Tamil Brahmans: The Making of a Middle-Class Caste 9780226152882

Citation preview

Tamil Brahmans

Tamil Brahmans The Making of a Middle-Class Caste

C. J. Fuller H a r i p r i y a N a r a s i mh a n

The University of Chicago Press Chicago and London

C. J. Fuller is emeritus professor of anthropology at the London School of Economics. He is the author of several books, including The Camphor Flame and The Renewal of the Priesthood. Haripriya Narasimhan is assistant professor of social anthropology and sociology at the Indian Institute of Technology Hyderabad. The University of Chicago Press, Chicago 60637 The University of Chicago Press, Ltd., London © 2014 by The University of Chicago All rights reserved. Published 2014. Printed in the United States of America 23  22  21  20  19  18  17  16  15  14   1  2  3  4  5 ISBN-13: 978–0-226–15260–8 (cloth) ISBN-13: 978–0-226–15274–5 (paper) ISBN-13: 978–0-226–15288–2 (e-book) DOI: 10.7208/chicago/9780226152882.001.0001 Library of Congress Cataloging-in-Publication Data Fuller, C. J. (Christopher John), 1949– author.   Tamil brahmans : the making of a middle-class caste / C. J. Fuller and Haripriya Narasimhan.     pages ; cm  Includes bibliographical references and index.  ISBN 978-0-226-15260-8 (cloth : alk. paper) — ISBN 978-0-226-15274-5 (pbk. : alk. paper) — ISBN 978-0-226-15288-2 (e-book)  1. Brahmans—India—Tamil Nadu— Social conditions.  2. Tamil (Indic people)—India—Tamil Nadu—Social conditions.  3. Middle class—India—Tamil Nadu.  4. Caste—India—Tamil Nadu.  5. Tamil Nadu (India)—Social conditions. I. Narasimhan, Haripriya, author. II. Title.   DS432.B73F85 2014   305.894’8110548208622—dc23 2013046104 a This paper meets the requirements of ANSI/NISO Z39.48–1992 (Permanence of Paper).

C ont e nt s

Acknowledgments / vii Note on Transliteration / ix

Introduction / 1 ONE

/ The Village: Caste, Land, and Emigration to the City / 29

TWO

/ Education and Employment in the Colonial Period / 61

T H REE

/ Education and Employment after Independence / 89 F O UR

/ The Changing Position of Women / 123 FIVE

SI X

SE V E N

/ Urban Ways of Life / 153

/ Religion, Music, and Dance / 183

/ Tamil Brahmans as a Middle-Class Caste / 211

APPE N D I X

/ Tamil Brahman Demographics / 231 Glossary / 243 Notes / 245 Bibliography / 259 Index / 273

Ac k now l e d g m e nt s

Since 2003, when we began the research on which this book is based, countless people have given us information and guidance, discussed our work with us, and commented on our writings. Unfortunately, it is impossible to thank them all here and we apologize to those whose names are omitted, but there is still a long list of people to whom we must express our gratitude. First and foremost, we are grateful to all the people who told us about themselves, their families, their communities, and their life and work, in the course of our research on the middle class in Chennai in 2003–5 and on Tamil Brahmans, particularly members of the Eighteen-Village Vattima community, in Chennai, the village of Tippirajapuram, and elsewhere in 2005–8. In this book, all our informants have been given pseudonyms, but we can still thank those who have been especially helpful to us. They include the late Rajagopal Natarajan, the late Jayam Vasudevan, the late Komalavalli Vedantachari, T. Vedanthachari and S. Parthasarathy, and V. L. Vijayaraghavan; T. Kannan Jagan and Indira Jagan, R. Ramkumar, P. Ramalingam, the late V. M. Ravikumar, Akila Satheesh, and Srikant Srinivasan; A. V. Ramamurthy and Thayyu Ramamurthy and their extended families, A. V. Swaminathan and Mangalam Swaminathan, N. Gopalaswamy, V. Pichu, Subbu Ramamurthy, Santha Raman, and Jayam Venkatraman. We owe special thanks for their help and hospitality to J. Pari, Akila Santhanam, and K. S. Sasisekaran. For their comments, criticisms, advice, and encouragement, we thank those who read an earlier draft of this book: Isabelle Clark-Decès, John Harriss (who also worked with us in Chennai in 2003–5), Mekhala Krishnamurthy, Johnny Parry (whose criticisms were characteristically acute), Nandini Sundar, and David Washbrook, as well as Mattison Mines and

viii / Acknowledgments

Sylvia Vatuk, who read the manuscript for the University of Chicago Press. For valuable conversations about Tamil Brahmans, we also thank S. Anandhi, Pushpa Arabindoo, Indira Arumugam, Janaki Bakhle, Mukulika Banerjee, Véronique Benei, André Béteille, Maurice Bloch, Nick Dirks, Henrike Donner, Peggy Froerer, Ramachandra Guha, Eugene Irschick, M. A. Kalam, David Ludden, David Mosse, V. K. Natraj, M. S. S. Pandian, Indira Peterson, Gyan Prakash, Arvind Rajagopal, Bhavani Raman, Velcheru Narayana Rao, the late Marie-Louise Reiniche, Charles Stafford, Sanjay Subrahmanyam, Carol Upadhya, Peter van der Veer, Rupa Viswanath, and Susan Wadley. We have benefited, too, from comments on our work at seminars and conferences in Britain, Canada, Germany, India, the Netherlands, and especially the United States. We also owe a general debt of gratitude to our colleagues in the London School of Economics for critical encouragement over many years, and thank the Asian College of Journalism, Chennai, where Haripriya was working at the time, for giving us space and time to discuss this book in January 2012. We are grateful for the support of David Brent and Priya Nelson, and the scrupulous copyediting of Susan Karani, at the University of Chicago Press. We thank Sanjeev Routray for research assistance in the University of British Columbia Archives, and Mina Moshkeri for preparing the maps in this book. Thanks are due to the Economic and Social Research Council, which financially supported all our research, and to the LSE for supplementary assistance. Last but not least, we owe a debt of gratitude to our families: Penny Logan and Alexis Fuller, and Amma, Appa, Nikhil, Raman, and Subha. Haripriya hopes Sunetra will read this book some day.

N ot e on T r a n s l i t e r a t i on

All Indian terms are systematically transliterated with diacritics and italicized. The proper names of Hindu deities, offices, and institutions, however, appear with diacritics only on first appearance and thereafter their spellings are adjusted in the usual way (e.g., Shiva for S´iva). Many Tamil terms in this book are Tamilized Sanskrit and are transliterated accordingly, instead of from the pure Tamil forms that only specialists would recognize (e.g., agraha¯ram, not akkiraka¯ram). Sanskrit religious terms and names are also transliterated from their more familiar Sanskrit forms, rather than their Tamil ones (e.g., dars´ana, not taricanam; S´iva, not Civan). A similar conven¯ ¯ tion is followed for revenue terms of Persian or Arabic origin (e.g., mira¯sida¯r, not mira¯cuta¯r). Personal names, caste names, and geographical and historical names, however, are spelled in their conventional, English forms. In many cases, there are several conventional forms, so that one has had to be selected—normally the commonest or most accurate transliteration. “Brahman” is preferred over “Brahmin”; for the Tamil Brahman honorifics, “Aiyar” and “Aiyangar” are preferred over variants such as “Ayyar,” “Iyer,” “Ayyangar,” “Iyengar,” etc. In general, though, personal names are spelled as individuals (apparently) preferred, so that when an honorific is part of a man’s personal name, the form he used is retained. For geographical names, the main difficulty is that many towns, cities, states, districts, etc., have been renamed since Independence. Thus, for example, the Tamil-speaking region of the Madras Presidency became Madras state after Independence and was renamed Tamilnadu (Tamil Nadu) in 1969; the city of Madras was renamed Chennai in 1996. In general, names of places appropriate to the historical context are used, but alternative names are indicated where necessary.

Map 1. South India

Introduction

Satyamurti Aiyar, a Tamil Brahman, was born in 1914 in Tindivanam, a small town about seventy-five miles south of the city of Madras. Satyamurti was the eldest son of Subrahmaniam, whose family had migrated from northern India in the eighteenth century and intermarried with local Tamil Brahmans, with whom they had become fully assimilated. The family had acquired land in Tindivanam, a small agricultural market town retaining many of the characteristics of a village in the early twentieth century, and they had a house in its Brahman quarter, the agraha¯ram. When Satyamurti was born, the family lived off the income from their land, whose cultivation was managed by Subrahmaniam, assisted by low-caste Vanniyar laborers, including members of one family hereditarily attached to his own. Around 1918, however, Subrahmaniam secured a minor administrative post in Tindivanam, because by then he had four children and needed more money. He gave up cultivation and rented his land to Vanniyar tenants. Satyamurti received an elementary education at home, where he learned a little English from a tutor. When he was twelve, Subrahmaniam sent his son to Madras, where he studied in a high school before entering a college run by Catholics. Because he had no relatives in the city, Satyamurti lived in a Brahman hostel, but he regularly traveled by train to visit his parents. The college was located in the suburb of Nungambakkam, which in those days was not very different from Tindivanam, so that when Satyamurti lived there it was much like living at home. At the age of fifteen, Satyamurti was married to a relative, an eleven-year-old girl he already knew, who had had some elementary education in the village school. But Satyamurti’s life hardly changed at this time because his wife remained with her parents and the marriage was not consummated until four years later.

 / Introduction

Beneath this quiet continuity, however, a profound and irrevocable break was occurring. Subrahmaniam was essentially a rural man, a hereditary landlord, who dressed in traditional style; although he had learned some English for his administrative work, his old-fashioned outlook on the world had hardly changed. Satyamurti, on the other hand, who knew nothing about agriculture and was uninterested in it, was imperceptibly transforming himself into a townsman, although his father was scarcely aware of it because he and his son rarely talked about such matters. Unlike his father, Satyamurti was completely ignorant of Sanskrit and could not draw much intellectual sustenance from his ancestral cultural heritage, but he was comfortable in English, the language current among students and in many other circles in Madras. On the other hand, in his noisy hostel, Satyamurti could not study very hard, so that he relied on traditional methods of memorization, which satisfied him and his teachers, but equipped him with only a mediocre education. A return to Tindivanam was impossible anyway, even if Satyamurti had considered it. In 1930, the Depression caused commodity and land prices to fall drastically. Subrahmaniam’s income from cash crops dried up and after a while the family jewelry ended up in the hands of a moneylender. Subrahmaniam was now almost fifty and responsible for four younger children born after Satyamurti, so he decided to solve his family’s financial problems by moving to Madras, which his administrative employment made possible. He rented a house in the city and Satyamurti, who was now twenty and had finished college, came to live there with his wife. Dreams of pursuing further study at Oxford were out of the question. Actually, in spite of the cost, Satyamurti’s father was not completely opposed to the idea, but his mother was adamantly against any breach of religious tradition or dharma, which prohibits Hindus from crossing the sea. Moreover, Satyamurti himself had lost his ambition, partly owing to his poor education, and—especially in uncertain economic times—he was happy to obtain a modest but secure administrative job. When the Second World War began, prices rapidly recovered and Subrahmaniam discharged his debts by selling some land. He rejected the opportunity to sell all his ancestral lands to invest in speculative businesses, as others were doing, and preferred to retain what was left of his reduced patrimony. Shortly after the war ended, he built a new house with a garden in Nungambakkam; his remaining land in Tindivanam supplied his family’s annual consumption of rice and, most importantly in his own eyes, guaranteed his high social rank. Subrahmaniam died before he was affected by land-reform legislation following Indian Independence in 1947.

Introduction / 

Not long after his death, his family split up and all their land was sold to a rich moneylender, so that their last links to a rural world were broken and nothing was left of an old Brahman landlord’s prestigious way of life. For Satyamurti, however, none of this mattered, because his life was in Madras and he was particularly attached to English, the colonial language of his education that now enabled him to work as a bureaucrat. From just one generation to another, the transition for this Tamil Brahman family was often difficult, but it was complete: Satyamurti, in contrast with his father, was fully an urbanite. Satyamurti Aiyar’s life history is recounted by Jacques Dupuis.1 This introduction could have begun with a more fascinating life history of a more distinguished Tamil Brahman. But we chose to describe Satyamurti because, insofar as any single person can typify an entire group, he does so by personifying the archetypal transformation from rural landlord into urban office worker that was so critical in the middle part of the twentieth century, even though it began in the nineteenth century and has continued to the present day. Most men like Satyamurti actually moved away from villages, rather than small towns like Tindivanam. In most other respects, though, Satyamurti is truly a Tamil Brahman Everyman for the modern age, one of countless members of his caste who have joined the urban middle class. Many Brahmans, like Satyamurti, were and are employed in subordinate posts and white-collar jobs that place them in the lower-middle class, whereas many others are well-paid, high-status, educated professionals and managers belonging to the upper-middle class. These people are often the children or grandchildren of men like Satyamurti who first made the move into the lower levels of urban, white-collar employment. Of course, there are still too many poor Brahmans: for example, landowners with tiny plots, village temple priests, domestic servants, or shop assistants. By the end of the twentieth century, however, the overwhelming majority of  Tamil Brahmans were living and working in urban areas as members of the middle class, in its lower or upper strata, and very few remained in the countryside. The Tamil Brahmans’ transformation from a traditional, mainly rural, caste elite into a modern, urban, middle-class community is the central theme of this book. It is a study of an unusual social group, but one whose history during the colonial and postcolonial periods throws a powerful light on structure and change in Indian society. As a case study, the Tamil Brahmans particularly help to explain, for example, the dynamics of both urban migration and regional, national, and transnational movement; the formation of the middle class; the interdependent relationship between

 / Introduction

caste and class; and the changing position of women in modern India. As a social group that is the antithesis of those within the fold of “subaltern studies,” the Tamil Brahmans also enable us to examine how and why privileged status within a hierarchical society can be perpetuated in the face of major social, cultural, economic, political, and ideological changes that might have been expected to undermine it completely. Last but not least, for making sense of the amalgam of ostensibly traditional and modern beliefs and practices characteristic of India today, the Tamil Brahmans provide an exemplary illustration. In recent years, India has acquired a new kind of visibility across the world, mainly because its economy has been growing so rapidly since it began to be liberalized in the early 1990s. Some sectors of the economy have seen spectacular rates of expansion, notably information technology (IT), which provides a key part of the infrastructure for contemporary global capitalism. Leaving aside wildly optimistic speculation that India may leap from a preindustrial, agrarian economy to a postindustrial, “informational” one, it is still very probable that by 2050 India’s economy, although smaller than China’s, will be almost as large as that of the United States. These developments and projections have led many American and European politicians, economists, and commentators not only to worry about the impact of outsourcing more and more work to countries like India, but also to recognize more generally that global economic power is decisively shifting, so that talk about the rise of the East and the fall of the West is now commonplace. All sorts of reasons are advanced to explain this transformation, and some of them are more plausible than others, but it is fair to say that they often consist of sweeping generalizations about India or China, as well as the West. One favorite generalization is about education. Thus international surveys commonly show that students in the West are falling behind their peers in the East in science, technology, engineering, and mathematics, and it is then concluded that the West’s decline is unstoppable unless its educational defects are corrected. In reality, the evidence for this conclusion is highly debatable, as an article in the influential American magazine Foreign Policy in 2011 argued in response to yet another announcement of a “massive wake-up call” by the US education secretary; Americans, the article’s headline proclaimed, should actually relax because “Chinese math whizzes and Indian engineers aren’t stealing your kids’ future.”2 This book does not try to join this debate or suggest policy initiatives that might improve American or European education. On the other hand, it is true that Indian engineers are playing an important role in both the national and global economies, and Tamil Brahmans are very well represented

Introduction / 

in their ranks. During the colonial period, caste traditions of learning were one important factor impelling the Brahmans to take up modern, Englishlanguage education, and many of them entered the three main learned professions of their day—law, medicine, and engineering. Law, like administration, was a traditional vocation for high-status Brahmans in precolonial times, but Western medicine and engineering obviously were not, and there was some particular hesitation about medicine owing to the dangers of ritual pollution. The essential point to make here, however, is that the case of the professions shows how “caste traditionalism”—frequently identified as one of the major brakes on economic growth in India—actually had a mainly positive effect, because it encouraged Tamil Brahmans to enter the modern, educated professions, while only briefly discouraging them from entering medicine and engineering. These developments have had profound, longrun consequences for the Tamil Brahmans and their transformation into a migratory, urban, middle-class group; they have also decisively shaped the Brahmans’ relationship with non-Brahmans in Tamilnadu in the social, cultural, and political domains in the colonial and postcolonial periods, from the early twentieth century to the present day. In the long run, too, the Tamil Brahmans’ early entry into engineering proved crucial—not only within Tamilnadu, but also in the rest of India and overseas—and it is one reason why so many of them, both men and women, are software engineers working in contemporary India’s booming IT industry. They aren’t actually stealing American kids’ futures, but they are active participants in India’s rise to global economic power, which can only be understood in any detail or depth by investigating the part played by people like the Tamil Brahmans.

Introduction to the Tamil Brahmans The Tamil Brahmans are a small group, by Indian standards anyway. Their total population can only be roughly estimated, however, because no Indian census after 1931 collected comprehensive data on caste. In 2012, however, people were asked to identify their caste in the first “socio-economic and caste census,” but no results were available when this book’s proofs were being corrected. In 1931, just under 2.5 percent of the population in the region corresponding to the modern state of Tamilnadu were Brahmans, of whom three-quarters were Tamil Brahmans: that is, Brahmans whose mother tongue is Tamil. The rest were mostly Telugu, Kannada, and Maharashtrian Deshastha Brahmans. Through extrapolation, the maximum size of the Brahman population in Tamilnadu in 2011 (the date of the last census) can be estimated at 1.78 mil­lion, 1.40 million being Tamil Brahmans. Around one

 / Introduction

quarter of all Tamil Brahmans are settled outside Tamilnadu, in the rest of India or overseas, and the estimated total size of the caste’s population, which is admittedly just an informed guess, is 1.85 million in 2011 (see appendix). The Tamil Brahman caste ( ja¯ti) is divided into two main sections: the Smartas or Aiyars (sma¯rta; aiyar) and the Sri Vaishnavas or Aiyangars (s´rı¯ vais·· nava; aiyan·kar). “Aiyar” and “Aiyangar,” like other titles such as “Acharya” (a¯ca¯rya) and “Sastri” (sa¯stri), are honorific terms that were formerly used by men as caste “surnames,” but this practice largely ceased in the mid-twentieth century, so that today Tamil Brahmans’ names rarely identify their caste unambiguously. The division between the two sections is primarily religious: Smartas traditionally worship both the great gods of pan´ Indian Hinduism, Vishnu (Vis·n ·u) and Shiva (Siva), as well as other deities, whereas the Sri Vaishnavas worship Vishnu alone. Smartas are segmented into four principal subcastes: Vadama (vat·ama), the largest and highest in rank; Brahacharanam (brahacaran ·am), second in size and rank; and Ashtasahasram (as··t asahasram) and Vattima (va¯ttima), the smallest and lowest, their mutual rank in dispute. Sri Vaishnavas are divided into two “sectarian” subsections: the larger, higher-status, “northern” Vadagalai (vat·akalai)—who accord priority to Sanskrit texts in worshipping Vishnu—and the smaller, lower-status, “southern” Tengalai (te˘nkalai)—for whom Tamil texts are pri¯ mary. All these various units, which are further subdivided, regularly dispute their ranking vis-à-vis each other. All of them are also traditionally endogamous, but during the last fifty years or so, marriages between Tamil Brahmans from different units have become commoner and more acceptable. Generally speaking, Sri Vaishnavas tend to be more conservative than Smartas—for example, about endogamy, ritual purity, or the proper way to worship the gods—but differences between them rarely matter in most social contexts today. The only census to collect figures on sections and subcastes dates from 1891, but their relative size has probably not changed much, so that about two-thirds of all Tamil Brahmans are still Smartas and the rest are Sri Vaishnavas (see appendix). Brahman priests are described in chapter 6; although Brahmans are often called a priestly caste, the vast majority of Tamil Brahmans have never been priests and have instead lived off the land or earned their living in secular occupations. In the classical, pan-Indian model of Hindu society, the population is divided into four varn ·as: Brahmans (bra¯hman ·a), Kshatriyas (ks·atriya) or warriors, Vaishyas (vais´ya) or farmers, traders, and herders, and Shudras (s´u¯dra) or servants. In south India, however, there were traditionally no Kshatriyas or Vaishyas, so that all non-Brahman castes were classified as Shudra. Outside the varn ·a system are the Untouchables, the Harijans or Dalits (as they

Introduction / 

increasingly call themselves), who have also been called Adi Dravidas (a¯di dra¯vid ·a, “original Dravidian”) in Tamilnadu. During the colonial period, the tripartite division into Brahmans, non-Brahmans, and Adi Dravidas or Dalits became a fixed, axiomatic feature of Tamil society, so that today almost everyone assumes that their membership of one of these three groupings matters and almost everyone knows whether their friends, colleagues, workmates, neighbors, and other associates are Brahman, non-Brahman, or Dalit, or alternatively Christian or Muslim. This is not normally a point of contention in ordinary daily life and may not greatly affect how people interact with each other. Moreover, in Tamilnadu like the rest of India today, people often avoid the words “caste” and ja¯ti, and prefer the euphemistic synonyms “community” and sama¯j instead. Even so, almost all Tamil Hindus (and many Christians and Muslims as well) still take it for granted that their caste identity—at least as defined by the gross categories of Brahman, non-Brahman, and Dalit—is a basic part of who they are. The social and cultural distinctions among the three caste groupings, and especially between Brahmans and the rest of the population, are drawn in manifold ways. Food, though certainly not the only marker, is very important. Tamil Brahmans, like most but not all Brahmans throughout India, traditionally adhere to vegetarianism, which is regarded as ritually purer than and morally superior to meat eating. Some other “Brahmanized,” high-ranking, non-Brahmans in Tamilnadu, especially Shaiva Pillais who mostly belong to the Vellalar caste, are also vegetarian. By contrast, all other non-Brahmans and Dalits, as well as Christians and Muslims, are traditionally non-vegetarian. In practice, most Tamil Brahmans actually are vegetarians who avoid meat, fish, and eggs, and sometimes other “heating” foods as well, such as onions and garlic. Not everyone is equally strict, however: for example, some parents give their children eggs for nutritional reasons and other people turn a blind eye to the ingredients in manufactured foods like cake. And some Brahmans, especially men living overseas, do eat meat and fish, at least when outside the home. Conversely, a considerable number of non-Brahmans, especially women, choose to be vegetarian, irrespective of their own caste’s traditional diet. Vegetarianism in India implies teetotalism as well, and Brahmans traditionally do not drink. In “Westernized,” upper-middle-class and elite circles, mainly in large cities and overseas, social drinking is acceptable and commonplace. In general, however, the consumption of alcohol attracts opprobrium and Brahman drinkers, nearly all men, tend to be evasive or keep quiet about it; “don’t ask, don’t tell” is common practice, so that guessing how many Brahmans drink is impossible. Many non-Brahmans avoid alcohol as well, of course,

 / Introduction

but it is not stigmatized to the same extent as it is among Brahmans. But whatever individuals belonging to different castes actually eat and drink, the opposition between vegetarianism and nonvegetarianism is consistently regarded as homologous with that between Brahmans and non-Brahmans in Tamilnadu. According to traditional, orthodox norms, Brahmans should eat only food that has been prepared by Brahmans complying with the rules of ritual purity. Other things being equal, Sri Vaishnavas have tended to be stricter than Smartas. Yet there has always been considerable variation in how purity rules are interpreted, and that is especially true today. Thus, for instance, some housewives let their non-Brahman servants prepare food and work in the kitchen, whereas others do not, and many people comply with purity rules quite strictly at home, but are less bothered when they eat outside, for example, in restaurants where the cooks are probably not Brahmans. A minority of Brahmans never eat food cooked away from home unless they have to and others do so only if they are certain that Brahman cooks have made the food properly. In towns and cities in Tamilnadu, orthodox Brahmans—as well as others who are less strict—patronize numerous “Brahman hotels” and “messes,” restaurants serving only vegetarian food cooked by Brahmans; there is also a growing trade in “carryout” food made by Brahmans. At weddings and other public events hosted and attended by Brahmans, Brahman cooks are always employed (although in Chennai today, it can be difficult to recruit them when required).3 Cooking, in fact, has always been a common source of employment for Tamil Brahmans, although it is neither very prestigious nor normally very remunerative. Dietary distinctions are salient among both divine and human beings. Vishnu and Shiva, the two great gods of pan-Indian Sanskritic Hinduism, are worshipped with Devı¯, the goddess, in temples served by Brahman priests and during worship they are offered only vegetarian food. Non-Sanskritic, Tamil, little “village deities,” on the other hand, are served by non-Brahman priests and some of them are offered animal sacrifice and then blood or meat (see chap. 6). Moreover, vegetarianism is so highly valued in Brahmanical Hinduism (as well as Buddhism and Jainism) because it is equated with ritual purity, coolness, bodily self-control, ascetical self-restraint, and especially nonviolence, whereas meat eating is equated with impurity, heat, aggression, energetic action, and loss of bodily control, and it also necessitates the immoral slaughter of living beings. To be sure, not all Brahmans are preoccupied by the symbolic and moral connotations of vegetarianism and meat eating, and non-Brahmans generally contest them anyway, insisting, for example, that butchery is not immoral and meat is necessary

Introduction / 

to build up vigor and strength. Nonetheless, from the Brahmans’ point of view, the ritual purity and ethical value of vegetarianism make it superior to nonvegetarianism, so that vegetarian Brahmans—like the great gods—are superior to all those who eat meat.4 Furthermore, vegetarianism, even if its superiority is disputed, is so firmly identified with Brahmans in Tamilnadu that—for Brahmans and non-Brahmans alike—the contrast in diets serves as a diacritical marker of their cultural difference and social distinction, especially, perhaps, in the modern world in which most visible signs of caste hierarchy and separation have faded or disappeared.

The Non-Brahman Movement and Politics in Tamilnadu The roots of the tripartite division into Brahmans, non-Brahmans, and Dalits, as it exists in present-day Tamilnadu, mainly lie in the emergence during the colonial period of the “non-Brahman” category within the context of the non-Brahman, Dravidian movement. This movement’s history, both before and after Independence, has been studied by many different scholars.5 In December 1916, the Non-Brahman Manifesto was published to protest against Brahman privilege.6 Issued by a new organization set up to promote non-Brahman interests, the manifesto quoted copious statistics to show that Brahmans, who formed only 3 percent of the Madras Presidency’s population, held an excessive number of official posts. (The presidency’s territory included most of present-day Tamilnadu, plus coastal Andhra Pradesh, north Kerala, and parts of Karnataka.) For example, in the Provincial Civil Service examinations between 1892 and 1904, fifteen out of sixteen successful candidates were Brahmans. In 1913, ninety-three of the 128 permanent district munsifs (lower-level judges) were Brahmans, compared with twenty-five non-Brahmans and ten from non-Hindu groups; in the Madras High Court, four of the five Indian judges were Brahmans. By 1914, the 650 registered graduates of the University of Madras included 450 Brahmans, 124 non-Brahmans and seventy-four from other communities; eleven out of twelve elected fellows of the university were also Brahmans. Not only did these and other figures show that Brahmans dominated the government service and the university, but the same group also dominated the nationalist movement, for fifteen out of sixteen members elected from Madras to the All-India Congress Committee were Brahmans and only one a non-Brahman. The manifesto acknowledged that Brahmans were ahead of other castes and communities in educational attainment, but it insisted that nonBrahmans had made significant progress and deserved a fairer share of

10 / Introduction

government appointments. It also made it clear that non-Brahmans could not trust Congress, so that they favored the continuation of British rule, which was more likely to look after their interests fairly. It concluded by calling upon all non-Brahmans to commit themselves to further educational improvement and to make their voices heard, and for each community “to put its own house in order” so that it could cooperate with others on equal terms. The Non-Brahman Manifesto marked a decisive moment in the modern history of south India. It postulated a clear-cut opposition, primarily in the Tamil country, between a small Brahman minority and a large, but extremely diverse, non-Brahman majority, which rejected the derogatory name “Shudra.” The manifesto’s publication was soon followed by the establishment of the Justice Party to represent the interests of non-Brahmans—or more exactly of their urban, middle-class minority—in opposition to the Brahman-dominated Congress. After the Justice Party won the Madras Legislative Council elections in 1920, the government introduced measures to ensure fairer recruitment among different castes and communities, which was the beginning of the reservations policy—“affirmative action” in favor of the lower castes—in the province. During the 1930s, the Justice Party went into decline and, when India became independent in 1947, Congress was in power, but by then Brahmans were being progressively displaced in the party by non-Brahmans. The Justice Party was eventually replaced by more radical Dravidian organizations representing non-Brahman interests, first the Dravida Kazhagam (DK: “Dravidian Federation”) and then the Dravida Munnetra Kazhagam (DMK: “Dravidian Progress Federation”), founded in 1949. In 1967, Congress lost the state elections and the DMK took power in Madras state, which it renamed Tamilnadu. Since 1967, with only brief interruptions, the DMK or its rival, the All-India Anna DMK (AIADMK, or ADMK for short), which was formed in 1972, has continually ruled Tamilnadu. By the end of the twentieth century, with very few exceptions—notably J. Jayalalitha, the ADMK leader—Brahmans had disappeared from state politics and the one or two who remained almost had to deny their own Brahman identity to win elections.7 Moreover, the DMK and ADMK governments have greatly expanded the reservations system, so that Brahmans are vastly outnumbered by non-Brahmans at all levels of the government service and public sector. In effect, everything the Non-Brahman Manifesto demanded eventually came to pass during the twentieth century, so that the “Brahman raj” (to quote Pamela Price), which sometimes looked indistinguishable from the British raj around 1900, had vanished by 2000.8

Introduction / 11

In 1977, the ADMK came to power for the first time and its populist policies were less stridently anti-Brahman than the DMK’s had been, despite the expansion of the reservations system. Indeed, anti-Brahmanism has steadily declined as a basis for government policy since the late 1970s. Yet the dichotomous division between Brahmans and non-Brahmans has not faded away; neither has the Dravidian movement’s broader claim that Brahmans were immigrants from the north whose hierarchical, Sanskritic, “Aryan” culture was imposed on that of the indigenous Tamil people. On the contrary, an opposition between the two caste groupings and all that they stand for has become more and more entrenched, so that it is now just taken for granted as a basic feature of Tamil society, culture, and politics. In the mid-1990s, Nicholas Dirks commented both that non-Brahman politicians keep alive “the threat of the internal other, the Brahman,” even though Brahman dominance has vastly diminished, and that “caste—and specifically the divide between Brahmans and non-Brahmans—therefore always seems to be the rhetorical point of over-determination” in intellectual political debate in Tamilnadu.9 One outcome of this situation, as M. S. S. Pandian has argued more recently, is that because the polar opposition between Brahman and non-Brahman has become “the naturalized language of politics in the region, [it] blocks the emergence of other inferiorized identities,” specifically that of Dalits, who now confront non-Brahman hegemony, much as non-Brahmans earlier fought against Brahmanism.10 In Tamilnadu today, notwithstanding this “naturalized language,” caste politics are in reality defined primarily by competition among non-Brahman groups and, of course, between them and the Dalits, whereas Brahmans themselves have left the fray.

Tamil Brahmans and Anti-Brahmanism Let us now return to the early twentieth century when the non-Brahmans’ opposition to Brahmans was first forged. The Non-Brahman Manifesto focused on unfairness in government appointments, but the anti-Brahmanism it heralded was always fueled by visceral resentment as well. The manifesto, in discussing why Brahmans were better educated than others, referred to their position as “the highest and most sacred of the Hindu castes,” and sarcastically mentioned their own belief that “they are so many divinely ordained intermediaries without whose active intervention and blessing the soul cannot obtain salvation.”11 Marguerite Ross Barnett, who interviewed former Justice Party members in the 1960s, was struck by how they described

12 / Introduction

the party’s origins. First, they mentioned non-Brahman backwardness and Brahman dominance in government and politics. “But then,” she said, “the tone of the respondent would change, and with emotion he would recount some personal incident, insult to his dignity, or psychic injury suffered at the hands of the Brahmins, and would relate this to being designated a [Shudra].”12 Another telling illustration comes from the autobiography of A. N. Sattanathan, who became a high-ranking Indian civil servant, but was born into a poor family belonging to the low-status Padaiyachi caste in southern Tamilnadu. In the early twentieth century, most schoolteachers in the region were Tamil Brahmans. Sattanathan vividly recalled that they did not believe that education “should be given impartially to all their students,” and their “lack of humanity” and “traditional arrogance” held back non-Brahman pupils for a long time, so that “hatred of the Brahman became almost natural” to non-Brahmans. Brahman pupils “also kept apart” from the rest. Sattanathan therefore became keenly “aware of the Brahmans’ claim to intellectual superiority and social aloofness.”13 There can be little doubt that humiliating personal incidents of the kind Barnett and Sattanathan describe were common and that non-Brahman resentment against Brahmans was widespread.14 The non-Brahman leader E .V. Ramasamy (known to his followers by the honorific “Periyar”), who founded the “Self-Respect” movement in the 1920s, called upon non-Brahmans to reject their Shudra status; he also ferociously castigated Brahmanism and mocked Brahmans. In the 1940s, Ramasamy and his supporters transformed the Self-Respect movement into the Dravida Kazhagam, which was virulently anti-Brahman.15 From time to time, DK members took to the streets to demonstrate their opposition to Brahmans and all they stood for. In the early 1950s, they desecrated temples and smashed deities’ images, a “politics of heresy” that made the DK unpopular with many Hindus, irrespective of caste.16 In the late 1950s and again in the early 1970s, DK members attacked Brahman houses and restaurants, and assaulted individual Brahmans. Elderly Brahmans now recall those incidents with a shudder, although cases of violence were actually few and far between, and not many people were seriously hurt. But the incidents were humiliating, especially when DK militants tried to cut off Brah17 man men’s topknots or tufts (kut·umi) and sacred threads (pu¯n ·u¯l). (Sacred threads, which are worn over the left shoulder and across the chest, signify the initiated, “twice-born” status of adult male Brahmans.) By the 1950s, many Tamil Brahmans living in urban areas had already given up their topknots and distinctive clothing, but the DK’s activities encouraged more to

Introduction / 13

do so to protect themselves in public; by the 1960s, as André Béteille’s village ethnography shows, rural Brahmans were following suit.18 In the past, Brahman men tonsured the front of the head and tied the remaining long hair in a topknot (sometimes hidden by a turban), whereas other men cut it short in conventional style. To cover the lower half of the body, Brahmans also wore longer cloths (ves··t i) than non-Brahmans and tied them between their legs in a special way, rather than simply wrapping them round the waist. South Indian men, who customarily draped a cloth across the shoulders, did not wear shirts, so that Brahman men’s sacred threads were clearly visible. When Brahman men abandoned their traditional hair and dress styles, they largely copied those of non-Brahmans. Especially in urban areas, more and more men of all castes also started to wear a European-style shirt and trousers, which tended to hide the Brahmans’ sacred threads. Today, virtually the only Brahman men retaining the traditional style are priests when they are at work, in temples or their clients’ homes, and even some priests now cut their hair. Brahman women’s dress has changed as well. Married Brahman women formerly wore nine-yard saris elaborately tied between the legs, whereas other women wore six-yard saris draped in the conventional way. The Brahman woman’s attire was deemed to be purer than the non-Brahman’s, although in practice the nine-yard sari always mainly served as a visible marker of caste distinction. A Brahman bride wears a nine-yard sari on her wedding day and a few conservative women stick to the old style, but the great majority of Brahman women now dress like everyone else. Middle-aged and elderly women usually wear six-yard saris; younger women, like their peers throughout India, generally prefer the shalwar kamiz (trousers and tunic). Particularly in the modern, “Westernized” cities of Bangalore, Mumbai, and Delhi, quite a lot of young women wear Western clothes, but fewer do so in Chennai or elsewhere in Tamilnadu. Sometimes, men and women wear distinctive jewelry or religious emblems that reveal them as Brahmans. On the whole, though, the overall effect of sartorial change is that Brahmans and non-Brahmans cannot normally be distinguished by their appearance in public places today. It is true that caste can often be identified by speech because Brahman and non-Brahman Tamil dialects vary considerably; Brahmans, for example, tend to use more words of Sanskritic origin than non-Brahmans. Nonetheless, the elimination of visible differences between Brahmans and non-Brahmans in public places because they now all dress alike is a significant change, which is intertwined with the long-term impact of the nonBrahman movement and anti-Brahmanism in Tamilnadu, as well as with

14 / Introduction

the broader erosion of the legitimacy of caste throughout India since Independence. The dropping of caste “surnames” like Aiyar and Aiyangar reflects the same trend; in Chennai, too, the numerous streets named after famous Brahmans—such as C. P. Ramaswamy Aiyar Road in Alwarpet—have been renamed to delete their honorific suffixes, so that caste titles are officially absent in the public space. Moreover, in all manner of contexts during the second half of the twentieth century, Brahmans and non-Brahmans have been mixing with each other much more frequently than in the past, even if dietary customs in particular regularly differentiate them, for example, when they go to the office canteen. Among some Brahmans, there is nostalgia and regret about the disappearance of their old way of life. Such sentiments are not confined to diedin-the-wool conservatives; they are also voiced by those who support social progress and the eradication of caste inequality, but still wish that traditional Brahman customs had not been so overwhelmed by modern change and non-Brahman tastes, as has happened, for instance, in some Hindu festivals. A small but still telling example is the autumnal Navara¯tri (“Nine nights”) festival, when Brahman women set up elaborate displays of dolls in the likeness of gods and goddesses in their homes, but because nonBrahmans now do so as well, the style of dolls on sale is increasingly designed to appeal to them.19 An especially eloquent expression of this nostalgia is The Toss of a Lemon, Padma Viswanathan’s novel about the history of a Tamil Brahman family from 1896 to 1958, which is based on her grandmother’s memories. The context for the novel, which is ethnographically and historically accurate, is urban migration, bureaucratic employment, anti-Brahmanism, the changing status of women, and all the other developments of the period. Viswanathan sensitively evokes both the impact of these modern changes and the sometimes confusing sense of loss that they cause for her characters, especially the rural women at the center of the family.20 From time to time, in both urban and rural areas, Brahmans still put other people down, but they do so far less often and much less blatantly than in the past, especially in public. Just as importantly, non-Brahmans, whether or not they make a fuss about slights, hardly ever suffer any “psychic injury” nowadays, because they do not see themselves as inferior to Brahmans, as so many of their ancestors, albeit ambivalently, had done. But paradox and ambiguity abound in the contemporary regime of caste, as David Mosse perceptively explains in his ethnography of a southern Tamilnadu village inhabited by non-Brahmans and Dalits, both Christian and Hindu. Not only in this village, but generally in Tamilnadu and probably most of India as well, a “clear concomitant of the equalized public space”—and the ille-

Introduction / 15

gitimacy of caste inequality—“is a privatization of caste distinction.” “Caste feeling,” which is seen as a deplorable weakness of mind, still exists, however, but it has gone inside the mind or heart. Many Dalits, Mosse reports, assume that caste feeling afflicts non-Brahmans who discriminate against them; many non-Brahmans throughout Tamilnadu probably assume that it also underlies Brahman attitudes toward all lower-caste people.21 Tamil Brahmans today know full well that non-Brahmans do not regard them as superior; they know, too, that they are not very popular with many people and are often called pa¯rppa¯n, the Tamil term for “Brahman,” which ¯ they see as insulting. Yet Brahmans, whether they really care about the insults or not, typically still regard themselves as superior to others and freely say so when talking privately among themselves—or to sympathetic foreigners—even though they normally keep quiet in front of non-Brahmans. Even without using the phrase, they tacitly admit to their “caste feeling.” Most commonly, Brahmans claim superiority on the grounds that they are innately more intelligent and intellectually able than people from other castes, which is supposedly proved—though of course it isn’t—by their generally higher standards of educational attainment, as well as their extensive participation in the learned professions and in science and technology, most recently information technology (IT). Since education and learning have always been Brahman traditions, the contemporary claim is actually a transformation of an old one, but other grounds for asserting social superiority, particularly ritual purity, are not as pivotal as they were fifty or a hundred years ago. Tamil Brahmanhood—the attributes deemed to define a Tamil Brahman and his or her identity and status—was never fixed in any epoch, but it has certainly changed considerably during the colonial and postcolonial periods, as an old caste elite has been transformed into a modern middle-class group.

