah.4 C arrying an umbrella had once been forbidden to D a l i ts (V iswanath 2014b , 30, 78), and even today in some villages for a D a l i t to carry one is seen as a deliberate provocation. But to truly understand caste— for Umbrel la Preacher, as for the people o f A nbu N agar—was to understand that it was a lie. Lik e Umbrella Preacher, Jesus was born into a part icular caste. But his caste identity meant nothing; what

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

mattered was that he had been made h u m a n . A s the extract we examined earlier explains J e s u s came out o f his caste and “ blessed al l people.” Likewise, Umbrella Preacher prayed not for his own salvation but for those who remained trapped with in the sin o f caste, his attackers. L ik e Jesus, he was w il l ing to sacrifice h im se lf so that others might see. A n d in the mystical kernel o f the story, they did. In his will ingness to suffer and even to die for others, Umbrella Preacher ceased to be a Pariah in their eyes and was recog � nized as the Son o f M an . H e was transformed, in an instant, from the lowest o f beings to the most high.

R E M A K I N G R E L A T I O N S

Slum C hrist ianity promised complete transformation , a moral revolution it tended to portray as sudden and total, as occurring at an unknown moment some time in the future, and as brought about by divine agency alone. This transformat ion occurred not in the everyday but in what has been called revolutionary-messianic time (Ben jamin 1968; Low y 2005; cf. Robbins 2007). For al l the prominence the slum church accorded to relations between the suffering poor and their oppressors, it was equally concerned with strife among slum dwellers themselves. Because it was not just members o f the dominant society that sinned. H usbands, wives, neighbors, and friends al l at times failed to treat one another with the love that they deserved. U nlike the expected transformation in their relation with the dominant society, how � ever, the transformation o f intraslum relations was typica lly portrayed, not in the language o f divinely orchestrated revolutionary justice, but as a series o f incremental improvements in what we might cal l quotidian-progressive time. That is to say, change was understood as a matter o f cumulative improvement that depended, at least in part , on human effort. Yet the gap between these two dimensions should not be overstated. Slum preachers themselves never in fact presented the two as distinct. O n the contrary, the conceptua l distance between quotidian transformation and millennia l upheaval was obscured in the discourse o f the slum church. First, revolution � ary transformation was evoked using highly personal language. Second, the everyday triumphs in face-to-face relations that we w i l l examine next were invested with the significance o f revelation. In this way the millennia l seemed almost to erupt into the everyday. Cou ld it be that the small battles o f their da i ly lives and the suffering they experienced collectively as slum dwellers

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

were ultimately one and the same? In the discourse o f the slum church both were described as manifestations o f one and the same problem, sin.

N o one knew precisely what the apocalypse would look like. W ha t slum Christ ians were sure of, however, was that it could happen at any time. It might already have begun . The promise o f total victory, as we have seen, was inseparable from the in junct ion to endure. Slum Christ ians were told to be patient and to w ill ingly suffer insults without fighting back, secure in the knowledge that their tormentors would one day be punished . A s they identi � fied their own suffering with that o f C hrist , their apparent defeat in life became the p roo f o f their own inevitable victory. But suffering was not sim � ply a passive experience for slum Christ ians. I t was a potent social and spir � i tua l practice and a technique for persona l and micro-soc iological transformation .

In chapter 3 we examined two hidden fault lines in the moral constitution o f slum life. M ora l community in the slum depended on the free sharing o f material resources among married women , even as these same women were pit ted against one another at another level by the discourses o f auspiciousness and va n k a n avi. Moreover, while sharing freely most o f the time, slum women also flagrantly violated slum moral codes by lending money to one another at exorbitant rates o f interest. The other relation in which moral expectations were routinely violated was in marriage. H ere again women bore the brunt o f moral censure, in part, I argued, because o f the way prevai ling ideas about auspiciousness held women ultimately responsible for the well-being o f their individual households. C hrist ian prayer practices, recall, eased these burdens by redistributing spiritual responsibili ty for each woman ’s misfortunes throughout the community o f prayers. W e w il l now see how suffering was actively employed as a technique for remaking relations among women and between women and their husbands.

The in junct ion to share in the suffering o f others and to embrace ones own suffering as the suffering o f C hrist was supplemented by specific instruc � t ions on how to conduct one’s relations w ith others. A major theme in slum sermons was women ’s frequent quarrels and how women needed to learn to endure others’ enmity without strik ing back.

We must not bear any grudges—no wrongs, disappointment, or irritation at each other. We cannot have all th is.. . . I f I bear anger I must get rid of it. Because then God cannot be inside us . . . . So what is your salvation [going to be] ? Anger, jealously, and doubts? Enmity, arrogance and fights?. . .

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • Z 3 1

Show a gentle attitude toward everyone. The Lord is coming soon—Amen! So there should be no secrets within the community.. . . There should only be Christian love within us . . . [for] there are no divisions in the Kingdom of G od . . . . This year what did you do for the Lord? Let us see. “I started a fight with another lady.” The love [anpu] we have been given must never go to waste [like this]!

Instead o f quarre ling w ith other women , Christ ians should manifest G o d s love. Acrimonious relations caused G o d ’s love to “go to waste.” Just as caring for others was in fact caring for C hrist , bickering with others squandered the love H e had shared with believers. I t short-circuited the spiritua l dynamo in which love begot love, the very source o f C hr ist ’s miracles.

Women were in this way inducted into a practice o f emotiona l self-regu � lation in order to better manage their relations with others. Consider the instructions women were given, on another Sunday, to refra in from retaliat � ing in k ind when subjected to other women ’s malice or gossip.

Do not pay back evil with evil or cursing with cursing. Instead, pay it back with a blessing, because a blessing is what God promised to give you when he called you . . . . Christ taught Anna [a member o f this pastor’s congregation] to improve herself. But how did this other woman react? She thoughtlessly insulted Anna. She insulted everyone in Anna’s family. But what did Anna say? “You may insult my mother and father, but my Lord will never insult me. He will place me on the highest peak. He will wipe away my tears.” . . . I f today you are suffering then I can tell you with great surety, right now, that he w ill wipe away your tears! Hallelujah!

Instructions o f this sort were repeated in different ways, with different exam � ples, week after week , and they did not fa ll on de a f ears. The pastor did not present these tales as models o f extraordinary goodness; the forms o f self- restraint and will ingness to suffer that slum C hrist ian ity en joined in the fa i th ful were not optiona l but absolutely required o f them. This was what it took to tap into the supernatural power o f Christ , the only power that could save them and their families from misery.

In my own time in A nbu N agar I saw the effects. I saw women ca lmly refuse to get angry or speak il l o f those who were maligning them. O thers too remarked on the effects that C hrist ianity had on people; in the slum there was no doubt in anyone’s mind that becoming a Christ ian made both men and women better people. A s Karuppan , an atheist D a l i t act ivist we met in chapter z, explained, “ C hrist ian ity def in itely does good for some people, women especially. M y mother is an example. She used to fight with the

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

woman across from her house all the time. They would just look at each other and for no reason start shouting. She was always angry. N ow there is no more fighting. She’s so busy praying and reading the Bible, she doesn’t even have time for fighting! She used to use horrible, foul language— it was frightful to hear. N ow she says only good things. That is what I call ‘development.’ ” To be sure, not everyone wanted to be transformed in this way. Indeed, few slum dwellers had any interest in goodness for its own sake. O ne woman I knew quite we ll con fided her ambivalence about church. She had stopped attend � ing because, as she explained to me, she really enjoyed using foul language and chewing betel, which the people in her church said was not allowed. She remained persona lly devoted to Christ , however, and had no interest in ever going back to her former gods or wearing jewelry or other adornments again. A n d Christ , she told me cheerfully, continued to perform miracles in her life even without her going to church. “ H e supports me no matter wha t!”

This woman was an unusual case. She was the only person I knew who believed one could receive C hrist ’s blessings without giving up bad habits. But she was typical in another respect. W ha t was o f foremost importance to her was the worldly benefits her religion provided. In this way Christ ianity and H induism were identical. W here the two religions differed was that the k inds o f worldly benefits C hrist provided included not only such things as curing illnesses, helping children pass their exams, and so forth , but also transforming women ’s relationships with one another and with their hus � bands. The ac tual work o f remaking these relations was, quite obviously, being done by women themselves. But that was not exactly how they saw it. They were o f course aware that their own forbearance was having an effect, but the fact that it had an effect— that others’ hearts were also moved—was credited to divine intervention . Their endurance o f suffering and refusal to strike back was seen as a sacrificial offering to God , who then reciprocated by softening the hearts o f their significant others.

Let us now turn to the other key relationship in women’s lives, their mar � riages. O ne way slum churches sought to strengthen marital relations was paradoxical ly by relat ivizing the importance o f marriage and the household. Compared to their relation with C hrist , women were told, the household simply did not matter. Yet the household did matter—because its well-being was sti ll held out as a reward for women’s devotion . Similarly, by elevating women ’s relat ionship w ith C hrist above their relationship w ith their husbands, slum C hrist ianity in fact sought to improve that relationship, not displace it.

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2 3 3

[When the Holy Sp irit is in your life] you will not [need to] concentrate only on familial chores. You will not [only] be looking after your children. You will be able to say your household doesn’t matter. For the one and only important thing is that you are the pure bride o f Christ—hallelujah! I f you are a pure bride for [even] one day, at once there will be enough in your house, and your debts will be cleared. Your children will become pure. And your husband, too .. . . Such is the . . . might o f Christ in preparing this congrega � tion as his bride.

W hen women were invited to see themselves as the brides, not o f their hus � band, but o f Christ , the heaviest burdens o f worldly responsibility could be downgraded in significance. “You w il l be able to say your household does not matter.” These were revolutionary words. Yet, they were told, as soon as they stopped worrying about their family and just trusted in C hrist , “ debts w il l be c leared” and their household would not want for anything. In other words: family life was downgraded, not in the name o f some otherworldly goal, but ultimately in order to make family life better. The difference was that now the success o f the household depended not on women themselves but on Christ . Church discourse in this way supplemented women ’s prayer networks in shifting ult imate responsibility for the household o f f women ’s shoulders.

In pushing women to recognize their true partner not as their husband but as C hrist himself, the church did not seek to undermine worldly mar � riages. In fact the goal was quite the opposite: to promote marita l harmony. The first step, however, was f i l l ing women with the idea that they were wor � thy o f better treatment, that they were always beautiful and new in the true vision o f Christ: “ The Lord . . . is your husband. H e loves you just as a hus � band loves his wife at the beginning [of their marriage]___A f ter the wed � ding, when the girl is sent to her husband ’s place, what does the husband do? H e loves her a lot. H e w i l l definitely love her, because she is the n ew bride, the n ew wife. But for Christ , the wife never gets old! Every time he sees her, she is new!” Slum C hrist ian women were often described as “pla in” looking because they did not wear jewe lry or otherwise adorn themselves. A ccording to them, however, makeup, flowers, and gold were in fact only a false beauty and a way o f “showing off.” People were so accustomed to seeing these adorn � ments, men especially, that they were unable to perceive woman ’s true beauty and inner goodness. O ne churchgoer, a woman who lived in a nearby slum tenement and was therefore better o f f than most slum women , explained to me her own decision to put her jewe lry away forever, in a box. She had had

Z 3 4 • S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

lots o f jewe lry by local standards and had been quite proud o f it. But some time after coming to Christ , she had a vivid dream in which her nose and ear rings, the rings on her fingers, and the chains around her neck were revealed to be horrible spiders and snakes that were sucking her blood and slowly strangling her. Women must come to see themselves as C hrist sees them, she said. They must have that vision first, and only then, gradually, w i l l their husbands come to see it too.

Very few churches, however, actual ly required women to give up jewelry. It must be their own choice, Pastor Yesudasan told me. O nly when they were truly ready should they remove their jewels, and not before. Even then, he went on to say, they should wait. “ I tell the women in my church that they should not remove their jewels i f their husband opposes it .” H e too had to approve. “Jewe lry must not be allowed to become a source o f tension between husbands and wives.” A l l decisions in a marriage, according to Yesudasan , had to be made together by the husband and the wife. The idea that husbands and wives should work together was a common theme in the preaching o f slum churches. Women were thus filled with a sense o f their own worth and at the same time provided with a template o f what marriage ought to be like. D rawing again on the image o f newlyweds, another pastor explained, “A t the beginning o f a marriage a husband w il l say to his wife, ‘Even i f we die, let us die together!’ That is how a relationship should be! There must be no separa � t ion. W e must live as one\ W e must bring up the children as one—whether it is in death, life, hell, or liberation . W hether we are starving or feasting. For without that love—without C hrist ’s love— no task can be done. W h a t k ind o f relationship [am I ta lk ing about] ? A n inseparable relationship! That is the proper relationship between man and wife .” W e see clearly in this extract that C hrist ’s love for women was not intended to compete with that o f the husband but served instead as a resource for women to draw upon in their ongoing efforts to remake marita l relations in a new image.

W h a t is also important to bear in mind here is that even when sermons were ostensibly addressed to married women , the true target audience was in fact much wider. For al though only a hand ful o f husbands were with in the congregation itself, electronic amplif icat ion ensured that pastors’ words traveled throughout the surrounding slum and into the ears o f men enjoying their day o f rest at home. Keeping this wider, implic itly male audience in mind , consider the following account o f M ary ’s impregnation by the H oly Spirit. In this pastor’s te ll ing o f the story, M ary was accused o f adultery and the survival o f her marriage was in doubt.

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • Z 3 5

What did God do to Mary? He chose her. He put the seed o f God inside her. Can you even imagine all the difficulties Mary had to endure [because o f that] ? So many harsh words! So many insults! What would they have done to her? She would have pleaded with her family and her neighbors that she was an innocent woman__ Joseph too would have been rebuked and mocked!

What would Joseph do? He secretly thought o f sending her away. But did he? He did not.

The whole community was against her. A n d yet Joseph , who was called a cuckold, stood by his wife and endured for her sake the stigma o f not being the father o f his own w ife ’s son. Slum pastors consistently and authoritatively endorsed the idea that women should expect their husbands to maintain faith in them through th ick and thin . I f a husband did not— and in rea lity most didn ’t— it was not the woman ’s fault. That was slum churches’ message. To women it said, do your best, endure in the knowledge that your true, heav � enly husband w il l always love you. If, despite a l l that you have w il l ingly endured, your husband sti ll rejects you, you are not to blame. You are not a failure. It is he who has fai led you.

Slum pastors spent a great deal o f their time counsel ing women , fasting for them, and work ing to solve their problems. It was because o f their long hours l istening to women— their extraordinary dedication to the women in their flock— that they were able to so convinc ingly preach to women and to incorporate elements o f women ’s li fe experience into biblical exegesis. Seeing the world through women’s eyes shaped the way they understood C hrist ianity itself. Even in my own private, man-to-man conversations with pastors, they often drew on female-centric examples in their effort to explain to me C hrist ’s message to humanity.

Aga in with a dual m a le-fem a le audience in mind , consider the following account o f the marriage o f A dam and Eve— a marriage understood by slum Christ ians as a model o f ideal con jugality. The pastor here recalled the story o f the Fall, from which he derived the surprising lesson that husbands should always listen to their wives.

What did Eve do? She called to him, “Adam, Adam! Taste this fruit—it is very sweet, very wonderful! My eyes were opened, you too should eat it.” She said this, and gave the fruit to Adam. Adam tasted it. Why? Because for Adam, Mrs. Adam’s words were very sweet. He ate it, and his eyes were opened. Hallelujah! Listening to one’s wife’s words is important__ Why? Because what she says is very important! Hallelujah! The [very] sound o f a wife’s voice is essential! . . . [So] you should lend your ears [cevicdykka

Z 3 6 • S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

ventuni\ . . . to your wife. A wife is not a slave! You must listen to her! I f you keep shaking your head to everything she says, we will all be ruined!

The ideal o f the loving, companionate marriage that Christ ian pastors repeat � edly and publicly promoted was not seen by anyone in the slum as specifically Christ ian . N or was it perceived as uniquely modern , or W estern. A marriage o f this type was seen by men and women alike as self-evidently appealing, and everyone claimed to strive for this ideal in principle, even as they recognized that marriages rarely lived up to it in practice.

The specific innovations o f the slum church were, first, in providing the fa i th ful with a framework o f advice and techniques with in which to pursue this goal, and second, in m aking the success o f its members’ marriages a col � lective spiritua l responsibili ty and a benefit to the slum community as a whole. Thus i f the man kept “shaking his head to everything she says,” it was not just their marriage that would suffer: “ W e w i l l a l l be r u in e d !” A nd the “we” that would be ruined was not just the “we” o f the congregation , or the “we” o f Christ ians. I t was everyone who was currently suffering. It was the slum as a whole. For the marriages that churches worked to support were rarely marriages in which both parties were Christ ian . The majority o f church members were married to nonbelievers, and these marriages too had to be cult ivated and respected, as the following extract makes clear.

As the scripture says, “A man will leave his father and mother and unite with his wife, and the two will become one. Hallelujah! . . . How must they be? Like one flesh! . . . It is the same way among us----This congregation must be the best congregation and be like a wife to Christ--- [But] we each have a family separately! We each have children separately! There might be a husband who is showering you with his love. Hallelujah! The Lord has told that you must love him too. What does he tell a wife? You must love your husband truly. What has he told husbands? That they must love their wives truly. Only by this kind o f loving can the congregation be blessed by Christ! Hallelujah!

The well-being o f the congregation as a whole rested upon that o f every indi � vidual marriage, marriages that tied Christ ians to the larger slum commu� nity. Strengthening the ir marriages was thus not about strengthening the Christ ian community, since, as we have seen, there was no “ Christ ian community” in the slum in the common Indian sense o f a distinct collective identity and interest group. Because Christ ians were more likely than not to be married to non-Christ ians, “strengthening Christ ians’ marriages”

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2 3 7

meant resolving social tensions that undermined the slum community as a whole.

R E D E M P T I O N : P A S T A S F U T U R E , O T H E R A S S E L F

Jesus is with m e!. . . Now I am different! I am in everything, oh Father! I am myself! I am myself!

In this chapter I have attempted to make sense o f some o f the more impor � tant themes in the discourse o f the slum church. I have focused on extracts that I am able to interpret. But as I warned at the outset, much o f what was said during the Sunday service was difficult and even impossible to parse. Consider the previous extract, words screamed out by a C hrist ian woman in the midst o f ecstatic prayer. W h a t did she mean? Slum C hrist ians claimed that they did not know what they were saying in such moments and that they just spoke whatever came into their heart. But while the exact meaning o f their words was often obscure to them, they were not insignificant. Believers valued highly these moments o f self-expression and were frequently overcome by the emotions their words called up. Their eyes became f illed with tears, and they gestured emphatically, as i f trying to explain something extremely important. I chose the above fragment as an example, not only because it is typical, but because I see in it a commentary on the genre itself.

“ N ow I am different.” Transformat ion was a fundamenta l goal o f slum Christ ianity . In messianic-revolutionary language, the entire world was to be remade. A t other times it was believers’ relationships with their neighbors or their husbands. But in al l these cases it was not just a relationship that was transformed but the believer herself. Into what? Believers were not turned into something completely new, I argue, but restored. “ I am myself! I am myself!” These were people who felt systematically misrecognized, who said that members o f the dominant society did not “see” them. “ They say, ‘You are a shameful man ,’ ‘You are worthless.’ ” But they were wrong. Those who had been discarded as useless, treated as trash, were shown to contain hidden vita l i ty and to be an “unseen garden .” The umbre l la-carrying Pariah was revealed as an embodiment o f C hrist himself. The insulted and neglected wife was in fact, in the true sight o f C hrist , eternal ly new. W h a t believers “ became” was in real ity what they had been all along. Women did not remove their jewe lry in order to “ become” beautiful; they were beautiful already, but others fai led to see it because their vision was clouded by sin. A n d thus like

2. 38 • S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

caste i tse lf—the very paradigm o f sin— it was an illusion or lie that prevented members o f the dominant society from recognizing the poor as what they in fact always had been: fellow human beings worthy o f love and care.