Tamil Brahmans and the Indian Middle Class The Indian middle class grew slowly during the colonial period, more quickly following Independence, and a lot more rapidly after economic liberalization began in the early 1990s. The postliberalization middle class has attracted the attention of countless journalists, commentators, and market researchers, as well as a growing cohort of social scientists. In introducing two edited volumes on the middle class, Henrike Donner and Geert De Neve, and Amita Baviskar and Raka Ray, outline much of the recent research, and both pairs of authors emphasize the middle class’s diversity, which has become more evident as empirical studies have multiplied. Because

16 / Introduction

of this diversity, some writers—such as Baviskar and Ray, and B. B. Misra earlier—prefer to describe the “middle classes” in the plural.22 Béteille, who pays due attention to developments before liberalization, points out that middle-class Indians are defined “not only by occupation, education and income,” but also “by language, religion and caste.” And hence “the co-existence of these two sets of divisions, new and old, . . . gives the Indian middle class its distinctive character.”23 Following Béteille and others, it is now well recognized, as Donner and De Neve reiterate, that middle classness has always been shaped by preexisting values and practices, which are important sources of diversity, alongside variations in income, wealth, or consumption patterns.24 The Indian middle class’s diversity is probably increasing as more people from different social backgrounds are drawn into it. On the other hand, there are common markers to middle classness, notably “an emphasis on respectability, specific gender relations, and ideas about the family, and the value attached to education.”25 Of course, middle-class norms themselves change too, as shown by attitudes to the desirability of women’s employment, which have often been affected by pressure to uphold “traditional” values. As this book will show, all these observations by Donner and De Neve are consistent with our evidence on Tamil Brahmans. One reason why middle-class lifestyles “have acquired a hegemonic role” in contemporary India, Donner and De Neve comment, is that they “are closely entwined with the aspirations and discourses of modernity.”26 Baviskar and Ray also emphasize the connection between modernity and the middle class, but they specially look at its political identity, especially since liberalization, which supposedly “unleashed” the Indian middle class “from the chastity belt of Nehruvian socialism and Indira [Gandhi]-era austerities, finally able to savour the fruits of its disciplined and diligent work.”27 The modernity of middle classness, though tempered by commitment to traditional values as well, is clearly important for Tamil Brahmans. We do not discuss politics very much in this book, but it is worth noting here that the middle-class people we interviewed, whether Brahman or nonBrahman, were mostly strong supporters of economic liberalization for the reasons given by Baviskar and Ray.28 Notwithstanding the stress often placed on it, consumerism is not equally important for all sections of the postliberalization middle class.29 Indeed, it is scarcely valued at all by middle-class Tamil Brahmans. Sara Dickey, who did research in Madurai in the 1980s and 1990s, observes that Brahmans are said by themselves and others “to avoid public display of their wealth and to have muted tastes in personal and domestic fashions,”

Introduction / 17

whereas some subcastes of Chettiyars, for instance, prefer more showy jewelry and expensive fashions, especially when attending weddings.30 Our findings fully accord with Dickey’s. In general, instead of spending money on consumer goods, Brahmans prefer to save and invest it, most commonly in children’s education and housing. Of course, education and property can be forms of consumption, and this entire discourse is really about stereo­ types, rather than actual economic behavior. Moreover, compared with the poor majority, all well-off, upper-middle-class Brahmans lead luxurious lives; there are also many extravagant Brahmans, just as there are plenty of plain-living, thrifty non-Brahmans. The key point, however, is that middleclass Tamil Brahmans, by avoiding ostentatious consumerism, uphold their caste’s traditional value of ascetical self-restraint—symbolized by vegetarianism—which is transmuted into the modern, middle-class virtues of self-discipline and moderation.31 The “world-rejecting” asceticism of Brahmanical Hinduism has been converted into something remarkably close to the “inner-worldly” asceticism of the Protestant ethic.32 A fundamental argument of this book, to which we specifically return in the concluding chapter, is that the Tamil Brahmans have become a middle-class caste. Critical in this development is their systematic tendency to equate traditional Brahman and modern middle-class values, as if they were one and the same. In other words, middle-class values are regarded by Brahmans as paradigmatically their own, so that the Tamil Brahman middle class—especially the upper-middle class—is, so to speak, a natural dyadic unit. In Max Weber’s terms, it is both a “social class” and a “status group,” which is built on an “isomorphism” of Tamil Brahmanhood and upper-middle classness, to borrow Sherry Ortner’s term from her discussion of the “isomorphism of ‘Jewishness’ and ‘middle classness’” in the modern United States.33

The Sociology of Class Because class is so important for our study of Tamil Brahmans, we will say a little more about the theory of class and explain our analytical terms. Weber drew a now familiar distinction between “economic classes,” which define people’s life chances in the market, primarily according to whether they own property or earn a living by offering services, and “status groups,” which depend on social estimations of “honor” and are normally communities sharing a “common style of life”; the fullest development of statusgroup stratification is a hierarchical system of “closed,” endogamous castes limiting interaction with each other.34 Weber also separately defined “social

18 / Introduction

classes” as groupings of people whose economic “class situations” are clustered together, especially because they have similar opportunities for social mobility, whether intragenerationally (within a person’s career) or intergenerationally. In modern, Western capitalist society, the four main social classes are the manual working class, the petty bourgeoisie, propertyless technicians and white-collar employees, and “the classes privileged through property and education.”35 Anthony Giddens defines the “structuration” of classes as “the processes whereby ‘economic classes’ become ‘social classes,’ and whereby in turn the latter are related to other social forms.”36 Structuration is facilitated by the linkage between “market capacities” and “distributive groupings” based on common consumption patterns. Since the early twentieth century, these market capacities have primarily been ownership of productive property, possession of educational or technical qualifications, and manual labor power; they in turn are strongly associated with “closed patterns” of relatively restricted mobility and hence lay the basis for a tripartite division into the upper, middle, and lower or working classes (rather than the four originally set out by Weber). As structuration proceeds, relatively open classes tend to turn into relatively closed status groups as well and class “awareness” becomes the norm, so that members of a class recognize that they share a common style of life, largely determined by common consumption patterns, which may be evaluated as more or less prestigious. This entire process is most likely to occur if social classes coincide with other types of status group, such as ethnic groups or castes, so that clear-cut differences in beliefs, values, and styles of life are reinforced. In the modern capitalist system, the middle class has enormously expanded (whereas the working class has contracted) and it has become more diverse; especially significant is the relatively rapid growth of an upper stratum of educated professionals and managers—the “upper-middle class”—that is usually distinguished from a lower stratum of subordinate, white-collar staff, the “lower-middle class.”37 It is the neo-Weberian definition of “social class” as outlined here that mainly guides our discussion in this book, not least because it is heuristically appropriate to the Indian and specifically the Tamil Brahman case. Plainly, notwithstanding rapid economic growth since the early 1990s, India remains very different from Western European countries or the United States. Thus, for example, hundreds of millions of Indians still live in dire poverty; more than half the country’s population is still rural, with many dependent on agriculture; the industrial working class is proportionally smaller than in Western societies, but it is still expanding and some segments of it are aptly described as socially and culturally “middle-class”;

Introduction / 19

and the upper echelons of India’s professional and managerial class earn salaries high enough to put them in the country’s rich elite. Hence the Indian urban middle class is, relatively speaking, a small, highly privileged section of society whose position within the class structure is very different from that of, say, the British middle class, which has historically been defined primarily by its relationship with the large but now shrinking working class. It therefore follows that much of the Western sociology of class cannot be applied to the Indian case without major modifications. On the other hand, the expansion of the Indian middle class before and after Independence, including the postliberalization period, has closely resembled its equivalent Western trajectory, inasmuch as it has also been driven by the growth of professional, managerial, and white-collar occupations that require educational and technical qualifications. Theories of class, it should be emphasized, are about class structures and are not mere descriptions of occupational stratification, but obviously the relationship with occupation is a key part of the sociological analysis of class.38 This relationship, in much writing about the Indian middle class, has not been examined very carefully, as if lifestyles alone were all that really mattered today. One significant factor complicating analysis is that the Indian middle class includes large numbers of businessmen and traders owning varying amounts of capital, as well as rural landowners who can be categorized as “rich farmers.” Quite commonly, too, one business or farming family includes both sons with outside professional employment and others who run the business or farm with their father. Among the Tamil Brahmans, there is a minority of large capitalists, and small- and mediumsize businessmen, some clearly belonging to the propertied upper class, as well as a dwindling population of landlords and farmers. Nonetheless, the vast majority of Tamil Brahmans have been drawn into the urban middle class—understood as a “social class”—because more and more of them have taken up nonmanual, white-collar, or professional and managerial occupations, which mostly have formal qualifications and “job descriptions” very similar to their Western counterparts.39 Although some people derive rental income from urban property or other investments, salaries are the main source of livelihood for most of the middle class. All in all, therefore, the market capacity of the Tamil Brahman middle class corresponds closely to Giddens’s neo-Weberian definition as framed for the analysis of modern, Western capitalist societies. Before and after Independence, many Tamil Brahmans—like Satyamurti at the beginning of this introduction—were and are employed as low-ranking government officials (“non-gazetted officers”), petty railroad

20 / Introduction

bureaucrats and officials at similar organizations, as well as clerks, stenographers, shop assistants, medical representatives, and so on; at least in relation to income, most priests and cooks also fall into the same category. In the past, but less so today, a lot of Brahmans worked as schoolteachers, a socially intermediate category because salaries are fairly low, but teaching is well respected and confers personal status. Many other Tamil Brahmans— often the descendants of men like Satyamurti—were or are educated professionals and managers, such as high-ranking government officials (“gazetted officers”), lawyers, doctors, engineers, bank officers, company managers, college professors, and most recently software engineers and other IT professionals. Within this broad category, however, there is a considerable variation in how much authority professionals and managers exercise, as well as in their remuneration and occupational status; doctors, for instance, tend to wield authority, are particularly well paid, and enjoy high esteem. Some people engaged in these occupations belong to wealthy, “old elite” families in Chennai (or other major cities) in which several generations of men, and often women as well, have pursued high-status professions; yet others, such as senior IT professionals, now belong to a postliberalization “new” middle class and their very high incomes (by Indian standards) place them in a “new-rich” category.40 In our research, it became clear that a fairly sharp division exists empirically between the white-collar, lower-middle class and the professional and managerial, upper-middle class. As chapters 2 and 3 show, this division is reinforced by restricted mobility, because in spite of some upward mobility from the lower-middle to the upper-middle class, there is very little downward mobility between them. Furthermore, although almost all Tamil Brahmans now regard themselves as members of an urban middle-class caste, they also readily recognize the division between the upper- and lowermiddle strata—and sometimes the elevated position of the old elite or new rich—so that these internal divisions are part of their class awareness. Here we can refer to Lionel Caplan’s study of Protestant Christians in Madras, undertaken in 1974–82, which is one of the first coherent ethnographies of the Indian urban middle class. According to Caplan, most Christians in Madras called themselves “middle class” using the English words, or nat·uttara makkal·, “middle people,” in Tamil. How they conceptualized the “middle” varied, however, between the “well-educated and well-to-do” upper-middle class of educated professionals and managers, on the one hand, and the lower-middle class, some in skilled manual, as well as nonmanual, occupations, on the other. For upper-middle-class Christians, there is a tiny rich elite above them, while the less fortunate majority below are

Introduction / 21

“lower middle,” “poor middle,” or even just “less educated”; for lowermiddle-class Christians, in contrast, the professionals and managers also belong to the “rich” or “elite.” (All these English categories have Tamil equivalents.) Caplan emphasizes, too, that these alternative senses of social class do not depend solely on occupational and economic criteria. Thus upper-middle-class Christians not only have better jobs, and more money and property, than people below them; they also distinguish themselves by their university education, English skills, more prestigious consumption patterns, and a different religious outlook. Marriages, moreover, normally take place between men and women of the same class status.41 Dickey, who worked with Hindus in Madurai, found that the salience of middle-class identity steadily grew after liberalization began. By the end of the 1990s, many people in the city were calling themselves “middle class” (in English) or “middle people” (in Tamil). Like Caplan, although some details differed in Madurai, Dickey found that upper-middle and lower-middle perceptions of the class system varied.42 In general, though, which is why Caplan and Dickey are cited, the categorical distinctions they clearly describe are essen­ tially the same as those used by Tamil Brahmans when talking about themselves as upper- or lower-middle class. This book, as we have said, is about the Tamil Brahmans’ transformation into a modern, urban middle class—a social class that is also a status group. It is, in brief, about the making of this caste-cum-class, as well as its internal stratification, through the interaction of economic factors, especially occupation, with social and cultural ones, including education, women’s status, and religious orientation.

Research Data and Statistical Absences This book is primarily based on ethnographic research carried out between 2003 and 2008, although it also relies heavily on the scholarship of other authors. Our first research project in 2003–5 investigated the middle class in Chennai and particularly compared IT professionals with managers in engineering manufacturing companies. The second project in 2005–8 focused on the Eighteen-Village Vattimas, members of a small Smarta subcaste whose ancestral villages lie in the Kaveri delta region, south of Chennai. Since Vattimas are atypical in some respects—for example, they stayed in their villages as landlords for longer than many other Tamil Brahmans—we have to be careful in using them as a case study. But this does not pose particularly difficult analytical or methodological problems. Much more difficult is the problem of upper-middle-class bias in the material we collected

22 / Introduction

in both projects, because we have relatively little data on the white-collar, lower-middle class. It would be tedious to qualify the term “middle class” over and over again, but we try to ensure that we do not carelessly generalize about all Brahmans on the basis of evidence about the upper stratum. The hazards of generalization are exacerbated by the dearth of useful statistics on the Tamil Brahmans, which has its parallels throughout India because, as Satish Deshpande points out, the “post-independence backlash against caste” has ensured the “paucity and poor quality” of data. He argues that the lack of sound data allows the urban middle class to deny “the strong link between social privilege and upper-caste status.”43 In fact, a recent statistical analysis clearly demonstrates that the high castes are overrepresented in the professional-cum-clerical occupational class.44 In Tamilnadu, too, the persisting opposition between Brahmans and non-Brahmans hardly allows Brahman privilege to escape notice. All the same, Deshpande’s main point about meager data is correct. The situation may change with the publication of data from the 2012 socioeconomic and caste census. But at the time of writing, mainly because caste populations have not been counted since 1931, there are no statistical data, for Tamilnadu or any other region, on the Tamil Brahmans’ migration patterns, educational and employment profiles, income distribution, household composition, and so on. Public- and private-sector employers, and educational institutions, collect no information on their employees’ or students’ caste identities, unless they are eligible for reservations quotas; sometimes, lists of names are available, but nothing much can be deduced from them because Brahman and non-Brahman names are often the same. This statistical vacuum is in itself a significant sociological fact. Thus, for example, everybody who knows anything about the IT industry in Chennai—and many who don’t—agree that Brahmans are very numerous in it, so that the proportion of Brahman employees is much higher than their share of the general population. Brahmans may mention this to demonstrate their community’s success in a modern, hi-tech industry, their critics may do so to complain about bias in the IT companies’ recruitment pro­ cesses, and company representatives may deny that caste is an issue at all. But the endlessly repeated wisdom about Brahman overrepresentation in IT is really just a factoid, because nobody ever cites any figures to prove it. Exactly the same also applies to all claims about the number of Brahmans in other occupational or educational fields, or indeed in almost any sphere of life. Factoids, in other words, persistently substitute for facts on all sides of the debate about the Brahmans’ position in contemporary Tamilnadu—as is also true for most other castes, but usually less polemically. This situation

Introduction / 23

is completely different from that in 1916, when the Non-Brahman Manifesto could quote firm figures to support its case. Actually, we, too, are sure that Brahmans are overrepresented in the major IT companies (see chap. 3), on the basis of our own and other researchers’ ethnographic evidence, and some survey data from Bangalore.45 We are also sure, on the basis of similar evidence, that Tamil Brahmans are overrepresented in professional and managerial occupations in general, and hence in the upper-middle class as a whole. But let us acknowledge at this early stage that we will be unable to cite many figures to reinforce the qualitative evidence that we present.

Research, Writing, and Collaboration For our first project on the middle class in Chennai in 2003–5, Haripriya Narasimhan (HN) carried out field research for about twelve months in total, mostly through semistructured interviews with people in their offices or homes; Chris Fuller (CJF) worked with her for about two months altogether; and John Harriss was also part of the research team. For the second project on the Eighteen-Village Vattimas in 2005–8, HN conducted fieldwork in Tippirajapuram, a Vattima village, for about six months in 2005–6 and revisited it several times afterward. Her research among Vattimas in Chennai, mainly through interviews, was mostly done in 2007 and 2008 for about six months in total. Vattimas in Kumbakonam and some other towns in Tamilnadu, as well as in Bangalore and Mumbai, were interviewed on short visits to these places at various stages of the project. HN also updated some of our earlier information in 2009 and 2010 during further research visits to India. She went to the outer suburbs of several American cities during two weeks in 2006. CJF accompanied HN for about three months altogether and went to each of these research sites in India and the US. CJF carried out most of the library and archival research. This book has been written by CJF, but we had discussed its contents extensively beforehand while undertaking the research and writing earlier papers. After HN returned from England to India permanently in October 2010, our face-to-face conversations had to stop, but we continued to discuss the book, mainly through e-mail communication. In Chennai in January 2012, HN read a complete draft of the book (except for the conclusion), and we went through it in detail together. Since then, she has read the later revisions. Throughout our work together, we have usually agreed about how to carry out the research and how to write it up. But sometimes we have not and then we have discussed the matter until we reached agreement or, occasionally, found a compromise allowing us to put the disagreement aside.

24 / Introduction

Most importantly, though, the text of this book is the product of our collaboration since 2003 and it represents our joint views. The advantages and disadvantages of doing ethnographic research in one’s own society, rather than among “others,” have been debated for a long time. So too have the problems of bias in highly stratified societies such as India, where the researcher has to try to avoid seeing everything from the perspective of some social groups, rather than others. In an early intervention by an Indian anthropologist, M. N. Srinivas addressed both these questions after being provoked by E. R. Leach’s criticism of his “Brahmanocentric” point of view. Srinivas, who was a south Indian Brahman, insisted that his important concepts of “Sanskritic” versus “non-Sanskritic” Hinduism, and “Sanskritization” (that is, emulation of Sanskritic forms), were not the product of a Brahmanical perspective, although his caste status, he acknowledged, had influenced some of his intellectual preoccupations, such as the significance of purity and pollution, or the workings of the reservations system. On studying one’s own society, Srinivas judiciously remarked that there are both advantages and disadvantages, but the latter must be minimized as far as possible and the ethnographer must obtain some emotional and intellectual detachment, which—at least in India—is always possible by studying a section of society very different from one’s own.46 It is certainly arguable that Srinivas’s Brahman background influenced his understanding of Hinduism and affected his work as a whole more than he recognized. Moreover, since Srinivas first wrote about his own research, numerous other anthropologists of India, both Indians and foreigners, have done so as well. The debate about fieldwork within the discipline as a whole has also become more complex, but sometimes just more contentious. Thus Nita Kumar, in her evocative “fieldwork memoirs” of Banaras, rightly objects to insistent, polemical generalizations about “the Other.” She frankly explains, too, that as a well-educated, middle-class woman whose first language is English, she felt “foreign” in Banaras, where she found that “in meeting a new person seemingly as Indian as you, you may actually confront a culture distant from everything you take for granted.”47 Our own research among Tamil Brahmans has been, unequivocally, a collaboration between an insider and an outsider. HN’s first fieldwork was among poor, lower-caste women in Chennai, but for our research what mattered most is that she is a middle-class Tamil Brahman, an Aiyangar, who was brought up in Chennai, so that she fully belongs to the section of society described in this book. HN found some other Brahmans’ customs unfamiliar, especially the Vattimas’, but minimally so compared with the culturally distant ones discovered by Kumar in Banaras, by Srinivas among

Introduction / 25

the Coorgs, or by numerous other Indians studying social groups to which they do not belong. Especially in Tippirajapuram, HN’s position resembled that of T. N. Madan, who did fieldwork among his own Kashmiri Pandit community in a village and sometimes found informants puzzled by his urban ignorance of basic social norms: “‘Surely you must know?’ would be the usual query.”48 For CJF, as an Englishman, the Tamil Brahman way of life has always been culturally distant, but he periodically surprised people because he did “know” about it as a result of earlier research among Adishaiva (a¯dis´aiva) temple priests in Madurai, even though they are atypical Tamil Brahmans in many respects. During our research, we consistently felt that our complementary differences—insider-outsider, female-male, younger-older—were an asset for reasons so obvious that we hardly need to spell them out: HN understood things, particularly spoken comments, that passed CJF by, but he noticed distinctively Brahman or Indian customs and attitudes that she tended to take for granted; or women spoke much more freely to HN, whereas senior men tended to take CJF’s inquiries more seriously than hers. Probably the most telling consequence of HN’s caste status is that our informants talked freely about Tamil Brahmanhood and were not apparently constrained by CJF’s foreignness because he was assumed to be sympathetic, as we have already implied. Conversely, though, our most serious blind spot is that we have very little firsthand information about nonBrahman perceptions of Brahmans; we did have some non-Brahman informants, especially in the research on Chennai’s middle class, but they almost certainly did not express their views about Brahmans frankly to either of us and, ideally at least, a non-Brahman colleague would have enhanced our study. The climate of caste politics in Tamilnadu makes accusations of bias hard to avoid; nevertheless, we hope that our attempt in this book to make Tamil Brahman society and culture intelligible is not adjudged Brahmanocentric.

Outline of the Book This book, as we have said, is first and foremost about how the Tamil Brahmans—who used to make up a traditional, mainly rural, caste elite—have become a modern, urban, middle-class community. Education and employment are the most critical factors in the story, but we begin, in chapter 1, with life in the villages, where Brahmans resided in their own exclusive quarter, the agraha¯ram, apart from non-Brahmans and Dalits. The Brahmans’ partial separation from the rest of Tamil rural society facilitated their

26 / Introduction

migration to towns and cities, where they moved to take advantage of educational and employment opportunities. Moreover, Brahman village landlords had a detached, unsentimental attitude to land, which helped them to quit their land relatively easily and further facilitated urban migration. These two “push” factors together help to explain why the Tamil Brahmans’ urban migration and transformation into fully fledged urbanites have been unusually rapid and complete. Once they entered the modern educational system and found employment during the colonial period, Tamil Brahmans rapidly established the dominance to which the Non-Brahman Manifesto objected. Brahmans were particularly well represented in law and administration, but they were also entering new professions, such as medicine and engineering. Many others were finding lower-level, white-collar jobs. All these developments meant that the Brahmans joined the south Indian, urban middle class—especially the professional and managerial, upper-middle class—earlier and faster than people from other social groups, and their progressive incorporation into this class continued throughout the twentieth century. Furthermore, Brahmans remain overrepresented in the upper-middle class, partly because downward class mobility has been minimal. This social stability is not primarily caused by persisting caste privilege, but by class reproduction operating within a modern society and economy. Most of the discussion about education and employment in chapters 2 and 3 concerns men; chapter 4 examines the changing position of women. In the mid-twentieth century, prepuberty marriage for Brahman girls ended; their standard of education also rose and, by century’s end, it was virtually equal to boys’. Especially in the family setting, women mostly remain subordinate to men. Nevertheless, whereas the Brahmans’ claim to superiority over other castes partly depended on extreme gender inequality in the past, the opposite is now more or less true, and the near parity between the sexes in education is important for the status of the Tamil Brahman middle class, particularly the upper-middle class. Chapter 5 is a comparative ethnography of urban ways of life in Chennai, Bangalore, Mumbai, and the outer suburbs of American cities. It looks at how Tamil Brahmans evaluate these different localities and how they compare them with villages, particularly with respect to caste-based social control, which is a pervasive concern, especially for women. This chapter also explores the impact on people of the American dream, whose promises of freedom and opportunity are counterbalanced by worries about overseas emigration damaging family ties and children’s socialization.

Introduction / 27

Religion and culture are the topics of chapter 6. The Brahmans’ role as the custodians of Sanskritic Hinduism has always been salient for their caste status. In the early twentieth century, south Indian Carnatic music and the dance style known as Bharatanatyam (bharatana¯t·yam), which both have a strongly religious ethos, were mainly fashioned into their current “classical” forms by Tamil Brahmans in Madras. Today, Brahmans dominate the two arts and middle-class Brahman girls are strongly encouraged to learn them. One result of these developments is that the Brahmanical, Sanskritic “great tradition” incorporates classical music and dance, as well as religion. This great tradition, which metonymically represents the “high culture” of upper-middle-class Tamil Brahmans, is vital in sustaining their collective identity and constituting them as a social class and status group. The concluding chapter 7 begins by comparing Bengal, Bombay, and Madras during the colonial period. In comparison with Brahmans in either Bengal or Maharashtra, Brahmans in Madras became a more dominant group within the new urban middle class, and they were not seriously challenged until the rise of the non-Brahman movement after 1916. Although they reacted more slowly to modern ideas and institutions than the highcaste middle classes of Bengal and Bombay, Tamil Brahmans eventually adapted more easily and saw modernity as less threatening to their social and cultural traditions. In the long run, too, Tamil Brahmans have come to see themselves as both fully modern and authentically traditional, although that self-perception certainly does not make them unique in contemporary India. But because Tamil Brahmans almost all belong to the urban middle class, the caste’s “traditional” values and practices are now intertwined with modern middle-class ones. Furthermore, Tamil Brahmanhood and middle classness have become mutually constitutive of each other, so that an isomorphism exists between them. Although relevant comparative evidence on other communities in India is fairly thin, it mostly shows that Brahmans have generally been more fully transformed into modern, urban, middleclass groups than non-Brahman, dominant landowning castes have been. Partly owing to their marginality within Tamil society, that transformation has been unusually far-reaching among the Tamil Brahmans, especially for the upper-middle class, whose privileged status and distinctive identity have been further reinforced by developments since economic liberalization. For all these reasons and others discussed in this book, Tamil Brahmanhood, far from fading into anachronistic unimportance, still matters a great deal in the modern world.

Map 2.  Kaveri Delta, Tamilnadu

Chapter One

The Village Caste, Land, and Emigration to the City

The delta of the river Kaveri in central Tamilnadu (see map 2) is one of the state’s most fertile and beautiful regions. The Kaveri and its subsidiary the Kollidam, as well as numerous smaller rivers, streams, and irrigation channels, flow through the flat terrain of the delta and provide the water for its crops, of which the most important is rice. Much of the delta’s beauty comes from its extensive patchwork of paddy fields, coconut groves, banana plantations, and shady trees, which ensure that even in hot, rainless months the land retains a greenness that disappears from Tamilnadu’s extensive dry plains. The natural charm is enhanced by an architectural heritage that includes some of India’s biggest, oldest, and most elegant temples, notably those built by the Cholas, whose great kingdom dominated the region from the ninth to the twelfth centuries. Of course, there is plenty of ugliness and pollution too, especially in the fast-growing towns, but in the countryside the delta’s attractiveness remains largely unspoilt. Most of the Kaveri delta lies in the old Tanjore District, which is now divided into Thanjavur, Thiruvarur, and Nagapattinam Districts, although it stretches upriver westward to Tiruchirappalli (Trichinopoly; Tiruchi or Trichy, colloquially) and beyond. Mainly owing to its fertility, the delta’s population density has always been high; the region has also long been characterized by extremes of inequality, especially between the landed rich and landless poor and members of the highest and lowest castes. According to the 1871 census, Brahmans comprised 6.4 percent of the old Tanjore District’s population, more than double their proportion in the Tamil region as a whole and higher than in any other district.1 A report on Tanjore in 1805 stated that 28 percent of its 62,048 mira¯sida¯rs or landowners were Brahmans;2 these figures may not be entirely reliable,

30 / Chapter One

but they are roughly consistent with data from the early twentieth century showing that 38 percent of the wealthiest mira¯sida¯rs in the district were Brahmans.3 At that time, according to David Washbrook, “the mirasidar elite [in Tanjore] ruled its agrarian dependents with a rod of iron and established a great social distance between itself and them.”4 As we shall see, that distance had been reduced, but had still not disappeared in the early post-Independence years. The Kaveri delta’s terrain is closely covered in villages, which can be divided into two main categories, according to whether Brahmans or nonBrahmans were the main landowners enjoying economic and political dom­­ inance. In “Brahman villages” there is an agraha¯ram, a street or quarter that is or was exclusively inhabited by Brahmans. The agraha¯ram stands apart from the main settlement area occupied by non-Brahman castes. In non-Brahman villages, of course, there is no agraha¯ram and usually no Brahman residents except, perhaps, for one or two families of priests. Separated from both Brahman and non-Brahman settlement areas are the ceris (“cheri”), the “colonies” exclusively inhabited by untouchable Adi Dravidas or Dalits. In some Brahman villages, the ceris are near the main village and are in practice part of it; in others, they are a considerable distance away and are regarded as distinct settlements with their own names. Apart from the Kaveri delta, Tamilnadu’s principal wet-zone regions are the smaller Palar valley (mostly in modern Kanchipuram District) in the north, and the Vaigai valley (in Madurai, Sivaganga, and Ramanathapuram Districts) and the Tambraparni valley (in Tirunelveli District) in the south. In these last three valleys, Brahmans were less numerous, rich, and powerful than in the Kaveri delta, but they were similarly dominant in their own Brahman villages, also partitioned into agraha¯rams, non-Brahman residential areas, and Dalit colonies. Although the great majority of  Tamil Brahmans used to live in the countryside, there were always considerable numbers in urban areas. The ancient temple city of Madurai, for instance, the political and cultural capital of the southern Tamil country, has had a strong Brahman presence for centuries and many Brahmans formerly lived in agraha¯rams near the Minakshi (Mı¯na¯ks·¯ı) temple in the city center or beside the river Vaigai. In all the older towns and cities, however, including small market towns like Tindivanam where Satyamurti lived before he migrated to Madras, as described in the introduction, there were agraha¯rams and settled Brahman communities. The Tamilized Sanskrit word agraha¯ram is one of several terms that referred to villages and land donated to Brahmans by Hindu kings between the medieval and late precolonial periods; the most widely used of these

The Village: Caste, Land, and Emigration to the City / 31

terms was probably brahmadeya.5 Early British reports on land rights refer to agraha¯rams as villages granted to Brahmans;6 by the mid-nineteenth century, H. H. Wilson’s glossary of revenue terms defined an agraha¯ram as a village occupied by Brahmans held rent-free or at a reduced assessment rate.7 By the early twentieth century, however, the term agraha¯ram appears to have acquired its modern dictionary meaning as an exclusively Brahman street or residential quarter. Agraha¯rams also exist in coastal Andhra Pradesh, Karnataka, and Kerala. In Andhra, only “orthodox” Vaidika (Vedic) Telugu Brahmans have lived in agraha¯rams, like the two in the Godavari delta where David Knipe did research in 1980–95.8 “Secular” Niyogi Brahmans, by contrast, did not have their own agraha¯rams and in the Godavari District in the early 1900s they “[did] not mind dwelling side by side with [non-Brahmans] and [did] not always have their own distinct streets.” F. R. Hemingway, who also wrote the Tanjore gazetteer, noted that these Telugu Brahmans did not “hold themselves as severely aloof” from non-Brahmans as Tamil Brahmans further south.9 An agraha¯ram in disarray in Karnataka is the setting for U. R. Anantha Murthy’s famous novel Samskara, but data about agraha¯rams in this region are very thin.10 In Kerala, Palghat Brahmans—who migrated from the Tamil country centuries ago—lived in agraha¯rams in parts of Cochin and Malabar, as did Tamil Brahmans settled in southern Travancore, which only became part of Tamilnadu after Independence.11 Colonial reports on land rights and revenue from the nineteenth and early twentieth centuries, as well as census reports and gazetteers, sometimes mention Brahman village streets and agraha¯rams in Tamilnadu and elsewhere in south India, but never with much ethnographic detail.12 Dusi and Gangaikondan, in the Palar and Tambraparni valleys, respectively, were the first two Brahman villages in Tamilnadu to be investigated sociologically, originally in 1916–17 and again in 1936–37, but not much detailed information about their agraha¯rams was collected.13 Quite possibly, “the separatist practices of Brahmans”—including residential segregation in agraha¯rams—were strengthened considerably when colonial rule consolidated the caste system and reinforced the Brahmans’ secular power in the Tamil country.14 By the early twentieth century, agraha¯rams were a contentious issue for the growing non-Brahman movement, which strongly objected to restrictions on public access and sought to “de-sacralize agraharam space.”15 Most probably, though, very similar agraha¯rams had existed for a long time, even though we cannot take it for granted that the same “traditional” rules and restrictions described by anthropologists in the midtwentieth century had always been in place.

32 / Chapter One

Brahman Villages in the Mid-Twentieth Century The most detailed ethnographies of Brahman villages in the mid-twentieth century were written by Kathleen Gough and André Béteille, whose material provides an important baseline for the evaluation of later change. Gough carried out fieldwork in Kumbapettai, northeast of Thanjavur, in 1951–52 and returned there in 1976.16 Béteille studied Sripuram, northwest of Thanjavur, in 1961–62.17 The majority of the Brahmans in Kumbapettai in the 1950s were Brahacharanam Smartas (Aiyars) belonging to a particular local subcaste. Compared with some mira¯sida¯rs in the Kaveri delta, the Kumbapettai Brahmans were not particularly wealthy. In 1952, their agraha¯ram, which was origi­ nally built in the 1780s, consisted of one street lined on both sides by rows of interconnected houses; by local standards, these houses were spacious and comfortable. The agraha¯ram contained thirty-five occupied and fourteen unoccupied houses, with a total population of 174 people. (One Telugu Brahman priestly family lived outside the agraha¯ram.) A temple dedicated to Ra¯ma (Vishnu) stands at one end of the agraha¯ram, farthest from its entrance and the main road; near the other end, there is a Shiva temple. Kumba­ pettai’s non-Brahman area, across the main road from the agraha¯ram, was inhabited by 285 people belonging to eighteen separate castes, and 354 Pallars lived in its outlying ceri.18 When the village lands were surveyed in 1827, Brahmans owned virtually all of them, although individual family shares varied greatly.19 By the early twentieth century, although some Brahman families had prospered, most had not and as a whole the community had lost wealth and land, so that by 1952 several households owned no land.20 In 1871, 412 Brahmans lived in Kumbapettai and made up almost half its population.21 They began to emigrate to urban areas before 1900, so that the agraha¯ram’s population started to decline. In 1952, Gough found that many of Kumbapettai’s Brahmans had left to take up salaried employment, although quite a lot of them still owned some village land that they rented out to non-Brahman tenants.22 Sripuram’s agraha¯ram, which was established in the mid-nineteenth century, consists of one street parallel to the main road; on the other side of the road runs the river Kaveri. The agraha¯ram street is longer than Kumbapettai’s and the houses, also arranged in rows on each side, are mostly bigger; in 1961–62, there were ninety-eight houses, of which ninety-two were occupied by 341 Brahmans. Vadagalai Sri Vaishnavas (Aiyangars) made up over half the population, with the rest belonging to range of mainly Smarta subcastes, including Telugu Brahmans.23 A Vishnu temple stands

The Village: Caste, Land, and Emigration to the City / 33

at one end of the agraha¯ram; not far away, but beyond the Brahman street, are an impressive, ancient Shiva temple beside the main road and the village goddess’s temple. Behind the agraha¯ram, away from the road, lies the non-Brahman area, where 688 people from twenty-six separate castes and subcastes lived, and beyond it are the ceris inhabited by 332 Pallars and thirty-nine others from three separate Adi Dravida castes.24 In the past, Sripuram’s Brahmans—together with one family related to the former Ma­ ra­tha rulers of Tanjore—owned almost all the land and some of them had been very wealthy mira¯sida¯rs. By the 1960s, many Brahmans had emigrated to towns and cities; some landowners rented out their land to non-Brahman tenants, but many sold it.25

The Hierarchy of Caste and Class On Tamil New Year’s Day in mid-April 1952, Gough watched the celebrations to mark the beginning of the agricultural year in Kumbapettai. In the evening, all the Brahman mira¯sida¯rs assembled in front of Rama’s temple at the end of the agraha¯ram street. The ritual in the temple was paid for each year in turn by one Brahman family group, whose members were all expected to be present, along with their non-Brahman tenants and laborers, who had to stand outside the temple gates. No Adi Dravidas were invited because they were not allowed to enter the agraha¯ram. A Brahman temple priest performed worship for Rama and his consort Sı¯ta¯ before images of the god and goddess were taken in procession up and down the agraha¯ram. When they had returned to the temple, the most senior domestic priest, who was also a landlord, read out the almanac for the new year in Sanskrit and Tamil, and concluded by worshipping Vina¯yaka (Gan ·es´a), the god of beginnings and obstacles. Later in the evening, another ritual took place at the temple beside the main road dedicated to the village goddess, Uridaichi¯ ritaicciyamman), who looks after Kumbapettai and its people. yamman (U · ¯ This ritual, which had been started seven years earlier by the Brahman pres­ ident of the local council or panchayat, was paid for by the non-Brahman village headman. The temple’s non-Brahman priest performed worship for the goddess and various minor deities, and then waved before them a flaming candelabra, brought from Rama’s temple by his priest. Some of the village’s most prominent Brahmans, with their wives and children, sat next to the goddess’s shrine, while leading non-Brahmans and their families stood at the back. When the ritual had finished, the headman distributed food, first to the Brahmans who got most of it and afterward to the nonBrahmans. Once again, Adi Dravidas were not invited.26

34 / Chapter One

The rituals of New Year’s Day were fairly minor events in Kumbapettai’s annual round. Yet they displayed with a special clarity the principal axis of its social structure, namely, the tripartite distinction between the Brahman landlords who placed themselves closest to the deities in both temples, the non-Brahman tenants and laborers who stood behind them, and the Adi Dravida laborers who were absent. Accordingly, the ritual at Rama’s temple, sponsored by a Brahman and performed by a Brahman priest, preceded the one at the goddess’s temple, sponsored by a non-Brahman and performed by a non-Brahman priest, although Brahmans were actually closer to the deities on both occasions. Through these rituals, which expressed the hierarchical norms of their society, Brahman and non-Brahman villagers sought the blessings of their deities—but they did not think it necessary to invite the untouchable Adi Dravidas or worship their deities. The Brahmans’ alleged superiority depended on their status as mira¯sida¯rs, as well as their high rank in the caste system. Brahman landlords never worked in their fields and, if they did not rent out their land to tenants, they confined themselves to managing and supervising cultivation. Manual work, as we discuss below, was and is consistently seen as demeaning. Agricultural labor, therefore, was all carried out by the lower castes and the hierarchy of caste-cum-agrarian class was repeatedly made visible in one of the Brahman villages’ most characteristic scenes. In Kumbapettai, writes Gough, “in each agricultural season one could see Pallars or Non-Brahmans toiling in the fields, the men stripped naked except for a narrow loin cloth, the women in old, looped saris with bare arms and legs. Standing on bunds beside the paddy fields or sitting on rope beds, benches, or chairs under shady trees, would be several Brahmans wearing white shawls and long lower white cloths. . . . [These] landlords would stand at intervals each near his own field, prompting and supervising.”27 Even more strikingly, when the transplanting of paddy seedlings began—a job that was (and still is) done only by women—a landlord stood near the fields and before starting work each Pallar woman laborer “approached him with a small bundle of seedlings in each hand and bowed three times, touching the ground with her seedlings about six feet away from him.”28 In just this little interaction between a Brahman landlord dressed in white, standing in the shade, and his female Untouchable laborer in a worn-out sari, about to toil under the sun, the extreme inequality of caste and class, and gender too, that prevailed in Tamilnadu’s Brahman villages in the 1950s was starkly displayed. In Brahman villages, the agraha¯ram’s exclusiveness, and the caste hierarchy more generally, were expressed and reinforced by codes of ritual purity and pollution. In 1952, all residents of Kumbapettai’s agraha¯ram were Brah-

The Village: Caste, Land, and Emigration to the City / 35

mans and any attempt to sell one of the houses to a non-Brahman would have been strenuously resisted.29 The purity and pollution rules observed in the village meant that people were normally prohibited from touching or even closely approaching those belonging to a higher caste, and these rules particularly applied to contact between Brahmans and members of the lowest castes. “Clean,” higher-ranking non-Brahmans could enter Brahman houses, but not their kitchens, and since the 1920s some Brahmans had employed non-Brahman women as servants. “Polluting,” lower-ranking non-Brahmans could walk along the agraha¯ram street, but not enter the houses; Adi Dravidas were forbidden to enter the street and came to the back door of Brahman houses to collect their wages or get instructions from their masters.30 When in the agraha¯ram, non-Brahmans hitched up their anklelength cloths to leave their lower legs bare and they could not wear footwear, although in 1952 some non-Brahmans were challenging these rules by walking to the post office located inside the agraha¯ram, dressed in shirts and with their legs covered; a couple of assertive students had even dared to wear shoes.31 Adi Dravidas, of course, could not enter the Rama and Shiva temples and they did not, even after they were legally permitted to do so by temple-entry legislation in the 1940s.32 It is unclear whether non-Brahmans could freely enter the two temples, but some probably went to Shiva’s more accessible temple and next to none to Rama’s inside the agraha¯ram. Brahman men could enter non-Brahman houses—as some did to have sex with non-Brahman women, for instance—but they did not eat there and had to bathe before reentering their own homes. Brahman men did not go into the ceri, although this was not only to avoid pollution but also because Adi Dravidas believed that if they went there, the whole colony would suffer serious misfortune; the same belief prevailed in Sripuram too.33 In Kumbapettai, Adi Dravidas could enter the streets, but not the homes of nonBrahmans; the latter could enter the Adi Dravidas’ streets, but would not go into their homes to avoid becoming polluted.34 Brahman women lived almost entirely within the agraha¯ram and only went out to bathe in the nearby tank or to walk to the main road, accompanied by male relatives, to catch a bus or visit a temple. Women never went to non-Brahman houses and had never even seen the Adi Dravidas’ colony. In fact, married Brahman women rarely even left their own homes, although neighbors could converse through holes in house walls; unmarried women and children were less restricted in their movements within the agraha¯ram. In general, the rules of purity, pollution, and caste interaction—both inside and outside the home—were much stricter for Brahman women, especially married women and widows, than for men. When suffering from menstrual

36 / Chapter One

or birth pollution, women were secluded in a room at the rear of the house or in the backyard; they were not allowed to touch other adults (though children were exempt from pollution) and were prohibited from entering the kitchen and from preparing or touching the food of others.35 Brahman children, both boys and girls, received elementary education from a Brahman tutor at his house in the agraha¯ram. Non-Brahman and Adi Dravida children, though not many of them, attended government elementary schools in nearby villages.36 Among Kumbapettai’s Brahmans, there was a lot of quarrelling, factional dispute, and personal rivalry. Yet most of them were related by kinship, they regularly cooperated to maintain their control over village affairs, and they could display considerable internal solidarity, as well as unity against outsiders.37 Brahmans also maintained a strong sense of exclusiveness toward and superiority over all the other, lower castes. As Gough makes clear, lowcaste attitudes toward Brahmans were ambivalent. Thus in private, nonBrahmans and Adi Dravidas “selectively approved certain features of the religious and caste ideology of the Brahmans, but challenged others,” usually in relation to what might be advantageous for them.38 In public on the other hand—though change would soon come—the Brahmans’ superior status and prerogatives, as well as their economic and political dominance, were still largely undisputed in Kumbapettai in the early 1950s. Béteille’s ethnography is less detailed than Gough’s, but hierarchy in Sripuram was clearly much the same as in Kumbapettai in all important respects. Our discussion of Sripuram is therefore confined to a few distinctive points. Sripuram’s agraha¯ram remained an exclusively Brahman space in the early 1960s and the sole exception was Béteille himself, as he was constantly reminded. The ostensible reason why no non-Brahmans had been allowed to live in the agraha¯ram was that they eat meat and fish, whose noxious smells and discarded bones would make the street filthy and polluted. But, according to Béteille, this was not really the most significant objection; after all, vegetarian non-Brahmans were excluded from the agraha¯ram as well. Rather, the admission of non-Brahmans “would upset a pattern of life which has been established over the centuries” and would be “a threat to [the Brahmans’] tradition and their cultural heritage.” For the Brahmans, “the agraharam is not only a physical unit; it is a community and a way of life.”39 The rules governing access to the agraha¯ram in Sripuram closely resembled those in Kumbapettai ten years earlier, as Béteille’s account of restrictions on his own movements illustrates.40 But more dramatically than in Kumbapettai when the students walked to the post office wearing footwear,

The Village: Caste, Land, and Emigration to the City / 37

in Sripuram a large group of non-Brahmans and Adi Dravidas, led by a village political leader, marched through the agraha¯ram on January 26, 1950, the day on which India became a republic. They went all the way to the gates of the Vishnu temple inside the agraha¯ram, which was the preeminent temple for the Brahmans and was never visited by non-Brahmans. This unprecedented event “was not only ritually polluting, but it also brought social humiliation” to the Brahmans, who did nothing about it because they realized that “both the law and the balance of political power were against them.”41 Nevertheless, although Brahman supremacy was plainly crumbling in Sripuram, it was doing so slowly and many features of the old hierarchical system were still in place in the 1960s, as they were in Kumbapettai a decade earlier. Let us now turn to Béteille’s theoretical argument about hierarchy in Caste, Class, and Power, which was inspired by Weber’s analysis, published in English as “Class, Status, Party.”42 Béteille starts with the premise that in Sripuram in the past the caste structure and the agrarian class system were closely consistent with each other, because the tripartite division among Brahmans, non-Brahmans, and Adi Dravidas was closely correlated with that among landlords, tenants, and laborers. The distribution of political power was segmented in the same way as well. Since the late nineteenth century, however, the structures of caste, class, and power have become increasingly divergent as the correlation between them has steadily weakened. The Brahmans’ loss of land and their emigration from the village; improving economic opportunities for other castes; the non-Brahmans’ growing political assertiveness; land-reform legislation after Independence to improve the condition of tenants and laborers, as well as to place a ceiling on the size of landownings: these and other changes all served to undermine the old social order and, in particular, to dissolve the conjunction of caste and class in Sripuram. Béteille presents a powerful thesis about the changing relationship between caste, class, and power in Sripuram that continues to inform debate about contemporary change in Indian villages. On the whole, it is more persuasive than Gough’s analysis of change in Kumbapettai in her two books, partly because her ethnographically detailed attention to caste undermines the Marxist class theory on which she ostensibly relies.43 Yet Béteille tends to overstate the extent to which his ideal-type categories of caste, class, and power can match the social reality. Thus, in particular, the non-Brahmans belonged to twenty-six different castes of variable status and occupation and, although most were tenants, they did include some small landowners and laborers.44 Among non-Brahmans, therefore, there was more class

38 / Chapter One

variation than among Brahmans or Adi Dravidas and the correspondence between non-Brahmans and tenants is rather overdrawn. Furthermore, the Adi Dravidas’ position was scarcely changing, for they were mostly still land­ less laborers as they had been in the past. Particularly salient for this book, too, is that Béteille’s argument is much more convincing about the Brahmans, for whom the conjunction of highcaste status and landlordism was plainly breaking down, than about the relatively heterogeneous non-Brahmans or the permanently exploited Adi Dravidas. At the same time, though, the Brahmans were being transformed from an agrarian class of landlords into an urban middle class, the product of their migration to the towns and cities. Indeed, even more so than Béteille implies, their large-scale urban migration has probably been the single most important cause of the twentieth-century decline of Brahman status, wealth, and power in the countryside. Moreover, although the increasingly powerful non-Brahmans did progressively displace them, it was for the most part the Brahmans themselves who first decided to quit their village agraha¯rams and thereby bring about the demise of the steeply hierarchical village social structure.

Brahman Villages in the Early Twenty-First Century In Kumbapettai in 1976, the number of Brahmans resident in the agraha¯ram had fallen slightly, but more significant was that eight non-Brahman families were living there and all non-Brahmans could freely move along its street. Indeed, as a result of a political dispute in the village in the 1970s, the Kallars showed their contempt for the Brahmans by driving their cattle through the agraha¯ram.45 By 2005, when we visited the village, the agraha¯­ ram housed only six Brahman families, who were outnumbered by nonBrahmans. In Sripuram, non-Brahmans began to move into the agra­ha¯ram around 1980, but in 2011 it still contained only twenty-five non-Brahman households compared with sixty-five Brahman ones (with seventeen houses unoccupied).46 Hence Sripuram belongs to the relatively small number of agraha¯rams in which Brahmans remain preponderant. In some rare agraha¯rams, Brahmans remain completely dominant; for example, in Kal­ yanapuram, which is close to Sripuram and locally renowned as a Brahman stronghold, most of the ninety houses in the agraha¯ram are occupied by Sri Vaishnava Brahmans, while only about six non-Brahman families live there and there are not many empty buildings. Almost everywhere in Tamil­ nadu, however, Brahmans have been continually leaving their agraha¯­rams,

The Village: Caste, Land, and Emigration to the City / 39

so that their population in the rural areas has fallen to around 4 percent of what it was in 1871 (see appendix). In some agraha¯rams, emigration occurred earlier and more fully than in Kumbapettai or Sripuram. Thyagasamuthiram, for example, is another Brahman village in the Kaveri delta, located west of Kumbakonam. In 1871, it had a Brahman population of 116.47 In 1957–58, its single-street agraha¯ram was still exclusively Brahman, but had only fifty-two residents belonging to thirteen households, for thirty to forty households had disappeared in the preceding half century or so and most of the agraha¯ram was in ruins.48 By 2007, when we visited Thyagasamuthiram’s agraha¯ram, only two Brahman households were left and other houses were occupied by non-Brahmans, although there were also a lot of ruined buildings; moreover, the Vishnu temple at one end of the agraha¯ram had completely collapsed and part of the Shiva temple at the other end had fallen down as well. The available evidence suggests that the scale of emigration from Thyagasamuthiram’s agraha¯ram is not particularly unusual anywhere in Tamilnadu or indeed outside it, such as a village in East Godavari District, Andhra Pradesh, where the majority of Brahmans had emigrated by the late 1960s.49 The overall picture is therefore mixed: a few agraha¯rams are still mainly occupied by Brahmans, but in many agraha¯rams Brahmans are a minority, often a small one, and in many others there are just one or two Brahman households or none at all. In almost all of them, too, ruined houses and other buildings are a common sight. Furthermore, the old restrictions on lower-caste access can no longer be enforced in any agraha¯ram, although they have not always vanished in practice. Much of our own research, as mentioned earlier, has focused on the Eighteen-Village Vattimas. As their name indicates, this subcaste is presumptively defined by eighteen villages in which its members have mainly lived; almost everyone agrees about sixteen of them, but disagrees about the other two, though four places are most often mentioned. The villages, which are listed in table 1 in the appendix, are all located in the Kaveri delta, most of them a few miles east and south of Kumbakonam. Anandatandavapuram, Konerirajapuram, Mudikondan, Sengalipuram, Tippirajapuram and, less certainly, Vishnupuram—or more precisely their agraha¯rams—count as the most important owing to the size of their dominant Vattima populations and the landed wealth of their leading families. Konerirajapuram was the most extensive Vattima agraha¯ram, with 160 houses and a Brahman population of 422 in 1871.50 Konerirajapuram also has an ancient Shiva temple, famous for its enormous bronze Nat·ara¯ja, which is said to have been

40 / Chapter One

built by the Chola queen Sembiyan Mahadevi in the tenth century.51 The temple’s grandeur is testimony, too, to the collective power and prestige of the Brahmans in the past, but today, reflecting their departure and decline, the temple’s fabric is deteriorating; according to an elderly Vattima landlord in Konerirajapuram who showed us round, its rituals have also been much reduced and few devotees now attend them. The other important Vattima agraha¯rams contain between about eighty and 130 houses; in 1871, their population ranged from 589 in Mudikondan to 324 in Anandatandavapuram.52 Today, except in Tippirajapuram (which we discuss below), the Brahman population has fallen by two-thirds to three-quarters in all of them, although the mix of Vattima and non-Vattima Brahmans varies considerably, as does the number of non-Brahman agraha¯ram residents. The Vattima population was always smaller in the less important agraha¯rams, except for Tediyur, and today they each contain only a dozen Vattima households, or sometimes only a handful, and there are not many other Brahman households either. In most agraha¯rams, too, a lot of old houses are empty or ruined, although others have been sold or rented to non-Brahmans. Tediyur is exceptional, partly because it is as big as the more important Vattima agraha¯rams, but mostly because nearly one hundred Vattima houses have been sold to Muslims, who are regarded by the majority of  Tamil Brahmans as unacceptable residents of an agraha¯ram (in contrast with Christians, who are generally tolerated). In Tediyur, though, as the Brahmans and Muslims we met were keen to emphasize, each group attends the other’s ceremonies and relationships between them do indeed appear cordial. The Vattimas started to emigrate to towns and cities in sizable numbers later than many other Brahmans living in the Kaveri delta and only a minority left before the early 1950s. Nowadays, though, most people of working age have gone, so that nearly all the Vattimas still living in villages are el­ derly; some, who continued to cultivate their land, never left, while others have retired there after working in urban areas. Tippirajapuram, which we studied in detail, consists of a large agraha¯ram on one side of a main road and a non-Brahman settlement area on the other. The two ceris linked to Tippirajapuram—both called “Pallar streets” although most of their inhabitants are Paraiyars—are about half a mile and one mile away, respectively, further from the main village site than the ceris in Kumbapettai and Sripuram. Tippirajapuram’s agraha¯ram is unlike the other Vattima agraha¯rams because Brahmans, both Vattimas and nonVattimas, still make up a large majority of its population. It is also unusual because, despite some empty houses, none has fallen down and no buildings are in ruins. Tippirajapuram’s agraha¯ram is a square formed of North,

The Village: Caste, Land, and Emigration to the City / 41

South, East, and West Streets; a short extension of South Street, Sannidhi Street, leads to the Vishnu temple in the southwest corner. In 1871, 403 Brah­mans lived in Tippirajapuram.53 In 2005–6, the agraha¯ram, which contained 129 houses, had a population of 285 made up of eighty-five Vat­timas (thirty-four households), 148 other Brahmans who were mostly Aiyars (forty-eight households), and fifty-two non-Brahmans (eleven house­ holds). (No information was available on two households; thirty-four houses were empty.) Brahman numbers have therefore fallen by 42 percent since 1871, considerably less than in other Vattima agraha¯rams. On South Street (plus Sannidhi Street), there were twenty Vattima households, ten of other Brahmans and one non-Brahman. South Street (together with the adjacent ends of East and West Streets) is the preeminent section of the agraha¯ram and a key feature of  Tippirajapuram is the opposition between it and North Street, where the four Vattima households were outnumbered by twenty-seven other Brahman and five non-Brahman households. South Street’s preeminence is reinforced by the location of the Vishnu temple, which is visited by very few non-Brahmans and no Dalits. Tippiraja­ puram’s Shiva temple is in the opposite corner of the agraha¯ram; its side entrance is on North Street, but its front entrance is on the village’s main road. Many non-Brahmans (and possibly some Dalits) worship in Shiva’s temple. One reason why so few non-Brahmans go to Vishnu’s temple is because they have to walk along South Street to reach it. Local people correctly assume that many Brahmans living on this street would prefer nonBrahmans to avoid it, even though they cannot openly object nowadays, and most non-Brahmans do not want to be discomfited by disapproving Brahmans watching them from the front of their houses. As far as most Vattimas are concerned, South Street is their street and because they are the ancestrally dominant group in the agraha¯ram, it is right and proper that this street should be kept as traditionally Brahmanical as possible. The nonVattima Brahmans on North Street harbor considerable resentment against the South Street Vattimas and have mixed feelings about whether North Street is a better or worse place to live. Some people prefer it because it has a more heterogeneous population and they can be more relaxed about purity and pollution rules, whereas others dislike it for those same reasons. One Brahman woman living on North Street even described it as “adulterated” by non-Brahmans. The difference between the two streets is symbolized by their surfaces, for South Street is made of sand and North Street is a metaled road. Partly because it is a through route, North Street was surfaced with tar several years ago, but then the local panchayat decided to do the other streets as well.

42 / Chapter One

However, an influential Vattima successfully objected to the plan to alter South Street. In 2007, the panchayat actually started to tar West Street from its northern end, but when the road workers neared the southern end, the Vattimas living in the houses there got the work stopped. Thus the newly tarred surface covers only part of West Street and none of South Street (or East Street) has been altered. Sand is preferred partly because it is just traditional, but also because it deters outsiders’ vehicles and supposedly encourages people to walk respectfully through the agraha¯ram. All sorts of people use North Street as a thoroughfare, but the incidents that have caused outrage among both Brahmans and non-Brahmans are Dalit funeral processions accompanied by the sound of drums. Many years ago, such processions would have gone along the riverside path behind the houses on North Street, but it is now blocked with rubbish. It is, however, the drumming, which apparently started in the 1990s, rather than the processions themselves, that is regarded as most outrageous. A Brahman living in North Street was discussing the matter with a non-Brahman who is influential in Tippirajapuram; the latter said that surely no Dalit would play a drum in an agraha¯ram, but the Brahman retorted that he had heard someone in the procession shout out for the drums to be played, which he assumed—probably correctly—to be deliberately provocative. When asked by another Brahman why he had said nothing, he replied: “Those days are over. Times have changed. I have no power to say anything.” A key feature of these incidents is that both Brahmans and non-Brahmans are equally hostile to Dalits, who are accused of behaving aggressively, but they also believe they can no longer do anything about it. On the other hand, although this is speculation, given that North Street is a metaled thoroughfare with a fairly mixed population, the Dalits themselves may not regard it as a proper agraha¯ram street. On some land close to the Shiva temple in North Street, which had been sold to a non-Brahman, a developer planned to build a “marriage hall” to rent out to wedding parties. Influential Vattimas successfully blocked this plan, objecting that crowds of non-Brahmans would come to the hall and cook meat there for wedding feasts (although in fact they normally serve vegetarian food on these occasions). The condition of South Street, however, concerns the Vattimas much more and they have ostensibly agreed to sell none of its houses to non-Brahmans; similar agreements have been made in Sripuram and Kalyanapuram. Sceptical Brahmans, in Tippiraja­ puram and elsewhere, say that such agreements cannot last, because most people selling agraha¯ram houses want the best price and do not care about a buyer’s caste. Nonetheless, a non-Brahman on West Street did say that his

The Village: Caste, Land, and Emigration to the City / 43

attempt to buy a Brahman’s house on Sannidhi Street had been thwarted by the owner’s mother, who flatly declared that they would not sell to nonBrahmans, even if they were vegetarians. Sometimes, to mollify Brahmans, non-Brahmans who buy agraha¯ram houses claim that they will not cook meat in them so that they remain pure; in Tediyur, even some Muslims are said to observe this restriction. In agraha¯rams, most houses largely retain their original design, although some have been modified. A traditional Tamil house (Brahman or nonBrahman) is normally long and narrow; it has a relatively small frontage on the street, but extends a long way behind the front door.54 There are, of course, some variations in the internal structure, but this is a brief description of a typical Vattima house. At the front is a veranda; the door behind the veranda opens on to a courtyard partly open to the sky, surrounded on three sides by a hall. At the back of the hall is a cupboard—the domestic shrine or “pu¯ ja¯ [worship] shelf”—containing small images or pictures of Hindu gods and goddesses. Behind the hall is the kitchen and behind that is the “second portion,” a storeroom for food known as the “daughter-inlaw’s room” because it is where she used to spend most of her time. At the rear of the house is a backyard with a well, bathrooms, and an area used for cooking large festive meals. Behind the backyard are a cowshed and a garden extending to the far end of the site, where there is a door to which Dalit laborers used to come. In a large house, there may also be extra rooms off the hall, a separate shrine room, a separate granary, and so on; the biggest houses have an upper story as well.55 In the past, the space of a Brahman house was clearly graded according to its relative purity, as well as markedly gendered. The veranda, outside the front door, was the least pure area, where visitors, including non-Brahmans, were received. In Vattima houses in Tippirajapuram until fairly recently, men mainly stayed on the veranda and in the hall, and only went into the kitchen to eat, whereas women remained in the kitchen, the storeroom, and the back half of the house. The kitchen and the hall containing the domestic shrine formed the center of the house, in which a specially heightened state of purity (mat·i) should be maintained. This central portion was closed to anyone of lower status, including members of the family temporarily suffering pollution, particularly menstruating women. During her period, a woman is secluded for three days and takes a purificatory bath on the fourth day; she resumes her normal status and can reenter the kitchen on the fifth day. In the past, menstruating Vattima women were secluded in the cowshed or the backyard and could only leave through the back door. Today, the rules about purity and male and female space in Vattima

44 / Chapter One

houses are much less strictly observed than they were. Thus women as well as men now frequent the hall and courtyard, and sit on the verandas chatting and watching people go by, but, as older women recall, this never happened a generation or so ago. Most women regard their increased freedom of movement as a definite improvement on the past. Menstruating women are no longer confined to the rear of the house and can sit in the courtyard, although they do not cross the threshold into the street unless they have to. Neighbors usually know when women are observing menstrual pollution, so that it is not a private matter. Nowadays, girls attend school and women go to work during their periods, and they leave the house by the front door; in most houses, indeed, the back doors are now permanently shut and the lanes behind them are often blocked by rubbish. Many Vattimas today—in common with other Brahmans—also have non-Brahman servants who enter the kitchen, which they formerly did not, and they no longer restrict non-Brahman visitors to the veranda. In their own homes, Vattimas and other Brahmans can of course decide all these questions for themselves, but it is significant that how they observe purity and pollution rules is now mainly a private, domestic concern, whereas they could formerly enforce the rules throughout the semipublic space of their agraha¯ram streets. Nonetheless, however much relaxation has taken place, almost all Vattimas agree that, for their own identity as Brahmans, it is vital to observe menstrual pollution inside the home, which they do not only in villages, but also in modified form in towns and cities, even overseas. Not all Tamil Brahmans still observe menstrual pollution as strictly as Vattimas. All the same, by controlling their own bodies and spaces, specifically the home, Brahmans gain recognition as Brahmans among themselves; in other words, caste distinction has been “privatized,” in Mosse’s terms, so that it is their Brahmanhood in their own eyes, not in other people’s, that now matters most to them.56 In general, the same conclusion applies to the Brahmans’ observation of all purity and pollution rules today, including those about cooking and eating food. Notwithstanding the changes that have taken place, especially the abolition of restrictions on lower-caste access, Tippirajapuram’s agraha¯ram is still more visibly “traditional” than most others because it is still predominantly occupied by Brahmans who want to keep it that way. Quite a lot of non-Vattima Brahmans in Tippirajapuram used to live in agraha¯rams now taken over by non-Brahmans. They moved to Tippirajapuram because it is conveniently close to Kumbakonam, but they also want to live in a place where Brahmans are the majority; Brahmans, they say, are peaceful, intelligent, well-educated, clean in their habits and devout, and they care for their

The Village: Caste, Land, and Emigration to the City / 45

temples. Some non-Brahmans in Tippirajapuram also prefer to live there for the same reasons; thus a Chettiyar widow living on East Street said she liked the “peace” and “spirituality” of the agraha¯ram, and another Chettiyar couple on North Street praised Tippirajapuram as “calm” and “good for the children’s future,” although they added that the Brahmans “still hate us,” even if they pretend otherwise. On the other hand, not everyone in Tippirajapuram accepts all this talk about its Brahmanical virtues at face value. Many local critics of the Vattimas castigate them as hypocrites merely keeping up appearances as traditionally pure, orthodox Brahmans. Some people also blame the Vattimas, as well as other Brahman residents, for breaking their own purity rules—for example, by letting non-Brahman servants into their kitchens—and thereby allowing the agraha¯ram to decay. A few of these conservative critics are actually non-Brahmans, who say (unlike the indifferent majority) that Brahmans should know the importance of adhering to their own rules in order to preserve the agraha¯ram’s way of life. In sustaining an agraha¯ram, its temples count for a lot and in Tippiraja­ puram, as elsewhere, even urbanized Vattimas living far away normally retain familial links with the local temples and make regular contributions to the cost of their rituals. The temples’ importance was highlighted in Tippi­ rajapuram in 2008, when renovation rituals (kumbha¯bhis·eka) were held for both the Vishnu and Shiva temples during one weekend. The rituals were expensive, but more than enough money was raised through donations from members of Tippirajapuram Vattima families; young people in America gave particularly generously. Local Vattimas were largely in charge of the renovation rituals and several powerful men in the community took leading roles. The outcome, as we have explained elsewhere, was that Vattima control over the two temples and Vattima dominance over the agraha¯ram were publicly displayed and reasserted.57 This demonstration of Brahman status and power in an agraha¯ram was highly unusual, however. And although it showed that the old rural social order has not entirely disappeared, that is unlikely to remain so for long. As Vattimas all acknowledge, the greatest threat to Tippirajapuram’s survival is not broken purity rules, but emigration. It is true that Eighteen-Village Vattimas as a whole retain stronger village roots than most Tamil Brahmans. But not all of them do and many Vattimas living in urban areas, especially outside Tamilnadu, are not really concerned about the villages. Naturally, it is the Vattimas still living in them who are particularly preoccupied with the value of their agraha¯rams and rural life. Yet demography is against them. As one elderly woman in another Vattima village put it, her agraha¯ram has no daughters-in-law, so that (without any young families) it will be extinct

46 / Chapter One

in the next generation. In Tippirajapuram, the position was scarcely better; among the eighty-five Vattimas living there in 2005–6, there were only three children under sixteen, whereas fifty-one men and women were aged sixty or over, although there were more children and the age structure was more balanced among the other Brahmans. Brahmans moving into Tippi­ rajapuram and a few other agraha¯rams may ensure that they survive with sizable Brahman populations, but in the majority of agraha¯rams—Vattima and non-Vattima alike—the prognosis for the future is that the number of Brahman residents will continue to fall.