In ecstatic moments slum Christ ians spoke words that were not fully their own . A t the same time, the words they spoke were understood as issuing from a truer, more perfect self. Such speech existed on a continuum , the end poin t o f which was a n n iy a p d c a i (“ the foreign language”). W here other ecstatic outbursts merely flir ted with incomprehensibility, the “ foreign” lan � guage was absolutely unintell igible and tota lly “other.” N o one claimed to be able to understand it. Nevertheless, slum Christ ians said, a n n iy a p d c a i was in fact the true language o f one’s own innermost soul. It was a direct line o f communication from the soul to God . That which was most distant (in the sense o f foreign, unknown) was in reali ty what was most near.

A l l these various thoughts— o f foreignness, o f transformation , o f a true, more perfect self, o f repairing fraught relations—were brought together one Sunday by Pastor Yesudasan in a sermon on what he termed the “ancient path .” H e spoke o f this ancient path in h ighly positive terms, which I found surprising because Pentecostalism is a millennial cult, strongly oriented to the future and not the past. The past—whether the O ld Testament past or the past o f indigenous gods and cosmology—was generally associated in Pentecostalism with evil, or at least with something outdated that had to be left behind. In this sermon, however, Pentecostalism was described as “the ancient path .” The pastor introduced this idea through the popular slum Christ ian rejection o f superficial beauty in favor o f a true beauty, envisioned not as something new but as what had been there all along, only hidden, and revealed by stripping away traditional forms o f adornment.

Clap your hands—hallelujah—we must look to the good path and walk the good way. We must walk the ancient path! . . . [In this path] you do not use outward aids—such as the way you do your hair, or the jewelry you put on, or the dresses you wear—to make yourselves beautiful. For your beauty is your true inner self, the ageless beauty of a gentle and quiet spirit that is of the greatest value in God ’s sight. What is indestructible beauty? Knowing the ancient path and following it—that is indestructible beauty.

H aving characterized the ancient path as a restoration o f true beauty, the pastor went on to define it in such a way that what was most ancient was simultaneously the cutt ing edge o f modernity. A t this point I myself— and my modern companionate marriage in “ N ew York A m erica”—were

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2 3 9

enlisted as an example. “ The ancient pa th—what is it? Look at Robert, sitting right there---- D o you know what he and his fam ily do? I asked him what the husband and wife do in N ew York America, and he told me. W h a t they do is, they cook together! The chore o f cook ing— the husband and wife do it together! A n d then I asked him what they cook , do you know what he told me? B ee f,T H e pronounced the last word triumphantly. W hen the pastor had asked me about my li fe in A m erica the week before, the fact that my wife and I cooked together was just one o f many details I men � tioned. That he should have focused on the fact that I, a modern man and prestigious foreigner, cooked (normally seen as a woman ’s job) and that I did so w ith m y w ife , is surely significant. So too with the pastor’s decision to emphasize that my wife and I were beef eaters. In my conversation with the pastor I had told him about a great many different foods we consumed at home, and beef was only one o f them. Yet o f al l o f the foods I had mentioned, he chose to h ighlight beef—a h ighly stigmatized and quintessential ly D a l i t food. In this way the pastor wove together “ the ancient” and true, that which was wrongfully stigmatized, and what was most modern . The ancient path was thus figured as be ing at once the true and authentic past o f D a l i t slum dwellers and yet also located, paradoxically, in the modern W est (an egalitar � ian utopia).

In chapters 5 and 6 I claimed in very strong terms that when slum dwellers embraced C hrist ianity they did not do so for reasons o f doctrine. Their deci � sions to convert were not a vote in favor o f one set o f teachings over another. Christ ianity did not represent, for them, a superior “moral system”; according to slum dwellers, there was only one morality, and all religions subscribed to it equally. Those who converted did so, they said, for the blessings C hrist bestowed. The blessings converts cited as the reason for their conversions were, for the most part , o f exactly the same sort that H indu gods were sup � posed to provide (curing illnesses, ensuring that schoolchildren would pass their exams, helping husbands find work , m aking sure their daughters did not die in childbirth). A c cording to Christ ians, the difference between H indu gods and theirs was not a matter o f moral ity; it was simply that their god was real and therefore could ac tual ly help them. N ot even pastors pre � sented C hrist ianity as offering a superior moral system. They recommended it for the same reasons as anyone else did: C hrist was real, and he would do the things for you that you wanted him to do.

2 4 0 � S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

The materials presented in this chapter might seem to tell a rather differ � ent story. C hrist ianity is shown to involve a wide range o f ideas that simply did not exist in local forms o f H induism , among the most important o f which was the promise o f social transformation . Were these ideas not also part o f C hrist ianity ’s appeal? It is tempting to suppose that such ideas d id actual ly contribute to C hrist ianity ’s success in w inn ing converts in the slum and that slum dwellers were therefore converting for reasons other than (or in addition to) those they admit. The problem with speculating about secret or unconscious motives is that, however plausible it might seem, in the absence o f any clear evidence it remains just that: speculation . Though it is true that people are not always reliable judges o f their own actions, before resorting to the explanatory deus ex m a ch in a o f “ hidden motives,” we ought at least to try to take people at their word.

First, i f the real reason converts chose Christ ianity was that it offered a new, more appealing moral vision, what are we to make o f the fact that most slum dwellers did not convert? W as it that the moral ideas slum Christ ianity expressed— its critique o f the dominant society, its affirmation o f companionate marriage, its promotion o f peace between neighbors—were not equally appeal � ing to nonconverts? But these were ideals a l l slum dwellers endorsed. Although the Christ ian religion talked about them a lot in ritual contexts, they were not seen (either by H indus or by Christians themselves) as specifically Christian values. It was true that H indus did not talk about such things in ritual contexts, but neither did they talk about anything else in such situations.

Relatedly, when Christ ians talked about the ideas presented in Sunday services and other ritual contexts, they did not characterize what they were receiving as a new moral system. They called it k now ledge. Primarily it was knowledge about the past (i.e., what had happened in the Bible), about what was going to happen in the future, and about how one needed to act in order to ga in C hrist ’s blessings. The details o f what had occurred in the past were regarded as fac tual information , no different in principle from what school � children learned from their textbooks. N e i ther Christ ians nor H indus per � ceived these facts about the past as contradicting anything the H indu reli � gion taught (not least because H indu gods were sile n t on such matters). In form at ion about the future—prim ar ily concerning the fact that the C hrist ian god would save the suffering poor from their oppressors—was likewise not seen to contradict anything in H induism . A nd as I have already noted, poor H indus were also expected to benefit from whatever the C hrist ian god had in store w h e th er or no t they converted.

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G Z 4 I

In addition to millennial justice, which was portrayed as benefi ting all who were poor irrespective o f what they did, C hrist provided more immedi � ate blessings. For the most part these were o f exactly the same sort as H indu gods offered, though as we have seen they also included improving marita l relations and relations between neighbors. (It was for this reason that C hrist was understood as a god who specialized in “women’s issues.”) To gain these sorts o f blessings, believers had to do the part icular things C hrist required, and this was the third broad form o f “ knowledge” that Sunday services pro � vided . Though the specific requirements o f the Christ ian god were unique, the fact that he had such requirements was not.

In addition to being intrinsical ly moral, religions in the slum were under � stood as divinely orchestrated systems o f reward and punishment. A mong the most unusual things that the Christ ian god required, as we have seen, was that believers actively embrace suffering. W h a t might have been merely a necessity (“ I f you do not endure you cannot live in this world”) was fashioned by the slum church as a ritual requirement and a moral discipline, and mobi � lized in ways that sought to transform believers’ relations with others. Christ ians needed not only to be w il l ing to “ turn the other cheek,” but also, as we have seen, to be w il l ing to share actively in the suffering o f others. They had to actual ly take on others’ suffering as their own .

W e may certainly say that al l o f this knowledge (about the past, the future, and about what the Christ ian god demands o f believers) had a moral dimen � sion. But starting from the premise that moral ity was singular and universal, the moral content o f the knowledge C hrist ianity provided was valid for eve � ryone. Slum sermons were not introduc ing a new moral system but simply he lping their audiences to better understand what l iv ing moral ly in fact entailed. Take, for example, the requirement to share in others’ suffering. It drew upon the preva i ling slum ethos o f care, but it extended the meaning o f care in one key respect. O utside o f C hrist ianity , the in junct ion to care entailed both being concerned for and giving aid to those in need. C ar ing was at once a central moral value in the slum and a part o f what it meant to be human . The preva i ling slum humanism did not, however, require anyone to litera lly f e e l others’ pa in , to actively take on their su ffering as one’s own . Suffering was understood only as someth ing to be eliminated . For slum Christ ians, however, the ideal o f compassion (literally: suffering with) meant that the pa in o f others was taken on actively as a spiritual discipline, and in this way the experience o f suffering was i tse lf redistributed and deliberately socialized. The slum ethos o f care was thus retained in its entirety but com �

2 . 4 1 S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

bined with something new. Consider the following passage, in which two distinct images— one rooted in slum humanism , the other in the C hrist ian claim that sharing others’ suffering is a precondition to one’s own salvation— are superimposed to form something entirely new, even as the outlines o f each remain clearly recognizable: “ I f you do not feel compassion for people, G o d ’s protection w il l go to someone e lse . . . . Those who are firm in this love [for others], they are G od ’s children— H allelu jah! I f there is no love w ith in me, l a m no t a h u m a n b e in g ” By superimposing (cf. Fulton i o o i , 36-38) the C hrist ian in junct ion to share in others’ suffering on existing slum ideas about what it meant to be human , preachers did not challenge or in any way overturn the existing moral framework . O n the contrary, the message o f slum churches intensified slum dwellers’ existing moral understanding and radi � calized it. I t did so, however, by presenting itself, not as something new, but in a way that appeared simply to be bringing out implications that had in fact been there al l along.

Slum C hrist ianity addressed and built upon the existing moral community o f the slum, ameliorating to some extent the tensions with in it that remained otherwise concealed. That it thus reinforced a moral community that was not defined by confessional faith, and without buttressing one re ligious commu � n ity as against another, is a “miracle” that would be impossible to imagine from the standpoint o f anticonversion laws and the national discourse from which such laws emerged. Slum C hrist ia n ity ’s significant micropolit ica l achievements remain invisible to those who want to outlaw it.

S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G • 2 4 3

Conclusion

I b e c a m e A v e g e t a r i a n i n 1989 after reading a book that persuaded me industrial meat production was ecologically disastrous and ra ising and k il l ing animals for food was cruel and unnecessary. I was an undergraduate at the U niversity o f C a l iforn ia , Berkeley, and soon moved into a vegetarian cooperative. M any who lived there followed the ecological and ethical argu � ments to their logical conclusion and gave up eggs and da iry products as well. The living conditions o f chickens and da iry cows were no less cruel, after all, and perhaps even worse for how prolonged their misery was. M ost o f my friends by this time were either vegetarian or vegan . I could not understand why anyone, once presented with the arguments against meat eating, would continue the practice. H ow could anyone be so selfish? There was a clear sense, in my circle, o f our own superiority to meat eaters, and I was personally proud o f my own bio-moral purity. I also began to feel a visceral disgust for meat i tse lf that matched my growing moral disgust with meat eaters. These aesthetic and moral dimensions were not always clearly dist inct , at least not for me. W e often made disparaging remarks among ourselves about meat eaters and their disgust ing “meat breath” and body odor.

Sometime in 1993 I accompanied a co-op friend, Brandon , to an A I D S hospice across the bay, in San Francisco’s C astro district. The hospice was an offshoot o f the H art ford Street Z en Center, founded by ex-drug addict and drag queen-turned-Zen-master Issan Dorsey. W e had gone there to spend the afternoon with another o f Brandon ’s friends, Richard , a volunteer who lived and worked in the Center. A part from doing household chores and cleaning, Rich ard ’s job was simply to be present with the dying, to be fully conscious with them, to offer care and courage. O ne often thinks o f courage as an individual virtue, a k ind o f inner citadel. But it is also an inter �

1 4 4

persona l one. I t can, in fact, be transmitted from one human being to another. Carers and dying people, in different stages o f decline, moved silently or sat together. D eath, like Zen , is serious business. I was sti ll in my early twenties at the time, but old enough to know these were people who

knew things I didn ’t. A f ter some time Richard asked whether Brandon and I were ready for

lunch . W e went upstairs to a small kitchen , and he began to cook for us, chicken. I was confused. I told Richard that I thought Buddhists were sup � posed to be vegetarian . Some are, and some aren’t, he told me. In the lineage the Center was a part of, the practice was to accept whatever was offered to them. They survived entirely on the charity o f others, he explained, and the chicken had been donated by a local grocer. Most likely it was about to expire, and would simply have rotted otherwise. I thought it all through as I sat with Brandon , watching Richard cook . The chicken was already dead, and because the meat had not been purchased I would not thereby be contributing, even indirectly, to agribusiness i f I were to eat it. But there was st ill the aesthetic problem: meat disgusted me. Earlier we had been ta lk ing about pa in and other unpleasant things students o f Z en are encouraged to face with unblink � ing eyes. Unpleasant things like ooz ing sores, shit, and dying. A n d there I was, too squeamish or too proud to accept food cooked by a man whose whole life was devoted to acceptance. I decided then and there that I, too, would henceforth accept whatever food was offered me. I ate the chicken that Richard had cooked for me. H ad I refused, the research on which this book is based would not have been possible. Some truths can be learned simply through observation. O thers require interaction—pushing and being pushed back , touching others and a l lowing oneself to be touched. The giving and receiving o f food is among the most important ways common being is affirmed in India, and by refusing others’ food that commonal ity is denied. H ad I refused to eat what was offered me by the people o f A nbu Nagar, I would have at once marked myse lf as a caste person and treated them as pariahs.

This has been a book about care and the absence o f care. It has been about people who feel that they are not being cared for and that they deserve to be. I t has been about their efforts to make moral sense o f the world they find themselves in, to live w ith in in it the best way they can, and to find their way to something better. They believe the weak are entitled to the help o f the strong and deserve to be supported until they can stand on their own . This is what humans owe one another, according to these slum dwellers. I have

C O N C L U S I O N • 2 4 5

attempted to summarize this moral ideal as “care,” care not merely in the sense o f concern but in the sense o f ac tual connect ion and sharing.

In chapter 1 1 described another moral ideal, autonomy. Autonomy is highly valued in our own political and moral culture; it is a guiding ideal o f political liberalism, o f democratic theory, and also o f modern nationalism (Roberts 2012a). Ih ere is probably no society that does not recognize autonomy as a good, but it is perhaps only in the early modern period that it became para � mount in so many areas o f life (Schneewind 1998). A u tonom y comes from a Greek word, avrovo[iia , used to describe city-states that made their own laws rather than being dictated to by another. Etymological ly it means self-rule { avro [“se lf”] + vojxia [“rule” or “ law”]) (Lindley 1986, 5). More generally, au tonom y today connotes deciding for oneself and being responsible to oneself. It also implies being responsible f o r oneself, in one’s own life and morally. Autonomy and care are incompatible precisely in the sense that they cannot both be paramount in any given situation. A child or anyone else who is dependent on others’ care is not autonomous. N or is the parent o f that child, or anyone on whom others depend. W hen others depend on us and need our help, and when we feel ob liged to give that help, we are no longer autonomous.

W e have explored the problem o f care in the relationship between the slum and the dominant society, and between women and their men. W e have considered the ways women care for one another, and what happens when they do not. C are is central, too, to the way slum dwellers conceive the rela � t ionship between human beings and gods. A n d we have witnessed a contra � dict ion in the moral world o f the slum, a contradiction that is rooted in the discourse o f w ife ly auspiciousness. This discourse celebrates individua l women as the autonomous source o f their households’ moral and material well-being. It also holds them ultimately responsible for themselves and for the household they preside over. W hen they do not succeed, they are blamed, blamed for conditions over which they in fact have little control. Such women are hurt by this, sometimes very badly.

Pastors spend most o f their time counseling women , fast ing for them, and work ing to solve their problems. They are not autonomous but bound by their responsibilities to others, responsibilities they experience as inescapable and even crushing. Women who convert, likewise, become even less autono � mous than when they depended just on men and on the material aid o f other women . By becoming C hrist ian they disown the autonomy o f ultimate responsibility for themselves by al lowing others to assume that responsibility and by accepting that when other women fa i l they do too. A key argument

C O N C L U S I O N

o f the book has been that C hrist ianity ’s success among slum women derives from its abili ty to overcome one o f the slum’s systemic moral contradictions, bringing slum practices in line with slum ideals. By suturing the moral fault lines that covertly pit the people o f A nbu N agar against one another, slum C hrist ian ity integrates the slum as a w hole and irrespective o f religious affiliation .

The problem o f care and o f being cared for was central also to the myth o f the benevolent foreigner. Despite all that slum dwellers must face, and despite the great sense o f bereavement and in justice they express, their vision o f the world was ultimately an optimist ic one. Their optimism rested, however, on rather weak foundat ions. I t rested on the be l ief that something basic in human beings makes them care for those in need and that the impulse to care is embraced by all the world ’s people as not only natural but right. In this way they convinced themselves that the situation they found themselves in was an unnatural one. This was very disturbing for me because I was equal ly sure that what they believed was not true. I t depended on the false premise that India was uniquely bad and that people elsewhere in the world really would care about them and would attempt to help them i f only they knew.

A foreigner in the flesh, I was o f considerable interest to the people o f A nbu Nagar, both in myse lf and for what I could tell them about the world beyond India’s borders. I did not want them to nurture false hope, even i f that meant te ll ing them that the odds were even more stacked against them than they realized. I spoke bluntly. They were w il l ing to accept on my say-so that there are inequalities everywhere, and not just in India, but they dis � missed my claim that the rich elsewhere could be as indifferent to the poor as the rich were there. I didn ’t fully comprehend what the rich in India were really like, they often told me. Recall slum dwellers’ complaint that the rich and the privileged did not “see” them, by which they meant that they did not see them as fe llow human beings. I argued with them about this. I insisted that, while this might be true, their be lief that foreigners were any different was mistaken . These exchanges had a repetitive quality, insofar as the focus always turned sooner or later to me. “But you are here!” I f foreigners were really no different than Indians, I was asked, how could I account for my own presence in the slum? The very fact that I had come to do research in A nbu N agar was used to refute me. It showed that I cared, or at least that my university did.

“ I am here,” I told them again and again, “ because I am studying for a degree!” I explained that I wanted to be a teacher one day and that this was

C O N C L U S I O N • Z 4 7

part o f my training. A requirement o f my degree, I made clear, was that I go and live somewhere, w ith people very different from myself, for an extended period. “ I am here to study you,” I explained. “ I listen to you because that is part o f my research.” They were not convinced. They accepted my story, so far as it went, but they were sure there was more to it than that. “But you really do care,” they insisted. “You don ’t just listen . You see us.” They were familiar with what it meant to be studied. Government census workers would , every ten years, come to the slum to ask them questions; N G O workers had also conducted surveys there. N one o f these people accepted food from them or remained in the slum any longer than they had to. The fact that the way I was studying them entailed actual ly l iving with them for a long period o f time meant that it was a fundamenta lly different k ind o f “study” than they were used to, one intrinsical ly based on care. O r so it seemed to them.