The Brahmans’ Social Separation in the Villages Educational and employment opportunities in the towns and cities have always been the main “pull” factors underlying urban emigration by Tamil Brahmans. They are examined in the next two chapters, but here we discuss two “push” factors that have received little previous attention. The first of these has to do with the Brahmans’ position in rural caste society and the second with their attitude toward the land. From an external, sociological point of view, Brahmans have obviously belonged to village society like the members of all other castes. In the past, Brahmans played a key role in the rural economic system, primarily as landlords, and they heavily depended on the lower castes for the provision of goods and services. Politically, too, Brahmans used to dominate their villages and they largely controlled relationships with government and external authority, partly because the principal local office holders in Brahman villages were all Brahmans until they were progressively replaced by non-Brahmans after Independence. Brahmans also exercised considerable judicial authority over the rest of the population, as Gough’s case studies of punishment meted out to non-Brahmans vividly show.58 More generally, inasmuch as village society was constituted by the relationships among its castes, the Brahmans, numerous different groups of non-Brahmans, and the untouchable Adi Dravidas were all integrated within a hierarchical whole. Yet not all castes were integrated to the same extent; indeed, a significant measure of social separation was characteristic of the Brahmans, as Béteille particularly emphasized, although Gough’s (and other anthropologists’) evidence reveals the same phenomenon. For the Brahmans, Béteille writes, “Sripuram signifies generally the agraharam, the residential area and the community of Brahmins living within it.” Hence Brahmans from Sripuram, when asked how many people live there, normally answered with refer-

The Village: Caste, Land, and Emigration to the City / 47

ence to the population of the agraha¯ram only.59 The same is true in other Brahman villages such as Tippirajapuram, which for the Brahmans is just the agraha¯ram, whereas for non-Brahmans it also includes their settlement area on the other side of the main road. In Sripuram’s agraha¯ram, as in others, there was a strong sense of internal solidarity and community, but the other side of this coin was Brahman exclusiveness: “the physical separation between the agraharam and the rest of the village [is] both a cause and a consequence of the social separation of the Brahmins from the rest of the population.”60 Furthermore, although the agraha¯ram was closed to local non-Brahmans, it was open to all Brahmans from anywhere in India, even if they came from the north. Béteille admits that it would “perhaps be an exaggeration to say that Brahmins, on the one hand, and NonBrahmins and Adi-Dravidas, on the other, represent two cultures,” as illustrated by the Brahmans’ high valuation of Sanskrit and the pan-Indian Sanskritic tradition, as opposed to Tamil and the regional culture.61 This Sanskrit-versus-Tamil opposition may have mattered more in Sripuram than some other Brahman villages; furthermore, the opposition became increasingly politicized during the twentieth century. Yet even if Béteille exaggerated and Sripuram had some exceptional features, his key point holds good for all Brahman villages: Brahmans do not identify themselves with the majority of the local population and the agraha¯ram’s physical separation from the rest of the village has itself been a constitutive factor in the caste’s social separation. Brahmans in Tippirajapuram generally refer to it by the Tamilized Sanskrit word gra¯mam, whereas non-Brahmans use the pure Tamil word u¯ r. The Vattimas’ “eighteen villages” are gra¯mams too, whereas similar nonBrahman numbered sets of villages are u¯rs. It is unclear whether Gough reports a different usage when she states that a Brahman village is a gra¯mam, as opposed to a non-Brahman village that is an u¯r.62 Gough does not mention any term for the non-Brahman part of a Brahman village, but it is known as the kut·iya¯na te˘ru, “farmer’s streets,” in Tippirajapuram and Sripuram, ¯ which seems to be common in the Kaveri delta region.63 In Kalappur, a non-Brahman village near Tiruchirappalli, Valentine Daniel reports that the term gra¯mam (kira¯mam) refers to the “revenue village,” an administrative unit, whereas u¯r is used for the village as a social unit that people may identify as their home.64 This distinction, however, is not one that we have come across in Tippirajapuram or other Brahman villages. Dusi and Manjapalayam (MM), two Brahman villages in the Palar valley, have structures like those found in the Kaveri delta, but we have no information about the terms used for “village” there.65

48 / Chapter One

Further south, the village structure and nomenclature differ. Yanai­ mangalam and Mel Ceval, studied by Diane Mines and Marie-Louise Reiniche, respectively, are two Brahman villages quite close to each other in the Tambraparni valley. Their agraha¯rams, unlike those in the Kaveri and Palar valleys, are not sharply separated from the non-Brahman streets. Thus in Yanaimangalam’s central residential area, which is called the “big village” (pe˘riya u¯r), there are four main streets and the one on the northern side is the agraha¯ram. The spacious village of Mel Ceval is unusual because it has two agraha¯rams. The old agraha¯ram is the middle street bisecting the main village site, which forms a square, and the other agraha¯ram is the middle street of an adjacent extension of the residential area. In both villages, most people call a Brahman street a gra¯mam (kira¯mam), which is locally understood as a variant of agraha¯ram. Both villages have Dalit colonies—u¯ r in Yanaimangalam and ceri in Mel Ceval—which are located outside the main settlement.66 There is plainly some variation in the meanings and usage of the words gra¯mam and u¯ r—especially between Brahmans and non-Brahmans, and between residents of Brahman villages in different regions—but we have too little information to discuss it fully. Nonetheless, gra¯mam (unless it is an administrative label) is clearly used mainly by Brahmans and normally denotes a Brahman village or agraha¯ram; u¯ r, on the other hand, is the word for “village” generally used by non-Brahmans—who may mean to include an agraha¯ram when referring to a Brahman village—although u¯ r primarily refers to either a non-Brahman village or the non-Brahman section of a Brahman village. Daniel studied a non-Brahman village, but his examination of the meaning of u¯r is still salient here, notably because u¯ r means “home” in Tamil—or “native place” in Indian English—as well as “village,” or indeed “town” or other locality.67 Hence there is a definite sense in which a gra¯mam or an agraha¯ram cannot truly be part of the Tamil home that is equated with the village. By the same implicit logic, a Brahman in Kalappur told Daniel that Brahmans cannot prosper on its bitter, Shudra (non-Brahman) soil, while a blacksmith said that castes like his own and the Brahmans could not thrive in a village whose name ended in -u¯r, implying that such places belong to dominant, non-Brahman cultivating castes. Thus Brahmans, to paraphrase Daniel, cannot establish a “relationship of substantial compatibility” with an u¯r.68 In fact, plenty of Brahman villages’ names (such as Tediyur) end in -u¯ r. It is also unclear whether most nonBrahmans think that Brahmans cannot be compatible with an u¯r. Nonetheless, Daniel undoubtedly highlights an attitude of mind that is widespread, albeit often unformed. Furthermore, whatever connotations the term u¯r car-

The Village: Caste, Land, and Emigration to the City / 49

ries for different speakers or in different contexts, non-Brahmans—as much as Brahmans themselves—clearly perceive an agraha¯ram and its inhabitants as separated from the rest of village society, specifically the non-Brahman section that is actually equivalent to a complete non-Brahman village and is of course the settlement’s heart for non-Brahmans themselves. The Brahmans’ separation may be briefly compared with that of the Dalits, the former Untouchables. The gap between a ceri and a village’s main settlement area, even if it is quite small, mirrors the Dalits’ traditional separation from the rest of the community except as servile inferiors. The other castes, both Brahmans and non-Brahmans, need their labor and ser­ vices—for example, as scavengers or drummers at festivals—so that they are included within the social division of labor, but doing such work reconfirms the Dalits’ degraded status and their ambiguous exclusion from the social whole. Thus to put it schematically, in Tamil rural society the large majority of non-Brahmans have constituted the core, whereas Brahmans in their agraha¯ram have been above it and Dalits in their ceri below it. For the Dalits, who are still overwhelmingly confined to their segregated colonies in rural areas, their modern struggle for civil rights has been focused on the struggle over public space and on ending discrimination against them, so that they can join the society from which they have been excluded, often most vigorously by non-Brahmans whose local dominance is threatened.69 The Dalits, to oversimplify, want to be included in Tamil society alongside non-Brahmans, whereas the Brahmans—who never saw themselves as fully part of it anyway—do not.

The Brahmans’ Attitude to the Land The rural Brahmans’ social separation depended heavily on their economic independence as landowners, which enabled them to minimize their reliance on lower-caste people and hence their relationships with them. In the past, especially for mira¯sida¯rs owning a lot of land, the “tempo of life was slow and relaxed, and the burdens of agriculture were left mainly to be shouldered by tenants and cultivators.”70 These landlords spent long hours sitting on the verandas of their houses, gossiping with their friends and neighbors, reading newspapers, playing cards or, in some cases, studying religious texts as orthodox Brahmans ideally should. This way of life has almost vanished, but not completely, because it lingers on among a few old-fashioned landlords in agraha¯rams, such as Tippirajapuram, that are still predominantly Brahman. Particularly important is that Brahman landlords never worked in their

50 / Chapter One

fields. One conventional reason given is that they cannot do so for religious reasons, because agricultural work, especially ploughing, always involves killing insects or worms in the soil, which nonviolent, vegetarian Brahmans must avoid. The main explanation, however, as is generally recognized, is that manual work is demeaning, so that high-status landlords refuse to do it—even though poorer Brahman farmers may have to. Contempt for manual work has been more or less universal for centuries throughout India—as well as in other traditional, agrarian societies—so that Brahman landlords were not at all exceptional and many non-Brahman landlords, in south India or elsewhere, also avoided agricultural labor. Yet in Tamilnadu it mattered most of all for Brahmans; as David Ludden explains, Brahmans, over the centuries, “became the cultural model of elite behavior and style among dominant peasants, [so that] not putting one’s hands in the mud would have become a mark of entitlement to elite stature.”71 Brahman landlords, to put it simply, have always had more to lose than non-Brahman landlords if they get their hands dirty. According to Gough, Brahman landlords in Kumbapettai “knew a great deal about every detail of cultivation,” whereas Béteille says that in Sripuram “most men were acquainted with only the principles of agriculture and not its practice.”72 This difference may be because landlords in Kumbapettai were more likely to supervise agricultural operations than those in Sripuram, but it is hard to tell whether knowledge or ignorance was more typical, although Vattima Brahman landlords commonly admit their lack of expertise. Equally salient, though, is received wisdom about the topic. At least by the late nineteenth century, the high-ranking, nonBrahman “peasant” caste of Vellalars, many of whom were also mira¯sida¯rs, were portrayed as the Tamil country’s authentic agriculturalists, so that proverbially “farming is in the blood” for Vellalars, whereas Brahmans and the low castes “make bad farmers.”73 Brahmans do not usually dissent from this judgment. Conversely, at least today, Brahmans are stereotypically assumed by themselves, as well as others, to be much better at urban, nonmanual jobs. As the author of a report on Gangaikondan in the 1930s put it, more colorfully than impartially, the local Brahmans had lost interest in the land, so that “it looks as though the Brahmin thrives best in towns and the rural soil is uncongenial to his genius.”74 The ownership and control of land, preferably large amounts of it, plainly mattered greatly to Brahman mira¯sida¯rs for economic, political, and social reasons. It was the source of their wealth and income; it also provided their supplies of rice, the principal foodstuff for all Tamils that is ideally grown on a household’s own land and stored in its granaries. Land was the

The Village: Caste, Land, and Emigration to the City / 51

foundation of much of the mira¯sida¯rs’ power as well, particularly during the colonial period, when in practice landlords, both Brahman and Vellalar, “were the government in the wet zone, not only at the village level; they and their caste stratum peers comprised the subregional ruling class.”75 After Independence, democracy and the rise of the non-Brahman movement progressively undermined the Brahman landlords’ rural power. Even so, as we have seen, in the fifties and sixties the combination of landownership and caste supremacy was still crucial to the Brahman mira¯sida¯rs’ status and power within their villages. Yet Brahmans were always “rentiers” or “agrarian patriarchs,” not farmers committed to agriculture as a means of livelihood.76 Brenda Beck compares Brahmans and “left-hand” castes loosely linked to them with Gounders (Kavuntars), the leading “right-hand” peasant cultivator caste in Coimba­ tore District and northwest Tamilnadu. Beck explains that Gounders, unlike Brahmans, spend a lot of time interacting with their laborers and maintain their dominance through heavy involvement in local affairs, as well as a “willingness to perform technically polluting tasks such as plowing”; “instead of emphasizing withdrawal, they attempt to participate in, and hence to control, as much local activity as possible.”77 The northwestern upland region differs from the river valleys, the right-left division is absent in much of Tamilnadu, and Gounder peasant farmers are unlike Vellalar mira¯sida¯rs in numerous respects. But in spite of these differences, the empirical contrast between Brahmans and Gounders can also be understood as a kind of ideal-type opposition defining the poles of a continuum. Non-Brahman landowners and cultivators are very diverse, but the vast majority can be placed somewhere along that continuum, many toward the Gounder pole, and next to none of them are as preoccupied as Brahman landowners with avoiding participation in agricultural work and exercising their lordly power free of entanglement in local affairs, especially those of other castes. Linked to their agrarian patriarchal attitude, too, is the Brahmans’ sentiment about the land. Ludden writes that for those, like Vellalars, who have farming “in the blood,” “the land does not so much belong to its owner as constitute a farmer’s being and community.”78 That is exactly what was not true among the vast majority of Tamil Brahmans, who had far less emotional commitment and attachment to their land than non-Brahman agriculturalists. The one important exception to generalizations about Brahman rentiers is the small minority of landowners who have become modern, capitalist farmers by investing, innovating, and managing their land and labor effectively; Gough and Haruka Yanagisawa mention a few examples.79 These modern Brahman farmers, who normally own large amounts of land and

52 / Chapter One

make handsome profits, do not necessarily live in villages. Of the four Vattima farmers we have met, one lives in Tippirajapuram, but formerly worked for a bank in Chennai, and he and his wife still spend considerable time in the city; another man, always a full-time farmer, lives in Kumbakonam; a third, who used to be an engineer, lives in Tiruchirappalli; and the fourth, an unusual woman who farms her land organically, lives in Chennai, although her rural brother-in-law helps to manage their land. All of them scorn the managerial incompetence of old-fashioned mira¯sida¯rs and they epitomize Ludden’s contrast between new “agrarian citizens” and old patriarchs.80

Urban Migration and the Sale of Land Urban migration among Tamil Brahmans has had a characteristic pattern. Initially, often when young and single, men moved from a village to an urban area—usually Chennai or a town in Tamilnadu, but sometimes further afield—for education, employment, or both. Some men belonging to the first generation of migrants returned to their villages with their wives after several years of urban work or, eventually, after retirement, whereas others never went back. Now and in the past, migrants’ children, who usually spend all or most of their lives in an urban locality, hardly ever return to villages; on becoming adults, sons take up urban employment and daughters, whether or not they work outside the home, normally marry men who are also employed in towns and cities, like their brothers. Indeed, young women, even those raised in villages, have become increasingly unwilling to marry village landlords or farmers, and for these men it is notoriously hard to find wives today. In the second generation, even if it did not begin among the initial migrants, urban migration frequently inaugurates a rapid process of personal and familial urbanization whereby people largely detach themselves from rural life. One crucial aspect of urbanization is the disposal of village land and houses. In Kumbapettai in the 1950s, if some men in a family remained in the village while others had left, their land was commonly retained as the property of an undivided joint family; the rural brothers looked after the land and handed over part of the produce to the urban brothers, who reciprocally sent them cash from their salaries. If men eventually retired in the village, the land would be partitioned and brothers would establish separate households.81 In other cases, land was sold or rented out to non-Brahman tenant cultivators, but this became less frequent when legislation to protect tenants’ rights enacted in the 1950s began to make an impact. By the 1970s,

The Village: Caste, Land, and Emigration to the City / 53

many of Kumbapettai’s Brahmans had little or no land left and relied on salaried work.82 Subsequently, as we were told when we visited Kumbapettai in 2005, the Brahmans continued to sell their remaining land as they left the village, and the majority have also sold their houses or abandoned them. In Sripuram in the early 1960s, land was being sold steadily and most of the big mira¯sida¯rs were absentee landlords, who often decided to sell their land eventually. Moreover, absentee landlords’ sons brought up in urban areas were considerably less attached to land than those who stayed in the village. A law imposing a ceiling on land ownership passed in 1961 accelerated sales and land “ceased to have the same unique value” among Brahman landowners that it had had in the past.83 The evidence suggests that in all Brahman villages, urban migrants have been selling their land, sooner or later, in much the same way as they did in Kumbapettai and Sripuram. There are exceptions, of course, and some Brahmans—especially if they live in towns fairly close to their ancestral villages—try to keep enough land to grow rice for household consumption, or they may retain rights to collect coconuts or other garden produce. But most do not, whereas non-Brahman urban migrants are quite likely to retain family land and sell it only if their farms are small and unprofitable; some non-Brahmans, too, invest urban earnings in rural land, whereas Brahmans hardly ever do. Today, at least among those resident in Chennai and other large cities, or overseas, the vast majority of ordinary, middle-class Brahmans own no rural land at all and have no interest in doing so. They also no longer have a rural agraha¯ram house, unless they have failed to sell or rent it. Moreover, if they do have money to invest, from either the sale of rural property or their earnings, they generally buy urban housing and land: in the first place, a home for their own family and maybe another for their parents, but sometimes also other houses or flats to let, plots of land for building, or commercial properties. Compared with most other Tamil Brahmans, the Eighteen-Village Vattimas probably remain more attached to their ancestral houses, particularly when some family members still live in villages or did so until recently. Some people also keep their agraha¯ram houses even after selling their village land. But attitudes are often mixed or ambivalent, as an example from one family illustrates. A landlord living with his wife in one of the principal Vattima villages used to own a lot of land, although he has now sold most of it and has only enough to grow the family’s rice supply. His eldest son, a businessman, and his youngest, a doctor, both live in Chennai; the middle son works for a bank and lives near Chicago. The businessman hopes to return to the village to live in his parents’ house and look after them in their old age, although his wife (like most younger women) would prefer

54 / Chapter One

to stay in the city; the doctor is noncommittal about where he lives in the future, but his wife is interested in going abroad; their brother in Chicago has no intention of returning to India, let alone to his village, although his wife might like to go home, and he flatly stated that he had no sentimental attachment to his family house. But all three brothers see eye to eye about the land: they have absolutely no interest in agriculture and none in retaining the remaining family land, even to grow their rice. In this unanimity, they are—like Satyamurti from Tindivanam—typical of the vast majority of urban Tamil Brahmans. This detached, unsentimental attitude to land, which is so different from that of traditional peasant farmers, together with the relative separation from village society, which is so different from that of non-Brahmans, have greatly facilitated the Brahmans’ departure from the countryside and their progressive transformation into a predominantly urban caste in the modern era. Not only did Brahmans have much to gain by moving to towns and cities, as they still do, they also had little to lose by leaving the villages and their land.

Caste Status, Ritual Purity, and Moral Conduct in Urban Areas In the process of urbanization, migrants from villages have to adjust to their new environment and face problems pertaining to their caste status, ritual purity, and moral conduct. Discussing migration, Marguerite Barnett observed that for Brahmans social position and status were independent of where they lived, whereas for non-Brahmans, especially the highest castes, they depended directly on economic and ritual dominance in the village, typically corroborated by asymmetrical transactions. In other words, owing to their social separation within an agraha¯ram, it scarcely mattered to Brahmans whether they lived in villages or not, whereas non-Brahmans, who were enmeshed in “very specific localized transactional relationships and deference patterns,” tended to become deracinated when they went to urban areas, because they no longer had a firm basis for defining their rank and were instead “lumped” with all the other non-Brahmans.84 To be sure, the mutual ranking of non-Brahman castes in any one place was normally contentious, but disputes were actually vital parts of local social life, as was so in Kumbapettai where, in principle if not in practice, they might be settled at a temple festival, for example.85 But even if precedence at a festival could hypothetically resolve the relative status of Kallars and Konars in Kumbapettai, it could sort out nothing for them if they went to an urban area, particularly in relation to other non-Brahman castes absent in Kumbapettai. Thus non-Brahmans moving to towns and cities have typically

The Village: Caste, Land, and Emigration to the City / 55

suffered a dislocation in their status, especially before enough members of their own and allied castes have arrived in urban areas and managed to establish their own neighborhoods, like the Kontaikkatti Vellalars eventually did in Chennai.86 Non-Brahmans’ dislocation may be mitigated, though, by retaining significant links to the home village. Thus in Kalappur, an Aru Nattu Vellalar businessman from Malaysia, whose father had emigrated there as a laborer, returned to see his ancestral lands and to arrange his son’s six-month stay in the village, which he saw as important because his son would then drink the local water and eat the local rice; as the businessman said of himself, “the soil of this village . . . is, after all, part of me.”87 The majority of Nagarattars (Nattukottai Chettiyars), who include many wealthy bankers and merchants, live in urban areas, often overseas. Nevertheless, they continue to identify themselves with the villages containing their grand ancestral houses in Chettinadu in southern Tamilnadu. They return to these mansions, which are often maintained at great expense, to celebrate family rituals and they sell them only if they need to.88 In Randam village in North Arcot, where Agamudaiyan Mudaliars form the dominant caste, several men who have left the village still own land there and “they all remain intimately associated with the village and return to it for all important life cycle rituals.”89 In the town of  Tiruchengodu (near Erode), “educated and urbanised” Kongu Vellalars “have never completely cut their rural ties,” because their “Vellalar identity, as defined by some kind of authority over people in the rural area, is still meaningful.”90 Similarly, in the nearby town of Bhavani, members of the Seven-Village Vanniyar subcaste regard their urban neighborhood as an u¯ r that is still strongly connected to the ancestral “seven villages (ur).”91 Urban Brahmans, by contrast, do not go back to ancestral villages to absorb their food and water or hold family rituals—nor do they ever link their urban neighborhoods to rural ones—and one crucial reason why they do not do so is because they have not been deracinated by urban migration. For Brahmans, far more than for non-Brahmans, status and identity are independent of where they live. Thus even members of localized subcastes, like the Eighteen-Village Vattimas, can quite easily associate themselves with other Vattimas and hence with other Aiyars and Tamil Brahmans collectively, and they in turn belong to the pan-Indian Brahman varn ·a. Non-Brahman castes, by contrast, are almost all regionally specific, while their traditional varn · a identity as Shudras is both indiscriminate and demeaning. It is true that the “family name” traditionally used by Tamil Brahman men and inherited from the father is often their ancestral village’s name,

56 / Chapter One

although they rarely attach much significance to it. Some Brahmans retain their ritual affiliation to a “family deity” (kuladevam), whose shrine may be in the ancestral village, and others, as noted above, keep up familial links with the village’s Vishnu and Shiva temples, especially by making financial contributions to pay for their rituals, as most Vattimas do.92 Among some urban Brahmans, especially those whose forebears left their agraha¯rams several generations ago, a new “roots” movement is starting to develop, which is usually given shape by donating to the renovation of village temples. For the majority of urban Brahmans, however, links to ancestral villages and their temples matter little today, and many people are vague or ignorant about where their families originally came from. An important aspect of Brahman identity is that caste became more explicitly defined by birth and descent under colonialism, as Washbrook explains.93 This was partly a product of Anglo-Indian law and its preoccupation with the different customs, rights, and privileges of different communities, but it was also strengthened by racial notions of caste, particularly in south India by the claim that Brahmans were Aryans, not Dravidians like non-Brahmans. Whether caste status—for instance, as a Brahman— depends on birth or on adherence to the caste’s own moral norms and code for conduct, its dharma, is actually an ancient question, but change during colonial times reinforced the weight placed on birth. Thus even if a Tamil Brahman worked in a secular occupation in a city and deviated from his Brahman caste dharma, he did not “lose caste” and questions about his authentic Brahman status were stilled. What matters, too, is the moral evaluation of this Brahman status. The Last Brahmin is a Telugu biography of a Sanskrit pandit (who died in 2000), as well as an autobiography by one of his sons, Rama Siva Sankara Sarma, a Sanskrit schoolteacher who calls himself a “non-believer.” The pandit rigidly adhered to the view that everyone must follow his or her own caste dharma. He insisted that “it is dreadful to imitate other people’s dharmas”; moreover, “when a Brahmin deviates from his dharma, he is merely Brahmin scum.”94 For the pandit, therefore, a Brahman is born, not made, but he degrades himself by failing to follow his own dharma as transmitted to him by sages and teachers since Vedic times. Even among Vaidika Telugu Brahmans—as opposed to “secular” Niyogis—the pandit’s uncompromising outlook on Brahmanical morality was rare. It was still rarer among Tamil Brahmans, who may describe the religiously orthodox as vaidika, but do not divide themselves into “Vedic” and “secular” subcastes; neither do they systematically distinguish between vaidika Brahmans following religious

The Village: Caste, Land, and Emigration to the City / 57

vocations and laukika (“worldly,” “lay”) Brahmans with ordinary jobs, as Kannada Brahmans do. All in all, therefore, unlike the pandit—whose extreme opinion is precisely what makes it so telling—the overwhelming majority of latter-day Tamil Brahmans have consistently regarded urban migration and modern employment as morally acceptable, for they have never really thought they would become “scum” by deviating from their caste dharma. Yet migration does pose some dangers to Brahman morality. Traditional religious observances are often hard to combine with daily work routines, and purity and pollution rules are more difficult to observe in towns than villages. Thus, for example, an adult male Brahman who wears a sacred thread should perform a series of daily rituals, particularly the sandhya¯ in the early morning, which includes a ritual bath, recitation of the Ga¯yatrı¯ mantra, and worship of the sun. The sandhya¯ is in full an elaborate ritual taking several hours to complete, ideally beside a flowing river, which should also be done at midday and in the evening. Some urban Brahmans carry out fairly lengthy sandhya¯s, especially in the morning, but very few of them can do so by a river; furthermore, for most men who have to go to work long rituals are impractical, so they content themselves with shorter ones, perform them only on the weekend, or do not bother at all. To protect their ritual purity, rural Brahmans could mostly stay away from nonBrahmans and Dalits, and even today they can quite easily control the purity of their own agraha¯ram houses, together with their food and water, which comes from their own household wells. In towns and cities, however, all this normally becomes harder. Observation of customs and rituals that are traditionally fitting is a¯ca¯ram, a term that Mary Hancock glosses as “morally and socially appropriate actions and attitudes,” and it refers to an important aspect of Tamil Brahman life.95 Maintaining a¯ca¯ram in urban areas is troublesome even within one’s own home, especially for the majority of people who live in modern houses and flats, where traditional purity rules cannot be strictly maintained because, for example, the bathroom and lavatory are close to the kitchen instead of in a backyard. Water for drinking and cooking may be filtered, but it usually comes from the municipal supply or a shared borewell, not a household well whose fresh water is drawn by family members and guaranteed to be pure. In urban homes, as mentioned above, menstrual pollution rules are still observed by Eighteen-Village Vattimas and other Tamil Brahmans, but they have to be partly relaxed, if only because there are rarely enough rooms to reserve one for women’s proper seclusion. As a result, some village Brahmans are suspicious about urban

58 / Chapter One

ways even today; thus one elderly Vattima woman “saw Chennai as a city full of menstruating women roaming around in public” and she certainly did not want to live there.96 As we shall see in chapter 5, many Brahmans in Chennai live in streets or apartment complexes where they form a majority of the inhabitants. Yet complete residential separation from other castes has rarely been feasible in urban areas, even in Brahman-dominated suburbs of Chennai, and at work most Brahmans have non-Brahman colleagues. Public transport and other public facilities have always been used by all castes as well. None of this is necessarily irksome, even for conservative Brahmans, because a bath and change of clothes on returning home are enough to remove pollution caused by ordinary contact with other people, say at work or on buses. Since Brahmans, like other people, commonly take food prepared at home to eat during the school or working day, or rely on “carryout” food prepared by Brahmans, they never need to worry about its vegetarian purity, and Chennai and other south Indian towns also have numerous vegetarian restaurants with Brahman cooks, where all but the most rigidly orthodox can eat. For this minority of very persnickety Brahmans, who refuse to eat anything not prepared in their own home or an equally impeccable environment, urban life and work persistently pose difficulties. We have even heard one or two stories of Brahmans who gave up jobs in cities like Chennai, because they could not cope with the constant problem of finding completely pure food and water. But such individuals—like the Vattima woman worried about menstruating women in Chennai—are extremely unusual. In reality, almost all Tamil Brahman urban migrants, even very conservative ones, have been prepared to make compromises and adjustments, just as Hindus have perennially done with respect to purity and pollution rules, and other ritual observances. Brahmanical codes for conduct may look almost impossibly detailed and rigid on paper, so to speak, but how they are understood and implemented have always been flexible and changeable. In practice, therefore, the rules can normally be interpreted and applied pragmatically. Sometimes, indeed, they are just ignored, so that, for example, at the turn of the twentieth century, Brahmans started to wear leather shoes to go to the office and indulged in drinking coffee, which was actually condemned like alcohol in Madras in those days. But as A. R. Venkatachalapathy explains, conservative denunciation scarcely slowed coffee’s progress, so that by the thirties or forties among the growing urban middle class, especially Brahmans, serving coffee at home became the norm and “coffee hotels” were typically run by Brahmans.97 Tamil Brahmans still regard coffee as their special beverage and, even if they look like trivial examples, both

The Village: Caste, Land, and Emigration to the City / 59

coffee drinking and shoe wearing illustrate very well how traditional rules can be modified or even abandoned. Certainly, urban migration potentially endangers purity and morality, but that has not deterred the vast majority of Tamil Brahmans, who have been emigrating from their village agraha¯rams since the nineteenth century, “pushed” by their separation from village society and detached attitude to the land, as well as “pulled” by educational and economic opportunities in towns and cities. Tamil Brahmans are not the only south Indians who have been migrating extensively in the modern era. Deshastha Brahmans from Maharashtra and some Niyogi Telugu Brahmans (see chap. 2) actually spread across the south much earlier, but other Telugu and most Kannada Brahmans have migrated in broadly similar ways to Tamil Brahmans since the nineteenth century. One particularly mobile group is the small Chitrapur Saraswat Brahman caste from Kanara (on the Karnataka coast), whose members migrated to Madras, Bombay, and other urban centers in the colonial period, so that it became “a caste on the move,” in Frank Conlon’s apt phrase.98 Quite a few other groups could also be mentioned; one striking example is the Nagarattar merchants and bankers, who were crossing the sea to Southeast Asia by the early nineteenth century and almost certainly represent the only Tamil community whose overall mobility matches the Brahmans.99 Yet even if they are not unique, Tamil Brahmans—also a caste on the move—are genuinely distinctive in both the scale of their urban migration and the completeness of their urbanization, which strongly shaped their collective experience, especially during the twentieth century, and transformed them into a group of mobile urbanites.

Two

Education and Employment in the Colonial Period

Since the medieval period, the Brahman minority has constituted the core of the Tamil country’s social, cultural, and religious elite. Many scholars, teachers, bureaucrats, and jurists were Brahmans, although some other high castes, especially Vellalars, were also represented among the traditional literati. Education and learning were historically the special prerogatives of Brahmans, however, and a high proportion of Brahman males, unlike members of other castes, have always been literate in vernacular languages and sometimes Sanskrit. In the new city of Madras in the early eighteenth century, the dubash (“man of two languages”) was the key intermediary between the East India Company’s agents and local communities. The majority of dubashes were Vellalars; Kanakkuppillais, scribes (kan ·akkuppil··lai), who were Tamil nonBrahman village accountants; Idaiyars or Yadavas, traditionally pastoralists; and Telugu or Maratha, but not Tamil, Brahmans.1 Dubash power and influence faded later in the century, as the British expanded their control over the region that became the Madras Presidency. In the early decades of colonial rule, the British relied heavily on their predecessors’ administrative systems, particularly for land settlement and revenue collection. Scribal expertise was concentrated in various social groups by the eighteenth century; thus, for example, Maratha Deshastha and Tamil Brahmans were in charge in Tanjore, but Telugu Niyogi Brahmans were dominant in coastal Andhra and had a strong presence in the Arcot kingdom’s territory in and around Madras. In the new bureaucracy established by the British, different groups also came to occupy different rungs in the system. In the district collectors’ offices, the chief Indian officials were typically polyglot Deshasthas or Niyogis, who were also skilled in writing several linguistic scripts, whereas the lower-level

62 / Chapter Two

record keepers included non-Brahman Vellalars, Kanakkuppillais, and others, as well as Brahmans. Tahsildars in charge of taluks, the subdivisions of a district, were almost all Brahmans, commonly Sri Vaishnavas, and answerable to them were Kanakkuppillais working in villages, as well as village headmen, who were Brahmans in Brahman villages and Vellalars or other high-caste non-Brahmans elsewhere.2 In general, though, at the highest levels occupied by Indians in the Madras Presidency’s revenue administration, Deccani Brahmans—Maratha Deshasthas and Telugu Niyogis—were more prominent than Tamil Brahmans. Deshasthas had been both innovative and powerful in the bureaucracies of the Muslim states in western India and then in Shivaji’s Maharashtrian Hindu kingdom in the seventeenth century. From the fifteenth century onward, Deshasthas had spread across the south to serve local rulers, both Hindu and Muslim, and in the Tamil country they became particularly powerful in the Maratha kingdom of Tanjore (1674–1855). But Deshasthas were found all over the Madras Presidency—even though there were only about 30,000 of them in the nineteenth century—and in some places, such as Guntur District in Andhra, which Robert Frykenberg investigated, they were so entrenched that “they carved out, secretly and silently, a state within a state.”3 A bemused question about whether “there is anything about Madras Collectorates that makes them so difficult that only a Maratha Brahmin can fathom it” was posed in an official report, which also stated that half the highest cadres in the district administration in 1855 were Deshasthas (154 out of 305 officials), whereas eighty-three were drawn from all other Brahman groups.4 The Deshasthas formed a strong, tightly knit group because they had a long experience of administration and manipulated the levers of power very effectively, but it was also because they quickly added English to their other languages and adapted to the new regime. Telugu Niyogi Brahmans spread from Andhra across the region over a long period and were widely dispersed in the Tamil country and the rest of south India by the late eighteenth century. As already mentioned, they were numerous in Arcot and, at the turn of the nineteenth century, there was a solid cluster of Niyogis in Madras city associated with the Arcot kingdom’s court.5 Srinivasa Raghavaiyangar, a senior government official belonging to a Tamil Brahman family whose ancestors had served in the Tanjore Maratha court, wrote in the 1890s that by then landlords wanted “to give [their] boys an expensive English education and to marry [their] daughters to educated husbands,” whereas forty years earlier most were contentedly unconcerned with education.6 As R. Suntharalingam explains, when Brahmans began to

Education and Employment in the Colonial Period / 63

move to towns and attend colleges to learn English in the mid-nineteenth century, the “pace was initially set by the Mahratta [Maratha] Brahmins,” but Tamil Brahmans then emulated them.7 So successful were they that Tamil Brahmans, alongside Telugu Brahmans, quite rapidly displaced the Deshasthas, although the government also actively sought to reduce Deshastha influence in the 1850s.8 In the 1871 census report, Deshasthas were described as the “most intellectual and energetic division” of Brahmans in Madras and in 1881 the North Arcot District manual reported that they were well represented in the local bureaucracy, being “cunning in accounts as in everything else.”9 On the other hand, the Tanjore District gazetteer in 1883 contrasted the dynamic Tamil and Telugu Brahmans with the Deshasthas, who “till recently enjoyed a high reputation for intellectual superiority and administrative skill, as well as for moral character, but are now a depressed race and are losing their prestige”; in the Tanjore fort, from where they had run the Maratha kingdom, they had fallen into “a state of great depression and indigence.”10 To be sure, this was not true of all of them; for example, a Deshastha from Kumbakonam, Sir T. Madhava Rao, was a renowned dewan (prime minister) of Travancore described as “the outstanding statesman of South India during the nineteenth century.”11 In the late nineteenth century in much of south India, however, the bureaucratic power and preeminence of the Deshasthas clearly did decline. Yet it was Deshasthas who had pioneered the model of the modern, “secular” Brahman that Tamil Brahmans would follow, a “model [that] offered means of acquiring and utilising ‘foreign’ knowledges and skills, while still self-consciously serving religion and maintaining contact with Brahminical tradition.”12 In the transformation from an old elite to a new middle class that is a central theme of this book, it is important historically that the Tamil Brahmans actually reacted quite slowly to the challenges and opportunities of colonial rule until they started to follow in the footsteps of the Telugu Niyogis and especially the Maratha Deshastha Brahmans whom they then eclipsed.

Higher Education and Government Service In 1841, at the grand opening of Madras High School, the governor of Madras, Lord Elphinstone, told his audience that they were witnessing “the dawn of a new era rather than the opening of a new school.” The governor exaggerated, but “many contemporary observers” nonetheless saw the school “as the first serious attempt to disseminate Western education in South India,” and one of its main objectives was to “furnish the new generation of administrators capable of holding important positions in the

64 / Chapter Two

government.”13 In 1853, the school was raised to collegiate status and transformed into Presidency College, a teaching college of the University of Madras which, like its sister institutions in Bombay and Calcutta, was founded in 1857. In 1850, Madras Medical College was established and in 1858 a civil engineering college opened in the city, which was later expanded and in 1920 became Guindy Engineering College (now part of Anna University). Law was taught in Presidency College until a separate law college was set up in 1891. These professional colleges were affiliated to the university, which progressively regulated and improved the degree courses in arts and sciences, as well as in law, medicine, and engineering. The vast majority of students who took arts and sciences courses were taught in teaching colleges affiliated to the university, some of which were older than the university itself, like Madras Christian College and Pachaiyappar’s College, founded by a wealthy dubash. Other colleges were established later, in provincial centers as well as Madras city; one that has been especially important to Tamil Brahmans is Kumbakonam Government Arts College, founded in 1867, which in the 1880s was said to have “long had the monopoly of all the best Native teaching in the [Madras] presidency,” all its teachers and 90 percent of its students being Brahmans.14 At the Madras High School, thirty-six men had obtained the proficient’s degree by 1855 and twenty of them were Brahmans; the rest included twelve non-Brahmans, three Eurasians (Anglo-Indians), and one Indian Christian. Although they often had to accept lowly positions initially, twenty-five of the thirty-six graduates joined the Madras government service and several were appointed to the new post of deputy collector from 1859, including A. Seshia (or Sashiah) Sastri, a member of the Eighteen-Village Vattima subcaste. Seshia Sastri later succeeded Madhava Rao, another high school graduate, as dewan of Travancore before being appointed dewan of Pudukkottai (a princely state that is now a district in Tamilnadu); late in life, he was knighted by the British. Many of these men rose steadily through the ranks of the Madras bureaucracy, although others followed different career paths, including Madhava Rao, who first went to Travancore as tutor to the princes, and Sir T. Muthusamy Iyer, who decided to study law and eventu­ ally became the first Indian judge at the Madras High Court in 1878.15 The graduates of Madras High School, who had been educated in English, formed the core of a new administrative elite in south India and most though not all of them were Brahmans; significantly, too, the majority of the latter were Tamil Brahmans, rather than Deshasthas or Telugu Niyogis. Once the university had opened, Brahmans, especially Tamil Brahmans, established a decisive lead over other communities in higher education and

Education and Employment in the Colonial Period / 65

in government and professional employment. Some figures illustrating Brahman dominance included in the Non-Brahman Manifesto of 1916 were cited in the introduction, and other data from the late nineteenth and early twentieth centuries all show the same pattern. The overall male literacy rate among Brahmans was considerably higher than among other groups, but the inequality was even more marked for literacy in English, for which the rates in the Madras Presidency in 1901 were 17.9 and 10.8 percent for Tamil and Telugu Brahmans, respectively, but only 2.7 and 1.5 percent for Indian Christians and Malayali Nayars, the next two groups. Among the graduates of the University of Madras between 1870 and 1918, approximately 67 percent were Brahmans and 21 percent non-Brahman Hindus.16 In government employment, similar disproportions existed so that, for example, 55 percent of deputy collectors, 83 percent of subjudges and 73 percent of district munsifs were Brahmans in 1913.17 All these figures for education and employment provided the fuel for the Non-Brahman movement’s demand for greater equity as first set out in its manifesto. In the legal profession, Brahmans were similarly preponderant. In the late nineteenth and early twentieth centuries, most leading Indian lawyers or vakils in Madras were Brahmans and, after the distinction between vakils and barristers was abolished in 1926, Brahmans dominated the new Bar Council just as they had the Vakils’ Association.18 In this period, the Madras High Court bench was largely European, but of the first ten Indian judges (appointed between 1878 and 1916) eight were Tamil Brahmans, while the fourth, C. Sankaran Nair, was appointed in 1903 precisely because he was not a Brahman.19 Many of these Brahman judges and “leaders of the Bar,” who were knighted by the British, notably V. Bhashyam Iyengar and S. Subramania Iyer, formed a close network of influential friends and colleagues at the heart of the “Brahman raj” in south India.20 Yet the same men were also leading members of the powerful Mylapore group that dominated Congress and provincial politics at the turn of the twentieth century. Although the political power of Brahman lawyers subsequently declined, the legal profession in Madras remained a Brahman stronghold until the midtwentieth century. By then, though, medicine and especially engineering were becoming more attractive professions than law for Tamil Brahmans and this change is important for the caste’s modern transformation.