They found con firmation , too, in the sort o f questions I asked. I told them that I had come to study th em , but o f course this was just a shorthand. I was not really studying them, but their problems. This was clear from the sort o f questions I asked, and when someone pointed it out I immediately agreed. That I had been sent to help them in some way seemed obvious to them. W ha t they wanted to know was how al l this research was supposed to do that. “ H ow is this going to help us?” they demanded. “ W h a t good is any o f this going to do?” This was usua lly the first th ing anyone I spoke to wanted to know. W ith every new avenue o f inquiry I opened, I was once again forced to explain and just ify it in these terms. This presented a difficulty, because the truth o f the matter was that I had no idea how any o f it would help them. I thought about this a lot and developed a standard reply.

I told them that one day I would write a book . The book would be about the k inds o f problems people like them faced, about what they thought, and about what they were trying to do in their lives. I said I needed the ir help to do this. I said I did not yet know what exactly the book would say, but I hoped it would be able to teach people who read it something about our world they did not already know. But in answer to the question “ H ow is this going to help us?” I told them very frankly that it would not. “ Though readers may feel something when they read it, they w i l l not be able to do anything for you,” I said, “ but at least they w i l l know you exist. A n d maybe, somehow some good w il l come o f it. N ot for you, but maybe for someone who is fac ing similar problems.” I gave this lit tle speech, in one variant or another, more times than I can count. Those who heard it nodded and appeared to take what I was saying very seriously. N ever once did anyone refuse to help.

2 4 8 • C O N C L U S I O N

A f ter my research was over, and I was preparing to leave A nbu Nagar, I spent two fu ll weeks saying my goodbyes. I tried to sit and spend time with everyone who had cared for me and helped me during my time there. N early everyone made me promise not to forget them. M any also bought up my book . “You have to write that book , Robert ,” one o f the last men I met with told me. “You have to tell people. Tell them everything. You must m ake them understand.”

C O N C L U S I O N 2-49

A P P E N D I X

T am il N adu Prohibition o f Forcible Conversion o f Religion O rdinance, 2002

N O . 9 O F 2 0 0 2

Note: the O rdinance was superseded by an A c t (No. 56) later in 2002.

A n O rdinance to provide for prohibit ion o f conversion from one religion to another by the use o f force o f allurement or by fraudulent means and for matters incidenta l thereto.

W hereas the Legislative Assembly o f the State is not in session and the Governor o f T amil N adu is satisfied that circumstances exist which render it necessary for him to take immediate action for the purposes hereinafter appearing

Now , therefore, in exercise o f the powers conferred by clause (I) o f Art ic le 213 o f the Const i tut ion , the Governor hereby promulgates the following O rdinance:

1. Short title and commencement

(1) This O rdinance may be called the T amil N adu Prohibit ion o f Forcible Conversion o f Rel igion O rdinance, 2002.

(2) It shal l come into force at once.

2. D efin it ions In this O rdinance, unless the context otherwise requires:

a) “allurement” means offer o f any temptation in the form of:

(1) any gift or gratification either in cash or k ind;

1 5 1

(2) grant o f any material benefit, either monetary or otherwise;

b) “convert” means to make one person to renounce one religion and adopt another religion;

c) “ force” includes a show o f force o f a threat o f in jury o f any k ind inc luding threat o f divine displeasure or social ex-communication;

d) “ fraudulent means” includes misrepresentation or any other fraudulent contrivance;

e) “minor” means a person under eighteen years o f age.

3. Prohibit ion o f forcible conversion N o person shall convert or attempt to convert, either directly or otherwise, any person from one religion to another by the use o f force or by allurement or by any fraudulent means nor shall any person abet any such conversion.

4. Punishment for contravention o f provisions o f section 3 W hoever contravenes the provisions o f Section 3 shall, without prejudice to any civil liability, be punished with imprisonment for a term which may extend to three years and also be liable to fine which may extend to f ifty thousand rupees.

Provided that whoever contravenes the provisions o f Sect ion 3 in respect o f a minor, a woman or a person be longing to Scheduled Castes or Scheduled Tribes shall be punished with imprisonment for a term which may extend to four years and also be liable to fine which may extend to one lakh [one hundred thousand] rupees.

5. Intimat ion to be given to D istric t M agistrate with respect to conversion

(1) W hoever converts any person from one religion to another either by performing any ceremony by h im se lf for such conversion as a religious priest or by tak ing part direc tly or indirectly in such ceremony shall, with in such period as may be prescribed, send an intimat ion to the D istr ict M agistrate o f the district in which the ceremony has taken place o f the fact o f such conversion in such form as may be prescribed.

(2) W hoever fails, without sufficient cause, to comply w ith the provisions o f sub-section (1), shall be punished with imprisonment for a term

2 5 Z • A P P E N D I X

which may extend to one year or with fine which may extend to one thousand rupees or with both.

6. Prosecution to be made with the sanction o f D istric t Magistrate N o prosecution for an offence under this O rdinance shall be instituted except by or with the previous sanction o f the D istrict Magistrate or such other authority, not below the rank o f a D istric t Revenue O fficer, as may be authorized by him in that behalf.

7. Power to make rules

(i) The State Government may make rules for the purpose o f carrying out the provisions o f this O rdinance.

(z) Every rule made under this O rdinance shall as soon as possible after it is made be placed on the table o f the Legislative Assembly, and i f before the expiry o f the session in which it so placed or the next session, the Assembly makes any modification in any such rule or the Assembly decides that the rule should not be made, the rule shall thereafter have effect only in such modif ied form , or be o f no effect, as the case may be, so however, that any such modif ication or annulment shall be without prejudice to the va lid ity o f anything previously done under that rule.

Sth O ctober 2002 P . S. R A M A M O H A N R A O G ove rnor o f T a m il N a d u

E X P L A N A T O R Y S T A T E M E N T

i. Reports have been received by the Government that conversions from one religion to another are made by use o f force or allurement or by fraudulent means. Bringing in a legislation to prohibit such conversions w i l l act as a deterrent against the anti-social and vested interest groups exploit ing the innocent people belonging to depressed classes. I t may also be useful to nip in the bud the attempts by certain religious fundamental ists and subversive force to create communal tension under the garb o f religious conversion. The Government have, therefore decided to enact a law to prevent conversion by use o f force or allurement or by fraudulent means.

A P P E N D I X 2-53

z. "The O rdinance seeks to give effect to the above decision.

(By order o f the Governor) A . K rishnankutty N a ir Secretary to Government , Law Department .

2 5 4 • A P P E N D I X

N O T E S

I N T R O D U C T I O N

The chapter’s first epigraph is from Viswanath (2014b, xi). 1. The existence o f a singular structural division between Dalits and non-Dalits

does not preclude the possibility of there being individual castes whose position within this binary is uncertain. While all castes could be unambiguously assigned to one or the other category in the deltaic heartlands o f India’s rice-growing civiliza � tion, where the caste system existed in its most pure form, in marginal and dry areas there were castes that could be classified as either touchable or untouchable depend� ing on what criteria are applied (Fuller 2015,19).

2. In India an ordinance is a law enacted by executive authority when the legisla � tive body is not in session. When the legislative body returns to session it must approve the ordinance, at which point its official designation is changed from ordi � nance to law; i f not, the ordinance lapses (Philip 2014).

I . O U T S I D E R S

1. A Pentecostal woman I knew well told me she had once been friends with an A frican man, a Muslim student from Sudan. H is name was Jamal. She confided that they had been lovers, though no one else in Anbu Nagar knew. She would never forget him, she told me, and he would never forget her. She loved him still. Before leaving, Jamal had gotten her name tattooed on his arm, in Tamil lettering, and she had had his name tattooed on her arm in Arabic. This was so no one in their respec � tive worlds would learn of their affair. When people asked what her tattoo said, she told me, she always just told them it read “Peace.” What was he like? I asked. “ He was exactly like you,” she said, “ kind and gentle.”

2. Guests are shown m ariydtai, for example. Outside o f the guest-host context the guest might be the host’s social equal, but within it the guest is treated formally as i f he or she were a superior.

2.55

3. Employm ent d iscr im in a t ion aga inst D a l i ts is pervasive in Ind ia , and we l l docum ented (Thorat and N ew m a n 20x0; A . D eshpande 20x1; H arriss- W h ite 2014).

4. The transvestites in A nbu N agar referred to themselves as a lis rather than the H induized neologism aravdn i that some authors and activists regard as more poli t i � ca l ly correc t. I use a li because th is was the ir own preferred self-designat ion and because the term aravdni references a H indu origin myth that margina l izes non- H indu a lis. U n lik e their N orth Ind ian counterparts, loijras, a lis do not surgic al ly alter their genitals; at any rate, the penises and testicles o f ones I kn ew were fu lly intact.

5. T ittu differs from these English terms, however, in that it can also refer to criticism o f a person expressed to a th ird p ar ty when the crit ic ized person is not physically present. But tittu is never simply “ ta lk ing behind someone’s back ,” which implies a desire to keep the target in the dark about what one is saying about him or her. T ittu is always a imed at produc ing an effec t, ge t t ing targets to recognize that wha t they did was wrong. Thus even when someone tittu-fies someone else to a th ird party , it is always in some sense “public” and spoken w ith the in tent that wha t is said w i l l eventual ly reach the ears o f its target.

6 . In T a m il N adu and elsewhere in India , women ’s N G O workers and activists portray poor women as strongly and un iform ly opposed to a lcohol consumption and seek successfully to mobil ize them on th is basis. A long w i th gender norms that tend to see women ’s sexua l ity though the lens o f vic t im hood— about whic h more be low— this is a com mon thread in middle-c lass ac t ivism on poor wom en ’s behalf. Targeted by paterna l ist ic “upliftm ent” programs, slum women respond by public ly expressing absolute opposit ion to alcohol and in so doing lay c la im to a mora l purity that the dom inant discourse on caste denies them . A side effec t is that this emphasizes their vic t im status w i th respect to men . W h ile not denying that poor wom en are often vic t im iz ed by poor men , there is no basis for the common assumption that slum men are any more abusive than men o f other com � munit ies. A n d while systematic con f l ic t between women and their husbands is a reality, it is rather more complex than domin ant narrat ives allow. This is a major focus o f chapter 3. M y poin t here is simply that the polit ics o f poor women ’s move � ments are not stra ightforward and that the public posit ions slum women are induced to take belie the degree o f solidari ty and common ground between them and their men folk .

7. Christodas G andh i , persona l communication , 2001. 8. M eena K andasamy notes that ceris are norm a l ly posit ioned not only at a

safe distance from the m a in caste vil lage , or ur, but also downwind , in order to minim ize the risk o f fire spreading to the homes o f the attackers (Thirumaava lavan 2003, 7 n. 3).

9. This law does not protect C hrist ian and M usl im D a l i ts, who are exc luded from the Scheduled C aste l ist (V iswanath 2012c) in a m anner that the Government o f India has admit ted is both arbitrary and discrim inatory but has yet to rec t ify (V iswanath 2015).

2. 56 • N O T E S

2 . C A S T E , C A R E , A N D T H E H U M A N

1. A notable exception is the late M. S. S. Pandian, a social scientist and middle- class resident o f Chennai at the time, who told me, “ Quite frankly, we really have no idea at all what goes on inside these slums” (I paraphrase). Pandian’s statement notwithstanding, Paul Wiebe (1975) and Joop de Wit (1996) offer excellent ethno� graphic accounts o f Madras slums, though neither picks up on slums’ caste-specific character; Penny Vera-Sanso (1994) has written a sensitive account of two slums, with a specific focus on gender.

2. The use o f the term ceri was banned by the Government o f Madras in 1957. In a remarkable paragraph in the above-quoted Census o f India (1961) report on the slums o f Madras, the author’s discomfort over this politically sensitive issue is still evident. The paragraph begins by admitting that the slums the report describes are in fact known as ceris but hastily adds that in premodern Tamil ceri did not just refer to the place where “Pariahs” lived but was formerly used as a generic term for any “town, village, or hamlet.” It then acknowledges that the term ceri is currently used exclusively for the “ Harijan quarters” (thereby implicitly admitting what is other � wise denied in the report, i.e., that slums are in fact Dalit ghettos,) before quickly bringing the discussion to a close by reminding readers o f the government’s 1957 ban on the word in all official records and correspondence. Problem solved!

3. No official statistics are collected on the percentage o f Dalits in nonslum areas. My middle-class Dalit friends estimate that no more than 2 to 3 percent o f the residents in the apartment buildings where they have lived are Dalit and that they normally seek to keep their caste secret from non-Dalit neighbors. Virtually all have described being made to feel unwelcome when their identity was revealed, and sev� eral report having had to move elsewhere as a consequence. Even the 2 to 3 percent figure probably overestimates the total portion o f Dalits in non-Dalit areas, because some areas are exclusively high-caste settlements where no Dalits at all live. (In Tamil Nadu it is common for rentals to be advertised as “Brahmin only” or “vegetar � ian only,” which is understood to mean no Christians, Muslims, or Hindus outside the top castes. Housing discrimination is regarded as perfectly acceptable in India, so long as landlords do not explicitly say “no untouchables.”) On the other hand, housing for government employees would have a higher portion o f Dalits because o f compensatory discrimination policies (“affirmative action” in US par � lance) and because overt caste discrimination is impossible in government-owned housing.

4. Hook-swinging was a religious practice in which low-caste (but usually not Dalit) devotees would be “sponsored” by a wealthy patron to put hooks through the muscle in their backs and then be swung through the air to honor a deity. This was regarded as inhumane and was banned by the British, although the swingers insisted they felt no pain (Dirks 2001).

5. But the racism that has partially defined European notions o f humanity since the eighteenth century (Bernasconi 2010) is absent in slum dwellers’ concept of the human.

N O T E S • 2 5 7

6. Jonathan Parry ( 1 9 9 1 , 1 6 8 - 6 9 ) and Quentin Skinner ( 1 988 , 1 20) provide use � ful discussions on the distinction between words and concepts, and on languages in which no single word exists to name culturally important concepts.

3 . S H A R I N G , C A R I N G , A N D S U P E R N A T U R A L A T T A C K

i. I know of no official data to corroborate slum dwellers’ claims, and I would be inclined to dismiss them as exaggerated but for the fact that some twelve successful suicides occurred in and around Anbu Nagar in a roughly one-year period during my stay there, which locals treated as entirely ordinary. I did not know any o f the victims personally, and I did not have the stomach to interview their families after � wards. But by all accounts these suicides fit the pattern for female suicides this book describes, and that slum dwellers themselves endorsed when I presented my argu� ment to them. Male suicides were far less common in the slum, and locals estimated that they accounted for only between one in ten and one in a hundred suicides. Though rare compared to female suicides, male suicides were not uncommon. My next-door neighbor Anbu, for instance, had lost both a sister and an elder brother to suicide, and during my stay the son o f another immediate neighbor consumed poison and lay down on a train track at night. He was found unconscious and rushed to a hospital. Thankfully he survived and was returned to his mother after several days in intensive care.

4 . R E L I G I O N , C O N V E R S I O N , A N D T H E N A T I O N A L F R A M E

1. The Government o f India does not officially recognize the existence o f Dalit Christians and therefore does not collect data on either the percentage o f Dalits who are Christian or the percentage o f Christians who are Dalit. Nor does any Christian denomination, because o f the sensitivity o f this matter. But according to unofficial estimates by Dalit Christian NG O s, approximately 60 percent o f Indian Christians are Dalit, and around 6 percent o f Dalits are Christian (John Dayal, Secretary General o f the All-India Christian Council, personal communication, May 28, 2013) . On the Government o f India’s nonrecognition o f Dalit Christians, see Viswa- nath (2012c , 2015) .

2. In the judgment commuting the sentence of the murderer o f Staines and his sons, the Court further explained that “though Graham Staines and his two minor sons were burnt to death . . . the intention was to teach a lesson to Graham Staines about his religious activities, namely, converting poor tribals to Christianity” (Supreme Court o f India 20 1 1 , 72) . A fter an outcry by Christian groups and civil society activists, this sentence was supposedly “expunged” (“SC Expunges” 201 1 ) . In fact, it remains on the Supreme Court’s official website as o f August 6, 2015 .

2 5 8 • N O T E S

3. Shortly after the Hindu nationalist BJP party was defeated at the national elections in May 2004, Chief Minister Jayalalitha announced that the Tamil Nadu anticonversion law would be repealed. The repeal took effect on June 7, 2006.

4. The Sangh Parivar is a confederation of formally distinct political and cul � tural groups under the ideological leadership of the RSS (Rashtriya Swayamsevak Sangh). Some o f the most important o f these are the BJP (Bharatiya Janata Party), the national parliamentary party o f Hindutva; the V H P (Vishwa Hindu Parishad), a federation o f prominent Hindu religious leaders through which the strategic direc � tives of the RSS leadership are disseminated to local Hindu religious organizations; and the Bajrang Dal, a militant wing o f the V H P that openly espouses violence and that functions as a strike force against Muslims and Christians.

5. An important exception is Nivedita Menon (2004), whose argument about allurement is similar to my own.

6. This definition appears in the anticonversion legislation of Tamil Nadu, Rajasthan, Gujarat, and Madhya Pradesh. Earlier anticonversion laws also ban allurement by means o f material benefits but do not refer to such allurement as “temptation.”

7. “I f an attempt is made to raise communalpassions, e.g. on the ground that some one has been ‘ forcibly’ converted to another religion, it would, in all probability, give rise to an apprehension o f a breach o f the public order, affecting the community at large. The impugned Acts therefore [are upheld] . . . to avoid disturbances to the public order” (Supreme Court Reports 1977, 618, emphasis added).

8. The expert legal commentary on the Stainislaus (Rev. Stainislaus v. State of Madhya Pradesh and Orissa, January 19, 1977) judgment that I have consulted appears to have overlooked the crucial and peculiar role o f agents provocateurs in the Court’s reasoning, focusing instead on its overly broad conception o f public order (e.g., SA H R D C 2008, 67; Cossman and Kapur 1997; Stahnke 1999).

9. I discovered Viswanathan’s 2000 article while in the last stages of copyediting this book. I cannot do justice to her arguments in that essay, except to say that her observations on anticonversion law anticipate many o f my own.

10. It is not simply that, by banning conversions premised on the hope o f mate � rial gain, the secular state is imposing a particular and contested notion of what religion is, and thereby adopting a theological position. Weighing in on theological matters is to some extent unavoidable for secular states because in order to adopt a “ hands-off” policy (or remain neutral) toward religious practices states must first decide which practices are in fact religious, thereby authoritatively delimiting the category of religion (cf. Galanter 1971; Asad 2003; Sullivan 2005). My point is that the Indian secular state, in addition to determining religion as all secular states must, also applies certain definitional criteria inconsistently, deeming material gain antithetical to religion only with respect to conversion.

n. While the text of anticonversion laws defines fr a u d as “misrepresentation,” in everyday speech in India conversions brought about by hopes of material gain rather than an inward transformation of an immaterial soul are also commonly referred to as fraudulent. The discrepancy between the legal and popular understandings of

N O T E S • Z 5 9

fraudulent conversion is systematically obscured in at least one recent election mani � festo of the Hindu nationalist party the BJP, which defines fraud and coercion as refer- ring to “promises of social or economic benefits,” according to Nivedita Menon (1004).

12. The section epigraph from Butalia (2000) is quoted in Joel Lee (2015, 81). I thank him for directing my attention to the writings o f Butalia and Searle-Chatter- jee and for providing the analysis of them (Lee 2015, 81-82) that the next three para- graphs repeat and expand upon.

13. This contrastive usage remained commonplace in confidential Government o f Tamil Nadu reports on anti-Dalit atrocities as late as the 1970s. Caste folk respon� sible for these attacks are referred to in these reports simply as “the H indus” in contradistinction to their victims, who are distinguished as “the Harijans” (Rupa Viswanath, personal communication, September 2015).

14. The idea that untouchables were always regarded as beyond the pale of H in � duism has been challenged by Arvind Sharma (2015). Sharma argues that the com� mon understanding o f untouchables as being outside the fourfold varna system is wrong. According to him untouchables should instead be thought o f as a special subcategory within the shudra varna, a novel category he terms the excluded shudra. I f untouchables were formerly included within the varna system, the argument seems to go, then by definition they must have been Hindu. But Sharma has previ � ously argued that varna was a classificatory system that extended to the entire world, and therefore that all the world’s people were originally regarded as being within it (1992, 179). I f consistently followed, the logic o f Sharma’s argument would compel us to accept not only that untouchables were originally Hindus but that Chinese, Greeks, and Persians were too. Apart from this implausible implication, Sharma’s claim that untouchables were regarded as Hindus in ancient times rests on a faulty methodology. Rarified theoretical texts accessible to only a tiny cohort o f Brahmin intellectuals provide no direct window into ancient social reality and tell us nothing about how ordinary people classified themselves and others.

15. Hindus have always objected to missionary discourse that insulted their faith and have responded with highly effective counterpolemics, frequently trouncing the Christians in philosophical disputation (Oddie 1982). Hindus also have objected strongly to the conversion o f their own family members—typically a high-caste boy in the context of missionary schooling—or of members of their extended kin net � work (i.e., subcaste) and its allies (K. Gupta 1971,160; Oddie 1991; Bugge 1994; For � rester 1980; Grafe 1990; G. Viswanathan 2001). Such conversions were a problem because the converted boy was no longer eligible to contract marriage alliances with others o f his caste or to pass on property. Converts were permanently cut off and thus represented a social and material loss to their caste community. The only recorded examples of anticonversion sentiment extending to the conversion o f Dalits are ones in which missionization threatened elites’ control over servile laborers. Such conver � sions were never described as an attack on Hinduism. We find such objections being raised by zamindars and plantation owners (including Europeans) in the early nine � teenth century in Bengal (M. A li 1965) and by landed castes in the late nineteenth century in Madras Presidency (Viswanath 2014b), Kerala (Mohan 2014), and Punjab

• N O T E S

and the United Provinces (Lee 2015, 96-97) . As for Dalits themselves, mission records make plain that—in striking contrast to caste Hindus—they raised no objec � tions to the conversion of their own kith and kin (Mosse 2012; Viswanath 2014b).

16. It is sometimes claimed that Hindu conversion by means of sbuddh i (“puri � fication”) is not really conversion at all. This is a semantic argument that works only so long as conversion is defined so narrowly that only Christian and Muslim exam� ples count, and it is thus ultimately circular. It is also anachronistic; historical research by Joel Lee (2015) shows clearly that the Arya Samajists who were inducting Dalits into Hinduism by shuddhi understood what they were doing as conversion.

17. The Collected Works o f M ahatm a G andh i (Gandhi 1958-94) is hereafter cited as C W M G by volume and page number(s). C W M G 48 :297-98 , quoted in Zelliot (2010 ,166) .

18. Gandhi’s bodily metaphor naturalized the encompassment (Dumont 1980; Baumann 2004) o f Dalits within Hinduism, obscuring the intrinsically political nature o f this claim.

19. Gandhi’s Christian interlocutors are typically portrayed as full o f admira � tion for Gandhi, and from their lips come questions that seem almost designed to reinforce the saintly image he created for himself. One such dialogue begins with a missionary nurse addressing to Gandhi such questions as “You have the reputation of never being angry—is that true?” and “ When did you come to experience this great love for the poor?” (C W M G 61:45-46) .

20. Desai’s editorial notes are not reproduced in the C W M G ; I quote here from the original article as it appeared in the pages o f H a rija n .

21. The Christians that Gandhi featured in the pages o f H a rija n were invariably white outsiders to India who appear to have perceived him as the voice o f an authentic India they knew nothing about. These were soft targets, as Susan Billington Harper shows in a revealing account of the way Gandhi dealt with a very different kind of Christian, a bishop of Madras V. S. Azariah, an Indian Christian born o f a caste once regarded as untouchable, who had been working intimately with Dalits for decades longer than Gandhi himself. Like most Christians by this time, Azariah was as com� mitted to Indian independence as any Hindu. Finally, Azariah was intellectually and spiritually every bit Gandhi’s equal, though not a prominent politician or powerful public presence, and Harper’s account strongly suggests that Gandhi regarded him as a major threat and sought deliberately to marginalize him (2000 , 291-351).

22. Legal scholar Dieter Conrad (2010 , 207) also notices the correspondence between secular Indian law and Hindutva ideology, tracing it to the influence of M . K . Gandhi via a historical argument that anticipates my own, while also address � ing a different set of concerns (cf. 2010 ,198-202) .

5. T H E L O G I C O F S L U M R E L I G I O N

1. Viewing religiously driven mob anger almost as an elemental force of nature, the state makes little effort to prosecute angry mobs and instead focuses on

N O T E S ‘ 1 6 1

suppressing those said to have angered them (cf. Yiswanath 2014a). The state effec � tively exonerates perpetrators by treating their action as an uncontrollable Traction to a prior offense. Victims o f religious violence are assumed to have provoked their attackers and are thus held ultimately responsible for their own persecution. This creates what political scientists call “perverse incentives”: it rewards those who act violently, by criminalizing the words or deeds o f those they object to, while treating religiously motivated mob violence as self-justificatory. I f you riot, this proves that someone else has done something deeply offensive, whereas to offer dispassionate objections that do not threaten mayhem simply proves that the others’ words or deeds are not so objectionable as to require official suppression.

2. In classical metaphysics, realism refers to the view that abstract entities like God, love, the self, the color red, etc., have a reality beyond our perception of them (Loux 2006). The contrary position, nominalism, holds that such entities exist only contingently as a result o f our naming practices; we ourselves create gods or ideas like beauty simply by talking about them, and (the anthropologist would add) by relating to them in coherent ways through our culturally and socially embedded practices (cf. Bialecki 2014).

3. A ta li is a small medallion attached to a piece o f yellow string and is equivalent to a wedding ring. Because o f its symbolic value (being married, wifely chastity, respectability), it is the one item that is never (in theory) pawned, and losing one’s t a li is considered a great disgrace. W hile no such disgrace falls upon one who has offered her ta li to a deity, it is nevertheless considered a big sacrifice.

4. In slum Christianity, as we will see in chapters 6 and 7, worship and moral responsibility were to some extent collectivized. But collective worship can be seen as an extension o f the basically dyadic structure of slum religion. Petitionary prayers, for example, were still phrased as personal requests, albeit ones in which the out � come depended on the support o f other believers. Believers, furthermore, entered and left particular congregations as individuals, and these collectives were under � stood by them to be composed o f individuals rather than to be an ontologically prior whole from which individual identities derived, as envisioned by the likes o f Emile Durkheim (1995) and Louis Dumont (1980). A possible exception is the collectivity of those characterized as “the poor,” “the excluded,” and so on, for whom Christians pray, and who are promised justice in some unspecified future. But as we will see, the collective subject in this case is not understood as a specifically Christian one, but one that includes the poor o f all faiths.

5. Was it contradictory for slum Christians to hold that Hindu gods were “good” while denying their existence? No more than for us to assert that unicorns are white, single-horned equines that can only be tamed by virgins—and that they do not really exist. That slum Christians should see things this way makes them unusual among Pentecostals, who do not doubt the existence of other gods but see them as devils. But however much o f an oddity slum Christians might be in the world of Pentecostalism, there is nothing especially paradoxical or non-Christian in their seeing other gods as nonexistent yet notionally good.

2 6 2 • N O T E S

6. Compare Michael Herzfeld’s remarkable description of the beliefs and prac � tices surrounding intercessionary prayer and oath taking among shepherds on the island o f Crete, where shrines to St. George in the Cretan villages o f Diskouri and Selinaris, for example, are understood to have entirely different personalities. The “St. George at Diskouri” is regarded as more effective for some purposes and the “St. George at Selinaris” for others (Herzfeld 2004,157-60).

7. This observation held true not only for lay believers but also for slum pastors. In a private conversation with one pastor I asked whether, according to Exodus, H indu slum dwellers were sinning against the Christian god. This pastor rose angrily to his H indu neighbors’ defense. They were good people, he said, and their worship o f idols was wholly innocent. It was not a sin for them to worship Hindu gods, he explained, because they were not C hristian.

6 . P A S T O R A L P O W E R A N D T H E M I R A C L E S O F C H R I S T

1. I have not conducted ethnographic research on any o f these programs; what follows is based on what I gathered from interviews with slum pastors and lay evan� gelists who participated in them.

2. An a li is a transvestite homosexual; please see chapter 1, note 4, for why I use this term.

3. Women could not be ordained, but female lay evangelists (essentially pastors without congregations) periodically preached in slum churches.

7. S A L V A T I O N , K N O W L E D G E , A N D S U F F E R I N G

1. The terse, concrete, and socially circumscribed context o f H indu divine speech does not mean it is ultimately straightforward or unworthy o f serious study. Its unsuspected cultural and social complexities are brilliantly revealed by Isabelle Nabokov (2000) in a book on possession and related rituals in popular Hinduism.

2. This is my own contextual translation o f the original Tamil as it was spoken (standard Tamil Bible editions may vary):

C ir iyavan ai pu lu t iy il iru n t u tukki vituk iravare [Ps. 113:7] stottirom! . . . C irum aippat ta ja n a t t a i iratcittu mett ima iydna k a n k a la i tdltt uk iravare [Ps. 18:27] stott irom! . . . E n k a l catt urukkalinru iratcittu enk alai pa k a ik k irava rk a la i vetkap- p a t u t t u k ir ir [Ps. 44 :7] stott irom! . . . Jd t i k a l a i tant ik k iravare [Ps. 94 :10] stottirom! . . . C irum aippat tavar kal in at a ik kalam an avarestott irom [Ps. 9:9] stdtt irom!. . .E n k a l nuka tt at iyai muritta kar tta re stottirom [Lev. 26:13] sto t t irom l . . . enkalai nim ir thu natakka pa n n in a kart tave stottirom [Lev. 26:13] stot t irom l. . . perumaiyu U avanukku et irttu nirpavare stottirom [1 Pe t. 5:5] stottirdml . . . Catt uvam il ldtavanuk ku cat- tuva tta i perukappa.nnukira.vare [Isa. 40:29] stott irom! . . . C irumaippa ttavanai

n o t e s • 2 6 3

ava ni lum palavdnu ta iya kaikku tappivik k irava re stdttirom [Ps. 35:10] stdttirom! . . . Cirumaippattavarkalutaiya ventutalai ketpavare [Ps. 10 :17] stottirom! . . . E liyavar kal in vinnappa tt ai ketpavare stdttirom [Ps. 69:33] std t t irom !. . . E U yava rk a lin niyayattai vicdr ikk iravare [Ps. 140 :12] stott irdm! . . . N a n unna i polldt ava rk a lin k ai kku tappi- vittu u n n a ip a lava n t a r in k aikku n ih k a ldk k i vituvippen enra vdkkukkak a [Jer . 15:2.1] stott irom!

3. Tliis old Christian virtue (“turning the other cheek”) was instrumentalized as a technique o f programmatic social suasion by M . K . Gandhi (Gandhi 1944; Bondurant 1965).

4. Punitha Pandyan, the atheist editor o f D a lith M ura.su, a well-known Tamil Dalit magazine, once told me that in the eyes o f the Dalit community anti-Christian politics is perceived as driven ultimately by anti-Dalit sentiment. “Most Christians, especially the poor ones, are Dalits. So when you attack Christians you are really attacking Dalits. The Hindus [i.e., Hindutva activists] hate Dalits, but they cannot attack them openly, because they need Dalits. They need their votes. So instead they attack Christians. But it is the same thing. A n attack on Christians in India is an attack on Dalits.” David Mosse similarly observes that when the Tamil Nadu anti � conversion law was passed in 2002 it was protested most vigorously not by Christian churches (the leadership o f which is conservative and dominated by caste people) but by Dalit organizations (2012, 204-5).

Z 6 4 • N O T E S

R E F E R E N C E S

Adcock, Catherine S. 2007. “Religious Freedom and Political Culture: The Arya Samaj in Colonial North India.” PhD diss., University o f Chicago.

-------- . Z014. The L im its o f Tolerance: Ind ia n Secularism a n d the Politics o f Religious Freedom . Oxford: Oxford University Press.

Ali, Babar. 1986. “Which ‘Foreign Hand "t" Economic a nd Politica l Weekly 21 (48): 2087. Ali , Muhammed Mohar. 1965. The Benga li Reaction to C hristian M issionary A ct ivi �

ties, 1833-1857. Mehrub: Mehrub. Althusser, Louis. 1972. L e n in a n d Philosophy a n d O ther Essays. New York: Monthly

Review Press. Ambedkar, B .R . [1936] 2014. A n n ih ila tion o f Caste: The A nnotated C ritic a l Ed it ion .

Edited by S. Anand. New Delhi: Navayana. -------- . 1946. W hat Congress a n d G andh i H ave D one to the Untouchables. 2nd ed.

Bombay: Thacker. Anand, S. 2014. “A Note on the Poona Pact.” In The A n n ih ila tion o f Caste: The

A nnotated C ritic a l Ed it ion , by B. R. Ambedkar, edited by S. Anand, 359-76. New Delhi: Navayana.

Anand, S., and R. Thangarasu. 2006. “The Smell o f Dead Bodies.” O utlook, January 9. www.0utl00kindia.c0m/article/the-smell-0f-dead-b0dies/229708.

Anderson, Benedict. 2006. Im agined Communities: Reflections on the O rigin and Spread o f N ationalism . Rev. ed. London: Verso.

“Are the Conversions for Real?” 1999. Sunday Times o f In d ia , January 17. Arora, Sunit. 2012. “A Swamp o f Alien Nesses.” O utlook, October 28, www

.0utl00kindia.c0m/printarticle.aspx?282402. Asad, Talal. 2001. “Reading a Modern Classic: W. C . Smith’s The M ea n ing a n d E n d

o f R e lig io n ” H istory o f Religions 40 (3): 205-22. -------- . 2003. Form ations o f the Secular: Christianity, Islam , M odernity. Stanford,

C A : Stanford University Press. Balagangadhara, S.N ., and Jakob De Roover. 2007. “The Secular State and Reli �

gious Conflict: Liberal Neutrality and the Indian Case o f Pluralism.” Jo u r n a l o f Po lit ica l Philosophy 15 (1): 67-92.

2 6 5

Balagopal, K . 1989. “Drought and T A D A in Adilabad.” Econom ic a n d Po lit ica l Weekly 24 (47): 2587-91.

Barnett, Marguerite Ross. 1976. The Politics o f C u ltu ra l N a tionalism in South Ind ia . Princeton, N J: Princeton University Press.

Bashkow, Ira. 2006. The M ea n ing of W hitem en: Race a n d M odern ity in the O rokaiva C u ltura l W orld. Chicago: University o f Chicago Press.

Basu, Tapan, Pradip Datta, Sumit Sarkar, Tanika Sarkar, and Sambuddha Sen. 1993. K h a k i Shorts, Saffron F lags: A C ritiqu e o f the H in du R igh t. Delhi: Orient Longman.

Bate, John Bernard, zooo. “Mêtaittamih Oratory and Democratic Practice in Tamil Nadu.” PhD diss., University o f Chicago.

Bauman, Chad M. 2011. “Anti-conversion Authors, Altered Definitions o f Religion, and Conversion as an Altered State.” Paper presented at “Altered and Alternative States,” Conference on the Study o f Religions o f India, Los Angeles, June 23-36.

Baumann, Gerd. 2004. “Grammars o f Identity/Alterity.” In Gramm ars o f Iden tity / A lterity: A Structura l Approach, edited by André Gingrich and Gerd Baumann, 18-50. New York: Berghahn Books.

Bayly, Susan. 1998. “ Hindu Modernisers and the ‘Public’ Arena: Indigenous Cri � tiques o f Caste in Colonial India.” In Swam i V ivekananda a n d the M odern iz a � tion o f H indu ism , edited by William Radice, 93-137. Chennai: Oxford University Press.

Benjamin, Walter. 1968. “ Theses on the Philosophy o f History.” In Illum ina tions, 253-64. New York: Harcourt, Brace and World.

Bernasconi, Robert. 2010. N ature, C ulture, a n d Race. Stockholm: Sôdertôrn Uni � versity Press.

Berreman, Gerald D. 1971. “ The Brahmanical View o f Caste.” Contributions to Ind ia n Sociology, n.s., 5:16-23.

-------- . 1972. H indus o f the H im alayas: Ethnography a n d Change. 2nd, rev. and enl. ed. Berkeley: University o f California Press.

-------- . 1979. Caste a n d O ther Inequ ities: Essays on Inequ ality. Meerut: Ved Prakash Vatuk Folklore Institute.

Béteille, André. 2000. “Caste in Contemporary India.” In Caste Today, edited by C .J. Fuller, 150-79. New Delhi: Oxford University Press.

-------- . 2012. “India’s Destiny Not Caste in Stone.” H indu , February 21, www .thehindu.com/opinion/lead/indias-destiny-not-caste-in-stone/article2913662.ece.

Bhatt, Chetan. 2001. H indu N a tion a lism : O rigins, Ideologies a n d M odern Myths. Oxford: Berg.

Bialecki, Jon . 2014. “ Does God Exist in Methodological Atheism? On Tanya Lurh- mann’s W hen God Talks Back and Bruno Latour.” Anthropology o f Consciousness 25 (1): 32-52.

Bondurant, Joan Valerie. 1965. Conquest o f Violence: The G andhian Philosophy o f Conflict. Berkeley: University o f California Press.

Bourdieu, Pierre. 1990. The Logic o f Practice. Translated by Richard Nice. Stanford, C A : Stanford University Press.

2 6 6 • R E F E R E N C E S

Bronkhorst, Johannes. 1007. “Modes of Debate and Refutation o f Adversaries in Classical and Medieval Ind ia: A Preliminary Investigation.” A ntiquorum Philos- ophia 1:269-80.

Brown, Peter. 1981. The C u lt o f the Saints: Its Rise a nd Function in L a t in Christianity. Chicago: SM C Press.

Bugge, Henriette. 1994. M ission a n d T am il Society: Socia l a n d Religious Change in South In d ia (1840-1900). Richmond: Curzon Press.

Butalia, Urvashi. 2000. The O ther Side o f Silence: Voicesfrom the Partition o f India . Durham: Duke University Press.

Caplan, Lionel. 1989. Religion a n d Power: Essays on the Christian Community in M adras. Madras: CLS.

Census o f India. 1961. Slums o f M adras C ity. Edited by P.K . Nambiar. Madras: Government of India, 1961.