Old and New Professions Administration and law were old service professions in which Brahmans had always been prominent throughout India, at both higher and lower

66 / Chapter Two

levels. Indeed, according to classical authorities, righteous Hindu kings invariably depended on Brahman ministers, counselors, and jurists;21 Susan Bayly discusses their active role in the early modern period.22 Yet, as C. A. Bayly shows, “the spirit of Western administration wrought subtle changes” during the colonial period. The increasingly rationalized bureaucracy, which was staffed by graduates from institutions like Madras High School and operated at higher levels in English, ensured that “the mystique of the old scribal order was slowly undermined.”23 That in turn gave birth to a new “secretarial culture” in south India, which Bhavani Raman explores in detail in relation to changes in language and education.24 Parallel changes took place in the law as the colonial legal system developed, especially after the high courts were established in 1861 and the Hindu pandits formerly advising the courts were dismissed in 1864. To some extent, the law drew in Brahmans (or indeed men of other castes) simply because there was a strong demand for the services of lawyers, especially after 1861. Many posts in the colonial bureaucracy were also open to men with legal qualifications. The legal profession was remunerative, as well as prestigious, and many lawyers in Madras prospered.25 Some men, most famously Bhashyam Iyengar and Subramania Iyer, became immensely wealthy. Brahmans had always been teachers, too, and during the nineteenth century when a modern educational system was progressively established in Madras, it employed large numbers of Brahmans, although teaching, despite being respectable, was far less remunerative than law. A crucial aspect of all these modernizing changes was, of course, that formal qualifications, not caste status, became the requirement for any professional position. The Brahmans’ domination of the bureaucracy and legal system might suggest that they had an inherited affinity for them, and in some families administrative or legal service was a tradition. Yet Brahmans did not dominate colonial administration and law because they were actually constrained to pursue occupations traditionally associated with their caste, but because there were obviously good social and economic reasons to do so, many had gained the requisite qualifications, and many of them had kinship links or other close connections with men already in place. As Raman shows, a “document raj” also developed in the Madras Presidency, which demanded new skills in recording taxes or collecting evidence in written forms that would satisfy bureaucratic superiors. But it also stimulated novel kinds of corruption and chicanery, which, she argues, were the outcome of contradictions between documentary accountability and discretionary powers in the colonial state system.26 How to deal with new sorts of venality had to be learned as well, although we are certainly not suggesting

Education and Employment in the Colonial Period / 67

that Tamil Brahmans had a special talent for malfeasance. Nonetheless, the key point, which is crucial for the entire history of the Tamil Brahmans from the mid-nineteenth century to the present day, is that their skills were always flexible and their inheritance was advantageous primarily because it enabled them to seize new opportunities, rather than to continue in their old ways. In many respects, this is best illustrated by the other two principal learned professions, medicine and engineering. The data on the community background of students in the law, medical, and engineering colleges in Madras are particularly informative. Figures were published intermittently in annual reports on “public instruction” between the early 1880s and 1904/5; after a gap of several years, they were then provided systematically from 1916/17 to 1938/39.27 The number of students in the three colleges fluctuated considerably from year to year. Between the 1880s and 1904/5, there were always far more students enrolled in law than medicine or engineering, but the excess reduced thereafter. Between 1916/17 and 1938/39, the average enrollments were 665 students in law, 524 in medicine and 342 in engineering, but law student numbers peaked at just over 1,000 in 1923/24 and then declined to around 500, whereas medical student numbers rose steadily from under 300 to around 800, and engineering student numbers went up and down erratically, between 200 and 500 approximately. Turning to the students’ caste background, we find that among law students, the Brahman proportion varied between 70 and 80 percent in the two decades from the mid-1880s; between 1916/17 and 1938/39, it peaked at 78 percent in 1920/21 and then fell to 49 percent in 1938/39. Among engineering students, Brahmans were also preponderant by the late nineteenth century; they were 74 percent of the total in 1920/21, but down to 40 percent in 1938/39 along a path similar to the law students’. The graph for medical students is slightly different; the Brahman proportion grew slowly from a low base, reaching nearly 30 percent in 1900 and a maximum of 60 percent in 1921/22, before falling to 30 percent in 1938/39. The changing proportion of Brahman students in the professional colleges is comparable with that for students registered for the BA or other university arts degrees (which included both arts and science). Thus in 1901/2, 80 percent of the 3,379 university students were Brahmans; in 1916/17, the respective figures were 68 percent of 7,573 and, in 1938/39, 44 percent of 13,467. Brahman overrepresentation in higher education was far greater than in secondary or primary schooling, because 39 percent of pupils in secondary schools and 8 percent in primary schools were Brahmans in 1916/17, and 28 and 6 percent, respectively, in 1938/39.28 The decline in

68 / Chapter Two

Brahman numbers in the three professional colleges, and in the university and its teaching colleges, indirectly reflected the progressive impact of official measures to reduce their domination in employment, which started in 1921 with the introduction of community quotas for government jobs, although the process was slow and uneven at least until the late 1930s.29 The overwhelming majority of college and university students were male, an inequity that altered very slowly (see chap. 4). Incidentally, from the 1890s, the reports regularly noted—as one would expect—that the majority of students in law, medicine, and engineering were the sons of landholders or officials, and belonged to the “richer and middle” classes. In other words, continuities of class as well as caste were operating in the educated professions. In the Bombay Presidency, it is noteworthy that Brahmans were also greatly overrepresented among students in the professional colleges, though less extremely than in Madras; no breakdown by caste is available for Bengal.30 It is well known that Hindus, especially Brahmans, have or had a traditional antipathy to medicine, owing to pollution from corpses and bodily fluids; dissection, an obligatory part of medical education, was particularly problematic. In Bombay and Calcutta, as well as Madras, these problems arose. But in Bombay, for instance, “if such an antipathy did exist it was not long in being surmounted”; “by the mid-1880s nearly a third of students at Grant Medical College were Hindus, drawn mainly from the higher castes, especially Brahmins.”31 The negative attitude of the “higher classes of the native community” was commented on in a couple of medical college reports from Madras (and one from Bombay) in the 1850s.32 In Madras by 1884, however, the increasing number of Brahman students was said to be indicative of “the growing decline of the prejudice of the highest caste of Hindus” to modern medicine. A dip in Brahman numbers then occurred, but in 1896 another report noted “the gradual disappearance of the difficulties arising from caste or custom” that deterred Brahmans from entering medicine and its expression of satisfaction was probably justified, as the continuing growth in Brahman enrollment up to 60 percent shows.33 Engineering student numbers remained relatively low, mainly because employment prospects were worse than in law or medicine, and there were continual complaints from British officials, as well as Indians, about government departments that refused to give enough jobs to qualified Indian engineers. Proportionally, though, Brahmans studied engineering almost as readily as law. Early engineering college reports from Madras noted Brahman preponderance several times between 1884 and 1905, and regretted that there were so few students from the “artisan classes [that] should have

Education and Employment in the Colonial Period / 69

the greatest natural aptitude” for engineering, or from non-Brahmans as a whole.34 The Public Works Department (PWD) was the main employer for engineering college graduates in the late nineteenth and early twentieth centuries, although it dragged its heels about employing Indians. In addition, prejudice against Indian engineers being promoted to senior positions was widespread all over the country.35 In the PWD in Madras, however, Brahmans were of course prominent among the Indian engineers and, notwithstanding racial discrimination, when the first Indians took up senior positions in the 1890s, several Brahmans were included among them. In 1900, in all the senior grades, there were sixty-seven engineers, of whom fifty-eight were Europeans, eight were Brahmans, and one was a nonBrahman. Until 1940, the total complement of senior engineers remained fairly steady, and between 1925 and 1940—in spite of community quotas—Brahmans made up over 20 percent of the complement.36 The PWD was not the sole employer in Madras and engineers, including numerous Brahmans, also worked in other government departments or public-sector undertakings such as the railways, as well as some private firms. Many Brahman engineers also migrated to other places in India to find work, for example in Bangalore, Bombay, Calcutta, or Jamshedpur (in Jharkhand, formerly Bihar), where the Tata company’s iron and steel plant started production in 1911. Some of these men probably moved away from Madras in response to the government’s introduction of caste quotas, but these (as noted above) had little serious impact until the late 1930s; rather, employment opportunities in other places—particularly Bombay, whose economy was expanding faster than Madras’s—exerted a greater effect. Beyond the bare official figures just cited and the names in some genealogies discussed below, we know little about Tamil Brahman engineers in the colonial period. But there is some information about a few Brahmans who joined the British professional engineering institutions, presumably for reasons of career advancement or prestige, or owing to patronage by a British engineer.37 The first to do so was probably S. A. Subrahmanyar Aiyar, whose membership proposal states that he was born in 1862, studied in the civil engineering college in Madras and received his degree from the university in 1885. He was then appointed as an assistant engineer in the Madras PWD and by 1900 he had become an executive engineer, but—in common with all other Indians in his cohort—he was never promoted to a higher rank. In 1894, Subrahmanyar Aiyar was proposed for election as an associate member of the Institution of Civil Engineers by the British engineer who had supervised his early training in the PWD and he had clearly been given significant responsibility in the key field of irrigation by his superior

70 / Chapter Two

officers, because the proposal stated that he had been entrusted with both new and repair works on several irrigation projects on rivers throughout the Madras Presidency. The second member was probably V. Ganesh Iyer, born in 1885, who became a Member of the Institution of Mechanical Engineers in 1925. His proposal stated that Ganesh Iyer took his BA degree from Madras University in 1907 and a diploma in mechanical engineering from the city’s engineering college in 1911. In 1925, he was the “master mechanic” or chief engineer at Mysore Iron Works and was in England to purchase manufacturing equipment. Mysore Iron Works had been founded in 1918 by Sir Mokshagundam Visvesvaraya, then dewan of Mysore and previously the princely state’s chief engineer, as well as a member of the Institution of Civil Engineers. Visvesvaraya, a Telugu Brahman born in a village in present-day Karnataka, was the most famous of all India’s early engineers and had a long and distinguished career; he consistently advocated modern industrialization in India and made several overseas visits to inspect steel and engineering companies.38 Visvesvaraya was probably impressed by Ganesh Iyer’s own six-month tour in 1920–21, which started in Sheffield, then the center of England’s steel industry, where he attended lectures at the university, before going on to inspect factories in Britain, several European countries, the United States, and Japan. Ganesh Iyer’s career was more unusual than Subrahmanyar Aiyar’s, but it shows that early Tamil Brahman engineers were involved in modern industry as well as the old, though still vitally important, field of irrigation. In the twenties and thirties, several more Tamil Brahmans were elected as members of the Institutions of Civil Engineers and Mechanical Engineers, and also of Electrical Engineers. Their proposals show that most had studied engineering in Madras, although a few had done so in Bangalore or Poona. The civil engineers were variously employed by railway companies in south and central India, by the PWDs in Travancore and Burma, in Cochin harbor, and in a construction company in Madras; a mechanical engineer worked for the Indian government’s stores department in Bombay; the electrical engineers included men of various ranks in the electrical supply companies in Madras, Coimbatore, Mysore, and Bombay, as well as in the PWD in Madras. The civil engineers’ membership proposals, which are the fullest, show that almost all of them were recommended by British engineers who had often trained them; one public works engineer had also spent a year working in Manchester. The details about training confirm these men’s extensive, practical experience and no doubt the same was true of the mechanical and electrical engineers. We can be sure that all these Tamil Brahman engineers,

Education and Employment in the Colonial Period / 71

at least in their early careers, had got their hands dirty on dams, in found­ ries, or in power stations, and that none of them merely supervised their subordinates like bureaucrats in offices, let alone old-fashioned landlords sitting at the edge of their fields. Unlike medicine, engineering provokes no concern about ritual pollution, but it still lacks any traditional occupational precedent for Brahmans, in ostensible contrast to members of the artisan castes, as the public instruction report quoted earlier implied. Some Brahmans had always had practical and numerical skills in record keeping and astrology, for example; all the same, there is a real discontinuity between the Brahmans’ old roles as literate service people and the new professions of engineering and medicine, which contrasts with the relative continuity evident in law and administration. In his examination of how Hinduism and caste had prevented the development of modern, rational capitalism in India, Weber argued that the caste system’s “traditionalist,” “vocational ethic” impeded members of one caste from taking up the activities of another, although the system was sufficiently adaptable to allow people to take up new occupations closely related to their traditional ones. Brahmans in colonial India illustrated continuity between old and new occupations, because they “were and are infiltrating occupations, particularly administrative posts which demand writing skill and education.” Weber specifically mentioned, too, that in south India, “the Brahmans have maintained a monopoly of administrative positions into modern times,” but found it “difficult” to become doctors for ritual reasons and “are but sparsely represented in the field of engineering.”39 Yet for Brahmans themselves, such as Subrahmanyar Aiyar or Ganesh Iyer, Weber’s “caste traditionalism” was plainly not a serious impediment. The parallel between engineering and low-status artisans’ work has rarely if ever been seen to matter; nor, after a short period of hesitancy, has the ritual pollution attendant on medicine been a major difficulty. Instead—for this is the decisive point—as modern, English-language, credential-based professions that are well paid and prestigious, law, medicine, and engineering belong to the same category, and all three learned professions, like higher-level administration, have been deemed eminently suitable for Tamil Brahmans since the late nineteenth century, even if their bodies might get dirty or polluted. As much as any other change occurring at that time, the Tamil Brahmans’ decisive movement into these professions shows how they adopted new, alien forms of knowledge and skills that were, in their own eyes, compatible and acceptable extensions of their old ones. During the twentieth century, Tamil Brahman employment in government bureaucracy eventually declined and the law became less attractive.

72 / Chapter Two

Medicine became increasingly popular, but severe competition for places in medical colleges has in practice reduced its attractiveness to Brahmans in recent decades. All branches of engineering—civil, mechanical, and electrical—have vastly expanded in India since Independence and many Brahmans have gone into them; by the 1990s, more engineering graduates were finding work in the booming IT sector than in construction or manu­ facturing, and electronic, computer, and software engineering are now very pop­ular professions among Tamil Brahmans. These post-Independence developments will be discussed in the following chapter. Now, however, we turn to a few prominent individuals and their families, whose histories illustrate significant developments among the educated, professional Tamil Brahmans who joined the urban middle class in the nineteenth and twentieth centuries.

A Mathematician and Two Physicists Srinivasa Ramanujan (1887–1920), the self-taught genius who became one of the world’s greatest mathematicians, is arguably the most eminent Tamil Brahman of the modern era. Ramanujan was born into a poor family and brought up in Kumbakonam. His father, like his grandfather, was a shop clerk; his mother was a subordinate court official’s daughter. As a young boy, Ramanujan was obviously a mathematical prodigy, but his progress at school was erratic and he failed his college examinations in both Kumbakonam and Madras. Eventually, though, he managed to get a job as a clerk at the Madras Port Trust. By then Ramanujan’s exceptional ability was known to other south Indian mathematicians and, in 1913, he wrote his famous letter full of theorems to the English mathematician G. H. Hardy, in Cambridge, who recognized his genius. With a special scholarship approved by officialdom in Madras, Ramanujan went to Cambridge in 1914, where he and Hardy wrote the papers that established his brilliant and enduring reputation. In 1918, he was elected a fellow of the Royal Society, the first Indian to receive the honor. But by then Ramanujan was seriously ill and in 1919 he returned to Madras, where he died one year later.40 Notwithstanding his illustrious career, Ramanujan and his family were only extremely marginal members of the Tamil Brahman urban middle class. Ramanujan failed to get a degree and ended up as a badly paid harbor clerk, so that his social status in Madras was scarcely better than his father’s in Kumbakonam, although that would have changed after his return from England if he had not died prematurely before he could settle down as a professor. But compared with all the other notable Tamil Brahmans dis-

Education and Employment in the Colonial Period / 73

cussed in this chapter, Ramanujan’s lowly social standing makes him highly anomalous. Ramanujan, who took no part in the social, political, and religious debates of his time among middle-class Brahmans in Madras, always remained a conventionally pious Sri Vaishnava. He was interested in astrology and numerology, however, and claimed that his theorems and formulas were revealed to him by Narasim · ha (Vishnu) and his consort Na¯magiri, the goddess to whom his mother was devoted; Namagiri, Ramanujan said, wrote equations on his tongue and bestowed mathematical insights in his dreams.41 Ramanujan is sometimes said to incarnate a special Brahman or Indian bent for mathematics, not least by contemporary Tamil Brahmans, who see him as a kind of culture hero for their community and think of themselves as naturally talented mathematically. (Advanced mathematics, especially pure mathematics, it should be noted, has next to no overlap with even complex arithmetical calculation of the type taught to Brahmans and others in Tamil schools in the past.42) But what has made Ramanujan such a famously enigmatic figure is his combination of modern mathematics and traditional, demotic religion. No doubt Ashis Nandy—in comparing the Bengali scientist Jagdish Chandra Bose with Ramanujan—is right that Ramanujan’s personality formation partly explains the enigma. 43 Yet it is also salient that Ramanujan’s work probably never gave him any reason to question his religious beliefs. As Nandy recognizes, the pure mathematics, primarily number theory, that Ramanujan explored is entirely nonempirical and more closely related to logical philosophy than natural science.44 Hence theorems about prime numbers, for instance, have no direct bearing on our explanations of the material world, so that they never raise any questions about rival scientific and religious explanations of natural phenomena. To call Ramanujan a “scientist” and to conclude, with Nandy, that he was “a conservative but integrated scientist, for whom ancient meanings and modern knowledge were one” is therefore highly misleading.45 Moreover, for the larger topic of the Tamil Brahmans’ relationship with modern science and technology, it is the natural sciences, as well as medicine and engineering, that matter most, not pure mathematics. Ramanujan is indisputably fascinating, but in every respect he is so exceptional that he tells us next to nothing about Tamil Brahmans in general. Two of the most eminent of all Tamil Brahman scientists are Sir C. V. Raman (1888–1970) and his nephew Subrahmanyan Chandrasekhar (1910–95). Raman became a fellow of the Royal Society in 1924 and was the first Indian Nobel laureate, winning the prize for physics in 1930; Chandrasekhar won the same prize in 1983. Raman’s elder brother and

74 / Chapter Two

Chandrasekhar’s father was C. Subrahmanya Ayyar (1885–1960), a leading figure in the Madras Music Academy (see chap. 6). Raman’s grandfather was a nineteenth-century village landlord, who sent his son, R. Chandrasekhar, to be educated in nearby Kumbakonam; after graduating, Chandrasekhar worked as a schoolteacher and later became a mathematics professor and vice principal at a college in Vishakhapatnam in Andhra. Chandrasekhar’s first two sons were Subrahmanya and Venkataraman (who later changed his name to C. V. Raman); they graduated from Presidency College, Madras, in engineering and physics, respectively, and both then joined the government service to start work as accountants in Calcutta in 1907.46 Shortly before leaving for Calcutta, C. V. Raman married the fifteen-yearold daughter of a senior government official belonging to a different Aiyar subcaste, which scandalized the conservatives of his day, but his wedding was attended by two distinguished judges, S. Subramania Iyer and T. Sadasiva Iyer, who are said to have praised the moral courage of the two parties.47 Alongside his job in Calcutta, Raman pursued his scientific interests as an amateur, until he was appointed a professor of physics at Calcutta University in 1917. He did research into the scattering of light and discovered what became known as the Raman effect in 1928, which provided further proof of the quantum nature of light and won him the Nobel Prize. As a young man, Raman briefly became very religious and was impressed by Theosophy, which promoted the idea that purified Hinduism was both “scientific” and the quintessence of Indian spirituality, although he was apathetic toward religion for most of his life. On Raman’s views about the relationship between science and religion, however, only scanty, anecdotal information has been published. For example, there is a story that in the 1930s Raman used to go home to bathe before an eclipse; when asked why, he quipped that the Nobel Prize was science, but a solar eclipse is personal.48 This story, possibly apocryphal anyway, could simply mean that Raman observed standard Hindu purity rules, and there is also evidence that he declared himself to be a nonbeliever.49 Moreover, his few reported public statements indicate an unambiguously modernist outlook on science and a rejection of any notion that Hindu scripture can be a source of scientific truths. Thus in a speech in Banaras in 1926, Raman said: “We live to-day not in an age of the Vedas and the Upanishads; we live in a modern age; we live in an age of research; a period of intense striving to create new realms of thought, to penetrate the mystery of nature, by the use of all intellectual and material forces under human command.”50 Abha Sur, who calls Raman the “most renowned scientist in modern Indian history,” justly insists that he must be understood through his own sci-

Education and Employment in the Colonial Period / 75

entific work.51 Raman wielded great influence over Indian science, especially physics, in the mid-twentieth century, and trained more than one hundred physicists at the laboratories and research institutes he headed in Calcutta and later in Bangalore. Sur examines the protracted controversy about crystal structures between Raman and the German physicist Max Born, who had done pioneering research into “lattice dynamics.” The matter was eventually resolved in Born’s favor.52 Sur also argues plausibly that Raman’s position in this dispute, which was influenced by his Indian nationalist sentiments, was consistent, too, with a conservative, Brahman worldview and a predilection for ordered structures that derived from his appreciation of Carnatic music and a broader Brahmanical aesthetic.53 Yet irrespective of these external influences, Raman was indisputably committed to modern science and its creative practice, and his argument with Born—notwithstanding his selfrighteous dogmatism—was always a fully scientific one about experimental evidence and its theoretical explanation. Subrahmanyan Chandrasekhar had an upbringing characteristic of upper-middle-class Tamil Brahman families resident in Madras in the early twentieth century, although for many years his father, Subrahmanya Ayyar, was working in north India. Chandrasekhar, a mathematical prodigy, studied mathematics and science at Presidency College; inspired by Ramanujan, he wanted to specialize in mathematics, but his father persuaded him to study physics as a better career choice. He also wanted his son to join the Indian Civil Service (ICS), British India’s elite administrative service. Chandrasekhar agreed to study physics, but not to take the ICS examinations, and after graduating in 1930 he won a scholarship to study at Cambridge. In 1936, he married Lalitha, who was exceptional in her day as a female science graduate and also came from a notable Tamil Brahman family in Madras. Chandrasekhar spent his working life in England and the United States, where he particularly studied the evolution of stars. As Kameshwar Wali’s biography shows, Chandrasekhar became entirely immersed in the international professional world of a modern scientist, although he kept in touch with his family through regular letters and, in his own mind, never cast off his roots.54 But his father thought otherwise and, when Chandrasekhar became a US citizen in 1953, he wrote a bitter letter of protest, which is interesting because it reveals so much about how a Tamil Brahman of his class and generation perceived his own traditions at that time. Subrahmanya Ayyar reminded Chandrasekhar that he was “born of the traditions of the Cauveri [Kaveri] delta . . . with high intellectual aspirations for four generations,” was cared for by his father for twenty years, “ate . . . the salt of the Indian tax-payer,” and won a scholarship to Cambridge; he must

76 / Chapter Two

therefore realize that his “life has borne the fruit of the best of the Hindu tradition combined with the British tradition.”55 We return to the two physicists: Raman and Chandrasekhar plainly saw the world and their work in as consistently rational a way as any modern scientist ever does. They definitely did not believe that scientific discoveries came to them from Hindu deities, or that science could substantiate Hindu ideas about the nature of the universe. Raman, indeed, reputedly denounced as “mumbo-jumbo” J. C. Bose’s research in plant physiology, which sought to demonstrate that plants (and even inorganic materials) respond to external stimuli similarly to animals and human beings.56 None of this is surprising, perhaps, but we highlight these points about Raman and Chandrasekhar to emphasize that, unlike Bose, they were not interested in creating an indigenous Indian science, inspired by the Vedas or nondualistic Hinduism, which could challenge Western science and conventional scientific rationality.57 Raman’s scientific outlook may have been shaped by his Brahman worldview, but neither his nor Chandrasekhar’s work as physicists was influenced substantively by the religious culture that was part of their upbringing, or by religious ideas current in Madras at the time. One of these was Theosophy, briefly attractive to Raman in his youth, which was very influential among leading Tamil Brahmans, including members of the Mylapore group such as S. Subramania Iyer, partly because the Theosophical Society played an important role in the early nationalist movement. But Theosophy also developed, to quote Gyan Prakash, as “the most prominent vehicle for transmitting the belief in the indivisibility of science and Hinduism in South India.”58 One lesser-known Theosophist was N. K. Ramaswami Aiya, who, like other famous figures such as Swami Dayananda Sarasvati of the north Indian Arya Samaj, “blended positivism, classical Hindu texts, and modern science to authenticate a monotheistic Hinduism.”59 Another, more prominent Theosophist was G. Srinivasa Murti, who was qualified in Western medicine, but also wrote an official government report in which he supported indigenous medicine and argued that Hindu science was both radically different from and superior to Western science.60 Many Theosophists and like-minded Hindus saw their religion as validated by science, but—as Prakash persuasively argues—many of them, like Srinivasa Murti, also saw science as validated by Hinduism, for “hybridization” developed in both directions in “the explosive cross-hatching of mind with matter, Vedanta with positivism.”61 The Tamil Brahmans’ response to modernity will be examined again, especially in the concluding chapter. Concerning science specifically, though, we know too little about the typical thinking of Tamil Brahmans trained in

Education and Employment in the Colonial Period / 77

Western scientific disciplines, because relevant evidence is so scanty. Hence it is impossible to tell how many, in common with Srinivasa Murti, espoused the science-cum-religion hybrid thinking identified by Prakash, and how many, like Raman and Chandrasekhar, rejected it. On the other hand— although absence of proof is never proof of absence—no evidence suggests that any form of “traditionalist” caste ethic or religious mode of thought seriously deterred those who were intent on pursuing scientific careers, any more than those who trained as doctors or engineers. And conversely, too, it is clear that for Tamil Brahmans, modern science and technology became and remain prestigious fields in which they believe they can excel.

Dewans, Judges, and Their Descendants The three families we now examine are unusually interesting, but also exceptional, because they include men who served as dewans in princely states and as high court judges during the colonial era, when these were among the highest positions that any Indian could attain. The evolution of employment, alongside education, within these families is illustrated with genealogical material that stretches across the colonial and postcolonial periods. The first family is Seshia Sastri’s, the Eighteen-Village Vattima graduate of Madras High School who has already been introduced; the second, which includes a man who served as dewan of Baroda (in modern Gujarat) for many years, has a genealogy available on a website; the third is a family whose members were prominent lawyers in colonial Madras. Sir A. Seshia Sastri (1828–1903) is constantly mentioned by Vattimas as one of their most distinguished ancestors, especially by his descendants who can share in the prestige still emanating from his name. Seshia Sastri was the son of a poor priest living in Amaravati, a village near Kumbakonam, but he had an uncle—a dealer in precious stones in Madras—who took him to the city.62 Seshia Sastri graduated from Madras High School with first-class honors in 1848. His first post was as a clerk in the Board of Revenue, but he rose through the ranks until he was selected as one of the new deputy collectors in 1859. Seshia Sastri became dewan of Travancore in 1872, succeeding his friend and classmate Madhava Rao; after leaving Travancore in 1877, he served as regent and dewan of Pudukkottai from 1878 to 1894. In retirement, Seshia Sastri, who had become a rich man and acquired land in Amaravati, lived in the mansion he built in Kumbakonam. Compared with Madhava Rao, whose reforming zeal transformed it, Seshia Sastri’s impact on Travancore was slight, although it was much greater on Pudukkottai in both the palace and the state as a whole.63

78 / Chapter Two

Seshia Sastri was married in 1847 to Sundari, a girl from Konerirajapuram, the largest Vattima village. Sundari, whose mother had died, was brought up by her uncle, who was engaged in business in Madras. She joined her husband (and the marriage was presumably consummated) in 1853, when Seshia Sastri was the tahsildar of Masulipatam revenue division in coastal Andhra. Seshia Sastri and Sundari had no children, so he adopted his brother’s son, who in turn had two sons, Kalyanasundara and Rajagopalan. Whether Seshia Sastri’s adoptive son had a job or only looked after his property is unclear, but Kalyanasundara and Rajagopalan were landlords resident in Kumbakonam; Rajagopalan married the sister of a big landlord in Mudikondan, the richest Vattima village. In the dewan’s family, a sort of regression to the norm occurred, so that after Seshia Sastri’s rise to the top, his grandsons became conventional landlords just like many of their Vattima and other Brahman contemporaries in the early twentieth century. Why this occurred is not entirely certain, but because Seshia Sastri came from a poor priestly family and Sundari probably did not belong to one of Konerirajapuram’s rich landed families, his son and grandsons may have wanted to secure their status as landlords on a par with other wealthy Vattima families with whom they could intermarry, as their descendants have also done in succeeding generations. In the next generation in the 1950s, the family’s employment profile began to change. Kalyanasundara’s son worked in the state-owned Life Insurance Company in Kumbakonam; Rajagopalan’s two sons were the property agent for the Kanchipuram Shankaracharya’s monastery (an important institution discussed in chap. 6) and a bank officer in Tiruchirappalli. The generation after that includes two bank officers and, among men marrying into the family, a leading engineer and industrialist who worked in both north and south India, and a chemical engineer who worked in Kerala, as well as two more bank officers; all the bank officers served in local towns. In the youngest generation of Seshia Sastri’s descendants and their spouses who are now working, there is a different mixture of occupations, including three male bank officers, a female physiotherapist, and a couple who are doctors—all working in urban centers in Tamilnadu—and two couples working overseas as IT professionals. Today, only five acres of land in Amara­ vati still belong to Seshia Sastri’s descendants, although they own property in Kumbakonam. Despite Seshia Sastri’s exceptional status, his descen­ dants’ destiny during the twentieth century closely resembles that of many other Eighteen-Village Vattimas, such as the two cases we have discussed elsewhere.64 All these family trees reveal a similar pattern of employment

Education and Employment in the Colonial Period / 79

and migration, including young people now working in the United States, particularly in IT. The second family are Aiyangars from Vangal, a village beside the river Kaveri near Karur in central Tamilnadu. This family’s genealogy, available on a website, lists more than 1,000 individual men, women, and children.65 Thiruvengata Chari (1837–1934) was a big landlord with four sons and three daughters (all married to landlords), who invested heavily in his sons’ education. The eldest son, Srinivasa, became a government civil engineer in Madras and his three sons, one civil engineer and two bureaucrats, worked for the government in Madras. Thiruvengata Chari’s second son, Satagopa, became a lawyer in Salem and had four sons: one was an income tax official in Tiruchirappalli and two went to Madras, one as an oil company executive and the other working for the Reserve Bank of India; on the fourth son, there is no information. Thiruvengata Chari’s third son, Ragunatha, married a wealthy contractor’s daughter and, unlike his brothers, he lived without a job near Tiruchirappalli. Thiruvengata Chari’s youngest son was Sir V. T. Krishnamachari (1881– 1964), who studied law in Madras before entering government service. During his career, he served as dewan of Baroda for seventeen years and then as dewan of Jaipur, as an Indian delegate to the League of Nations and the United Nations, and lastly as deputy chairman of the Indian government’s Planning Commission. Krishnamachari had three sons: the eldest became advocate general for Madras, the second joined the railways in southern India, and the youngest, V. K. Ramaswami, became a senior economic advisor to the Indian government. Krishnamachari’s two daughters lived in Madras and respectively married an engineer and a railway officer. In this Vangal family, two sons of the landlord Thiruvengata Chari entered government service—one in the new profession of engineering—and one went into the law. Srinivasa kept a residence in Vangal, but Satagopa and Krishnamachari, and all their sons, lived and worked in towns or cities. The majority of the numerous descendants of Thiruvengata Chari’s professional sons had or have similar jobs, and today many of them—women now, as well as men—work in IT, banking, or other private-sector industries throughout India and overseas (mostly in America), while others are employed by the government or public sector in India. But there are some variations; Krishnamachari’s descendants and their spouses now include professors and journalists, as well as engineers and company managers, so that they include a wider range of occupations than his brothers’ descen­ dants, and more of them also work overseas.

80 / Chapter Two

The third family, also Aiyangar, came from Varahakulam, a village near Kumbakonam, and it intermarried with the Vembakkam family, originally from Chingleput, just south of Madras city. Vembakkam family members had been prominent in Madras since the early 1800s and there were several lawyers and two dewans among them; by the end of the century, the family “possessed an enormous presence in the local bureaucracy of the capital and its surrounding district.”66 The family produced a series of legal luminaries, including Sir V. Bhashyam Iyengar (1844–1908). One of Bhashyam Iyengar’s daughters married C. R. Thiruvenkatachari (1865–1935), eldest son of Rangaswami Iyengar, a wealthy landlord from Varahakulam; in addition, one of Bhashyam’s sons married Rangaswami’s eldest daughter. Like his father-in-law, Thiruvenkatachari became a high court judge. His wife Sittammal bore him twelve children, of whom eleven survived. The eldest son, Gopalsamy, had a BA degree, but may not have had a job, and one son died young. The other three sons were respectively a lawyer in Kerala, an insurance company agent in Madras, and an army doctor. Five of Thiruvenkatachari’s daughters’ husbands were lawyers in Madras and provincial towns, and two became judges in lower courts. The sixth daughter’s husband was a government engineer in Madras. Thiruvenkatachari had thirty-four grandchildren, but for brevity’s sake we mention only the descendants of Gopalsamy, whose mixture of occupations is typical. Gopalsamy’s son worked for the railways in Calcutta; his two daughters respectively married a doctor in Thanjavur and a landlord near Chidambaram. Their children and grandchildren, and their spouses, include a central government officer; two doctors (one in Pondicherry and one in America); two men employed in private companies (probably in Chennai), one of whom is married to a teacher; and one couple and another man, all with doctorates in science and technology, who work in America. Notable is that none of Gopalsamy’s children—or indeed any of Thiruvenkatachari’s grandchildren—became lawyers; this was almost certainly because, after Independence, growing competition from non-Brahmans made the law less attractive and other occupations looked more secure or better paid. Thiruvenkatachari’s descendants, who often mention Bhashyam Iyengar as well, wistfully acknowledge that they have come down in the world. One of his late granddaughters, in her nineties when she talked to us, reminisced about the governors and judges, British and Indian, who visited Thiruvenkatachari’s home; she also amusingly remembered that her grandmother used to play tennis at the elite ladies’ Willingdon Club clad in a nine-yard sari, then obligatory dress for respectable Brahman women. Elderly family members often remember, too, the mansions on Luz Church

Education and Employment in the Colonial Period / 81

Road in Mylapore where the city’s richest lawyers lived. Thiruvenkatachari’s home on this road, like Bhashyam’s vast mansion opposite, have now been replaced by nondescript buildings. Much the same fate befell almost all the big houses in Mylapore, which were sold for redevelopment after property partition. Especially from Bhashyam Iyengar’s lofty position, relative decline was almost inevitable. Yet it has been strictly limited, for nearly all Thiruvenkatachari’s descendants on whom we have information—like members of the Vangal family—were or are educated professionals and managers employed as government officials, bank officers, accountants, company managers, doctors, engineers, academics, and IT professionals in Chennai, other Indian cities, or America. Seshia Sastri and Krishnamachari, as dewans, both occupied highranking positions that were traditionally held by Brahmans in Hindu kingdoms. Judges in the superior colonial courts, like Bhashyam Iyengar and his colleagues, can also be seen as successors to the Brahman judges and asses­ sors who advised the kings in exercising their judicial functions. Hence for Tamil Brahman dewans and high court judges there were certainly ancient precedents. Moreover, some of them, such as Bhashyam Iyengar, had ancestors who had served as bureaucrats and lawyers under the precolonial or early British regimes. All the same, in the case of our particular families, discontinuity is more evident than continuity. The two dewans had fathers who were respectively a priest and a big landlord, and they had no family traditions of service; neither did the Varahakulam family, as far as we know. Varahakulam and Vembakkam were, of course, major landowning families as well. Thus Seshia Sastri, Krishnamachari, Bhashyam Iyengar, and Thiruvenkatachari—like their peers—were men who largely departed from their ancestors’ social role and way of life in order to join a new administrative or legal elite, which was educated in English and operated within a modern, rationalized, bureaucratic system of governance. Furthermore, Krishnama­ chari’s eldest brother was a civil engineer and, within one or two genera­ tions, members of all these men’s families entered other educated professions, including in particular engineering and medicine, as well as occupations such as banking or company management, which have little or no continuity with old Brahman service roles and traditional caste occupations.

Ordinary Brahman Urban Migrants The fullest ethnographic data available on ordinary urban migrants are about Brahmans from Kumbapettai, which Gough collected in 1951–52, although her published work says little about them as individuals. Gough

82 / Chapter Two

explains that the Kumbapettai Brahmans’ progressive loss of wealth and land was the main factor encouraging them to seek urban employment, so that their circumstances were quite different from those of rich landlords investing in their sons’ education. By 1952, fifty-eight Brahman men, many accompanied by their wives and children, lived and worked outside Kumbapettai. The majority were low-ranking government officials, petty bureaucrats, and clerks; the rest included a doctor and a professor, several men in business, a household priest, and a few university students. In addition, there were other emigrants who no longer maintained any connection with their ancestral village.67 In Gough’s unpublished fieldnotes, there is more information about individual emigrants, which we draw on for three illustrative examples.68 One Brahman family in Kumbapettai included five brothers and one sister, the adult children of a deceased landlord. One of the eldest brother’s sons worked as a clerk in Madras; so too did the second brother, whose daughter’s husband worked in Bombay. The widowed sister had been mar­ ried to a postal inspector, who had traveled widely across India; her el­ dest son was a draftsman in the government irrigation engineer’s office in Madras, her second son was a doctor, and her three daughters were respectively married to clerks working in Tiruchirappalli, Hospet (Karnataka), and Madras. A second family was headed by a retired station master in his late fifties, who had returned to Kumbapettai from Bombay after spending most of his life away. He had two sons, one working in the navy’s accounts department in Bombay and the other for the railways near the city, and one daughter who was married in Bombay; a second deceased daughter had been married to a police inspector in Mysore. The employment profile of a third family was rather superior to the others. A landlord’s eldest son, a graduate in his late fifties, was a retired assis­ tant engineer in the state government’s PWD. After retirement, he could not afford to buy a house in Madras, so he returned to Kumbapettai and restored his decrepit ancestral home. His younger son, also a graduate, worked in the accountant general’s office in Madras, and his elder daughter was with her husband in Delhi. The landlord’s second son was a doctor in Kumbakonam, the highest-status job of any man from Kumbapettai at the time; his third son (who died young) had been a schoolteacher in Thanjavur; his youngest son, also a graduate, worked in a town in northern Tamilnadu. The landlord’s two sons-in-law worked in the tax office in Madurai and the PWD in Madras. Other members of the extended family included an auditor in Thanjavur and another PWD employee in Madras.

Education and Employment in the Colonial Period / 83

As these three families and several others featuring in Gough’s fieldnotes show, the majority of emigrants from Kumbapettai had moved to Madras and other south Indian towns and cities, but several were in Bombay, as well as other places in the north; two men had also gone to Burma. Several genealogies reveal clusters of family members working in the same place, such as Bombay, and sometimes in the same organizations; these cases exemplify the “chain migration” and use of personal networks typical of urban migrants, whereby people from one village help relatives and neighbors who follow them. Taken as a whole, Gough’s material, alongside other available information, shows that the majority of Tamil Brahman urban migrants in the early to mid-twentieth century—like Satyamurti and his father Subrahmaniam from Tindivanam—found employment in government departments and organizations, often in subordinate positions but sometimes higher up, although some people also worked for private-sector companies and banks, and some ran businesses, usually on a small scale. Quite a few men worked as schoolteachers. A minority obtained professional and managerial employment, but this was probably commoner in the next generation or two among the sons and grandsons of men who first found work as clerks, petty bureaucrats, and so on. This progression cannot be traced for emigrants from Kumbapettai, because nothing is known about their descendants, but some information exists in the genealogies we collected from ordinary Tamil Brahman families, even though they are not as full as those of the elite families discussed above. A good example of family progress out of penury comes from a genealogy collected from Krishna, an IT professional running his own small company with his wife Indu in Chennai. Krishna, an Aiyangar, was born in 1953; he graduated in electronics from Guindy Engineering College and also took a master’s degree from the Indian Institute of Technology (IIT) at Kharagpur, West Bengal. Indu, a doctor’s daughter from Bangalore, has a science degree. Their only son is a university scientist, who works abroad. Krishna’s father’s father was a priest and landlord in Chingleput, who lost a lot of money (probably in the Depression), so that his family, which included six sons and one daughter born between 1911 and 1927, was reduced to poverty. All the sons managed to find jobs, however; the eldest became a medical representative in Madras, the second worked for a bus company in Madurai and Madras, the third joined the army, the fourth worked in the army’s civilian support service in Poona, the fifth for a publisher in Madras, and the youngest (Krishna’s father) for the railroad, mainly in Poona and Madras, while the daughter married a government

84 / Chapter Two

official based in Hyderabad. (Poona’s army base, incidentally, especially its accounts department, was a stronghold of Tamil Brahmans recruited through caste and family networks from the First World War until after Independence.69) Most of Krishna’s cousins secured lower-level, white-collar jobs in Tamilnadu, although two sons of his uncle in Poona became army engineers, also in Poona. Krishna’s sister is married to a professor in Chennai; her two married daughters are both IT professionals, one in Chennai and one in America. Krishna’s mother’s father, the son of a landlord, was a graduate, who— like his own father-in-law—worked as a stenographer, initially in the Madras High Court. In 1948, this grandfather moved to Delhi, where he eventually became chief secretary to the chief justice of India. He had two younger brothers who were bureaucrats and two sisters, respectively married to a doctor and a newspaper reporter, who all lived in Madras or other Tamilnadu towns. Krishna’s mother’s mother had two brothers: one, married to a college principal, became the manager of a British engineering company’s Delhi office and the other was a clerk in Madras. She also had two sisters, married to a clerk and a headmaster, who lived in Madras. Krishna’s mother had three brothers—one in advertising in Delhi and two airline managers, based in Delhi and abroad—and one younger sister, married to a company manager in Delhi. Krishna’s cousins on his maternal side are in a wide range of jobs, from academia to banking, but several are IT professionals in India or America. Taken as a whole, Krishna’s maternal relatives have a superior employment profile to his paternal ones, and more of them live abroad. Personal and family circumstances no doubt account for part of this difference, but also relevant is that the maternal relatives had a head start, so to speak, for when Krishna’s father’s father was an impoverished priest in the thirties and forties, his mother’s father was already working in Madras alongside his two brothers before going to Delhi. This pattern of cumulative upward mobility, whereby “success breeds success,” is evident in much—though not of course all—of our data on Tamil Brahmans whose village ancestors were poor, as well as those who were wealthy landlords.

Conclusion The first point to reiterate is that, since the nineteenth century, men in subordinate government positions and most other white-collar jobs—as well as high-ranking administrators, lawyers, learned professionals, and senior managers—have always worked within modern, rationalized, bureaucratic

Education and Employment in the Colonial Period / 85

organizations in which the primary operational language has been English, at least for written records, even if a lot of day-to-day work is in Tamil or other Indian languages. Furthermore, for most jobs a reasonable standard of school or college education has always been a necessary qualification, even though other factors, especially connections with people already in post, were or are important as well. Thus for all petty bureaucrats and clerks, as well as for dewans and judges, there was some continuity between their employment and that of Brahman literate service people in earlier times. Nonetheless, the majority of men in new occupations were actually the descendants of landowners or sometimes priests. Moreover, in almost all Tamil Brahman families, some men became doctors and engineers, or took up other occupations, which never fitted traditional caste roles. The second, crucially important point has to do with class mobility among Tamil Brahmans. In the introduction, we drew attention to the division within the middle class between its upper and lower strata, which is initially defined by the distinction in economic “class situations” between educated professionals and managers, on the one hand, and subordinate, white-collar employees, on the other. A standard sociological method for describing and measuring intergenerational mobility is to compare the class positions of fathers and sons, assuming that a family’s class is defined by its male head’s; this assumption is unproblematic for middle-class Brahman families because, even when a married woman works outside the home, she rarely has a better-paid or more prestigious job than her husband. Studies of social mobility normally investigate both upward and downward movement; in our material, there are plenty of cases of upward mobility as in Krishna’s family, but more striking is the virtual absence of any downward mobility from the upper-middle class. Thus our genealogies consistently show that after the initial stage of urban migration, the members of both elite and ordinary families have normally consolidated or improved on their predecessors’ educational and occupational status. Of course, when someone attains a very high rank, such as a dewan or senior judge, their sons rarely match them, but these are exceptional cases and, for the purposes of calculation, elite positions can be included with other educated professions. Among all the descendants (excluding landlords) of  Seshia Sastri, Thiruvengata Chari (Vangal) and C. R. Thiruvenkatachari (Varahakulam) on whom we have reliable data, there are a total of fifty-eight father-son pairs in which both men had or have professional or managerial occupations, whereas there are none in which the son took a lower-level job than his father, although the sons of professional fathers became businessmen in two cases. In the genealogies of Krishna and

86 / Chapter Two

his wife, there are eight father-son pairs in which both men had or have professional or managerial occupations, four with both in white-collar jobs, and five in which the son’s occupational status is higher than his father’s. Hence all the father-son pairs in these genealogies exemplify class stability or upward mobility, but not downward mobility. Another source of information, which is reliable for fifty-two fatherson pairs, derives from interviews with Vattimas in different localities. In thirty-three of these pairs, both father and son had or have professional or managerial occupations (including two fathers who were teachers); three sons of men with such occupations became businessmen, one became a writer, and one cannot work owing to disability, while three other sons took lower-level, white-collar jobs because, we were told, they did badly at school or college. In nine pairs in which fathers had white-collar jobs, four sons stayed at the same level and five secured professional or managerial employment. In two pairs, both father and son were in business. Apart from just three instances of downward mobility, therefore, the Vattima data show the same pattern of class stability or upward mobility. Cases of intergenerational downward mobility may sometimes be left out of genealogies or unmentioned in interviews, because people forget or ignore family members who have been unsuccessful in the world of work. Furthermore, our findings may be misleadingly partial, because most of our research was done with upper-middle-class informants or, in the case of some Vattimas, with well-off landlords. Our data on Brahmans with lower-level, white-collar jobs are thin, so that we do not know how frequently their position results from downward mobility, as opposed to stability within the lower-middle class and obstacles preventing their upward mobility. But even with all these caveats and qualifications, the rarity of downward mobility from the upper-middle class, as indicated by occupation, is a notable feature of the genealogical and interview material, which is also consistent with our other ethnographic data, so that we are convinced it is not a spurious finding. It may be placed alongside the calculations by Sanjay Kumar, Anthony Heath, and Oliver Heath, using large-scale survey data for India from 1996, which showed that 58 percent of the sons of men in the “higher salariat” class stayed in the same class. The authors reasonably interpret this figure as part of the evidence for high levels of class continuity and inequality in India.70 Obviously, our limited data are not truly comparable with those in the survey. Nonetheless, the comparison strongly suggests that the rarity of downward mobility among Tamil Brahmans is extreme. And that is sociologically important, because “mobility closure” in the downward direction, though not the upward, is critical to the devel-

Education and Employment in the Colonial Period / 87

opment or “structuration” of the upper- and lower-middle classes among Tamil Brahmans.71 The “class awareness” of Brahmans in the upper-middle class is strongly reinforced by minimal downward mobility. But so, too, is the upper-middle class’s consolidation as a social class that is also distinguished from the lower-middle class as a status group, if only because the offspring of “better” families are unlikely to “lower the tone” by dropping down the class hierarchy. The next chapter ends with a further discussion of class mobility and class reproduction. But to anticipate our later conclusion, we are sure that the overwhelming lead in modern education and urban employment that the Tamil Brahmans established by the end of the nineteenth century was a highly significant historical development. They thus became members of the urban middle class, especially the upper-middle class, earlier and faster than people from any other caste or community in south India. Because the Brahmans’ progressive incorporation into the urban middle class continued throughout the twentieth century, their economic position and social status became more and more determined by their capacity to sustain and reproduce themselves as middle-class within a modern society and economy, and less and less by the persistence of caste privilege itself.

Three

Education and Employment after Independence

Tata Consultancy Services (TCS), which was founded in 1968, is now In­ dia’s biggest software and services company. S. Ramadorai, who joined TCS in 1972, retired as its CEO in 2009. Ramadorai, who describes himself as the son of “a typical Tam Bram . . . family,” was born in Nagpur, Maha­ rashtra, in 1945. His mother’s father retired to Nagpur after working in the Posts and Telegraphs audit department in Bombay. His father’s father was a landlord and village accountant in Sengalipuram, one of the EighteenVillage Vattimas’ villages, and Ramadorai spent part of his childhood living there with his grandparents. His father worked for the Indian government and retired as accountant general of  Tamilnadu in Madras, but spent much of his career in Delhi. Ramadorai, who is fluent in Hindi, went to school in Delhi where, he admits, he was not a very diligent student, despite his stern father’s insistence on studying hard, his belief that “all that mattered . . . was being extremely good at mathematics or science,” and his hope that his son would become a doctor. At university in Delhi, where he finally began to take academic work seriously, Ramadorai obtained a degree in physics and mathematics; he then took a further degree in electronics and telecom­ munications at the Indian Institute of Science in Bangalore, graduating in 1968. In 1969, he went to the University of California, Los Angeles, where he completed a PhD in computer science, and afterward worked for a short time in Los Angeles. Ramadorai returned to India in 1972 and joined TCS as a systems pro­ grammer and analyst, becoming one of the early members of a small pio­ neering team that played a vital part in establishing India’s IT industry. At that time, TCS’s general manager was F. C. Kohli, “the architect of India’s IT industry,” who became Ramadorai’s mentor. Another young member of the team was S. Mahalingam, a Vattima, who eventually became TCS’s

90 / Chapter Three

chief financial officer. Back home in India, Ramadorai had a traditional ar­ ranged marriage to Mala, who had been brought up in Madras and was an accomplished vocalist in classical Carnatic music. During an eventful career at TCS, partly based in the United States but mostly in India, Ramadorai played a leading role in the rise of TCS from a small and rather shaky busi­ ness into a multinational company, whose workforce numbered around 30,000 in 2002 and has since been recruiting between 30,000 and 50,000 new people per year.1 In 1900, an ambitious young Tamil Brahman would probably have hoped to have become a high court judge, but by 2000 he— or even she—would be much more likely to aspire to the position of CEO in a major IT company. This change, and its causes and consequences, is at the heart of the Tamil Brahmans’ story in the second half of the twentieth century.

Tamil Brahmans and the Reservations System By the late 1930s, as we have seen, employment quotas had begun to re­ duce Brahman recruitment in the Madras government service. Some Tamil Brahmans were finding employment in private companies and some were emigrating from Tamilnadu to other localities, such as Bombay, where there was faster economic growth and industrialization. Since the 1930s, no of­ ficial statistics about the education and employment of Tamil Brahmans in their home state or elsewhere have been published, so that we have only some indicative information from surveys undertaken for the Ambasankar Commission, whose report was published in 1985. It showed that the overall “indices” for higher academic qualifications among Brahmans were higher than for any other caste in Tamilnadu at that time, thus confirming that their pre-Independence lead over the rest of the population had been maintained over four decades.2 It is highly unlikely that this position has changed since the 1980s. In 1951, the Madras state government, led by the Congress party, in­ troduced a 25-percent quota for Other Backward Classes (OBC) in college places and government jobs; the OBCs include the bulk of the nonBrahman castes. The DMK government increased the quota to 31 percent in 1973 and the ADMK government raised it to 50 percent in 1980, to which must be added 18 percent for the Scheduled Castes and Tribes (SC and ST). The OBCs are now subdivided into “Most Backward Classes,” “Deno­ tified Tribes,” and the rest.3 These quotas have progressively reduced the number of “open” college places and government and public-sector jobs in Tamilnadu for which Brahmans and members of other Forward Castes

Education and Employment after Independence / 91

(FC) can apply. As the quotas affect new recruits, but not those already in post, the reservations system reduced the number of Brahmans in the state government service only slowly, but it did so steadily, so that not many were left by the late twentieth century. Moreover, the system discourages Brah­ mans from applying for jobs with the government, because they feel that even if they manage to obtain them, they will not be treated fairly by nonBrahman superiors over matters like promotion. Lower-middle-class Brah­ mans from poorer families without good educational qualifications are actually the ones who most bitterly resent the reservations system, because they are convinced that it excludes them from secure jobs in the Tamilnadu government’s huge workforce. They form the core support of the Tamil Nadu Brahman Association (Tambras), a caste association that provides var­ ious social and charitable services to Brahmans in need, but also campaigns (without any success) for changes to the reservations policy that would benefit Brahmans from poorer families.4 Brahmans who are better off and better educated continually complain about the reservations system as well, but they are rarely active in Tambras. In the job market, moreover, they are not really affected by reservations very much, because they no longer apply for state government positions. In college education, on the other hand, the reservations system has a negative impact on large numbers of middleclass Brahmans, mainly owing to the high “capitation fees” that they often have to pay to get into private colleges if they cannot secure admission to a government college. This problem, which is acute in engineering colleges, is further discussed below. In central government services and public-sector organizations, as well as higher educational institutions, SC and ST quotas have been in place since shortly after Independence, but there were none for OBCs until the recommendations of the Mandal Commission began to be implemented after 1991.

Education Among Tamil Brahmans, in common with other Indians, parents or other senior kin are always involved in decisions about young people’s educa­ tion, at college as well as at school. For boys, educational decisions have al­ most always been made in relation to employment opportunities and many informants told us that their fathers or uncles had guided, encouraged, persuaded, or even forced them to take a particular kind of degree or profes­ sional training, and then an appropriate job. Thus some men said that their own personal preferences had been overruled within the family, so that they

92 / Chapter Three

had wanted to become research scientists, for example, and had ended up as engineers. For girls, marriage prospects have normally been the most critical factor in educational decisions; employment opportunities only began to be given much weight toward the end of the twentieth century, particularly after jobs in the IT industry became attractive. One hundred years ago, even in elite urban families, Brahman girls were rarely educated beyond primary school, but they started to go to secondary school and eventually to college as more and more educated professional men began to prefer wives who were better educated. By the end of the twentieth century, the gap between male and female education was minimal. The evolution of female educa­ tion in relation to marriage and paid employment, which is a crucial aspect of the Tamil Brahmans’ transformation into a modern middle class, will be examined in more detail in the next chapter; in this one, although some of the discussion of education and employment applies to both sexes, most of it is primarily about men. Many middle-aged and elderly Tamil Brahmans we spoke to went to government schools, where some or all of their education was in Tamil, but most of the younger ones went to private, English-medium schools. Indeed, these are the schools attended by the great majority of urban middle-class boys and girls from all castes and communities throughout Tamilnadu and much of the rest of India today. A lot of private schools are poor and their teachers can hardly speak English, but in Chennai and other towns and cities in Tamilnadu there are excellent ones as well; upper-middle-class par­ ents strive hard to make sure that their children get into the best schools and accept the financial burden that their high fees impose. In Chennai, admis­ sion to the top schools is particularly fraught, so that, for example, even pre­ kindergarten classes for three-year-olds may be assessed in relation to how well they prepare little children for primary school selection, and parents anxiously dread the interviews they must undergo as part of their children’s selection process. Chennai’s leading schools vary in their pedagogic ethos, so that some are very academically competitive and examination-oriented, whereas others place greater emphasis on extracurricular activities or are more “progressive.” Some of these schools are secular, some are Hindu in­ stitutions, including several dominated by Brahmans, and some are Chris­ tian or Muslim; many Tamil Brahmans are happy to send their children to reputed Christian or Muslim schools. In the Indian education system, children normally start school when five years old. Primary school has five grades, first to fifth standards, and second­ ary school five more, sixth to tenth standards; pupils successfully complet­

Education and Employment after Independence / 93

ing the final grade are awarded the Secondary School Leaving Certificate (SSLC). Two more years of study, now eleventh and twelfth standards in Tamilnadu, but formerly the Pre-University Certificate (PUC) course and still earlier the Intermediate course, are required before entering college or university to study for a bachelor’s degree. Irrespective of which sort of school they attend, and despite disapproval from many teachers, children in Tamilnadu are frequently sent to extra coaching classes before or after school hours, especially when they are ap­ proaching the SSLC examination. More coaching often follows in eleventh and twelfth standards when studying for college entrance examinations, especially for students who want to apply to medical schools, the Indian Institutes of Technology (IIT), or other extremely competitive, universitylevel institutions. Even coaching centers themselves may have selection processes, because those with good reputations receive so many applicants. Although almost everyone deplores it, a lot of young people do next to nothing but study for twelve or more hours a day, seven days a week, for months and months on end, in order to try to score the very high marks needed for admission into the top institutions. From early childhood, Brah­ man and other Forward Caste boys and girls are told that they must try even harder and do even better in exams than lower-caste children, because the reservations system discriminates against them. Most children learn that lesson well and the combination of parental encouragement and pressure undoubtedly contributes to the Tamil Brahmans’ scholastic success. Today as in the past, many Tamil Brahman young men and women study for conventional arts and sciences degrees, usually at one of the large number of government or private colleges affiliated with the state’s univer­ sities. Male students have long been biased toward science subjects, but fe­ males are increasingly following suit. Some vocational degrees, such as the BCom (bachelor’s in commerce), were more popular in the past than they are now. Since the nineteenth century, however, as we saw in the previous chapter, professional qualifications in law, medicine, and engineering have been held in highest esteem and that remains largely true today, although the attractiveness of law degrees, and the legal profession, was declining by the mid-twentieth century. Thus in Bangalore by 1970, and probably Madras as well, many lawyers admitted that they had hoped to become engineers, doctors, or civil servants instead.5 The medical profession is highly prestigious and remunerative through­ out India, and the number of medical students has risen continuously since Independence. Many young Tamil Brahmans, like other Indians, would like

94 / Chapter Three

to become doctors and, unlike law and engineering, medicine has always attracted a significant minority of women. But Tamilnadu’s reservations policy, coupled with the ferocious competition for places in government medical colleges, have severely reduced the number of Brahman medical students over the last few decades, so that in practice medicine has lost much of its appeal as a realistic career option. Very rich families may be able to send their children to reputable, private medical colleges in India or overseas, but for most salaried middle-class families, let alone the rest of the population, the fees are impossibly high. Engineering, however, presents a different picture. In 1949–50, India had around 12,000 students in engineering and technology (more than in law but fewer than in medicine) and produced approximately 5,500 gradu­ ates per year.6 After Independence, the prospects for engineers improved greatly. Jawaharlal Nehru’s “modern temples”—the dams, power stations, steel plants, and heavy engineering factories built by the state as the center­ pieces of India’s industrialization program, as well as those run by Tata and other private-sector firms—all required sizable cohorts of engineers, but many men also worked in smaller companies. Thus the number of en­ gineers in India began to grow markedly, so that by 1970 the annual output of engineering graduates was around 20,000; it rose more rapidly after the IT sector had started to grow, so that by 1997 the annual output was around 65,000 and by 2006 nearly 220,000. The exponential increase since the late 1940s is beyond doubt, even though the reliability of the figures cited is questionable.7 The Indian Institutes of Technology are generally recognized as the best institutions in the country for engineering education. By training highquality engineers, as well as other scientists and technologists, the IITs were designed to make a major contribution to India’s industrialization and the first one opened in Kharagpur (near Calcutta) in 1951; IIT–Madras, the third to be set up, opened in 1959. Because they are central government institutions, which had quotas only for SCs and STs, IITs have been particu­ larly attractive to Tamil Brahmans facing OBC quotas in colleges controlled by the Tamilnadu government; in 2008, however, quotas were also intro­ duced for OBCs in IITs. Competition for places in IITs has always been stiff and more than fifty applicants vie for every place, but Tamil Brahmans are widely believed, probably correctly, to be very well represented among both students and faculty. Below the IITs on lists ranking India’s top engineering colleges come a range of government and private institutions, many of them in south India;

Education and Employment after Independence / 95

they include, for instance, Anna University (successor to the old Guindy Engineering College) in Chennai and the private PSG College of  Technol­ ogy in Coimbatore. The top colleges, however, are greatly outnumbered by the rest, many of them established recently in response to the rapid growth of the IT industry. Thus Tamilnadu had twenty-nine engineering colleges in 1985–86, seventy-two in 1995–96, 254 in 2005, and 471—with 150,000 places—in 2010.8 Despite attempts to regulate private engineering colleges—which are the majority in Tamilnadu (and other states)—the stan­ dard of education in many of them is notoriously poor, they cannot recruit enough students to fill all their places, and their graduates cannot obtain jobs in reputable companies. To secure a place in the top government and private engineering col­ leges, very high marks in both the twelfth-standard examination and the state-wide professional college entrance examination are required. On the other hand, competition for places in engineering colleges is less severe than in IITs and medical colleges. Moreover, in private engineering colleges, un­ like government colleges, the reservations system does not apply to some of the places (“management seats”), which makes it easier for FC students to gain admission; these students normally pay higher tuition fees than those admitted through open competition, but the fees are still low compared with private medical colleges. Private colleges regularly demand capita­ tion fees or “donations,” however, which often amount to several hundred thousand rupees, far more than the annual tuition fees. Many middle-class families struggle to pay the capitation fees, which are widely condemned as an underhand racket. Other things being equal, the reservations system means that Brahman and other FC students are more likely to have to pay these fees than OBC, SC, or ST students. All the same, despite the financial and other difficulties facing them, large numbers of middle-class Brahmans who do well at school manage to study engineering. Brahmans are certainly not the majority of engineering students as they were in Madras a century ago, but they are still very well represented in the various branches of the discipline. The prestige of medicine and engineering, and related fields of tech­ nology, means that the vast majority of academically able young people, supported by their parents, want to study these subjects, or sometimes mathematics and natural sciences, rather than the arts, humanities, and social sciences. This is important for the modern IT industry, whose foun­ dations lie in post-Independence India’s scientific, technological, and in­ dustrial manufacturing base, because one critical condition for its growth

96 / Chapter Three

has been the flow of graduates from engineering colleges. In particular, a good supply of highly skilled computer and software engineers was vital in some sectors of the industry, such as TCS in its early years. But that is much less true today. Although TCS, and other major Indian software and services companies, recruit tens of thousands of graduates from the top engineering colleges each year, many of them have studied civil, mechanical, or elec­ trical engineering, rather than electronic, computer, or software engineer­ ing. Their specialization does not matter, because, as the CEO of one major software company explained to us, graduates are not actually employed for their engineering expertise, which is rarely needed for their work. Rather, his company recruits them because so many of the cleverest young people go to the top engineering colleges; there, too, they acquire good analytical and problem-solving skills, whereas students in other colleges usually do not, so that only a small number of their graduates, mostly in science, are taken on. Nonetheless, if history or sociology graduates, for example, were in general as clever as engineering graduates and were equally well trained, his company would recruit them and they would be just as good at software engineering. The CEO, a Tamil Brahman, whose own daughter was studying sociology in a Chennai college, thought that it was the right choice for her, so that he was not himself prejudiced against the subject. Instead, he was explaining the rationale of his and other IT companies’ recruitment poli­ cies, which are, to a large extent, both the cause and effect of engineering’s prestige and attractiveness to so many bright, young Tamil Brahmans and other Indians.

Employment and Business Unlike the Tamilnadu state government, whose reservations policy increas­ ingly discouraged Tamil Brahmans, the Indian central government was an attractive source of employment until quotas for OBCs were introduced. From 1947 onward, many Tamil Brahmans secured central government jobs and many of them therefore moved to Delhi. By the 1960s, 40,000 south Indians lived in Delhi and as many as 75 percent of them were prob­ ably Brahmans, especially Tamil Brahmans; another similar estimate is that about half of the approximately 55,000 Tamils in Delhi in the 1980s were Brahmans.9 In central government offices, some Tamil Brahmans rose to very high positions and were apparently glad to take revenge on antiBrahman politicians from Tamilnadu who visited Delhi by denying them central government support for projects in their home state.10 Bureaucrats and public-sector staff are employed all over India, however, and employ­

Education and Employment after Independence / 97

ment by the central government is one reason why Tamil Brahmans became so widely dispersed throughout India after Independence. Another source of employment is the banking sector and a lot of Tamil Brahmans became bureaucrats in banks instead of government offices. In banks, like the rest of the private sector, there were no reservations until the large banks were nationalized by the central government in 1969. Nationalized banks, like other central government organizations, regularly transfer staff between of­ fices so that they, too, have contributed to the Brahmans’ pattern of mi­ gration and circulation across the country. By the 1990s, however, mainly because employment opportunities expanded and salary levels rose in the private sector following economic liberalization—but partly also because of concern about extension of the reservations policy—Tamil Brahmans were turning away from the central government and its public sector, including the nationalized banks. Even before Independence, though, Tamil Brahmans found jobs in pri­ vate companies, large and small, in Madras and elsewhere in India. Some men were employed as qualified engineers or accountants, for example, but many others belonged to the general management or office staff, and they worked at all levels from senior managers to petty clerks. Indeed, privatesector company management, alongside government administration, is one field in which Tamil Brahmans have long been prominent. Most companies employing Brahmans are not of course owned by them, but some are and should be briefly discussed. There is a common belief—held even by Brahmans themselves—that they make bad businessmen compared with Chettiyars, traditionally the preeminent commercial caste in south India, but the reality is quite different, as John Harriss shows.11 Some Tamil Brahmans, such as the Eighteen-Village Vattimas, but most famously those from the Brahman village of Kallidaiku­ richi in the Tambraparni valley, have a long tradition of moneylending and banking; by the early twentieth century, Kallidaikurichi Brahmans had ex­ tensive banking interests throughout southern Tamilnadu. Brahman banks include the Kumbakonam Bank established in 1904, principally by Vatti­ mas, which later developed into the City Union Bank (CUB), and the IndoCommercial Bank established in Mayavaram (Mayiladuthurai) in 1933 with capital from the Kallidaikurichi Brahmans.12 S. N. N. Sankaralinga Iyer, a Kallidaikurichi Brahman, was one of the Indo-Commercial Bank’s found­ ing partners and its first manager; another partner was R. A. Viswanatha Iyer, a Vattima who studied law in Madras but never practiced. Sankaralinga Iyer also established an industrial company, India Cements, in 1946. In 1939, two other Kallidaikurichi Brahmans, K. R. Sundaram Iyer and his nephew

98 / Chapter Three

K. Iswaran, founded Easun Engineering, which manufactured Royal Enfield motorbikes. The owners and managers of India Cements and Easun feature among the industrial leaders of Madras studied by Milton Singer in the 1960s and restudied by Harriss in 2000.13 Not far from Kallidaikurichi is Alvarkurichi, the Brahman village home of S. Anantharamakrishnan, who in 1941 founded the Amalgamations Group, a very large, family-controlled, industrial conglomerate based in Tamilnadu. An even larger and much better known industrial group (also studied by Singer and Harriss) is TVS, founded in Madurai in 1929 by T. V. Sundaram Iyengar, which now incor­ porates several separate companies run by different branches of his family. In addition to big companies, Tamil Brahmans have established numerous smaller ones, such as the light engineering firms studied in the 1950s by James Berna, who noted the prominence of Brahmans among engineering entrepreneurs, and reported their awareness that times were changing and industry was a suitable field for them, partly in reaction to anti-Brahman discrimination in government employment.14 More recently, as Harriss shows, Brahmans remain prominent among south India’s established busi­ nessmen and new entrepreneurs as well, especially in IT; on the other hand, Tamil Brahman banks and businesses have not expanded across India or into neighboring countries like those of the Nagarattars (Nattukottai Chetti­ yars) or north Indian Marwaris. Brahman-owned companies do not employ Brahmans only, but they are usually well represented on the staff, more so among the top than the middle management.15 Turning to the learned professions, we have already noted that the law was declining in popularity from the mid-twentieth century. A sizable group of Brahman lawyers and judges still exists in Tamilnadu, but they are now a small minority and no longer dominate the high court and the rest of the legal system, as they did a century ago. Although entry into the medi­ cal profession has become very difficult, there are still quite a lot of Tamil Brahman doctors, who are probably more numerous than Brahman lawyers nowadays. Engineering, on the other hand, presents a very different picture by virtue of its much larger scale. During the colonial period, in Madras and elsewhere, employment opportunities for Indians were restricted because European engineers resisted “Indianization” in government organizations, like the Public Works Department, and there were few Indian employers, such as Tata or TVS, to provide alternatives. After Independence, all that changed as the demand for engineers expanded enormously, and in all three main branches—civil, mechanical, and electrical—Tamil Brahmans were and are conspicuously present.

Education and Employment after Independence / 99

The State and Private Sectors, and the “Old” and “New” Economies In our research in Chennai in 2003–5, one objective was to compare pro­ fessional engineers and other managerial staff in older, “core” engineering manufacturing companies with software engineers and other IT profession­ als in the new software and services companies. This was also a comparison between older and younger generations, because senior managers in manu­ facturing companies are normally at least fifty, whereas in IT companies they are often only in their thirties. But the difference between manufactur­ ing and IT companies is also partly a contrast between “old economy” and “new economy” firms. IT has the highest profile within the private-sector, new economy that has thrived since liberalization started, but other organi­ zations—such as new banks, airlines, or retail chains—are also significant sectors of it. Obviously, Tamil Brahmans, like other middle-class people, earn their living in a wide range of jobs, not just in engineering and IT, but these two industries are particularly important in modern Tamilnadu. One important point is that many people described below have salaries high enough to place them in the top bracket of Indian income earners, so that they can be accurately described as members of the “newrich” upper-middle class.16 In 2001–2, about 6 percent of Indian house­ holds were classified as “high-income” and, by 2009–10, the proportion had risen to about 20 percent; these households have an annual income of 180,000 ru­pees at 2001–2 prices, equivalent to approximately 325,000 ru­ pees at 2009–10 prices.17 In 2010, newly recruited engineers in the top soft­ ware companies were paid about 300,000 rupees per annum, and managers with ten years’ experience could earn 1.2 million rupees. In the manufactur­ ing companies, the starting salary was lower at around 100,000 rupees per annum, although it rose to about 300,000 rupees after three years’ training; it might double after ten years and a senior manager with twenty to thirty years’ experience could receive 1.2 million rupees per annum. As these figures show, IT professionals are paid more generously than managerial staff in manufacturing companies, but most of the latter still earn enough to put them among India’s highest earners, given that their personal incomes commonly equal or exceed that of “high-income” households. Senior managers in both the manufacturing and IT companies typically live with their families in two- or three-bedroom apartments that they own in Chen­ nai’s middle-class suburbs and they run an ordinary family car. By the stan­­ dards of the vast majority of their fellow citizens, they live very comfortably.