Chakrabarty, Dipesh. 1991. “ Open Space/Public Place: Garbage, Modernity and India.” South A sia 14 (1): 15-31.

Chatterjee, Partha. 1986. N a tion a list Thought a n d the C olon ia l W orld: A D eriva tive D iscourse? London: Zed Books.

-------- . 1999. The N ation a n d Its Fragments. Delhi: Oxford University Press. Claerhout, Sarah, and Jakob De Roover. 2005. “The Question o f Conversion in

India! ’ Econom ic a n d Po lit ica l Weekly 40 (28): 3048-55. Clastres, Pierre. 2000. Chronicle o f the Guayaki Indians. New York: Zone Books. Conrad, Dieter. 2010. “The Personal Law Question and Hindu Nationalism.” In The

O xford Ind ia H indu ism Reader, edited by Vasudha Dalmia and Heinrich von Stietencron, 187-230. New Delhi: Oxford University Press.

“Conversions May Have Been behind Staines Killing.” 1999. H indu , May 21. Cossman, Brenda, and Ratna Kapur. 1997. “Secularism’s Last Sigh: The Hindu

Right, the Courts, and India’s Struggle for Democracy.” H a rva rd In tern a tiona l L a w Jo u r n a l 38 (1): 113-70.

Crehan, Kate. 2002. Gramsci, C ulture a n d Anthropology. Berkeley: University of California Press.

Datta, Pradip Kumar. 1999. C a rvingB locs: Com m unal Ideology in E a r ly Twentieth- Century Benga l. New Delhi: Oxford University Press.

“Deadly ‘Foreign Hand.’” 1990. Econom ic a n d Po lit ica l Weekly 25 (17): 911. Deliege, Robert. 1997. The W orld o f the “ Untouchables”: Para iyars o f T am il N adu .

Chennai: Oxford University Press. -------- . 1999. The Untouchables o f Ind ia . Oxford: Berg. Derrett, J. Duncan M . 1968. Religion, L a w a nd the State in Ind ia . London: Faber

and Faber. Desai, Mahadev. 1935. “Interesting Questions.” H a rija n , May 11. -------- . 1953. The D ia ry o f M ahadev D esai. Edited and translated by Valji Govind

Desai. Ahemadabad: Navajivan. Deshpande, Ashwini. 2011. The Gram m ar o f Caste: Econom ic D iscrim ina tion in

Contemporary Ind ia . New Delhi: Oxford University Press. Deshpande, G . P. 1985. “The Plural Tradition.” Sem in ar 313:23-25.

R E F E R E N C E S • 2 6 7

de Wit, Joop W. 1996. Poverty, Policy a n d Politics in M adras Slums: D ynam ics o f Su rviva l, Gender a n d Leadership. Delhi: Sage Publications.

Dirks, Nicholas, 2001. Castes o f M in d: Colonialism a n d the M a k ing o f M odern Ind ia . Princeton, N J: Princeton University Press.

Dollard, John . 1957. Caste a n d Class in a Southern Town. New York: Doubleday. Dumont, Louis. 1980. H omo H ierarchicus: The Caste System a n d Its Implications.

Rev. ed. Chicago: University o f Chicago Press. Durkheim, Emile. 1995. The E lem en tary Forms o f Religious L ife . New York: Free

Press. Ebeling, Sascha, 2010. “Another Tomorrow for Nantanär: The Continuation and

Re-invention o f a Medieval South-Indian Untouchable Saint.” In Geschichten u nd Geschichte: H istoriographie u nd H agiographie in der Asiatischen Religionsge � schichte, edited by Peter Schalk, Max Deeg, Oliver Frieberger, and Christoph Kleine, 433-516. Uppsala: Uppsala University Press.

Eck, Diana L. 1998. D arsan: Seeing the D ivin e Im age in Ind ia . 3rd ed. New York: Columbia University Press.

Evans-Pritchard, E. E. 1937. W itchcraft, O racles a n d M agic among the A z a nde. Oxford: Clarendon Press.

Ferguson, James. 2.013. “ Declarations o f Dependence: Labour, Personhood, and Welfare in Southern Africa.” Jo u r n a l ofthe Roya l A nthropological Institute 19 (2): 223-42.

Fernandes, Leela, and Patrick Heller. 2006. “ Hegemonic Aspirations: New Middle Class Politics and India’s Democracy in Comparative Perspective.” C r itic a l Asian Studies 38 (4): 495-522.

“ ‘Foreign H and’ Again.” 1999. Econom ic a n d Po lit ica l W eekly 34 (21): 1228-29. Forrester, Duncan. 1980. Caste a n d C hristia n ity: A ttitudes a n d Policies on Caste o f

A nglo-Saxon Protestant M issions in In d ia . London: Curzon. Foucault, Michel. 1991. D iscipline a n d Pun ish: The B irth o f the Prison . London: Pen�

guin Books. Freeman, James M. 1986. “The Consciousness o f Freedom among India’s Untoucha �

bles.” In Socia l a n d Econom ic D evelopment in In d ia , edited by Dilip K . Basu and Richard Sisson, 153-71. New Delhi: Sage Publications.

Frykenberg, Robert. 1997. “The Emergence o f Modern ‘Hinduism’ as a Concept and as an Institution: A Reappraisal with Special Reference to South India.” In H in � duism Reconsidered, rev. ed., edited by Günther D. Sontheimer and Hermann Kulke, 82-107. New Delhi: Manohar.

Fuller, C . J. 2004. The Camphor F lam e: Popu lar H indu ism a n d Society in Ind ia . Rev. and expanded ed. Princeton, N J: Princeton University Press.

-------- . 2015. “Colonial Anthropology and the Decline o f the Raj: Caste, Religion and Political Change in India in the Early Twentieth Century.” Jo u r n a l o f the Roya l Asia tic Society FirstView: 1-24 .

Fulton, Rachel. 2002. “ History, Conversion, and the Saxon Christ.” In From Ju dg � ment to Passion: D evotion to C hrist a n d the V irgin M ary, 800-1200, 9-59. New York: Columbia University Press.

z 6 8 • R E F E R E N C E S

Galanter, Marc. 1971. “ Hinduism, Secularism, and the Indian Judiciary.” Philosophy E ast a n d West 1 1 (4): 467-87.

Gandhi, M. K . 1944. N on-violence in Peace a n d War. Ahmedabad: Navajivan. -------- . 1958-94. The Collected Works o f M aha tm a G andhi. 100 vols. New Delhi:

Publications Division o f the Government of India. -------- . 1997. H in d Sw ara j a n d O ther W ritings. Edited by Anthony J. Parel. Cam �

bridge: Cambridge University Press. Geuss, Raymond. 1999. “Nietzsche and Genealogy.” In M orality, Culture, a nd H is �

tory: Essays in Germ an Philosophy, 1-2.8. Cambridge: Cambridge University Press.

Gill, Timothy. 2007. “Making Things Worse: How ‘Caste-Blindness’ in Indian Post-tsunami Recovery Has Exacerbated Vulnerability and Exclusion.” United Nations Office o f the H igh Commissioner for Human Rights. http:/ /www2 .ohchr.org/english/bodies/cerd/docs/ngos/tsunami_report.pdf.

Goffman, Erving. 1981. “ Footing.” In Forms o f T alk, 124-57. Oxford: Basil Blackwell.

Gough, E. Kathleen. 1956. “Brahman Kinship in a Tamil Village.” A m erican A nthropologist 58 (5): 826-53.

-------- . 1979. “Caste in a Tanjore Village.” In Aspects o f Caste in South Ind ia , Ceylon a n d N orth-W est Pakistan, edited by E. R Leach, 11-60 . Cambridge: Cambridge University Press.

Gould, William. 2004. H indu N a tionalism a n d the Language o f Politics in L a te C olon ia l Ind ia . Cambridge: Cambridge University Press.

Grafe, Hugald. 1990. H istory o f C hristian ity in In d ia . Vol. 4, pt. 2, T am ilnadu in the N ineteen th a n d Twentieth Centuries. Bangalore: Church History Association of India.

Granoff, Phyllis. 1985. “Scholars and Wonder-Workers: Some Remarks on the Role o f the Supernatural in Philosophical Contests in Vedanta H agiographies.” Jo u r � n a l o f the A m erican O rien ta l Society 105 (3): 459-67.

Guha, Sumit. 2013. Beyond Caste: Iden tity a n d Power in South Asia, Past a n d Present. Leiden: Brill.

Gupta, Dipankar. 2004. “The Certitudes of Caste: When Identity Trumps Hierar � chy.” In Caste in Q uestion, edited by Dipankar Gupta, ix-xxi . London: Sage Publications.

Gupta, Kanti Prasanna Sen. 1971. The C hristian M issionaries in Bengal, 1793-1833. Calcutta: Firma K . L. Mukhopadhyay.

Guru, Gopal, ed. 1990. H um ilia tion : C laims a n d Context. New Delhi: Oxford Uni � versity Press.

Hamermesh, Mira, dir. and prod. 1990. Caste a t B irth . Film. Alexandria, VA: Film� makers Library.

Harper, Susan Billington. 2000. In the Shadow o f the M aha tm a: Bishop V .S. A z a- r iah a n d the Travails o f C hristian ity in B rit ish Ind ia . Surrey: Curzon Press.

Harriss-White, Barbara, ed. 2014. D a lits a nd A divasis in In d ia ’s Business Economy. Gurgaon, India: Three Essays Collective.

R E F E R E N C E S • 2 6 9

Herzfeld, Michael. 1004 . “Structural Nostalgia: Time and the Oath in the Moun � tain Villages o f Crete.” In C u ltu ra l Intim acy: Socia l Poetics in the N ation-State, 2nd ed., 177-82. New York: Routledge.

Hogeveen, Jeremy, Michael Inzlicht, and Sukhvinder S. Obhi. 2014. “Power Changes How the Brain Responds to O t h e rs” Jo u r n a l o f E xperim en ta l Psychol � ogy: Genera l 143 (2): 755-62.

Human Rights Watch. 1999. Broken People: Caste Violence against In d ia ’s “Untouch- ables.” Edited by Smita Narula. New York: Human Rights Watch. https:/ /www .hrw.org/reports/1999/india/.

-------- . 2002. ‘“ We Have Orders Not to Save You’: State Participation and Com � plicity in Communal Violence in Gujarat.” H um an Rights Watch 14 (3C): 1-68. https:/ /www.hrw.0rg/rep0rts/2002/india/gujarat.pdf.

-------- . 2005. “After the Deluge: India’s Reconstruction Following the 2004 Tsu� nami” H um an Rights Watch 17(30): 1-49. https:/ /www.hrw.0rg/rep0rts/2005 / india0505/ india0505.pdf.

Hume, David. [1757] 1957. The N a tu ra l H istory o f Religion . Edited by H .E . Root. Stanford, C A : Stanford University Press.

Irschick, Eugene F. 1969. Politics a n d Socia l Conflict in South In d ia : The N on-Brah � man M ovement a n d T a m il Separatism, 1916-1929 . Berkeley: University o f Cal i � fornia Press.

ISH R (International Service for Human Rights). 2007. Comm ittee on the E lim in a � tion o f R a c ia l D iscrim ina tion, 70th Session, Ind ia , 15 th-19 th Report. Treaty Body Monitor, Human Rights Series. New York: ISH R . http:/ /olddoc.ishr.ch/hrm /tmb/treaty/cerd/rep0rts/cerd_70/cerd_70_india.pdf.

Iyotheedas, C . [1892] 1985. “The Open Letter [to Srinivasa Raghavaiyangar, 1892].” In Scheduled Caste’s Struggle fo r Em ancipation in South In d ia , edited by T. P. Kamalanathan, 10-18. Tiruppatur, Tamil Nadu: South India Sakkya Buddhist Association.

Jaffrelot, Christophe. 1996. The H in du N a tion a list M ovem ent in Ind ia . New York: Columbia University Press.

-------- . 2005. D r A m bedk ar a n d U ntouchability: A n a lysing a nd F igh ting Caste. London: C . Hurst.

Jha, D. N . 2002. The M yth o f the H oly Cow. London: Verso; New Delhi: Navayana. Jones, Kenneth. 1976. A rya D h a rm : H indu Consciousness in 19 th-C en tu ry Punjab.

Delhi: Manohar. Jordens, J. T. F. 1981. Swdm i Shraddhananda, H is L ife a n d Causes. Delhi: Oxford

University Press. Kapadia, Karin. 1996. Siva a n d H er Sisters: Gender, Caste a n d Class in R u r a l South

Ind ia . Madras: Oxford University Press. Kaur, Naunidhi. 2002. “The Foreign Hand.” Fron tline, December 7-20 . Kaviraj, Sudipta. 1997. “Filth and the Public Sphere: Concepts and Practices about

Space in Calcutta.” Public C u lture 10 (1): 83-113. Khetan, Ashish. 2007. “Gujarat 2002—The Truth.” Tehelka, November 3. http:/ /

archive.tehelka.com/story _main35.asp?filename=Neo3iio7gujrat_sec.asp.

2 7 0 • R E F E R E N C E S

Kim, Sebastian C . H . 2003. In Search o f Iden tity: On Religious Conversion in Ind ia . Chennai: Oxford University Press.

Kolenda, Pauline Mahar. 1964. “Religious Anxiety and Hindu Fate.” Jo u r n a l o f Asian Studies 23:71-81.

Krishnamacharya, U. P. [1936?] Temple Worship a n d Temple E n try. 2nd ed. Nellore, India: n.p.

Kuper, Adam. 1999. C ulture: The Anthropologists’ Account. Cambridge, M A : Har � vard University Press.

Kuppuswami, Alladi. 1999. “Can Conversions Be Banned?” H indu , January 30. Lee, Joel. 20x5. Recognition a n d Its Shadows: D a lits a n d the Politics o f Religion in

In d ia . PhD diss., Columbia University. Leslie, Julia. 1991. Roles a n d R itu a ls fo r H indu Women. London: Pinter. Lindley, Richard. 1986. Autonomy. London: Macmillan. Loux, Michael J. 2006. M etaphysics: A Contemporary Introduction. Vol. 3. Abing�

don: Routledge. Löwy, Michael. 2005. F ire A la rm : Reading W alter Ben jam ins “On the Concept o f

H istory.” Translated by Chris Turner. London: Verso. Lynch, Owen M. 1969. The Politics o f Untouch ab ility . New York: Columbia Univer �

sity Press. -------- . 1977. “Method and Theory in the Sociology o f Louis Dumont.” In The N ew

W ind: Changing Iden tities in South A sia , edited by Kenneth David, 239-63. The Hague: Mouton.

Madan, T . N . 2003. “Freedom o f Religion.” Econom ic a n d Po lit ica l Weekly 38 (11): 1034-41.

Mamdani, Mahmood. 2000. Beyond Rights T alk a n d C u lture T a lk: Compara � t ive Essays on the Politics o f Rights a n d C ulture. New York: St. Martin’s Press.

Mantena, Karuna. 2012. “ On Gandhi’s Critique o f the State: Sources, Contexts, Conjectures.” M odern In tellectua l H istory 9 (3): 535-63.

Mauss, Marcel. 1966. The G ift: Forms a n d Functions o f Exchange in A rchaic Societies. Translated by Ian Cunnison. London: Cohen and West.

-------- . 1973. “Techniques o f the Body.” Economy a n d Society 2 (1): 70-88. Mencher, Joan P. 1972. “Continuity and Change in an Ex-Untouchable Commu�

nity o f South India.” In The Untouchables in Contemporary Ind ia , edited by J. Michael Mahar, 37-56. Tuscon: University o f Arizona Press.

-------- . 1974. “The Caste System Upside Down, or the Not So Mysterious East.” C urren t Anthropology 15 (4): 469-93.

-------- . 1980. “ On Being an Untouchable in India: A Materialist Perspective.” In Beyond the M yths o f C ulture: Essays in C u ltu ra l M a teria lism , edited by Eric B. Ross, 261-94. New York: Academic Press.

Mendelsohn, Oliver, and Marika Vicziany. 2000. The Untouchables. New Delhi: Cambridge University Press.

Menon, Nivedita. 2004. “A ll Ye Faithless.” Telegraph, May 6, www.telegraphindia .com/i0405o6/asp/opinion/story_3i97994.asp.

R E F E R E N C E S • 2 7 1

Metcalf, Barbara D., and Thomas R. Metcalf. 2006. A Concise H istory o f M odern Ind ia . Cambridge: Cambridge University Press.

Metcalf, Thomas R. 2007. Ideologies o f the R a j. Cambridge: Cambridge University Press.

Meyer, Birgit. 1999. T ranslating the D ev il: Religion a n d M odern ity among the Ew e in Ghana. Trenton, N J: A frica World Press.

Mitchell, Timothy. 1990. “Everyday Metaphors o f Power.” Theory a n d Society I 9 ;545~77-

Mohan, P. Sanal. 2014. M odern ity o f Slavery: Struggles against Caste Inequ a lity in K era la . New Delhi: Oxford University Press.

Mosse, David. 2012. The Sa in t in the Banyan Tree: C hristian ity a n d Caste Society in Ind ia . Berkeley: University of California Press.

MPC M A E C (Madhya Pradesh Christian Missionary Activities Enquiry Commit � tee). 1957. Report o f the C hristian M issionary A ctivities E n qu iry Committee, M a d � hya Pradesh o f 1956. Edited by M. B. Niyogi. Indore: Government Regional Press.

Mukerji, UpendraNath. [1909] 1929. A D y ing Race. Calcutta: Bhaskar Mukerjee. -------- . 1911. H indu ism a n d the Com ing C ensus: C hristian ity a n d H indu ism . Cal �

cutta: Srikali Ghosh Cotton Press. “Mutalvar Jeyalalita Tokutiyil ‘Matamarram’ ” [Religious conversion in Chief Min �

ister Jeyalalitha’s constituency]. 2002. T inam alar, May 11. Nabokov, Isabelle. 2000. Religion against the Se lf: A n Ethnography o f T am il Ritu a ls.

New York: Oxford University Press. Nandy, Ashis. 1983. The In tim a te Enem y: Loss a n d Recovery o f S e l f under C olonia l �

ism. New Delhi: Oxford University Press. -------- . 2004. “The Politics o f Secularism and the Recovery o f Religious Tolerance.”

In Secularism a n d Its C ritics, edited by Rajeev Bhargava, 321-44 . Chennai: Oxford University Press.

“Natu muluvatum matamarram totarntu atikarippu” [Religious conversion con� tinuing to increase throughout the land]. 1999. T inam alar, May 5.

Nietzsche, Friedrich. 1997. On the Genealogy o f M ora lity. Edited by Keith Ansell- Pearson. Cambridge: Cambridge University Press.

Oddie, Geoffrey. 1982. “Anti-missionary Feeling and Hindu Revivalism in Madras: The Hindu Preaching and Tract Societies, c. 1886-1891.” In Images o f M a n : R e li �

gion a n d H istorica l Processes in South Asia , edited by Fred W. Clothey, 217-34 . Madras: New Era Publications.

-------- . 1991. H indu a n d C hristian in Sou th-E ast Ind ia . London: Curzon. O ’Malley, L .S .S . 1935. Popu lar H indu ism : The Religion o f the Masses. Cambridge:

Cambridge University Press. Pandey, Gyanendra. 1983. “Rallying around the Cow: Sectarian Strife in the Bho-

jpur Region, c. 1888-1917.” In Suba ltern Studies I I , edited by Ranajit Guha, 60-129. New Delhi: Oxford University Press.