100 / Chapter Three

Nevertheless, they do not belong to the wealthy Indian elite who reside in air-conditioned mansions, own luxury cars, and patronize shops selling world-famous brands of clothes or jewelry. International comparisons are always problematic, but the material standard of living of the people we describe is mostly still below that of equivalent professionals and managers in North America or Western Europe. The division between the old and new economies, as well as the contrast between the state and private sectors, were very plain to all our informants, who typically placed a lot of emphasis on them when describing how the middle-class Brahmans’ world has changed since Independence. But many of the people we interviewed in our research in 2003–5 were non-Brahmans from a variety of castes, whose outlook on work and life more broadly often differed little from that of Brahmans with similar jobs, educational quali­ fications, and income levels—except in opinions about the reservations system. In many contexts, therefore, common membership of the profes­ sional and managerial upper-middle class matters more than different caste backgrounds. One key marker of class is competence in English; in all these companies, old or new, private or nationalized, English is the working lan­ guage and all managerial staff are reasonably fluent in it, as are those at the higher echelons of the government bureaucracy. Everyone old enough to have started work by the 1980s agrees that jobs in the central government administration or its public sector, such as the nationalized industries or banks, or services like the railways, were formerly regarded as the best. Employment in the Indian Administrative Service (IAS), to which entry is by competitive examination, was highly presti­­ gious, but so too were jobs in leading organizations like the Indian Space Research Organization and some technologically advanced companies like Bharat Electronics Limited (BEL) or Hindustan Aeronautics Limited (HAL). In the first place, all government and public-sector salaries were reasonably good, and steady promotion and salary rises could be expected, but most importantly there was security of employment, which mattered a great deal when worthwhile jobs were scarce. Thus among middle-class Christians in Chennai in the 1970s—though the same would have applied in other com­ munities—Lionel Caplan found that the perceived shortage of economic opportunities, after the rapid development of the post-Independence pe­ riod faded, had “encouraged a near siege mentality” among his informants, which meant that people used to “cling desperately to the most lucrative and prestigious kinds of employment.”18 In preferring jobs with the state, people also mentioned various other factors. Sampath is a Tamil Brahman in his fifties with a BCom degree and a

Education and Employment after Independence / 101

master’s degree, who is the personnel manager in Chakra Vehicles (a pseu­ donym), a major bus and truck manufacturer; he said that it was generally believed that government and public-sector jobs were better because they were obtained on grounds of merit, whereas positions in private compa­ nies were secured by people with the right connections. Some public-sector firms were regarded as very interesting places to work as well. Anil, a Kan­ nada Brahman in his fifties, who has an engineering degree from IIT–Madras and is a senior manager in Chakra, found his first job in Hindustan Machine Tools; he stayed there for five years in the late 1970s and remembers the company as one of the navaratnas (“nine jewels”) of the public sector, an excellent place to work. Krishnamurti, a Tamil Brahman in his early sixties who is a senior manager in one of the TVS manufacturing firms, obtained his first job in BEL after graduating in engineering. He worked for BEL in Bangalore and a town near Delhi for fourteen years, and said that publicsector firms like BEL were the most interesting in the late 1960s and early 1970s, because they had access to advanced technology through foreign collaboration agreements. Commitment to nationalist ideals could also count, especially during the first two decades of Independence, when Nehru was prime minister until his death in 1964. A Tamil Brahman retired accountant who spent his entire career in the Food Corporation of India (FCI) reported that he was offered a post in a private company in the early 1960s, but chose to work for the government because he answered “a call from Nehru”—which demonstrated his patriotism, said his friend who had also retired from the FCI. These two men epitomize the post-Independence, “Nehruvian” middle class, whose eclipse since liberalization has been widely discussed.19 Very similar is Viswanathan, a Tamil Brahman from Kerala, who worked in the nationalized steel plant in Rourkela, Orissa, from 1960 until he retired to Chennai in 1992. In Rourkela, Viswanathan had colleagues from all over India and he reminisced about his career in something akin to the Nehru­ vian style discussed for the Bhilai plant by Jonathan Parry.20 Several retired bank officers also explained how bank nationalization, partly because it led to many staff transfers between regions, positively enhanced a sense of commitment to national development. A considerable number of Tamil Brahmans used to work for the defense services, including Sampath’s father. Sampath was brought up on military cantonments, which made a deep im­ pression on him, because their inhabitants lived side by side in a strongly secular environment mostly unaffected by caste divisions, which helped to create a feeling of national unity among Indians of all social backgrounds from across the country. Last but not least, it matters that large numbers of

102 / Chapter Three

Tamil Brahman men, and often their wives and children as well, learned Hindi or other Indian languages because they lived and worked in differ­ ent parts of India. Viswanathan and his wife, for example, can speak Hindi, Bengali, and Oriya, as well as Tamil, Malayalam, and English. Such linguis­ tic skills were practically useful to many people, but they also encouraged the Brahmans’ disdain for Tamil-language chauvinism in their home state. Middle-class attitudes toward the government and public sector changed completely toward the end of the twentieth century, so that everyone we interviewed, irrespective of their age or whether they once worked for the state, now takes the view that private-sector jobs are much better. Most par­ ents do not advise their children to look for government and public-sector jobs and almost all the young people we met made it clear that they had no intention of doing so. And the retired FCI accountant bluntly said that government jobs are now for people who cannot get anything else. There are three principal reasons for this change. First, many people insist that government and public-sector employment have become com­ pletely politicized—partly because of the reservations system, according to Brahmans—so that the state itself has become corrupt and no longer serves the nation and people. This judgment is made with special vehemence by men who used to work for the government, especially as civil servants, but it is almost universally held, so that nobody still looks to the state to pro­ mote national unity and development. Instead, as is characteristic of today’s middle class, they put their faith in economic liberalization, and private initiative and enterprise. The second reason is the general assumption that low salaries, bureau­ cratic hierarchies, and stifling conservatism characterize government and public-sector jobs. Nationalized industries, it is said, also became sclerotic well before liberalization exposed their mismanagement, so that even com­ panies like BEL are no longer exciting places to work compared with many private companies. Thirdly, attitudes toward security of employment have altered radically. Although parents sometimes express caution, young people in their twen­ ties or thirties are unimpressed by secure jobs with reliable monthly salaries, annual increments, and slow but steady age-related promotion, because they just do not fear unemployment and poverty like many older people did at a similar age. The fundamental reason for this change in attitude is the high economic growth rate, especially in south India since the early 2000s, and the huge expansion in private-sector, professional employment that has accompanied it. The IT sector, which has expanded fastest, is highly suscep­ tible to global economic cycles. Within the last ten years it has experienced

Education and Employment after Independence / 103

the dotcom bubble bursting in 2001 and the downturn in 2008–10 linked to the worldwide recession, which both led to job losses and recruitment freezes. Yet neither downturn seriously dented people’s confidence that the Indian IT industry—as well as profitable manufacturing companies such as Chakra—have bright, long-term futures in an environment of irreversible economic liberalization and globalization. Hence those who work in these industries can be confident about their own futures, even if they sometimes have to change jobs. To be sure, this optimism may fade in future years and is largely restricted to members of the professional and managerial uppermiddle class with the qualifications required for good jobs, whereas the majority of the population, irrespective of caste background, cannot afford to be so disdainful about secure employment with the state. But for those who can, including a large number of Tamil Brahmans, there has been a sea change in attitudes toward government and public-sector jobs since the 1980s.

Working for Manufacturing Companies Old-economy industrial firms vary greatly in size, from the huge Tata con­ glomerate to the medium-sized Chakra to many smaller companies. There are some significant differences, too, between firms that are not family busi­ nesses, such as Chakra, and those that still are mainly under family owner­ ship and control, such as the TVS companies and indeed the majority of top companies based in Chennai.21 Before liberalization, all private-sector companies operated within an economy dominated by the state sector and the licensing system of the “per­ mit raj.” Some companies have adjusted to the postliberalization climate better than others, but all of them have had to change their business prac­ tices since the 1990s, especially to cope with increased domestic and foreign competition in their product markets. Manufacturing company managers have also had to keep up with innovations in the manufacturing process brought about by automation, the rise of the IT sector and the resulting competition for high-quality personnel, the introduction of modern staff appraisal systems, and so on. They tended to complain about stress, but also said that their work has become more challenging and interesting. All these changes mean, in brief, that companies that once resembled govern­ ment bureaucracies do so less and less. Tardy decision making because there was little reason to rush, extended managerial hierarchies regulating slow, age-related promotion, masses of paperwork undertaken by a large support staff, and subordinate peons wandering around with armfuls of files and

104 / Chapter Three

cups of tea and coffee—these are all vanishing features in successful manu­ facturing companies, whether family-owned or not. Manufacturing company managers mainly work in offices, of course, but during their training they must learn about their company’s products, how they are made and who buys them, and they also find out about work on the factory floor. Some managers regularly go to factories throughout their careers and interact with the workers there. In Chakra, in common with other vehicle manufacturers, automation and modern working meth­ ods have transformed factories, so that they are cleaner and safer than in the past. Factory workers, who are all male, are also better educated than their predecessors. Thus a generation ago, illiterate workers were common in Chakra’s factories, whereas today some young workers have engineer­ ing diplomas and often control automated equipment with computers in­ stead of machinery. Nonetheless, factories are relatively dirty, noisy places compared with offices, and the workers, many belonging to low castes, selfevidently belong to a lower social class than managers. Some managers, such as those in sales and marketing, may travel extensively to visit custom­ ers. Chakra’s domestic customers include many small, privately owned bus and haulage businesses, which are commonly based in small towns and run by local entrepreneurs from a wide range of social backgrounds. Talking to men of lower class and caste status in dirty or remote sites, such as factories and small-town garages, is therefore part of the job for many Chakra man­ agers, even if they do no hands-on engineering work themselves. All this means that virtually all managerial staff in Chakra, like other engineering manufacturing companies, are men, because women, it is widely assumed, would risk their moral reputations if they spent much time with male fac­ tory workers or transport company owners in unsafe places away from the office. Exceptions exist and we did meet one young female engineer in Chakra, but in manufacturing companies, unlike IT companies, women are still very rare.

Career Trajectories and Company Loyalty All over India, the Tata group has a good reputation for looking after its employees and providing job security, so that it can command their loyalty, and all the men we met who worked for Tata firms described them favor­ ably. In Chakra, senior managers were similarly positive about their em­ ployer. Ratnaswamy, a Tamil Brahman from Chennai in his early sixties, has an engineering degree and joined the company shortly after graduation. He compared Chakra to a Japanese firm building a strong, committed relation­

Education and Employment after Independence / 105

ship between itself and its employees, and indeed their families as well. His young colleagues, he said, are amused that he has spent more than thirty years in Chakra, but he favors “long innings” (a cricket metaphor) and the commitment that it produces. Ashok, a Maratha Brahman from Chennai in his fifties, has degrees in both physics and engineering, and had wanted to do research in physics. But his parents persuaded him to study engineering for career reasons and his first job was at Hindustan Motors, a nationalized car company. According to his contract, he should have stayed for ten years, but he refused “to spend the best years of my life” there and resigned after four years; one year later, in 1979, he joined Chakra as a sales manager. In his youth, he remembers, people just wanted to get a secure job and settle down; in those days, Ashok explained (using another cricket metaphor), “you played one ball at a time, one over at a time” and had no ambitious plans for the future, whereas frequently changing jobs is now normal. Sam­ path found his first job in a Tata-owned cement factory in Punjab, where he worked for several years. He moved to Chakra in 1982, attracted by the opportunity to set up new personnel management systems in a new plant. Sampath also favors long service since it creates a large pool of staff who subscribe to the “values” of the company, as well as easing communication because there is a collective memory about how problems were solved in the past. As a personnel manager, however, he also recognizes the need for flexible attitudes in the postliberalization environment. Krishnamurti talked about commitment to the TVS firm in which he has worked for twenty-one years in similar terms; it, too, takes care of its staff like family members and they in turn are expected to be loyal to the firm. But young people, he complained, often fail to reciprocate and senior man­ agers in Chakra spoke in the same vein. They typically assumed that their young colleagues are no more impressed by secure jobs in the private sector than the public sector, are too impatient to wait for slow promotions and salary increases, and are allergic to the bureaucratic conservatism lingering in old-economy manufacturing companies. Comparisons were repeatedly made with new-economy IT companies, which are disparaged, as Sampath put it in 2003, because they have “disturbed the social fabric” by giving their employees “quick promotions,” “fancy designations,” and “phenomenal compensation.” IT professionals, it is generally claimed, constantly move from one company to another in endless pursuit of higher salaries and display none of the loyalty found in old-economy companies; IT profes­ sionals also like the opportunities to work overseas, which manufacturing companies rarely offer. Sampath, who is responsible for recruiting gradu­ ate trainees for Chakra from the engineering colleges, was partly grumbling

106 / Chapter Three

about competition from IT companies and one of his colleagues specifically said that by the late 1990s, only fourth-rate college graduates wanted to join Chakra. By 2010, however, Sampath was more sanguine, commenting that attitudes to IT had become more realistic, and solid, long-term prospects for manufacturing had improved its reputation, so that recruiting around one hundred good engineering graduates per year was no longer so difficult. Notwithstanding the assumptions made by senior managers, young people working in Chakra and the TVS firm just mentioned largely agreed with Sampath and often compared themselves positively with friends in IT companies. Pradeep, a Tamil Brahman from Bangalore in his twenties, who has an engineering degree and also an MBA, liked Chakra because it shows “commitment” to its new recruits and “stuck to what it said,” unlike a lot of IT companies; Pradeep’s brother, who works in an IT company, also told him that his work is just boring debugging to “rectify someone else’s mistakes.” Suresh, a north Indian engineering graduate in his twen­ ties, enjoyed working for Chakra because the work is steady and he has job security, whereas his friends in IT either work extremely long hours or find themselves “on the bench” with no work at all. Joseph, a Tamil Christian and another engineering graduate in his twenties, works for the TVS firm; he felt he had some power over his own work, unlike his friends in IT com­ panies, and believed (too optimistically, of course) that manufacturing is an “evergreen industry” with “no boom and bust.” Particularly interesting are the opinions of Kumaran, a Tamil non-Brahman from Bangalore in his forties, who had recently joined the TVS firm when interviewed in 2005. Ku­ maran is an engineering graduate who worked in Bangalore for fifteen years in a manufacturing company, followed by six years in several IT companies, before moving to his present job in Chennai. He returned to manufactur­ ing, he said, because he was looking for “permanency” and job satisfaction, and his new firm has a good “mix of age and experience,” whereby young recruits can work alongside men whose careers have lasted thirty years. Ku­ maran was not uncritical, for he thinks that the loyalty inspired in a family firm tends to be inhibiting and makes employees assume that everything will be provided for them. In the IT sector, though, expectations are very high, staff suffer from a lot of stress and “burnout,” and there is hardly any loyalty to companies as people “job-hop” between them. IT companies, he concluded, have excellent working environments and belong to a glamor­ ous industry, but the “beauty is only skin-deep.” Among both older and younger managers, therefore, there is a wide­ spread view that old-economy manufacturing companies have some ad­ vantages formerly associated with the public sector, but following economic

Education and Employment after Independence / 107

liberalization they have also shed many disadvantages, without acquiring the defects of the new IT companies. Not surprisingly, those who actually work in IT companies usually disagree.

The Indian Information Technology Sector Within the new economy, the IT sector is the most prominent and it gen­ erates a substantial proportion of India’s export revenues, although its workforce is still relatively small. The sector can be broadly divided into IT companies themselves and IT-enabled services (ITES) companies, which include business process outsourcing (BPO) firms and call centers. Call centers have received considerable public and academic attention, but their employees handling overseas customers’ questions and complaints on the telephone do not actually work with IT systems. Among Indian IT companies, the largest in terms of revenue and num­ bers of employees are the software and services companies, which have now become multinational corporations. In 2011–12 (as in previous years), the list of “top twenty” companies was headed by TCS, Infosys, Wipro, and HCL Technologies; in the list of “top twenty employers,” Cognizant Tech­ nology Solutions joined these four.22 There are also numerous smaller Indian-owned firms pursuing many different lines of business, some of which are highly innovative and successful, whereas others are not and shrivel as quickly as they appear. Partly competing with the major Indian software companies are multinationals with sizable Indian operations, such as Accenture and IBM. The IT sector also includes multinational IT giants like Microsoft and Oracle, as well as many smaller foreign companies with offshore research and development centers in India. Some of these compa­ nies, notably Microsoft, recruit high-flying graduates from IITs and other top institutions to whom they pay exceptionally high salaries. ITES companies primarily carry out BPO or “back-office” operations for their customers, mostly firms based in North America and Europe, such as airlines, insurance companies, and retail chains. The main business of the software companies is setting up the IT systems that enable outsourcing, so that, for example, claims sent into a US medical insurance company can be processed in India by a BPO company. Nowadays, the BPO may actually be in a third country, such as Indonesia, partly because Indian software companies have themselves become multinationals; thus by 2013, TCS was operating in forty-four countries and had 276,000 professional staff, with a significant minority employed outside India.23 Bangalore is the Indian IT industry’s principal center, followed in south

108 / Chapter Three

India by Chennai and Hyderabad; some cities in the north, such as Delhi and Mumbai, also have sizable IT sectors. Bangalore has the most diverse IT sector, whereas Chennai’s is dominated by the top Indian software compa­ nies. In 2003–5, at the time of our research, these companies were expand­ ing rapidly and continued to do so for the rest of the decade. For example, in Cognizant, whose main center of operations is Chennai, employee num­ bers rose from around 10,000 in 2004 to 162,700 worldwide in 2013, six­ teen times as many as nine years earlier.24 Despite their global expansion, the top software companies still find a very large proportion of their new recruits from India’s leading engineering colleges, especially those in Tamilnadu and the south. All the companies have similar application and selection procedures for college graduates. Only those with a consistently first-class academic record can apply; for example, applicants to TCS in 2012 had to have achieved first-class marks in tenth and twelfth standards, as well as in all degree examinations, at their first attempt. The selection process includes an aptitude test and three interviews covering technical knowledge and other skills, especially com­ munications skills.25 Applicants to some companies also attend a “group discussion” to see how well they can express themselves in front of other people. The selection procedure is very competitive and there are perennial complaints that graduates from the majority of engineering colleges always fail, because, in particular, they lack “communication skills.” Plainly, appli­ cants who have professional or managerial, well-educated, English-speaking parents, and went to good English-medium schools, have a significant ad­ vantage, so that the software companies’ selection procedures favor the ur­ ban, upper-middle class.26 On the other hand, there is no evidence that caste background, such as being Brahman, matters at all. We return to this crucial question in conclusion.

Working for a Software Company The men and women who work in the software companies generally call themselves “IT professionals.” Most of them are young. In 2005, the median age of software professionals was around 27.5 and even senior managers are often in their thirties.27 A large minority are women; in 2005, around 25 per­cent of the major companies’ new recruits were women, and by 2012, in Cognizant and Infosys, for example, they made up between 32 and 35 percent of the workforce, although only about half that proportion in the management.28 We conducted interviews with IT professionals in all the top software companies in Chennai in 2003–5, although most were

Education and Employment after Independence / 109

employees of one that we call Indian Computer Services (ICS). Information was also collected in various smaller IT companies in Chennai and, because many of them are IT professionals, we have material from Eighteen-Village Vattimas as well. Although old-economy manufacturing companies have been changed by liberalization, the differences between them and IT companies have not vanished. Thus the contrast between the working environments in, say, Chakra and ICS was huge when we did our research. Like other software companies, ICS occupies office space in Tidel Park, an IT business park in south Chennai opened in 2000, but it has also constructed its own large, sophisticated buildings alongside a road designated as Chennai’s “IT cor­ ridor,” which runs south from the park. Inside Tidel Park’s main building, as well as in the new ones along the IT corridor, most of the space consists of open-plan offices of the standard modern type, although a small number of senior managers have their own offices, usually with glass walls. All the buildings are kept permanently cool by efficient air conditioning, however stiflingly hot it is outside, and they are always very clean, as are the spa­ cious areas surrounding them. These qualities are greatly appreciated by IT professionals, especially women, who often comment on the pristine en­ vironment, and describe Tidel Park and other software company offices as prestigious places to work. Chakra Vehicles has recently built a similar new headquarters building, but in its older premises—like those of other oldeconomy companies—managers occupy individual rooms along corridors, and in general the office space is neither particularly cool nor clean. Another striking feature is that software companies’ offices are virtually paperless. They also house no support staff—such as secretaries, reception­ ists, and peons—whose absence reflects the relatively flat hierarchies in soft­ ware companies compared with old-economy ones. One characteristic of flat hierarchies, at least ideally, is the open and equal exchange of views. Equality is part of the software companies’ corporate ideology and, although the reality can be rather different, it is true, for example, that younger people are often managers of their seniors, and the reciprocal use of names with­ out titles or kinship suffixes (such as ma¯ma, “uncle”; or ma¯mi, “aunt”) is the norm, even between managers and their subordinates. Furthermore, an important part of an IT professional’s job is effective communication on a telephone or videophone with foreign clients, most often Americans, who tend to be impatient with and confused by Indians who appear too defer­ ential. To work successfully in a major software company, therefore, people have to learn communication skills that are normally unfamiliar, and may be quite disturbing, to anyone from a “traditional” Indian family and social

110 / Chapter Three

milieu who is not used to an egalitarian style of social interaction conducted in English. Most work in software companies is done by teams of software engi­ neers, together with a few “domain specialists” (that is, experts in particular business sectors), who are supervised by team leaders and project managers. Teams vary considerably in size and some members may work abroad “on­ site,” while others stay in India. Most staff work a forty- to fifty-hour, fiveday week according to a flexitime system, but managers often work longer hours; pressure to meet deadlines is common, too, and extra hours are then expected. Casual observation of IT professionals in front of their computers may give the impression that they are just like other bureaucrats in modern offices, but this is erroneous, because the technical work they do is entirely distinct from administration. Gender relations in software companies will be examined in the next chapter, but we note here that male and female team members often work closely together. Conversations between men and women—as well as their body language in each other’s company—consistently appear relaxed and in semi-public areas away from the workspaces, such as the offices’ entrance halls or coffee shops, they mix freely when taking a break from work; it is also not unusual to see obviously affectionate couples talking together. All this behavior contrasts strongly with that in most workplaces or other social situations in Chennai, including those dominated by the young such as college campuses, although it would not be so unusual in Bangalore. It is salient that Chennai’s IT professionals, especially when they are inside Tidel Park or their own companies’ buildings, are more or less insulated from the outside world and interact almost exclusively with each other; they are also rarely observed either by outsiders—apart from occasional foreign visi­ tors from client companies—or by subordinates, except for a small number of catering staff, cleaners, and security personnel. These conditions plainly contribute to the relaxed interactions between men and women, as one young single woman in ICS explicitly noted; we know, she said, that rela­ tions between the sexes are quite free inside Tidel Park (where she worked), but outside it, even in the company of the same men—for example, at a team lunch in a restaurant—we have to revert to convention.

Career Trajectories and Professional Ambitions in IT Outside the IT sector and sometimes even within it, people commonly claim that all IT professionals are consumed with individual ambition and constantly move between software companies in pursuit of quick promo­

Education and Employment after Independence / 111

tions and higher salaries. This claim has some truth in it and “attrition rates” measuring the annual loss of staff—which roughly vary between 10 and 20 percent—are a permanent concern for the companies, although there is some evidence that they are lower in Chennai than Bangalore, for example. Even so, most stereotypes about IT professionals are grossly exaggerated and sorting out fact from fiction matters for understanding the characteristics of today’s Tamil Brahman upper-middle class. India’s IT industry has a longer history than is sometimes recognized. Thus TCS, the oldest major software company, dates from 1968, as already mentioned. Two of its young employees in the 1970s were S. Ramadorai and S. Mahalingam; a third was Lakshmi Narayanan, another Tamil Brah­ man, who eventually became Cognizant’s CEO. Krishna, introduced in chapter 2, who now runs his own software company with his wife in Chen­ nai, was also employed by TCS; he joined in Mumbai in 1978 and worked for the company in India and overseas for ten years. In the 1970s, Krishna recalled, TCS probably had about one hundred software engineers—fifty Tamils, twenty-five Telugus, and twenty-five others—and he remembers overhearing Kohli, the general manager, saying, “I get all these Tamils, they are cheap and work hard.” Krishna may have misremembered the size and mix of TCS’s staff, but a lot of the Tamils certainly were Brahmans. Another was Krishna’s friend Venkat, the son of an engineer and a graduate of IIT– Madras, who joined in 1975 and worked for TCS for ten years—in India and overseas—before moving to Citibank. Yet another was Ratnam, who came from a humble background in Srirangam, near Tiruchirappalli, graduated from IIT–Madras, and joined TCS in 1972. He worked for TCS for several years, mostly in the US, until he resigned and went to work for other com­ panies in California for fifteen years. In the late 1990s, much like Krishna, Ratnam returned to Chennai to set up his own small software firm. TCS pioneers such as Krishna, Venkat, and Ratnam were in their fifties when we met them in 2003–5. Like men of similar age working in manu­ facturing, they remember when almost everyone regarded government and public-sector jobs as the best. Krishna’s first job was in the nationalized steel plant in Bokaro, but because it was “going stale” he left to join TCS, despite his father’s opposition to his resignation from secure employment. Venkat’s first job was in the old Tata Electric company, rather than the public sector, but he “broke tradition” by joining a new venture and was also criticized by his father, who did not believe in so-called consultancy. But although the TCS pioneers we have mentioned left the company even­ tually (except for Ramadorai and Mahalingam), they all stayed for fairly long periods and still talk with respect about it; furthermore, none of them

112 / Chapter Three

subsequently changed jobs more often than is common in the manufactur­ ing sector, although Krishna and Ratnam were rather unusual in deciding to leave America to start their own businesses in India. It is worth remarking, too, that Krishna, Venkat, and Ratnam, when reminiscing about their early careers, not only emphasized how exciting the work was, but also that it was often very challenging, especially when they had to solve difficult technical problems quickly, sometimes in unfamiliar foreign environments.29 Com­ pared with much software engineering today, their work for TCS was rarely routine and it drew on knowledge they had acquired as IIT students, as well as through training at TCS. To some extent, there has been deskilling in the Indian software companies and, significantly, although some managers are IIT graduates, few recruits since the late 1990s have come from IITs. We now present ethnographic sketches of a few IT professionals em­ ployed by ICS to illustrate their outlook toward their work.30 The majority of senior managers in the major software companies are men and women in their thirties or forties. Ramakrishna, a personnel manager in ICS in his late thirties, was the first IT professional we interviewed in 2003. He is a nonBrahman Reddiyar, who has an engineering degree and an MCA (master’s in computer applications), a qualification that was a popular route into the IT industry in the 1990s. Ramakrishna, who looked seriously stressed-out when we met him, described his own job as “tough and challenging”; it was his responsibility to ensure that in the “war for talent” among the leading companies, ICS could recruit enough high-quality staff, especially by pick­ ing the best graduates from the top colleges. As far as Ramakrishna was concerned, his attitude to his work and family differed radically from his father’s. He “lives for the present,” while his father, a retired manager in a state government corporation, “lived for the future”; although Ramakrishna still lived with his parents, he put his own nuclear family and himself first, ahead of the extended family. Ramakrishna’s wife was not employed; his two daughters attended one of Chennai’s most expensive schools and he spoke of spending “quality time” with them. In 2003, he said, he was happy to work in India because prospects in the booming IT sector were so good, but if it were better for his career and for his wife and children, he would move to America. (Actually, we learned, he was then negotiating for a sal­ ary increase in ICS while threatening to leave for another job in Bangalore.) Ramakrishna also emphasized that ICS, like all successful IT companies, is a very competitive place to work in which those who want to succeed must often work long, antisocial hours under the continual pressure of tight deadlines; good people in the firm want “careers,” not just “jobs,” but they

Education and Employment after Independence / 113

also want high salaries and they know that pay and promotion depend on individual performance, not age and seniority. Ramakrishna conformed quite closely to the image of the driven, indi­ vidualistic executive that is an IT industry stereotype, although we learned in 2012 that, fed up with the stresses and strains of his job, he had recently resigned from ICS and returned to his family lands south of Chennai to farm them as a business. Interestingly, too, Ramakrishna’s outlook turned out to be rather unusual in ICS. Partha, an ICS project manager in his early thirties, is a Tamil Brahman from Chennai, whose father had an engineer­ ing diploma, although we do not know about his employment. Partha has a BSc in computing science and an MCA. He taught in a computer-training institute for a year before joining ICS, where he had been for ten years in 2005. He was then in charge of a team of twenty staff. Partha did not com­ plain about hard work, but mentioned flexible work timings as one advan­ tage of his job. He spent most weekends at home with his wife, who did not have a job, and their child. ICS sent him to the United States for project work, which he liked, but he did not want to stay there; he preferred living in Chennai, which has been transformed and “put on the world map” by the IT industry. Partha favored the “working culture” in ICS—exemplified by the options available for overseas postings and the openness of the se­ nior management—and he saw no point in moving to another company unless it was clearly better. He did not anticipate problems in gaining pro­ motion within ICS in the future. Partha said that he would not move just for more money, as lots of IT professionals allegedly do, although many of our informants actually made an identical comment. Anuradha, like Partha, was an ICS project manager in her early thirties in 2004. She is a Tamil Brahman from Chennai, who is married to another IT professional and they had one daughter; they all lived with his parents, and her own parents lived in the same apartment block. After taking her engineering degree, Anuradha successively worked for about six years in two major software companies, one of which sent her to Britain and the United States. After marriage, Anuradha moved with her husband to the US, where she joined ICS in 2000. But she never wanted to live abroad permanently and in 2002, after making some money, she and her husband returned to Chennai. In 2004, Anuradha was managing a team of twenty staff in Chen­ nai. Because she had a young child, Anuradha was refusing to go abroad for on-site project work and said that it was not a problem, because she did not need any more “exposure”—a commonly used term denoting a wide range of new opportunities and challenges. She did not find her job stressful and

114 / Chapter Three

did not work on weekends; she is close to her family and could always rely on them and her own parents to care for her daughter during the working day. Anuradha firmly believed that the IT industry has helped to empower women and, all in all, she was content with the comfortable life that work­ ing for ICS in Chennai had given her. Anuradha’s arrangements for her daughter’s care were crucial to her hap­ piness at work and other married women who are struggling with childcare are in a different position, as we shall see in the next chapter. Otherwise, though, the prevailing attitude of software company managers, male or fe­ male, toward their work and careers, more often resembles Partha’s and Anuradha’s, rather than Ramakrishna’s. They certainly enjoy their high sala­ ries and the rapid promotion that is possible, but although they commonly talk about attrition rates and “job-hopping,” they have rarely done it them­ selves and some of them specifically say that promotion may come more easily by remaining in one company. They are glad to have opportunities to work abroad and some have spent quite long periods outside India, but those now working in Chennai mostly want to do so and are not striving to emigrate to America or anywhere else. They also see the modernity and dynamism of the software companies operating in global markets as far more attractive than the supposed staidness of old-economy manufacturing companies like Chakra, and none of them, Brahman or non-Brahman, has ever seriously contemplated working for the government or public sector. But very few IT company managers, pace Ramakrishna, talk about “tough” jobs with excessive hours or describe themselves as competitively individu­ alistic, and none of them would accept Sampath’s criticisms about “quick promotions,” “fancy designations,” and “phenomenal compensation.” Project managers do very little of the coding work and other routine tasks carried out by junior staff on their teams. Some managers even admit that they can no longer do much of the basic work, because software systems have changed and they have not kept up with them. Two young IT profes­ sionals employed by ICS are Sarojini and Parvati. They are friends who were working together on the same project team in 2004, when they were both in their mid-twenties and unmarried. Sarojini belongs to the matrilineal Nayar caste and comes from southern Tamilnadu; she has an engineering degree and entered ICS as a trainee in 2002. She told us she was happy in ICS and found her work exciting and challenging, although her first ambition had been to study medicine. Both Sarojini and Parvati said that they normally worked about forty or forty-five hours per week, but never on the weekend, and they laughed about a bunch of young men who worked at night and on weekends, pretending they have to do so, whereas in fact they are just

Education and Employment after Independence / 115

geeks obsessed with computers. In 2004, Sarojini went to the US on an onsite project assignment and by then she had acquired a boyfriend in ICS, a Tamil Brahman from Mumbai, who was also sent to the US; they wanted to marry and, although everyone else in both families eventually gave their assent, his mother did not because Sarojini is not a Brahman. According to Parvati (in late 2005), Sarojini and her boyfriend were hoping to stay in America until his mother changed her mind, but unfortunately we do not know whether their story had a happy ending. Parvati is a Tamil Brahman and a native of Chennai. She has a BSc in computer science and an MSc in information technology. In 2004, after six months in ICS, Parvati was finding her work boring, although in early 2005, in spite of working longer hours and complaining about her project manager, she was much more enthusiastic and ambitious. She had become a team leader and was determined to get an early posting in the US. But later in the year, after one transfer to an American city had failed and a sec­ ond was looking dubious, Parvati was thoroughly discontented and com­ plained again about her manager, uncooperative team members, and too much stress pervading ICS. But then a posting came through and she finally set off to the US in a mixture of excitement and trepidation. Our final sketch is of Raghu, an unmarried Tamil Brahman in his early twenties, who was brought up in Kalpakkam, south of Chennai, where his father is an engineer in the nuclear power station, a public-sector undertak­ ing. Raghu has an engineering degree and joined ICS after graduating in 2003. In 2005, he was working on a British client’s project and said that every day a new problem had to be solved, so that he never got bored. He did not mind working long hours, saying that anyone of his age should be able to do so, but his job required none of his engineering knowledge and he did not think he was learning much—except how to say no to demand­ ing, foreign clients. Hence Raghu did not expect to stay in the IT industry for a long time, where “the term loyalty has no value,” and he intended to start studying for an MBA fairly soon. He preferred to live in Chennai close to his family and had no real desire to go abroad. For the time being, though, he was enjoying the money and commented that after one year he earned 200,000 rupees, compared with his father’s annual salary of 400,000 rupees after thirty years, although even now, said Raghu, his father tended to spend less money than him. To conclude this section, let us recall Shivakumar’s comments on the IT sector, which he compared unfavorably with manufacturing, even though software companies have excellent working environments and IT is a glam­ orous industry. Shivakumar’s decision to leave IT for manufacturing was

116 / Chapter Three

very unusual, and his opinions were predictably not shared by the IT pro­ fessionals we met, although some of them do complain about boredom, stress, or a shortage of technical challenges. Nonetheless, it is a source of self-esteem, as well as family pride, to be employed in the youthful, rapidly expanding, technologically advanced, export-oriented IT industry, which has done more than any other to enhance India’s visibility on the global economic map. Moreover, the very fact that IT is widely seen as a glamor­ ous industry should not be dismissed as trivial, because it is important in attracting young men and women into it. And even if most of them do not seriously aspire to become CEOs of top software companies, industrial lead­ ers like Ramadorai are widely admired and represent role models for many educated young people, including a lot of Tamil Brahmans. Of course, the high salaries and fast promotions available in the ma­ jor software companies are very attractive features, as all IT professionals acknowledge, more or less openly. On the whole, too, people agree that company loyalty is weak in the industry, but many IT professionals still talk positively about their own companies, where they have often stayed for fairly long periods; in any case, only a few people, such as the TCS pio­ neers, could possibly have had a “long innings” of thirty years in one firm, because most software companies were tiny or nonexistent so long ago. In fact, “job-hopping” is less endemic than is widely supposed and, when people do move, improved job satisfaction or family reasons, such as prox­ imity to parents, are just as likely to be motives as more money. Working hours for most IT professionals are not abnormally high, unless deadlines are imminent, and complaints about stress levels are very variable and not necessarily worse than those made by managers in companies like Chakra. Another positive feature of the IT industry is the opportunity to go abroad, either on project assignments or for longer periods to earn a lot of money before returning to India, or indeed to emigrate permanently. But an ad­ ditional attraction for IT professionals is that they usually do not need to make a final choice between staying in India and emigrating, because they can fairly easily go away and then return, like Krishna and Ratnam did after several years in the US. A lot of hyperbole perpetually surrounds the Indian IT industry and solid evidence about those working in it is still sparse. In most respects, however, the IT professionals we have met quite closely resemble those in Bangalore, Mumbai, and elsewhere described by other ethnographers.31 Because the IT companies are the quintessential new-economy, privatesector firms—though by no means the only ones—they are always in peo­ ple’s minds when contrasting the old and new economies, as well as the

Education and Employment after Independence / 117

public and private sectors. And despite the myths and exaggerations, it is plain, too, that IT has brought real changes to the lives of many middle-class people, both Brahmans and non-Brahmans, since the 1990s.