-------- . 1993. “ Which o f Us Are Hindus?” In H indus a n d O thers: The Q uestion o f H indu Identity in Ind ia Today, edited by Gyanendra Pandey, 238-72. New Delhi: Viking.

Z 7 1 • R E F E R E N C E S

Panini, M . N . 1996. “ The Political Economy o f Caste.” In Caste: Its Twentieth C en tury A va tar, edited by M . N . Srinivas, 28-68. New Delhi: Penguin Books.

Parel, Anthony J. 1997. Editor’s Introduction. In H in d Sw ara j a nd O ther W ritings, by M .K . Gandhi, edited by Anthony J. Parel, xii i-lxii . Cambridge: Cambridge University Press.

Parry, Jonathan. 1991. “A Hindu Lexicographer? A Note on Auspiciousness and Purity.” Contributions to Ind ia n Sociology 25 (2): 267-85.

Peabody, Norbert. 2001. “Cents, Sense, Census: Human Inventories in Late Preco� lonial and Early Colonial India.” Comparative Studies in Society a nd H istory 43 (3): 8x9-50.

Philip, Jo ji Thomas. 2014. “ Decoding India’s Ordinance System.” L ive M in t , January 10. www .livemint .com/Spec ia ls/ZRtVJMBf O LoQ 4l9ZoM A2wK /Decoding-Indias-ordinance-system—Shubhankar-Dam.html.

Prashad, Vijay. 1996. “The Untouchable Question.” Econom ic a n d Po lit ica l Weekly 39 ( 9): 551- 69 -

-------- . 2000. Untouchable Freedom : A Socia l H istory o f a D a lit Community. Chen � nai: Oxford University Press.

Preus, J. Samuel. 1979. “Machiavelli’s Functional Analysis o f Religion: Context and Object.” Jo u r n a l o f the H istory o f Ideas 40 (2): 171-90 .

-------- . 1996. E x p la in in g Re ligion : C riticism a n d Theory Jrom Bodin to Freud. Atlanta, G A : Scholars Press.

Radhakrishnan, S. 1940. Eastern Religions a n d Western Thought. 2nd ed. Oxford: Oxford University Press.

Raghavaiyangar, S. Srinivasa. 1893. M emorandum on the Progress o f the M adras Presidency du ring the L ast Forty Years o f British A dm inistration. Madras: Govern � ment Press.

Raj, Selva J., and William P. Harman. 2006. D ea ling with D eities: The R it u a l Vow in South Asia . Albany: State University o f New York Press.

Ramani, Rukmani. 1985. Slums o f M adras C ity. Interns Report. Madras: M IDS. Rao, Anupama. 2009. The Caste Q uestion: D a lits a n d the Politics o f M odern Ind ia .

Berkeley: University o f California Press. Rawat, Ramnarayan S. z o u . Reconsidering U ntouchability: Chamars a n d D a lit H is �

tory in N orth In d ia . Bloomington: Indiana University Press. Renold, Leah. 1994. “ Gandhi: Patron Saint of the Industrialist.” Sagar: South Asia

Graduate Research Jo u r n a l 1 (1): 16-38. Robbins, Joel. 1998. “Reading ‘ World News’: Apocalyptic Narrative, Negative

Nationalism and Transnational Christianity in a Papua New Guinea Society.” Socia l A nalysis 42 (2): 103-30.

-------- . 2004. Becom ing Sinners: C hristian ity a n d M ora l Torment in Papua N ew Guinea Society. Berkeley: University o f California Press.

-------- . 2007. “Continuity Thinking and the Problem o f Christian Culture: Belief, Time, and the Anthropology o f Christianity.” C urren t A nthropology 48 (1): 5- 17-

R E F E R E N C E S • Z 7 3

Roberts, Nathaniel. 2008. “Caste, Anthropology of.” In In tern a tion a l Encyclopedia o f the Socia l Sciences, 2nd ed., edited by William S. Darity, 461-63. New York: Macmillan.

-------- . 2009. “Ethnographic Knowledge and the Government o f Religion: Conver � sion in India.” Paper presented at the Department o f Anthropology, University o f Virginia, March 20.

-------- . 20x2a. “Is Conversion a ‘Colonization o f Consciousness’ t” A nthropological Theory 12 (3): 271-94.

-------- . 2012b. “Meanings o f Monotheism: Ethnographic Evidence and the Intoler � ance Thesis.” Paper presented at the University o f Edinburgh, Centre for South Asian Studies, March 22.

-------- . 2015a. “From Village to City: Hinduism and the ‘Hindu Caste System.’ ” In H andbook o f Religion in the Asian C ity, edited by Peter van der Veer, 237-53. Berkeley: University of California Press.

-------- . 2015b. “Setting Caste Back on Its Feet.” A nthropology o f This Century, no. 13, May. http:/ /aotcpress.com/articles/setting-caste-feet/.

SA H R D C (South Asian Human Rights Documentation Centre). 2008. “A nti � conversion Laws: Challenges to Secularism and Fundamental Human Rights.” Econom ic a n d Po lit ica l W eekly 43 (2): 63-73.

Sampath, E .V .K . Sulochana. 1987. “ The Housing Problems o f the Low-Income Groups: Tamilnadu Experience o f Slum Improvement.” In Poverty in M etropoli � tan C ities, edited by S. Manzoor Alam and Fatima Alikhan , 159-68. New Delhi: Concept Publishing.

Saraswati, Swami Dayananda. 1999. “Conversion Is Violence.” Ind ia n Express, Octo � ber 29.

Sarkar, Sumit. 2007. “Christian Conversions, Hindutva, and Secularism.” In The C risis o f Secularism in In d ia , edited by Anuradha Dingwaney Needham and Rajeswari Sunder Rajan, 356-67. Durham, N C : Duke University Press.

Sartori, Andrew. 2003. “ ‘Culture’ in Bengal, 1870s to 1920s: The Historical Genesis o f an Ambivalent Concept.” PhD diss., University of Chicago.

-------- . 2008. Benga l in G loba l Concept H istory: Culturalism in the Age o f C apita l. Chicago: University o f Chicago Press.

Savarkar, V . D . 1969. H in du tva : W ho Is a H in du? Bombay: Veer Savarkar Prakashan.

“SC Expunges Its Objectionable Remarks on Staines.” 2011. Econom ic Times, January 26. http:/ /artic les.ec0n0mictimes.indiat imes.c0m /2011-01-26 /news/28424945_i_graham-staines-dara-singh-minor-sons.

Scherl, Richard. 1996. “Speaking with Mariyätai: A Linguistic and Cultural Analy � sis o f Markers of Plurality in Tamil.” PhD diss., University o f Chicago.

Schneewind, J. B. 1998. The Inven tion o f A utonomy: A H istory o f M odern M ora l Philosophy. Cambridge: Cambridge University Press.

Schwab, Raymond. 1984. The O rien ta l Renaissance: E u rope’s Rediscovery o f Ind ia a n d the East. New York: Columbia University Press.

Z 7 4 • R E F E R E N C E S

Scott, James C . 1990. D om ination a nd the A rts o f Resistance: H idden Transcripts. New Haven, C T : Yale University Press.

Searle-Chatterjee, Mary. 2008. “Attributing and Rejecting the Label ‘ Hindu’ in North India.” In Religion, Language a n d Power, edited by Mary Searle-Chatterjee and Nile Green, 186-201. New York: Routledge.

Sen, Dwaipayan. 2012. ‘“ No Matter How, Jogendranath Had to Be Defeated’: The Scheduled Castes Federation and the Making o f Partition in Bengal, 1945-1947.”

. Ind ia n Econom ic a n d Socia l H istory Review 49 (3): 321-64. Sen, Ronojoy. 200 7. Lega liz ing Re ligion: The Ind ia n Supreme Court a nd Secularism.

Washington, D C : East-West Center Washington. Sharma, Arvind. 1992. “Ancient Hinduism as a Missionary Religion.” N umen 39

(2): 175. -------- . 201$. Review o f The P a ria h Problem , by Rupa Viswanath. In tern a tion a l

Jo u r n a l ofD h arm a Studies 3 (1): 8. Sherlock, Stephen. 1998. “Berlin, Moscow and Bombay: The Marxism That India

Inherited.” South A sia : Jo u r n a l o f South Asian Studies 21 (1): 63-76. Singh, Tavleen. 2004. “ The Return o f the ‘Foreign Hand.’ ” Ind ia n Express,

Ju ly 18. http:/ /expressindia.indianexpress.com/news/budgeto4/fullestory .php?type=ie&content_id=5i237.

Skinner, Quentin. 1988. M ean ing a nd Context: Q uentin Sk in ner a nd H is Critics. Edited by James Tully. Princeton, NJ: Princeton University Press.

Srinivas, M . N . 1956. “A Note on Sanskritization and Westernization.” F a r Eastern Q uarterly 15 (4): 481-96.

-------- . 1962. “ Hinduism.” In Encyclopaedia Brita nn ica , 14th ed., 574-77. Chicago: Encyclopaedia Britannica.

-------- . 2003. “An Obituary on Caste as a System.” Econom ic a n d Po lit ica l Weekly 38 ( 5): 455-59-

— ——. 2012. The Remembered V illage, 2nd ed. Delhi: Oxford University Press. Stahnke, Tad. 1999. “Proselytism and the Freedom to Change Religion in Interna �

tional Human Rights Law.” Brigham Young U niversity L a w Review 1999 (1): 251-349.

“Staines Was Not Converting Tribals.” 2003. Ind ia n Express, January 17. Subramanian, Narendra. 1999. E thn icity a nd Popu list M obiliz a tion: Po litica l P a r �

ties, C itizens a n d Democracy in South Ind ia . Delhi: Oxford University Press. Sullivan, Winnifred Fallers. 2005. The Impossibility o f Religious Freedom . Princeton,

N J: Princeton University Press. Supreme Court of India. 2011. Criminal Appeal No. 1366 of 2005, Rabindra Kumar Pal

@ Dara Singh v. Republic of India, January 21. http:/ /judis.nic.in/supremecourt /imgsi.aspx?filename=37394.

Supreme Court Reports. 1977. Rev. Stainislaus v. State o f Madhya Pradesh and Orissa, January 17.

Taylor, Charles. 1992. M ulticu ltura lism a n d “The Politics o f Recognition. ” Edited by Amy Gutmann. Princeton, N J: Princeton University Press.

R E F E R E N C E S • 2 7 5

Tejani, Shabnum. zoo8. Ind ia n Secularism : A Socia l a n d In tellectu a l H istory, 1890- ipso. Bloomington: Indiana University Press.

Terchek, Ronald J. zon . “Conflict and Nonviolence.” In The C ambridge Companion to G andh i, edited by Judith Brown and Anthony Parel, 117-34 . New York: Cam � bridge University Press.

Thapar, Romesh. 1983. “The Foreign Hand.” Economic a nd Politica l Weekly 18 (zz): 943. Thapar, Romila. 1985. “Syndicated Moksha?” Sem in ar 314 :14 ^ 4 . Thirumaavalavan, T. Z 003 . T alism an: E x trem e Emotions o f D a lit L ibera tion . Trans �

lated and annotated by Meena Kandasamy. Kolkata: Samya. Thomas, V. V. zoo8. D a lit Pentecostalism: Spiritu a lity o f the Em powered Poor. Ban �

galore: Asian Trading Company. Thorat, Sukhadeo, and Katherine S. Newman, eds. zoio. Blocked by Caste: Econom ic

D iscrim ina tion in M odern In d ia . Delhi: Oxford University Press. T N SC B (Tamil Nadu Slum Clearance Board). 1971. T a m il N adu Slum Clearance:

F irst Year in M adras. Madras: Tamil Nadu Slum Clearance Board. -------- . 1975. Socio-Economic Survey o f M adras Slums. Edited by Rama Arangannal.

Madras: Tamil Nadu Slum Clearance Board. Trautmann, Thomas R. 1981. D ravid ia n K insh ip. Cambridge: Cambridge Univer �

sity Press. University of Madras. 19Z9 . T a m il Lexicon . Vol. 3. Madras: University o f Madras. van der Veer, Peter. 1994. “Syncretism, Multiculturalism and the Discourse on Tol �

erance.” In Syncretism / A nti-syncretism , edited by Charles Stewart and Rosalind Shaw, 196-211. New York: Routledge.

Vasudevan, R. zou. “ Widow o f Murdered Graham Staines Shows Compassion to Killer.” Asia n T ribune, January zz, www.asiantribune.com/news/zoii /oi /zi /widow-murdered-graham-staines-shows-compassion-killer.

Vera-Sanso, Penny. 1994. “ What the Neighbours Say: Gender, Personhood and Power in Two Low-Income Settlements o f Madras.” PhD diss., University o f London.

Vincentnathan, Lynn. 1993 . “ Untouchable Concepts o f Person and Society.” Con � tributions to India n Sociology irj (1): 53-8 Z .

Viramma, Josianne Racine, and Jean-Luc Racine, zooo. V iramm a: L ife o f a D a lit . New Delhi: Social Science Press.

Viswanath, Rupa. zoio. “Spiritual Slavery, Material Malaise: ‘“ Untouchables’ and Religious Neutrality in Colonial South India.” H istorica l Research 83 (Z19): 1 Z 4 -4 5 .

-------- . zoxza. “ Dalits/Ex-Untouchables.” In B r i l l Encyclopedia o f H indu ism , vol. 4, edited by Knut A . Jacobsen, Angelika Malinar, Helene Basu, and Vasudha Naray� anan, 779-87. Leiden: Brill.

-------- . zoizb. “Making Way: Two Faces o f Religion and the Liberal Concept o f Abstract Space.” Paper presented at the annual meeting o f the American Acad � emy o f Religion, Chicago, November 19.

-------- . zoizc. “A Textbook Case o f Exclusion.” Ind ia n Express, July 13, www .indianexpress.com/news/ a-textbook-case-of-exclusion/973711 /o.

1 7 6 • R E F E R E N C E S

-------- . 2013- “ The Emergence of Authenticity Talk and the Giving o f Accounts: Conversion as Movement o f the Soul in South India, ca. 1900.” Comparative Studies in Society a n d H istory 55 (1): 120-41.

-------- . 2014a. “Economies o f Offense: Hatred, Speech and Violence in India.” Paper presented at roundtable on controversy surrounding Wendy Doniger’s The H indus: A n A ltern a tive H istory, Annual Meeting o f the American Academy of Religion, San Diego.

-------- . 2014b. The Pa ria h Problem : Caste, Religion a n d the Socia l in M odern Ind ia . New York: Columbia University Press.

-------- . 2014c. “Rethinking Caste and Class: ‘Labour,’ the ‘Depressed Classes,’ and the Politics o f Distinctions, Madras, 1918-1924.” In tern a tion a l Review o f Social H istory 59 (1): 1-37.

-------- . 2015. “CommissioningRepresentation: The Misra Report, Deliberation and the Government of the People in Modern India.” South A sia : Jo u r n a l o f South Asian Studies 38 (3): 495-511.

Viswanathan, Gauri. 1995. “Beyond Orientalism: Syncretism and the Politics of Knowledge.” Stanford H um an ities Review 5 (1): 19-34.

-------- . 2000. “Literacy and Conversion in the Discourse o f Hindu Nationalism.” Race a n d Class 42 (1): 1-20.

-------- . 2001. O utside the F o ld: Conversion, M odern ity a n d Belief. Chennai: Oxford University Press.

Viswanathan, S. 2003. “A Decree on Animal Sacrifice.” Fron tline, September 27- October 10, www.fr0ntline.in/static/html/ fl2020/st0ries/20031010001205000 .htm.

Volosinov, V . N . 1986. M arxism a nd the Philosophy o f Language. Cambridge, M A : Harvard University Press.

Vyas, Neena. 1999. “The Sangh Parivar’s Sinister Game.” H indu , January 17. -------- . 2002. “ When Their Gods Fail Them.” H in du , October 20. Weber, Max. 1978. Economy a nd Society: A n O utline o f Interpretive Sociology. Edited

by Geunther Roth and Claus Wittich. Berkeley: University o f California Press. Webster, John C .B . 2009. The D a lit C hristians: A H istory. 4th, rev. and enl. ed. New

Delhi: ISPC K . Wiebe, Paul. 1975. 7he Socia l L ife o f an Ind ia n Slum . Delhi: Vikas Publishing

House. Zelliot, Eleanor. 2010. “Gandhi and Ambedkar: A Study in Leadership.” In From

Untouchable to D a lit: Essays in the A m bedkar M ovement, 3rd ed., 150-83. Delhi: Manohar.

-------- . 2013. A m bedkars W orld: The M a k ing ofBabasaheb a n d the D a lit Movement. New Delhi: Navayana.

R E F E R E N C E S • 2 7 7

I N D E X

A d coc k , C a th er in e S., 118 -19 , 13 9 adultery, 87, 89 ,198-9 9 , 235, 255m alcohol , 163; al leged use by slum wom en of,

51; as cause o f su ffering and m ari ta l strife , 10 5 ,19 8 -9 9 , 208 , 212, 256n6; conversion as a way o f stopping hus � bands from dr in k ing , 204 , 206 ; e l im i � na t ion o f use, as p ar t o f H in d u iz a t ion o f D a l i ts , 13 9 ,14 0 ; as focus o f prayers in church , 212, 213-16 ; as necessity for slum men , 30, 85-86 , 88; not used by slum C hrist ia ns , 152 ,157 , 200 ; used by slum dwe llers, 60, 8 4 ,10 2

alis, (transvestites), 27-29 , 202-3 , 203fig, 256n 4, 263n 2

a llurem ent , 113—15 , 12 2 ,14 5 , 251-53 , 259ns- 6. See also fore ign money

A m b ed k ar , B . R . , 2 1-2 2 ,13 5 ,13 6 ,13 7 , 14 3 A n b u N agar (pseud.), xv ; caste composit ion

of, 56; l ife in , 34-4 5 ; m ean ing o f name, 71. See also slum dwe llers; slums

ann iyapasa i (fore ign tongue), 12, 219, 239 anpu (love), xv , 68, 71, 78, 232 anticonversion laws, 8 -9 ; a l legedly needed

to m a in ta in public order, 1 1 4 , 1 1 6 - 2 3 , 150 ,16 7 ; a llurem ent defined in , 259n6; asym m e try in , 1 1 5 - 1 6 , 1 2 3 , 1 3 4 , 14 5 , 1 4 7 ; depend upon argum ents deve loped by G a n d h i , 15 0 -5 1 , 16 7 ; fraud defined in , 2 5 9 -6 o n n ; genera lly render any conver � sions to C h r is t ia n i ty or Isla m il lega l, 113; slum dwellers at t itudes toward , 162; in T a m il N an du , 7 ,113 , 251-54 , 264n 4 ;

theologica l underpinn ings of, 115, 259nio

A n t i-U n tou c h ab il i ty League . See H ar i jan Seva k Sangh ( H SS)

apoca lypse , 231. See also m il len arianism apostasy, 158 ,18 3 ,19 8 -20 0 aravänis. See alis (transvestites) A ry a Samaj, 12 7 -2 8 ,13 0 ,13 1 ,13 5 , 26 im 6 auspic iousness: concept of, 8 3 ,10 5-6 ,

10 8 - 10 ,17 9 ; female, 82, 9 0 -9 1 ,10 4 , 10 5 , 212, 231, 246

authentic ity, not ion of: cultura l , 9 ,12 1 , 14 8 -4 9 ; G a n d h i as emblemat ic o f I n d ia ’s 2.6in2i; persona l, 1 1 9 , 1 2 0 - 1 ; re ligion as source o f 147 ; spir i tua l 7

autonomy, 5, 8 - 9 , 16 , 17 , 30, 246 ; D a l i t , 130, 138; na t iona l, 5. See also swaraj

A z ar ia h , V . S., 26 in 2 i

bac ksl id ing, 15 8 ,18 3 ,19 8 -20 0 beef, 34 , 59, 6 6 , 13 9 ,14 0 , 17 3 , 240 , 245 Berrem an , G era ld , 43 bh a k t i (religious devotion), 92 ,145 Bh ar t iya Ja n a ta Par ty (BJP) , 2 5 9 ^ -4 ,

260 m l Bib le , 16 4 ,18 7 , 226, 228, 241; meetings,

197; N e w T estament, 176 ; O ld T esta � ment , 17 0 , 17 6 -7 7 , 226 , 239; study groups, 16 4 ,19 3 , 210 ; verses of, 217, 2 2 0 -2 1 , 228

Bir la , G . D ., 131,138 B o d in ,Je a n , 116 Bourd ieu , Pierre, 25

2 7 9

brideprice, 104 Buta l ia , U rvash i , 12 4 , 12 5 ,14 2 , 2601112

care: contrasted w ith autonomy, 17, 246 ; he ightened perceptions re la ted to, 74-7 5 ; to be hum an is to be worthy of, 6, 78; im puted to fore igners, 17, 7 6 -7 7 , 24 7 -4 8 ; on ly “ the poor” do, 71, 75; “ the r ic h ” don ’t, 65-66 , 69, 7 8 -7 9 ; seen by slum dwellers as som e th ing basic to hum an beings, 6, 21, 75, 77, 79, 245-46 ; o f slum dwellers for the author 7 2-7 3 ; T a m il words for, 78

caste (jdt i): as complex soc ia l order, 62; as conceptua l template th rough w h ic h sin is understood , 2 22-24 ; de f in i t ion of, x i i i ; as denia l o f com mon hum an ity , 6, 64; as d ifference in t iny deta ils, 3; as an effect o f self ish ac t ions, 65; embarrass � ment about , 57; endogamy as defin it ive of, 70 ; idea o f “no caste in the ci ty,” 57; as inde lib le , 68; not expla in ed by ri tua l or re ligious prac t ices, 6 6 -6 7 ; not made by G od , 6 7 ,17 4 ; not seen as connec ted to H in d u re ligion , 67; not spec if ic a lly m entioned by slum C hrist ia ns , 2 23-24 ; not ta l k ed about in A n b u N agar , 45; Scheduled , x i i i -x iv ; soc ia l boycot t ing among most pow er fu l sanc t ions of, 7 8-7 9 ; as som e th ing tha t is elsewhere,

54- 55. 57 caste people ( jd t i m ak ka l), xiv, 60; ca lled

“ the r ic h ” by slum dwellers, 32, 63, 75; D a l i t stereotypes about , 6 7 -7 0

C a tholic ism , 176 census: C ensus o f Ind ia , 24 , 56 ,127 , 248 ,

25702; colon ia l 1 2 5 , 1 2 6 - 7 , IZ8; precolo � n ia l 126

céri (D a l i t ghetto), xiv, 33, 35, 53-4 , 58, 59, 224 , 256n8, 25702

C é t tus (N or th In d ia n businessmen), 94-95 , 96. See also paw ning; pawnshops

C h a krabar ty , D ip esh , 41 charism a , o f pastors, 186 -92 C henna i: discriminat ion in , 62; majori ty o f

slums are il legal in , 35; slums predomi � nantly occupied by Scheduled Castes, 52-53; theological t ra in ing centers in , 194

C h er ia n , D ivya , 126 chit fu nd , 9 9 - 10 0 , 10 4 , 216 C hrist ia n i ty , slum , 4 -5 ; “ ac t ion at a d is �

tanc e” in 11 , 206 , 212; acts o f fa i th in , 17 9 ,180 ; appea l to wom en of , 1 0 - 1 1 , 2 0 4 -6 , 2 1 1- 12 ; b e l ie f in and fe ar o f the same evil spir its as H in d us , 174 -7 5 ; churchgoing view ed as source o f k n ow l � edge , 217, 224 , 241; church names, 186; collec t iv i ty of, 202n4 ; conce ives sa lva � t ion as a world ly revolu t ion in wh ich caste is abolished , 5; crit ic ism o f H in d us in , 152; crit ique and transform a t ion o f exist ing soc ia l re la tions centra l to discourse of, 221; defined , x iv -xv ; d iffers from H in d u ism in tha t i t can transform re la t ionships, 233; d iffers from Pente � costa l C h r is t ia n i ty in o th er par ts o f the world , 170 ; d ist ingu ish es m agic and re ligion , 17 5 -7 7 ; ecstat ic prayer in , 238, 239; extension o f slum ethos o f care by, 2 4 2 -4 3 ; glossola lia in services of, 12, 207 , 219; God’s voice in, 2 19 -2 0 ; highly interactive church services of, 220; Hindu gods not seen as malevolent by, 173; holds that Hindu gods are “good” but don’t exist, 262^; ideal of loving, companionate marriage in, 235-37; identifies Christianity as victor, not victim, 226; identify own suffering with that of Christ, 231; idol worship an error, not a sin, 175 ,183 ; integrates slum as a whole, 11 , 232-33 , 247 ; interpreta� tion of Adam and Eve’s fall in, 4 , 236- 37; interpretation of divine speech in, 218 -20 ; maintains clear distinction between Hindu gods and demons, 17 4 -7 5 ; male converts frequently drift away from, 157 ,199 ; members concerned about getting saved from sins of others, 222; pastor-centric interpretation of, 18 5 ,18 9 -9 0 , 19 5 ,19 7 -9 8 , 201; prayer is for the benefit of all, 222, 225-26 ; prayer requests in, 210 , 2 13-16 ; prayer teams in, 1 0 - 1 1 , 2 10 - 1 1 ; presence of Holy Spirit in services of, 154 ,19 1 , 207 , 208 , 234 , 235; rumors of foreign money supporting, 13-15 ; shifts ultimate responsibility for

2 8 0 • I N D E X

household o f f wom en ’s shoulders, 234; slum H in d u crit ic ism of, 19 0 -9 1; soc ia l technology o f prayer in , 2 1 1- 1 2 ; spat io- tem pora l horizons of , 1 1 - 1 2 ; stric t m enta l d isc ip l ine required by, 181; test imonies in , 2 0 7 - 10 ; typ ic a l church in terior of, 14 ; v iew o f backsliders by, 183; v iew o f A n b u N agar men of, 152; views su ffering as transform a t ion ,

226-30 , 2 4 2 -4 3 ; vo lun tary prayers in , 207 ; wom en induc ted in to emotion a l se lf-regula t ion by, 231-33 ; women perform divin e pow er in , 20 6 -12 . See also B ible; conversion ; God /gods; Jesus; pastors; re ligion , world ly versus oth erworld ly; sin

d r (dowry), 10 3 -4 , i o 9 class, versus caste, 4 4 , 6 1-6 5 , 82. See also

caste (jdt i) com m unist party , 222 Congress Party , 3 5 ,12 8 ,13 1 ,13 5 ,138 C on ra d , D ie ter , 26m 22 C ons t i tu t ion o f Ind ia , 1 1 1 , 1 1 4 - 1 7 , 1 2 2 ,

135. 1 3 8 conversion : affec t on autonomy o f women

of , 2 4 6 -4 7 ; anthropologic a l v iew of, 120 ; forced , 1 1 4 , 1 1 5 , 1 2 1 ; as a form o f violence , 112 ; fraudulent , 123, 259- 6o n n ; G a n d h ia n discourse on , 8 ,116 , 1 3 0 , 1 3 1 , 1 3 3 , 1 3 4 , 1 4 2 , 1 4 4 - 5 1 ; idea tha t i t is soc ia l ly disruptive , 1 16 - 1 7 , 1 1 8 - 2 2 ; issue o f motive for, 12 2-2 3 , Z4 I > l aws aga inst , 7 ,113 -2 3 , 2 51-54 ; as a m a t ter o f v i ta l na t iona l concern , 8, m - 2 3 ; na t ion a l d iscourse on , 1 2 2 - 2 3 , 15° : or igin o f H in d u opposit ion to, 134 ; serves to un ite slum com m unity , 11. See also anticonversion laws

cow, 139 ,146 . See also b e e f culture , concept o f 26; not ion o f a nat iona l,

xiv, 7 ,16 , 26, 9 0 , 12 1 ,1 3 2 , 14 8 ; re ligion as, 8, 9 - 1 0 , 1 1 9 , 1 2 0 , 1 4 8 . See also authen � tic ity, not ion of; re la t ivism , cultura l

D a l i ts (untouchables; Paraiyars): access to w a ter con trolled by castes, 4 2-4 5 ; charac terize non-D a l i t castes (“ the r ic h ” ) as rejec t ing fe l low hum an ity of,

65-66 , 69, 75, 222, 225; c lassif icat ion o f is race like , 3; consumption o f b e e f by, 66 ,139 ; def in i t ion of, x i i i , 2-3 ; division from non-D a l i ts premised upon system � at ic dehum aniz a t ion , 3; do not ob jec t to conversion , 26in i5; early H indu iz a t ion attempts of, 13 0 -3 1 ; most live in slum areas, 2 5 7 ^ ; not considered H in d u u n t i l tw entie th century , 125-26 ,128 , 129 , 26o n i4 ; not targets o f H in d u / M us l im violence during part i t ion because they belonged to neither group , 124 ; or igin myths of, 64 ; as outsiders, 3; percentage in A n b u N agar of, 56; per � centage tha t are C h r ist ia n , 258m ; recorded as H in d u by Bri t ish census offic ia ls, 12 6 -27 ; rura l to urban migra � t ion of, 55-56; se lf-abnegat ion required of, 4 2 -4 3 ; se lf-def in i t ion of , 5-6 ; stere � o types o f “ the r ic h ” of, 6 7 -7 0 . See also C h rist ia n i ty , slum ; mora li ty , slum; re ligion , slum ; slum dwellers

D a t ta , Pradip K um ar , 129 debt: as form o f investment or savings, 83;

instrumenta l in reproduct ion o f moral community , 109; o f sources of, 93-94 ; two types of, 9 4-9 5 ,109 ; is widespread, but not cause o f poverty, 93. See also loans

demographic m ajori tarianism , 8 ,128 -2 9 D erre t t , J . D u n c a n M . , 150 D irks , N icholas, 57 D ravid ian ist : ideology , 67; poli t ica l

part ies, 32 D urk h e im , E m ile , 116 , 1 18 - 1 9 , 12 1 , 262n4

elites, xiv, 16 ,18 , 31, 57; discourse on religion of, 1 1 4 - 1 2 3 , 1 3 2 , 1 4 4 ; discourse on slums of, 93; urban 4 7 ,54 ,57 . See also m iddle class

endogamy, 70 ethnographic knowledge: bod ily presence

in re la t ion to 24-25 ; 33-34 ; contrasted to mere observa t ion 245; field notes as a pow erfu l technology in , 25-26 ; issue o f discrepancies expressed by in form ants, 31-32 ; m e thodologica l re la t ion to care, 24 7 -48 ; outsider status o f researcher in re la t ion to , 2 4-2 7

i n d e x • 2 8 1

Evans-Pri tchard , E . E . , 91 evil, supern a tura l (vankandvi or cey-

vinaiydvi), 4 1 ; assumed to come prim a � r i ly from fema le envy, 105; care repro � duced and m a in ta in ed by ta l k of, 108; as dark underside o f care, 7; pastors strug � gle w ith , 188-89

Fern andes, Lee la , x iv f i l th , 5 3 , 1 4 0 . See also garbage fore igners: credited by slum dwe llers as

c ar ing for others, 6 , 1 7 , 1 8 , u - z z , 7 6 - 7 7 , 1 4 7 - 4 8 ; money from , 5 , 1 3 - 1 5 , 1 8 - 1 8 , Z O - Z I , 4 5 , 7 6 , I I I , 1 5 Z

Fuller , C hristoph er , 1 6 5

G a i t , E . A . , i z 8 , 119 G a lan ter , M arc , 150 G a n d h i , Ind ira , 15-16 G a n d h i , M . K .: anticonversion laws depend

on argum ents deve loped by, 151; a t tempt to change H in d u att itudes about untouchables of, 1 4 0 - 4 Z ; c la im ed to represent D a l i ts , 1 3 5 - 3 6 ; conversion o f un touchables as threa t to nat ion , 1 4 1 ; created sa in t ly image o f h im se lf, z 6 in i9 ; on H in d u ism as most tolerant o f al l re ligions, 1 4 8 ; identif ied H in d u in terest w ith tha t o f In d ia n na t ion 1 4 1 , 1 4 4 ; l in k ed persona l autonomy to n a t iona l autonomy, 1 7 ; low opin ion o f un touch a � bles of, 1 4 5 - 4 6 ; m ission o f ending un touch ab il i ty of, 1 3 0 , 1 3 1 ; opposi t ion to conversion of, 1 3 4 , 1 5 0 ; on prose ly t ism as form o f in tolerance , 1 4 7 , 1 4 8 - 4 9 ; redef in ed D a l i ts as H in d us by de f in i � t ion , 1 3 9 - 4 0 , 1 4 Z , 1 4 4 , 1 5 0 ; suic ide threa t by, 1 3 6 ; on un touc h ab il i ty and the surviva l o f H in d u ism , 1 4 3 ; on un touch abil i ty as a poison , 143; v ie w o f M usl ims of, 1 4 3 - 4 4 ; world ly in terests in worship seen as in fer ior to spir i tua l ones by, 1 6 5

G a n d h i , Sonia , n z garbage , 5 , 3 4 - 4 1 , 4 8 - 4 9 , 1 9 7 , Z Z 4 - Z 5 , Z 3 8 gift , M aussian , 7 3 God /gods: anthropologic a l v ie w of, 160 ;

def in i t ion of, xv; d ifference between

H in d u and C h rist ia n , 1 1 7 - 1 9 , z zo ; G a n d h i on , 1 4 9 ; as in tr ins ic a l ly mora l, w h e ther th ey exist or not , 1 5 7 , 1 6 4 , 1 7 Z , 1 7 5 , 1 8 3 - 8 4 ; “Puran ic ,” 1 7 1 - 7 3 ; re la t ion � ship between hum ans and gods is in tr ins ic a l ly d isc ip l in ary and reform a � t ive, 1 6 6 ; slum H in d us speak o f a l l gods as d i fferen t form o f G o d , 1 7 7 , 1 8 3 . See also Jesus

gold , 1 0 3 - 4 , 1 0 9 , n o , Z 3 4 guna. See kunarn (inherent nature) G upta , D ipan k ar , 3

H a r i ja n Sev a kSa ngh ( H SS) , 1 3 4 , 1 3 8 , 1 4 0 H arper , Susan B il l ington , z 6 i n n H e l ler , Pa tr ic k , x iv H erz fe ld , M ich a e l , z63n6 hijras, Z56n4. See a lso alis H in d us / H in d u ism : an im a l sacri fice by,

i6 i -6 z ; at tacks on C h rist ia ns by are rea l ly at tacks on D a l i ts , z 64 n 4 ; chau � vin ism , 1 4 4 ; consulta t ion o f m ediums by, 118 ; convers ion by shuddhi to, 131, i6 i n i6 ; D a l i t conversion as threa t to, 8, 1 3 3 - 3 4 , z 6 o -6 i n is ; defined in Ind ian law, 1 5 0 ; d ivin e possession in , Z 1 7 - 1 9 ; effec t o f vow bre a k ing to, i8z ; hook- sw inging by, 60 , Z57n4; m ajoritarian- ism , 8 , 1 1 4 - 1 5 , 1 1 9 , 1 1 5 ; persona l roster o f deities shif ts th rough t im e based upon tangible evidence , 1 7 7 - 7 8 , 1 8 1 - 8 3 ; r i tua l tests perform ed by, 1 7 9 ; sa id to be inh erent ly tolerant , 13 1 ; slum dwellers take d ifferent v ie w o f m ora l i ty and gods of, 1 7 1 , 1 7 1 , 1 7 3 ; standard in terpre ta t ion o f gods as m ora l ly ambiguous, 1 5 6 - 5 7 , 1 7 0 - 7 1 ; un iversa list m ora l i ty a t tr ibu ted to by slum dwe llers, 1 7 3 ; world ly in ter � ests in worship seen as in fer ior to spir � itua l ones in , 1 6 5 . See also under D a l i ts

H in d u na t ion a l ism : considers D a l i ts H in d u by def in i t ion , 1 5 0 ; “ fore ign in f luenc e” defined by, 16; need to incor � porate D a l i ts as stra tegic imperat ive for, 1 1 9 ; view o f re ligious convers ion of, 1 1 1 - 1 3 , 1 1 7 , 1 3 1 . See also G a n d h i , M . K .; Saraswa t i , Sw am i D ayan and a; Shrad- dh an and , Sw am i

1 8 1 • I N D E X

H in d u tva , 1 1 4 , 1 1 7 , 2.59114, z6inzz, 2 6404 H indutva (Savarkar) , 150 H o ly Sp iri t . See under C hrist ia n i ty , slum H um e , D avid , 200

identity: commonplace to see religion as persona l and collec tive, 119 -2 2 ,12 3 ; consumption o f b e e f as marker d ist in � gu ish ing D a l i ts from H indus, 139; G a n � d h i on , 13 4 ,14 7 -4 9 ; slum C hrist ians id en t ify themselves w ith C hrist , 226, 228; to slum dwellers religion is not a m at ter of , 9 -10 ,15 3 ,15 4 , 237; slum dwell � ers use class, not status, to define, 63

Ind ia : agrarian slavery in Bri t ish , 2; housing segregat ion acceptable in , 2 5 7 ^ ; im por � tance o f E nglish in na t iona l d iscourse in , 118 ; m iddle class in , xiv, 47; state re ac t ion to re ligious violence in , 2 6 1- 62m ; trope o f “ fore ign h a n d ” in poli t ics of, 15 -17

Iyotheedas, Pandit , 133-34

jd t i. See caste (jdti) Jesus: came out o f h is caste, 226 , 230; cares

for poor , 18 ,152 ,16 9 , 204 ; compared w i th K a r l M a rx by pastor, 222; d is � m issed by H in d us as we a k god , 9 ,152 , 154 ; viewed as spec ia l i z ing in “wom en ’s issues,” 177 , 20 3-6 , 208-9

jew e lry: frow n ed upon by slum C hrist ia ns , 152, 200 , 233, 234-35 , 238-39 ; as savings, 103; as wom en ’s property , 89 ,100

K andasamy , M een a , 256n8 K avira j , Sudipta , 4 0 - 4 1 , 53 kirdmattdr (villagers), 224 . See also caste

people kunam (guna; inherent nature): o f fore ign �

ers, 7 6 -7 7 ; o f “ the r ic h ” (caste people), 63, 68, 69, 76; o f wom en , 92

K uppusw a m i , A l la d i , 123

Lee , Jo e l , 7 5 ,12 5 ,13 0 ,13 8 , 26on i2 loans: fights over, 96-98 ; mora l force o f

repaying, 98-99 ; pawnshop vs. inslum , 9 4 -9 5 ; as soc ia l prac t ic e for creat ing solid ar i ty among wom en , 96-97 ,

9 9 -10 1 ; taken out even when not needed, 96. See also debt

M adras, 2, 53, 55-56 ,126 , 228. See also A n b u N agar; C h en n a i; T a m il N adu

mariydtai (respect), 23, 255n2 m arriage: arranged , 7 0 , 10 4 ; endogamy in ,

70 ; fin an c ia l troubles greatest treat to, 99; love, 7 0 , 10 4 ; as mora l fault line in wom en ’s lives, 6 -7 ; slum churches seek to improve , 233-37 ; supern atura l a t tack most com mon cause o f trouble in , 105. See also wom en

M a rx , K ar l , 222 men: dependence o f women on wages of,

82, 87-88 ; descript ion o f the ir work by, 85-87 ; d r in k as necessity for , 30, 85-86 , 88; employed as “coolies,” 84; view o f C h r is t ia n i ty of, 152 ,185; work in juries of, 86. See also pastors

M enon , N ived i ta , 2 5 9 ^ , 260 m l m iddle class, xiv, 4 0 - 4 1 , 4 7 ,118 ; discourse

on slums of, 4 8 -5 1 ,53 , 87, 88, 92, 93, 256n6, 257m ; on religious conversion , 123.

m il len arian ism , 12, 230 , 231, 238, 239, 242 mockery: o f author by slum dwellers, 23, 41;

o f C h rist ians by H in d us 152; 19 0 -9 1 ; o f H in d us by C hrist ians 152; o f jos e p h 236; o f m a instream C hrist ia ns by slum C hrist ians, 176 ; as pervasive feature o f slum l ife , 23 ,154 ; o f slum dwellers by o ther slum dwellers 4 1 ,18 0

mora li ty, slum : erroneous middle-c lass ideas about , 49; extended by slum C hrist ians, 24 2 -4 3 , 2 4 6 -4 7 ; prim it ive un iversa lism of , 16 9 -7 0 ; re lations betw een lender-debtor and husband- w if e viola te core precept of, 6 , 10 4 -5 , 212, 246 ; reproduc t ion and m a in te � nance of, 10 8 -10 ; view ed as un iversa l proper ty tha t does not vary by re ligion , 9 - 10 , 15 6 -5 7 , 16 9 ,17 0 , 242. See also care

M osse , D avid , 264n 4 M uk er j i , U . N . , 128-2 9 ,138

N abokov , Isabe lle , 263m N iyogi Com m iss ion Report , 122 ,150

I N D E X • 2 8 3

nongovernm enta l organiza t ions (N G O s) , 248; d iscourse on urban poor of, 4 7 -4 8 , 5 0 , 5 4 , 117 ; employees considered elites, 31 ,32 ; por tray poor wom en as u n iform ly opposed to a lcohol , %s6n6; report sexua l harassment o f D a l i t wom en by caste men , 55

O ’ M a l ley , L . S. S., 171

p a k t i (bhak t i; re ligious devotion) , 92 ,152 . See also bh ak t i

Pandey, Gyan , 126 Pandian , M . S. S., 257m Pandyan , Pu n ith a , 26404 pariah , xi i i , 1- 2 , 54 , 228, 229-30 , 238, 245,

25702; “ Par iah C h rist ia ns ,” 12 6 ,13 4 ; “ Par iah M oh am m adans,” 12 6 ,13 4

Para iyars. See D a l i ts Parry , Jon a th a n , 258n6 Pasca l, Bla ise , 14 0 pastors, 185; ab il i ty to take wom en ’s poin t

o f view of, 236-37 ; ambiguous language used by, 12; ba c ks l id ing as a m ajor worry of, 19 8-20 0 ; church as extension of, 186-87 ; com pe t it ion among, 195-97 ; d ifference in ma le and fema le views of, 185-86 ; as d iff icu lt pa th to follow , 195; H in d u view of, 190 ; im portanc e o f charism a of, 186 -9 2 ; as independent opera tors, 194 ; in tent scru t iny o f by residents, 20; m a k ing o f a, 19 2-9 4 ; negat ive view of, 19; pastor-hopping as m ajor w orry of, 2 0 1-2 ; question o f charisma of, 19 6-9 2 ; quest ion o f mon � e tary gifts from , 19 -2 0 ; rumors about receipt o f fore ign money by, 13-15 , 20; struggle w ith evi l spirits of, 188-89 ; su ffered physic a l ly for th e ir flocks, 18 8 -8 9 ; two spa t iotempora l horizons evoked by, 1 1 - 1 2

paw ning, 8 9 ,10 3 -4 , J°9> 2 6 2 ^ pawnshops, 9 4 -9 5 , 10 1 , 10 3 , 10 9 , 1 10 .

See also C e t tus Peabody, N orber t , 126 Pentecosta lism , 4 . See also C h rist ia n i ty ,

slum ; pastors peru m a i (pride), 68-69 , ^ n x

“poor , the .” See slum dwe llers; D a l i ts pura kk anippu (social boycott), 45-46 ,

78-7 9

Raghava iyangar , S. Srin ivasa , 133 Ra ja h , M . C . , 136 -37 Ram asa m y Periyar , E . V ., 21 Rash tr iya Sw ayamsevak Sangh (RSS) ,

259n 4 re la t ivism , cu ltura l , 14 8 ,15 6 ,16 0 , 16 7 , 16 9 re ligion , slum: afterlife not telos of, 166;

basic un it of, 15 6 ,16 2 -6 3 ,16 4 ; C h r is t i � an i ty and H in d u ism m ora l ly identical in , 170 ; constant movement across and w ith in tradit ions in , 157-58 ; defin it ion of, 155; di fferences be tween H in d u and C hrist ia n worship in , 16 3 -6 4 ,16 8 ; as d ivin e ly orchestrated systems o f reward and pun ishm ent , 242; does not va lorize pure ly spir i tua l worship , 165; fun c t ion a l � ist mode l of, 116 ; as in trinsic a l ly moral prac tice , 183; keen observat ion o f omens and signs in , 17 9 -8 1 ; mora l universa lism of, 156 ,16 9 -7 0 ; is not a m at ter o f iden � tity , 9 - 10 ,15 3 ,15 4 ,2 3 7 ; outside money viewed positive ly, 17 - 19 ; persona l dependenc e on gods in , 17 8 -7 9 ,18 1-8 2 ; as source o f soc ia l cohesion , 118-2 2 ; stress on rules in , 17 8 -7 9 ; supern atura l evil in , 7; theologica l rea lism of, 10, 158-62; two fundam enta l ways o f th in k � ing about , 16 0 -6 1 ; women are pu n c t i l i � ous in observanc es of, 92; world ly inter � ests paramount in , 156 ,16 5-6 7 ,168 , 233. See also conversion ; mora li ty, slum

re ligion , world ly versus oth erworld ly: asym m e try o f anticonversion discourse favoring otherworld ly , 1 1 5 , 12 3 ,13 4 , 14 5 , 147 ; G a n d h i on , 17 ,18 ; world ly interest o f slum , 8, 9 ,19 , 6 7 ,15 6 , 16 2 -6 7 , !68, 218 , 220 , 233, 234

re ligious conversion . See conversion Rev. Sta in islaus v. Sta te o f M ad hya Pradesh

and O rissa , 116 , 259ns revolu t ion , 5 ,12 , 2 21-2 2 , 226 , 230, 234 “rich , the .” See caste people; m iddle c lass r i tua l dom in a t ion , 43 Rou n d T able Con feren c e , 135 ,138 ,14 3

2 8 4 • I N D E X

Sangh Parivar , 1 14 , 1 4 4 , 1 5 9 1 1 4 Sarkar , Sum it , 112 Saraswa t i , Sw am i D ayan and a , 1 1 2 , 1 1 9 - 2 0 ,

13 1-3 2 , 16 0 Savark ar, V . D ., 150 Sawaswat i, Jayendra , 116 scold ing, 29-30 , 256115 Scot t , Ja m es C . , 31 Sear le-C h a t ter jee , M ary , 124-25 Sh arm a , A rv in d , 2601114 Shraddh an an d , Swami , 130 ,13 1 ,13 8 ,13 9 ,

1 4 0 - 4 1 sin , 222; be ing “saved from ,” 222; caste

view ed as a, 222-23 ; defined as harm done to others, 164 ; idol worship is an error, not a, 175 ,183

Sk inn er , Q uent in , 258n6 slavery, 2 ,18 , 60 , 237 slum C hrist ia n i ty . See C h rist ia n i ty , slum slum dwellers: a lleged la c k o f persona l

hygiene of, 4 8 -4 9 ; are acutely aware o f o th er ’s su ffering, 7 2 -7 4 ; asym m e try in d iscourse about nonslum dwellers, 32-34 ; care for one another because th ey do not prac t ic e caste, 79; centra l i ty o f concept o f care to, 78, 79; charac ter � ize themse lves as warm and affec t ion a te towards a l l people , 71, 72; c la im fu n d a � m enta l equa l i ty and va lue o f a l l persons, 77 ; conceptua li ze resource monopoliza � t ion and dom in a t ion as essence o f caste, 65; deny tha t they possess any inheri ted tra i ts, 75, 77 ; fee l most Ind ians do not care for them , 38 -4 1 , 4 3 -4 4 , 45-46 , 66; hum an ism of, 24 2 -4 3 ; imperat ive to share as basic axiom in , 21; in tent scru � t iny o f pastors by, 20; k in d o f class d ivision between men and women among, 82; lie about or downplay th e ir w ea lth , 10 7 -8 ; m any never experience r i tua l un touchabil i ty , 55; mora l com m u � n i ty among, 81; negat ive view o f po l i t i � cians of, 18; no clear d ichotomy in economic terms between nonslum dwellers and , 34; not ions o f ci t izenship of, 38, 4 1; offic ia l ignorance of , 92; posit ive image o f fore igners among, 6, 17 ,18 , 2 1-2 2 , 7 6 -7 7 , 24 7 -48 ; prefer to

describe themselves as “ the poor ,” 63, 64; refer to caste people as “ the rich ,” 32, 63, 75; reject endogamy, 70 ; reluctanc e to ta l k about the ir own caste status, 45, 54 , 58; supposed lavishness o f l i fe cycle r i tua ls of, 4 9 -50 ; supposed violent crim e among, 51-52; trash as potent symbol o f condit ion of, 40 ; treatment o f transvest i tes by, 28-29 ; uncanny abil i ty to “see” pa in and discom fort , 74-75 ; urban m iddle class stereotypes about, 4 7 -5 4 ; v iew o f form er vi l lage l ife of, 55, 60. See also care; C hrist ian ity , slum; D a l i ts; mora li ty , slum; re ligion , slum

slum hum an ism , 24 2-4 3 slums, urban : fires in , 35; mora l ethos of,

6 -7 ; mora l fault lines w ith in , 6 -7 ; no longer referred to as ceris, 58, 2 5 7 ^ ; st i l l referred to as ceris 53, 54 ,58-59 ; sup � posed violent crim e in , 51-52 ; trash in , 3 7 -4 1 , 224 ; wa ter in , 4 1-4 5

Sta ines, G ladys, 112 -13 , 238n2 Sta ines, G rah a m , 112 -13 , 258n2 suicide: com mon in slums, 81, 258m ; danger

o f spirits o f female, 105; debts and m ari ta l str ife as fac tors in , 83, 99; G a n � d h i threatens to com mit , 136 -37 ; “ the r ic h ” drive the ir c hildren to, 70

Supreme C o u r t o f Ind ia , 1 1 2 - 1 3 , 1 1 4 , 116 - 18 , 12 1- 2 2 , 1 5 0 , 258n2, 259n8

swaraj (self-rule), 17 ,131 syncret ism , 9

T a m il N adu : anticonversion ordin ance of, 7 ,113 , 251-54 , 264n 4 ; law bann ing an im a l sacrifice in , 161; mar iya ta i (respect) shown to outsiders in , 23; men and wom en lead very separate lives in , 27; slum dwe llers’ op in ion of, 32

T a m il N adu Slum C le aranc e Board ( T N S C B ) , 47, 4 9 ,5 0 -5 2 ,9 2

theologica l rea lism , 10 ,15 8 -6 2 tittu or t ittutal (scolding) , 29-30 , 256ns transvest ites. See alis (transvestites) trash . See garbage

U m bre l la Preacher, 228-30 un touchables. See D a l i ts

i n d e x • 2 8 5

van der Veer, Peter, 127 vankandvi. See evil, supern a tura l V ishw a H in d u Par ish ad ( V H P) , 259114 V isw an an a th an , G auri , 12 0 -2 1 V isw an a th , Rupa , 1, 2 , 12 6 , 12 7 Vyas, N een a , 120

W eber, M ax , 191 wom en : auspic iousness of, as source o f and

responsib ili ty for household we ll-be ing, 82, 90 , 9 1 ,10 5 ,10 6 ,10 8 , 212, 246 ; bla m ed for th ings th ey have no control over, 88, 105; cash surpluses of, loaned to keep out o fh a n ds o f husbands and friends, 10 1-2 ; condemn vic t im o f spouse neglec t, 90; debt as a way o f art icu la t ing friendship between , 82; debts and m ari ta l str ife as fac tors in suic ide of, 83, 99; dependenc e on men ’s wages among, 82, 87-89 ; have

less control o f lives, thus more devoted to re ligion , 92; induc ted in to prac t ic e o f emotion a l se lf-regula t ion by slum C h r is � t ianity, 232; involved in f inanc ia l schemes, 9 9 -10 1 ; lac k o f oppor tu n i ty for employment in A n b u N agar of, 83-84 , 91; nega tive view o f men of, 88-90 ; outsiders tend to emphasi ze vic t im hood of, 256n6; police one anoth er ’s behavior, 29-30 ; poor wom en unable to par t ic i � pa te in mon e tary schemes that ar t icu � late fema le soc ia l ity, 91; tend to st ick w ith C h r is t more t h an any oth er god , 157; urban stereotypes about slum , 51-52; use o f chit fu n d scheme by, 9 9 -10 1 ; use o f gold jew e lry as savings by, 103; v iew world as cauldron o f tempta � t ions, 87. See also C h rist ian i ty , slum ; m arr iage

2 8 6 • I N D E X

T H E A N T H R O P O L O G Y O F C H R I S T I A N I T Y

E d ited by Jo e l Robbins

1. Christian M oderns: Freedom a n d Fetish in the M ission Encounter, by Webb Keane

2. A Problem o f Presence: Beyond Scripture in an A frican Church, by Matthew Engelke

3. Reason to Believe: C u ltu ra l Agency in L a t in American Evangelica lism , by David Smilde

4. C hanting D own the N ew Jerusa lem : Calypso, Christianity, a n d C apitalism in the C aribbean, by Francio Guadeloupe

5. In G od’s Im age: The M etaculture o f F ijia n C hristianity, by Matt Tomlinson

6. Converting Words: M aya in the Age o f the Cross, by William F. Hanks

7. C ity o f God: C hristian C itizenship in Postwar Guatem ala, by Kevin O ’Neill

8. D eath in a Church o f L ife : M ora l Passion du ring Botswana’s T ime o f A I D S , by Frederick Klaits

9. Eastern Christians in A nthropological Perspective, edited by Chris Hann and Hermann Goltz

10. Studying G loba l Pentecostalism: Theories a n d Methods, by Allan Anderson, Michael Bergunder, Andre Droogers, and Cornelis van der Laan

11 . H oly H ustlers, Schism, a n d Prophecy: Apostolic Reform ation in Botswana, by Richard Werbner

12. M ora l A m bition: M obiliz a tion a n d Socia l Outreach in Evangelica l Megachurches, by Omri Elisha

13. Spirits o f Protestantism: M edicine, H ealing, a n d L ib e r a l Christianity, by Pamela E. Klassen

14. The Sa in t in the Banyan Tree: C hristian ity a n d Caste Society in In d ia , by David Mosse

15. God ’s Agents: B ib lic a l Publicity in Contemporary E ngland, by Matthew Engelke

16. C ritic a l C hristian ity: Translation a nd D enom in ation al Conflict in Papua N ew Guinea, by Courtney Handman

17. Sensational M ovies: Video, Vision, a nd C hristian ity in Ghana, by Birgit Meyer

18. C hristianity, Islam , a n d O risa-Religion: Three Traditions in Comparison a n d Interaction, by J. D. Y. Peel

19. Praying a n d Preying: C hristian ity in Indigenous A m a zon ia, by Aparecida Vila^a

20. To B e C ared For: The Power o f Conversion a n d Foreignness o f Belonging in an Ind ia n Slum , by Nathaniel Roberts

To Be Cored For offers a un ique vie w into the conc e p tu a l and moral world o f slum-bound Da lits (“ un touch ab les”) in the South Indian city of Chenna i. Fo cusingon th e d e c ision by many women t o embrace loca l ly spec if ic forms of Pentecosta l Christ ian ity, Nathan ie l Roberts cha llenges dom inant a n th ro p o lo g ic a l u n d e rst a n d ings of re ligion as a m a t t er of cu lture and ident ity, as we ll as Ind ian na t iona list narrat ives of Christ ian ity as a “fore ign ” id eo logy that d isrup ts loca l commun it ies. Far from be ing a d ivisive force, conversion integrates the slum com m un ity—Christ ians and H indus a lik e— by addressing hidden moral fault lines that subt ly pit residents aga inst one another in a na t iona l context that renders Dalits outsiders in the ir own land.

“ To Be Cared For is a rich ly layered crit iqu e of e lite d iscourses and anxie t ies about caste and conversion in India and a m ov ing e thnography of re ligious pract ices and mora lity am ong the

profound ly d ispossesse d . At once about caste , gender, hunger, injust ice , and c ar ing, this be au � t ifu lly written book carries an an a lyt ic a l heft rare ly seen in such grounded e thnograph ies.”

- R A K A R A Y , PR O FESSO R OF SOCIOLO GY AND SO UTH AND SO UTHEAST ASIA STUDIES , UNIVERSITY OF CALIFORNIA , BERKELEY

“Th is book tak es a huge ly comp lex sub ject and gives us new po ints of departure . It uses p a ins �

t a k ing e thnograph ic research to show how an thropo log ica l appro ach es to re ligion , ident ity, and conversion need to ch ange . It h e lps the reader re th ink re lig ion , cu lture , and tru th—

m atters urgent to peop le fac ing d iscr im ina t ion , poverty, and acute uncerta inty.”

- D A V I D M O S S E , PRO FESSO R OF ANTHROPOLOGY, SCH O O L OF AFRICAN AND ORIENTAL STUDIES , UNIVERSITY OF LONDON

“Th is remarkable e thnography te lls a fascinat ing story with crit ica l sk ill and compassion. Roberts gives us a comp licated picture of moral contrad ict ions and re ligious insights. Th is book is essen �

t ia l for anthropo logists and others interested in the roles of re ligion in the modern world.”

- T A L A L A S A D , DISTINGUISHED PR O FESSO R OF ANTHROPOLOGY, THE GRADUATE CENTER , CITY UNIVERSITY OF NEW YORK

N a than ie l Roberts is Research Fe llow at the Max P lanck Inst itute for the Study of Re ligious and Ethn ic Diversity, Germany.

A Ph ilip E. Li l ie n th a l Book in Asian Stud ies | The Anthropo logy o f Christ ian ity , 20

UNIVERSITY OF CALIFORNIA PRESS w w w .ucpress.edu Co v e r d esign by Dom in i Dragoon e Co v e r p h o t ogr a p h by N a t h a n ie l Ro b e rts

I S B N T 7 ñ - 0 - 5 2 0 - 5 f l ñ a 5 - T

9 7 8 0 5 2 0 2 8 8 8 2 9