Caste, Class, and the IT Industry IT professionals in Chennai, as well as other people, generally assume that Brahmans are very well represented in the industry, especially in the major software companies. Carol Upadhya and A. R. Vasavi’s survey of 132 soft­ ware engineers in Bangalore in 2004–5 revealed that 48 percent of them were Brahmans; despite its limitations, this carefully conducted survey is the most extensive to have been done among the employees of major IT companies.32 In Chennai, too, it is very possible that about half the IT pro­ fessionals in the top companies in Chennai are Brahmans, although most of our informants guessed that it was fewer than that. Almost certainly, however, as all the circumstantial evidence suggests, the proportion of IT professionals who are Brahmans is much higher than their share in the gen­ eral population. When Tamil Brahmans are asked why they are so numerous in the IT industry, they often just assert that they are cleverer than people from other communities, but sometimes, too, they specifically claim that they are bet­ ter at mathematics. Krishna actually disputed this and recalled a Christian boy at school who was outstanding in the subject. Yet he also said that many Brahmans think they possess a special mathematical intelligence and expect their children, both boys and girls, to score very highly in all math tests. Sometimes, parents tell their children stories about Ramanujan, as if his mathematical brilliance was the hereditary birthright of all Tamil Brah­ mans. In reality, of course, not all Brahman schoolchildren shine at math­ ematics, although they do not have the “fear of math” so widespread among American schoolchildren, as a Tamil Brahman professor of mathematics in Chicago noted. But almost all parents are anxious about their children’s progress in mathematics, so that any boy or girl who fails to score 100 per­ cent or close to it in a test may be told off for not trying hard enough or sent for extra coaching to overcome the handicap. Competitive examination pressure partly explains the emphasis on mathematics, because test scores of 100 percent are possible, whereas they rarely are in other subjects; math test results, too, are thought to be completely objective and uninfluenced by teachers who are supposedly biased against Brahmans. Mathematics, though, is also widely regarded as the supreme academic subject and it is assumed that “if you do well in math, you are a bright kid,” to quote a Tamil

118 / Chapter Three

Brahman software company executive, who also recalled that if he was not in the top five in his class at school, life became very difficult at home. In fact, the idea that mathematics is the preeminent discipline is common among Indians, not just Tamil Brahmans. Moreover, the claim that any one caste has a natural talent for it is obviously unfounded, although the Brah­ mans’ inclination to think that they excel at mathematics, combined with parental support and pressure, undoubtedly contributes to their children’s scholastic success. In truth, even if Brahmans were exceptionally good at mathematics, it would not explain their large presence in the IT industry today. In the early days of computing, senior IT professionals have told us, a good knowledge of mathematics was sometimes helpful. But for the day-to-day business of contemporary software and services firms, as well as many other IT compa­ nies, mathematical skill no longer matters much, because software systems are usually adaptations of current standard systems, a software engineer’s work is largely routine and, although coding skills are needed, mathematics itself rarely is. To be sure, a solid grasp of mathematics provides a good basis for the competencies desired by IT firms because the discipline is rigorously logical. All the same, people can learn to think logically by other means and being able to do mathematics is no more crucial for IT than for conven­ tional engineering or other fields of applied science and technology. Leaving claims about the Tamil Brahmans’ cleverness or mathematical skill to one side, it is plain that the most immediately critical factors un­ derlying their prominence in the IT industry are the caste’s generally high standard of education and its strong representation in professional and managerial employment over many years. That in turn takes us to the issues of class mobility and reproduction, in particular the reproduction of the upper-middle class. As we have seen, graduates with first-class academic records from the leading engineering colleges who are competent in English make up the vast majority of the major IT companies’ new recruits. Moreover, nearly all the IT professionals we interviewed have fathers (and occasionally mothers) who are or were educated professionals, or senior managers and adminis­ trators in government departments, or in public- or private-sector firms; the only exceptions were a few fathers who were lower-level bureaucrats, small businessmen, or farmers. Despite some upward mobility, therefore, the great majority of these IT professionals belong to urban, upper-middleclass families whose class position is stable; exactly the same findings have been reported by researchers in Bangalore.33 Caste traditions of education and learning, as well as the Brahmans’ con­

Education and Employment after Independence / 119

fidence in their own superior intelligence and intellect (not only in math­ ematics), undoubtedly contribute to how well Brahman children do in school and college. Nonetheless, for educational success, class as mediated by the family matters much more than caste, as Béteille has convincingly argued.34 Professional and managerial upper-middle-class parents in Chen­ nai, and other large towns and cities, ensure that they send their sons and daughters to the best English-medium schools. They can afford the high fees—as well as extra coaching fees—and, if necessary, the costs of mov­ ing to live reasonably close to a good school. Through relatives, friends, colleagues, and other acquaintances, such parents have access to reliable information about alternative schools and may also have contacts with influential figures who could help to secure admission to oversubscribed institutions. Upper-middle-class men, as well as many women (especially in the younger generation), normally speak good English and may encour­ age their children to use it at home. Good English, of course, is one key marker distinguishing the upper- from the lower-middle class among Indians in general, not just Tamil Brahmans. Upper-middle-class families are normally well informed, too, about higher education, the quality of engineering colleges, and company recruitment systems, often because a father or another relative is an engineering or IT company manager or has contacts among college teachers. Securing a college place and finding money for the fees, especially capitation fees, are serious problems for some students and their families. But many or most of them manage somehow. Thanks to their family background and high-quality education in school and col­ lege, upper-middle-class children living in major urban centers are therefore more than likely to acquire good communications skills in En­glish, as well as personal self-confidence, which are both needed to succeed in selection tests and interviews for the top IT companies and, indeed, leading manu­ facturing companies such as Chakra. Upper-middle-class families in large towns and cities, in other words, have access to forms of social and cultural capital—in Pierre Bourdieu’s terms—that give them a decisive advantage over other families, especially those living in villages or small towns.35 The latter usually send their sons and daughters to inferior schools, know little or no English, and lack the networks needed to improve their situation, unless their children are fortunate enough to be helped—as some are— by charities for poor Brahmans or by richer, better-placed, urban relatives. In Indian English parlance, this kind of social and cultural capital is often called “exposure.” “Exposure,” which is usually mentioned in conver­ sations about improving educational or career prospects, mainly denotes the process of enhancing social skills and cultural knowledge through new

120 / Chapter Three

opportunities, challenges, experiences, and sources of information, as well as the improved state that ensues. It follows that people who have acquired exposure are far more likely to understand its importance than those who have not, so that they are better able to improve their position still further. Inequalities in social and cultural capital therefore tend to widen. Exposure can be gained by various means, but immigrants to Chennai consistently say that the city itself offers excellent exposure for people like them, who were at a disadvantage because they were brought up and educated in small towns or villages. Dipesh, for example, is a young man working in a major IT company in Chennai, whose father was a professor in Gandhigram ru­ ral university, near Madurai. Dipesh comes from a highly educated, uppermiddle-class family, but he nevertheless remarked about his schooling that “exposure is 9 percent in Gandhigram and 90 percent in Chennai.” However Dipesh calculated the percentages, his basic point was that pu­ pils in Chennai’s schools have a huge advantage because, as he put it, “the way people are trained to think from seventh standard itself is different in Chennai.” Like his friend Murugan in the same company, who has a rural background, Dipesh came to Chennai for one month’s intensive coach­ ing before taking the engineering colleges’ entrance examination, so that he could gain the “awareness” necessary for success that candidates in the city already had. It matters, too, that English is much more widely used in Chennai than elsewhere. Thus Murugan, who is a non-Brahman, said that what struck him most on first arriving in Chennai was the extensive use of English—a “prestige thing,” rather than just “a medium of instruction,” as it always had been for him in his village and at college in Coimbatore. He also told Haripriya—implicitly referring to communication skills as a whole, not just use of the English language—“Brahmans know how to talk, Madam; they know good English.”36 Murugan’s comment goes to the nub of the issue. One of the Brahmans’ decisive advantages lies in their “good English,” which serves as a meto­ nym for all the forms of social and cultural capital that are mainly owned by the professional and managerial upper-middle class of Chennai and other cities. On the other hand, although a large proportion of Brahmans belong to the upper-middle class, whereas only a minority of the huge nonBrahman population does so, the upper-middle class today is in fact pre­ dominantly non-Brahman. Upper-middle-class, non-Brahman applicants are just as likely to obtain jobs in top IT companies as Brahman applicants with a similar background; conversely, Brahmans who belong to the lowermiddle class are far less likely to get these jobs, and Brahmans from poor,

Education and Employment after Independence / 121

badly educated backgrounds have no chance at all. The plain fact is that caste background is irrelevant for securing jobs in the major IT companies, where candidates are selected in their thousands through standardized pro­ cedures, although caste-based patronage networks do sometimes matter in family firms and smaller organizations, where personal contacts count for more. For the most part, though, young people want to work for the top companies and, if they do, their caste does not count. Hence the first and foremost reason why so many Tamil Brahmans are recruited by Chennai’s top software companies—as well as major manufacturing companies—is that so many of them have “good English”; more precisely, to a far greater extent than non-Brahmans (with the possible exception of some Chetti­ yar or Vellalar subcastes), Brahmans are very well represented in the urban upper-middle class, as they have been for more than a hundred years. All of this may look like caste privilege persisting throughout the colo­ nial and postcolonial periods and to some extent it obviously is. But it is so to a far lesser extent than people in south India typically assume, even if some Brahmans do favor their own in certain contexts. Furthermore, it has nothing to do with any putative Brahman ability to control the command­ ing heights of the economy for their own caste advantage, nor of course is it a product of their supposedly innate cleverness. Rather, Tamil Brahmans as a group remain privileged because so many of them have professional and managerial occupations, and therefore possess the economic, social, and cultural resources to ensure that their sons—and nowadays many of their daughters as well—are very well placed to secure the same kind of employment in open recruitment systems. In the upper-middle class, social stability is therefore the norm and, although there is some upward mobility from the lower-middle class, downward mobility is rare. The Tamil Brah­ mans’ disproportionately large share of professional and managerial em­ ployment is maintained, in other words, by a process of class reproduction within a modern society and economy that is, despite its extreme character, fundamentally the same as the process in class-stratified Britain or other Western capitalist countries.37 This conclusion is consistent with that of Ku­ mar, Heath, and Heath, whose statistical analysis of data on social mobility shows that class origin matters more than caste or community background for access to higher-level occupations; the “historical legacy of caste” is sig­ nificant, they write, but “class inequalities cannot . . . be explained by the current operation of caste in Indian society.”38 Our material also lends sup­ port to Béteille’s strong claim, mainly based on his analysis of the modern Indian middle-class family’s role, that “caste has ceased to play an active

122 / Chapter Three

part in the reproduction of inequality, at least at the upper levels of the social hierarchy.”39 Social class, of course, is about status, as well as occupa­ tion and economic position, and much of the rest of this book is about the social and cultural status of the Tamil Brahman middle class. In concluding our examination of education and employment, however, it is important to emphasize that, in the final analysis, there are so many Tamil Brahman IT professionals not because they were born as Brahmans, but because they come from families in the upper-middle class.

Four

The Changing Position of Women

“In childhood a woman should be under her father’s control, in youth under her husband’s, and when her husband is dead, under her sons’. She should not have independence.”1 As this famous verse in The Laws of Manu—written about two thousand years ago by Brahman lawgivers—illustrates, gender inequality has been a vital part of Indian society since ancient times. Manu, of course, is normative prescription, not sociological description, but the inferiority and subordination of women within the family and beyond it have long been key features of hierarchical society and culture among Hindus (as well as other communities) in India; in many contexts, they still are. In general, though, gender inequality was and is more pronounced among high castes. Hence the conduct of daughters, wives, mothers, and widows is most severely constrained by purity and pollution rules, seclusion practices, moral rigidity, and obsessive control of female sexuality among Brahmans and other high-ranking castes. Lowcaste women, by contrast, are more relaxed about purity, commonly work outside the home so that they cannot be secluded, are less influenced by Brahmanical norms, and are often less subject to male authority, although these features themselves contribute to the low status of their families and communities. In the past, in accordance with this general pattern, gender inequality was extreme among Tamil Brahmans and markedly more so than among the vast majority of south Indian non-Brahman groups. A key aspect of this inequality was prepuberty marriage for girls. In Tamilnadu, prepuberty marriage was largely confined to Brahmans and was the practice not only in old-fashioned rural families, but also in those headed by the educated professional men who formed the new urban middle class of colonial Madras.

124 / Chapter Four

Nita Kumar explains that the fathers of members of this “new intelligentsia” in British India were usually “pillars of society [who] typically honoured caste and sectarian rules, practised gender and age hierarchies, and believed in child marriage.”2 Among Tamil Brahmans, the new men were scarcely less conservative than their fathers, so that even the daughters of highranking lawyers and administrators in Madras city were normally married before puberty. In the early twentieth century, child marriage became less prevalent and, by the late 1940s, it had largely ended. Moreover, female literacy rates and educational standards steadily improved over the same period. The 1891 Madras census report stated that it was a “wonder and regret” that Brahmans, like traders, did not educate their daughters, and 96.3 percent of female Brahmans were recorded as illiterate.3 V. Nagam Aiya, a senior Tamil Brahman official in the Travancore government, contended that illiteracy was merely a “mechanical” defect, not an “intellectual” one, and that Hindu women—meaning Brahman women—were not in fact uneducated, because they learned the scriptures in detail by listening to religious teachers and senior relatives. “The Hindu woman,” he insisted, “appears . . . to be the most cultured of her sex in any part of the world.”4 Nagam Aiya’s attempt to defend tradition by eulogizing Hindu women, who “ennobled ‘domestic duty’ into a religion,” was probably not confined to him. Yet even in the nineteenth century in the Madras Presidency, when only very small numbers of Brahman girls went to school, at least as many were receiving basic education at home; in some Brahman families, indeed, there was a strong tradition of female education imparted within the home.5 By 1901, the Brahmans’ female literacy rate in Tamil stood a little higher than a decade earlier at 5.8 percent, and it went up to 38.5 percent in 1931 (the comparable male rates were 73.6 and 87.6 percent, respectively), while in English it rose from virtually zero to 3.1 percent (17.9 and 44.8 percent for males).6 Throughout the twentieth century, the overall standard of Tamil Brahman girls’ education continued to improve, so that more and more of them, like their brothers, completed schooling and went to college. By the end of the century, girls’ educational standards more or less matched boys’ and are nowadays generally higher than those of non-Brahman girls. Most Brahman women today—like other middle-class women—marry in their early twenties after finishing their education. Many Brahman women are also employed outside the home, which was previously uncommon, and they have a significant presence in the information technology industry. The family of one woman, Uma, illustrates the changes that have taken place. Uma belongs to the Eighteen-Village Vattima subcaste, which was

The Changing Position of  Women / 125

unusual in continuing to practice child marriage until the 1970s, so that in her family the rate of change was slowed by Vattima conservatism. Uma was born in 1936 and brought up in a village, whose school she attended. She was married at eleven to a sixteen-year-old law student. When she came of age and reached menarche at fourteen, Uma was taken out of school, but, during the next three years prior to joining her husband, her parents arranged private tuition for her at home, so that she reached ninth standard (one short of the final tenth standard). Once she had moved to her in-laws’ house, though, her education ceased because, leaving aside all her domestic duties, young wives like her “weren’t even allowed to come to lighted areas of the house.” Uma, however, was far better educated than her mother, who completed only third standard, and her grandmothers, who were probably illiterate. Uma and her husband, who qualified as a lawyer and practiced in Chen­ nai for fifteen years, had two sons and two daughters, who were not married before puberty. Uma’s elder daughter, born in 1958, married after she came of age at fifteen, during her final year at school; she went to live with her husband, a bank officer, two years later, and in the meantime she completed her Pre-University Certificate (PUC). The couple live in Chennai and have two daughters. Uma’s younger daughter, born in 1970, married at twenty-one after completing her BCom degree; she and her husband, an engineer, live in Kerala and have one son. The difference between Uma’s two daughters reflects the changes that occurred between the mid-seventies, when finishing school was more than enough for a typical Vattima girl married in her teenage years, and the early nineties, when girls were obtaining degrees to improve their matrimonial prospects. In the next generation, women’s educational qualifications moved steadily closer to men’s, as Uma’s elder daughter’s daughters illustrate. Uma’s first granddaughter, born in 1978, has an engineering degree from a prestigious technology institute in north India and an MBA from a leading university in California; unlike her forebears, she also has a job—in a multinational auditing and consulting company in Los Angeles—and she married a non-Vattima Brahman in 2006 when she was twenty-eight. The younger granddaughter, born in 1984, studied for her MSc in biomedical genetics at another prestigious institution in Tamilnadu and then became a PhD student at a university in Texas. She married a Vattima at the age of twenty-four in 2008. The dramatic changes across five generations in Uma’s family, from her probably illiterate grandmothers to her highly educated granddaughters, are common among Vattimas, as well as other Tamil Brahmans. Of course, comparatively huge improvements in female educational standards have

126 / Chapter Four

occurred in much of the world during the last century or so. Over the same period, child marriage has also become far less common, not only in India, but in numerous other countries as well. In many respects, therefore, the Tamil Brahmans’ story is far from unique. And across India, as well as much of the rest of the world in which gender inequality has been the norm, there are countless women like Uma who are amazed when they reflect on how women’s lives have been transformed, particularly in relation to marriage and education. Here, however, we will focus on this transformation for Tamil Brahmans specifically and look at how the improved position of  women is regarded today as one important sign of their caste’s modern middle-class status.

The Age of Consent, Child Marriage, and Social Reform in the Colonial Period According to Brahmanical tradition, marriage is a kanya¯da¯na, a “gift of a virgin” who is ritually pure, and parents failing to marry their daughters before the onset of polluting menstruation commit a grave sin. On this matter, scriptural authorities were never entirely unanimous, but prepuberty marriage for Brahman girls became general from the sixth or seventh centuries.7 A marriage was consummated and a girl went to live with her husband after she came of age. In the late nineteenth century, social reformers throughout India began to campaign against child marriage. In 1891, the Age of Consent Act be­ came law in British India and raised the age of consent for sexual intercourse from ten to twelve to protect young girls from older husbands (or other men). In the 1920s, the age of consent was reconsidered and the outcome was the Child Marriage Restraint Act, popularly known as the Sarda Act after its principal sponsor, Har Bilas Sarda. This act, which was passed in 1929 and came into force in 1930, fixed the minimum age for marriage at fourteen for girls and eighteen for boys. (Since 1978, the respective ages have been eighteen and twenty-one.) In the debates about reform, the primary issue was the health and welfare of girls who became wives and mothers when very young, and one important consideration was their education, or lack of it. Another issue was the harsh fate of child widows, who were not allowed to remarry and were therefore condemned to remain inauspicious celibates for life. Yet there was also a lot of argument over Indian “public opinion,” especially about whether ordinary people would accept reform, as well as questions about, for example, the scriptural veracity of Brahman orthodoxy. Particularly con-

The Changing Position of  Women / 127

tentious was the question of how reform enacted by a colonial government would affect the honor of the nation and its women.8 Prior to the 1920s, nearly all public debate over reform was conducted by men and, as many writers have commented, they were often really arguing about colonialism and nationalism, or tradition and modernity, not about the condition of women actually living in their country. Controversy over the “woman question,” in which marriage and female education were always to the fore, first arose in the mid-nineteenth century, particularly in Bengal. Urban, middle-class, nationalist reformers—almost all belonging to the upper castes—overwhelmingly favored improved education for girls, mainly to make them better wives and mothers, but reform was colored by deep anxiety. As Sanjay Seth puts it, “The middle-class and nationalist project was to educate women so that they would become modern and yet able to represent Indianness; the anxiety that was the underside of this project was the fear that Indianness might be effaced by the pursuit of modernity.”9 It was vital that Indians, not foreign overlords, should control this modernizing project for women, which—to employ Partha Chatterjee’s conceptual dichotomy—affected the “inner,” spiritual domain of “essential” Indian values, as opposed to the “external,” material domain. 10 Marriage, belonging to the inner domain, was a serious matter for reformers, but they disagreed about it, for progressive liberals opposed child marriage, which precluded the emergence of modern educated women, whereas traditionalist conservatives defended the custom as part of true Indian culture and a guarantor of female purity and virtue. In the Madras Presidency, the age of consent and child marriage reform were debated from every angle at one time or another. Argument about reform in the late nineteenth century was probably less vociferous than in Bengal or Bombay, but it was more lively than the received wisdom suggests, as recent research on newspapers and magazines in Madras shows. In these publications, mainly but not entirely produced for Madras’s new middle class, debates on many social and political issues were aired, including child marriage, widow remarriage, and female education. The new genre of the novel also played a significant role. Vedanayagam Pillai, a Vellalar government official and one of the first “modern” Tamil authors, was “an ardent social reformer” who condemned Brahmanical marriage practices, and his second novel, The History of Suguna Sundari (1887), has a lot to say about the evils of child marriage. By contrast, in The Fatal Rumour (1893–95), the Brahman author B. R. Rajam Aiyar described the loving marriage of a child bride and pointedly opposed the reformists.11 Debate in the new public sphere opened up by printing and publication

128 / Chapter Four

was not confined to educated Brahmans, but they did take a leading role. On the age of consent and child marriage, one vocal advocate of reform was G. Subramania Iyer, editor of the Hindu newspaper (founded in 1878) and a leading nationalist in Madras. Some men supported him, but others were vehemently opposed, and yet others—including elite figures like the lawyers V. Bhashyam Iyengar and S. Subramania Iyer—either remained silent or expressed serious reservations about using legislation for social reform. Moreover, notwithstanding G. Subramania Iyer’s position on the age of consent, much of the hostility directed at him actually arose when he remarried his twelve-year-old daughter, whose husband died in 1889 before their marriage was consummated. For members of his family and other Brahmans, Subramania Iyer’s violation of the traditional prohibition on widow remarriage was the most serious outrage. All over India, arguments about the age of consent sharpened the division between liberals and conservatives within the nationalist movement. In Madras, the controversy also dispelled the myth that “the Westerneducated class were allies in the cause of social reform.”12 This was particularly true of Tamil Brahmans, who mostly opposed any change that might undermine their caste superiority within south Indian society because, as Charles Heimsath observes, “social innovation by Madras Brahmins . . . could mean the extinction of the privileges which separated them, and their entire families, from the benighted lower orders.”13 In Madras, therefore, Chatterjee’s inner-outer dichotomy was significantly modified by the division between Brahmans and non-Brahmans, which made marriage a controversial topic in the public domain of caste politics.14 The controversy over the age of consent and child marriage in the 1920s occurred in a political climate that had changed since the nineteenth century. It was heightened, too, by the lurid criticisms of child marriage made in Mother India (1927), a book by the American journalist Katherine Mayo, who unsparingly condemned Tamil Brahmans.15 In her examination of events surrounding the Sarda Act, Mrinalini Sinha argues that the most significant new feature was probably “the mobilization of a collective identity for and by women in the public realm.” Female voices were heard, as they had not been earlier, and women campaigning for reform claimed—more or less convincingly—to represent the views of all women, irrespective of caste or community. By the early twentieth century, too, marriage “had become an especially dense site for indigenous reformist public concerns” and there was widespread nationalist support for modern marriage reform, even if it did interfere with supposedly essential Indian values.16 Hindu conservative hostility to the Sarda Act was strongest in Bengal

The Changing Position of  Women / 129

and Madras.17 In Madras, much of the fiercest opposition came from Tamil Brahmans, including Congress supporters and nationalists, although some liberal Brahmans did support the bill.18 In the Indian Legislative Assembly in 1929, the Sarda Act was most strenuously resisted by two Brahman members from Madras, M. K. Acharya (a member of the Tamil Nadu Congress Committee) and M. S. Sesha Ayyangar, who insisted that raising the girls’ marriage age to fourteen contravened Hindu orthodox opinion and violated religious law, the dharmas´a¯stra, by which they plainly meant Brahman opinion and Brahmanical law.19 The arguments against reform resembled those put against the 1891 Age of Consent Act, but the rise of the non-Brahman movement and the Justice Party had drastically changed Madras politics by the 1920s. As Mytheli Sreenivas explains, there was now a non-Brahman, Dravidian nationalist debate about marriage, as well as an Indian nationalist one.20 Brahmans did not care how non-Brahmans chose to marry, but the political changes meant that unlike forty years earlier the Tamil Brahman legislators’ “rather desperate defense of pre-puberty marriage had lost its claims to nationalist authority,” so that they could not block the Sarda Act.21 On paper at least, the new act prohibited most prepuberty marriages, so that an ostensibly vital and cherished custom of Tamil Brahmans was eradicated. Yet by 1929 more Tamil Brahmans probably favored marriage reform than a generation earlier and more wanted to improve the welfare of girls. One person who did a lot to change attitudes was R. S. Subbalakshmi (1886–1969), the first notable Tamil Brahman woman reformer in Madras.22 Subbalakshmi was married in 1898 when she was eleven, but widowed soon afterward. Her unconventional parents and widowed aunt decided that Subbalakshmi should resume her education, instead of being secluded as a widow, and in 1911 she became one of the first women to graduate from Madras University. After graduating, Subbalakshmi—who never remarried and was affectionately called “Sister Subbalakshmi”—dedicated herself to helping other young Brahman widows by establishing a home, supported by the government, where they were educated and then trained as teachers. When the government opened a teachers’ training college for women in 1922, Subbalakshmi became its principal and eventually a leading educationalist in south India. Brahman child widows’ welfare was therefore not the sole focus of Subbalakshmi’s life’s work, but it probably was the most important for a former child widow who exemplified the role of widows in pioneering social reform and girls’ education in India.23 Subbalakshmi, though restrained by her official position, also campaigned in support of the Sarda Act.24

130 / Chapter Four

Subbalakshmi’s school for widows admitted a few other girls and its evident success encouraged some Brahman parents to think about educating their daughters.25 Moreover, as Sita Anantha Raman comments, Subbalakshmi’s “success in freeing widows like herself from mental torpor was a more powerful argument for liberalism than vociferous debate.”26 Ame­ liorating the lives of widows and improving female education were increasingly seen as closely related issues. As time passed, Subbalakshmi gained the support of other powerful liberals in Madras—mostly men and most but not all Brahmans. They also favored better education for all girls, which would simultaneously encourage the postponement of marriage, because girls were normally withdrawn from school after marriage or, if married, as soon as they reached menarche. Actually, even in elite circles where support for girls’ education was most vocal, some of it was probably rhetorical and did little to dispel the social pressure on parents to marry daughters before puberty in accordance with “respectable” norms. Thus an elderly Brahman widow told us in 2010 that she was married aged ten, one month before the Sarda Act came into force in 1930. Her father, an engineer, objected to the marriage, but he was overruled by senior relatives; his own father, too, pointed out that he (the engineer) had married his wife, a high court judge’s daughter from one of Madras city’s most illustrious families, when she was only eight. The girl (our informant) was then married to a subordinate judge’s son, a law student at the time, although her father flouted convention by letting her stay at school after the wedding. She came of age at thirteen and went to her husband’s house a year later. As soon as it was passed, the Sarda Act’s weaknesses were obvious to both supporters and opponents. Its provisions were “so toothless . . . as to have little practical effect” and its first result was actually “a drastic spate of child marriages” immediately before the law came into force.27 Thus the 1931 census reported a “rush” to get girls married in Madras.28 On the other hand, it is also worth noting that prepuberty marriage, albeit the ideal norm, was far from universal among Tamil Brahmans. The 1911 and 1921 censuses contain data on the civil condition of Tamil Brahmans in Madras and Tanjore districts, and the 1931 census on all Tamil Brahmans in Madras Presidency.29 In 1911, 80 percent of girls in the twelve-fifteen age cohort were married or widowed, but in 1921 the proportion had fallen to 60.5 percent, so that the frequency of child marriage was declining and a large minority must have reached fourteen without getting married. By 1931, 84.1 percent of girls in the seven-thirteen age cohort, as well as 13.1 percent in the fourteen-sixteen cohort, were recorded as unmarried. Unfortunately, developments after 1931 cannot be traced, because later censuses provide

The Changing Position of  Women / 131

no relevant data. This lacuna is not filled by other sources and later ethnographic studies hardly mention child marriage. Thus Fuller briefly reports that child marriage has not occurred among the Brahman Adishaiva temple priests of the goddess Minakshi’s temple in Madurai since the 1940s, while Gough, who must have met many former child brides in Kumbapettai in the early 1950s, merely noted that “marriage formerly took place before puberty but now occurs about the age of fifteen.”30 Obviously, people may have misled the census enumerators to disguise child marriages; the 1931 figures, following the Sarda Act, look distinctly unreliable. Furthermore, statistics on child marriage tell us little about attitudes toward it, although the downward trend after 1911 suggests declining popularity. Yet it is also salient that the majority of postpuberty marriages almost certainly did not occur in elite progressive families, but in poor families unable to pay bountiful dowries and wedding expenses. Sometimes, too, desperate parents would try to pass girls off as prepubertal, even though they had started to menstruate. For poor Brahmans especially, postpuberty marriage was usually a misfortune, not a progressive choice. In truth, though, there is next to no solid evidence about “public opinion” concerning child marriage among ordinary Tamil Brahmans in the early twentieth century, notwithstanding the claims made by both sides in the Sarda Act debate. Maybe the new law, albeit ineffectual, subsequently had an “educational impact” among Tamil Brahmans.31 Or maybe the debate surrounding it and the advocacy of reformers like Sister Subbalakshmi changed conservative minds, influenced by growing concern for the welfare of child brides and especially widows, as well as increasing support for girls’ education. But in the final analysis, although prepuberty marriage definitely disappeared among the mass of Tamil Brahmans within twenty years of the passing of the Sarda Act, the full explanation for the custom’s demise is far from clear.

Women’s Experience of Child Marriage In the debate about child marriage, when not dealing in generalities, reformers cited cases of horrific abuse of young girls and conservatives countered that these were exceptions in an otherwise benign institution. But rarely if ever was it recognized that women’s experience of child marriage could vary considerably. Certainly, Tamil Brahman widows, who were not allowed to remarry, of­ ten suffered appallingly. Widows personify inauspiciousness as aman·galı¯s, whereas conversely married women with living husbands are auspicious

132 / Chapter Four

suman·galı¯s. A widow’s condition was displayed by her tonsured head, plain white or ocher sari, and lack of any jewelry, specifically a ta¯li, the ornament tied round the neck during the wedding and never removed by an auspicious married woman. Widows were also objects of pity and fear, who were often secluded within their homes and prevented from attending auspicious rituals such as marriages. Child widows were the most unfortunate of all. Girls bereaved before menarche—who were often too young to understand marriage, let alone widowhood, and hardly knew who their husbands were—had their ta¯lis removed immediately. When menstruation began, their girlish clothes were replaced by a plain sari and their heads were shaved; instead of celebrating their new maturity and status as suman·galı¯s, they would instead begin to observe the lifelong austerities of widowhood. Among the Eighteen-Village Vattimas, a handful of elderly widows who were bereaved as children are still alive. They attract the sympathy of most other people and nobody now defends the way they were treated. An el­ derly Vattima man recalled his elder sister who was widowed when fourteen and died aged ninety-one, having spent most of  her life as an aman·galı¯. “What did my sister do wrong? Why was she made to suffer like that?” he asked, before castigating such “orthodoxy” as “stupid, unwanted, and unacceptable.” Compared with widows, the experience of other Tamil Brahman wives, both as child brides and mature women, was more mixed, although egregious cases definitely existed. For example, Leela Gulati’s grandmother Seetha (1885–1964), who was born into an impoverished family, became the second wife of a man twenty years older than herself when she was ten. She moved into his house when she was eleven, gave birth to four children, and was severely exploited by her mother-in-law and the rest of the household. Seetha’s youngest daughter, Saras (1913–93), who was Leela Gulati’s mother, was married at twelve and pregnant at fourteen, and always regretted having to leave school on marriage, which was a condition laid down by her husband’s family. Saras was not exploited like her mother had been, but she envied friends who could stay at school, as well as her own sisters, who married into “more enlightened families” and were able to complete their schooling after marriage.32 Mythily Sivaraman’s mother Pankajam belonged to a “more enlightened family,” but only as a result of painful struggle by her own mother, Subbalakshmi (1897–1978). Subbalakshmi’s father was a surveyor working for the Travancore government, who suddenly died when his wife Kamakshi was twenty-nine; Kamakshi, with Subbalakshmi and her other children,

The Changing Position of  Women / 133

went to live with her own father, a landlord, where she escaped the worst restrictions imposed on widows. Subbalakshmi was married at eleven to Gopalakrishnan, a government official aged twenty-eight, and started to cohabit with him at thirteen. Subbalakshmi’s first child was Pankajam; her next two children, both sons, died in infancy and she started to become chronically ill. Her relationship with Gopalakrishnan also deteriorated and he flatly opposed sending Pankajam to school. Subbalakshmi then took Pankajam to Madras to live with her brother, who supported female education. Pankajam eventually graduated from the teachers’ training college headed by Sister Subbalakshmi and married an engineer in 1929, when she was nearly eighteen. Educating her daughter and marrying her when she was “old” was “an unthinkable sin” for most Brahmans at the time, “but for Subbalakshmi, this was a matter of pride.”33 Yet not all child brides suffered and struggled. Kanaka was born in 1911 and we interviewed her in 2005, shortly before her death. Kanaka’s father was a lawyer, who was a son of C. R. Thiruvenkatachari, the judge whose family was described in chapter 2. Kanaka’s father was practicing in Chidambaram town and sent her to a local convent school. Kanaka was married at twelve to an eighteen-year-old student at Madras Law College and started to live with him when she was seventeen. After her wedding, she was withdrawn from school, but she studied on her own at home and used to read English novels. Kanaka visited her husband’s home, or he visited hers, at each major festival; she said that shyness prevented open conversation with her husband, but they managed to talk nevertheless, so that during the five or six years before she moved to his home they “courted,” as she put it while recalling the experience fondly. Kanaka’s mother-in-law and the rest of her husband’s family treated her kindly. Kanaka compared her own happy marriage with modern married life, which begins so “late” that all its “charm” has already vanished. Kanaka recalled that improved girls’ education was a key factor in ending child marriage, but, for her at least, it had been a joyful experience that young women today can never share. Generalizing from just a few examples is impossible, although we also have information from several Vattima women who married as children that has been set out elsewhere.34 Vattima cases, like the three just outlined, show that the experience of prepuberty marriage varied because it depended so much on an individual girl’s or woman’s circumstances. Most crucial was how a wife was treated in her husband’s family. If a young bride moved into a joint family, her fate overwhelmingly depended on whether she was well treated by its members, especially her mother-in-law, like Kanaka, or badly

134 / Chapter Four

treated like Leela Gulati’s grandmother. If, alternatively, a couple lived separately, like Mythily Sivaraman’s grandmother, a young wife’s relationship with her husband alone mattered most. Also important was a wife’s age when her marriage was consummated. In general, her prospects improved with delay and were best if she had reached her late teenage years, especially if she were also allowed to continue her education until she joined her husband. Consummation was sometimes further postponed after a girl started to menstruate, because, for example, a husband was away studying as in Uma’s case described earlier, but also out of parental concern that a wife was still too young for sexual intercourse and motherhood.35 Moreover and more certainly, it is clear that the welfare of all child brides primarily depended on their age when cohabitation started, not when they were married. To that extent at least, conservatives objecting to the Sarda Act were right to insist that the age of sexual consent was much more critical than that of marriage. Neither this conclusion nor evidence about women’s variable experiences can be used to justify prepuberty marriage, of course, but it would also be a mistake to exaggerate the custom’s evil by assuming that vulnerable Tamil Brahman girls always suffered miserably when married as children.

Child Marriage and Female Education One reason for lack of clarity about the disappearance of prepuberty marriage among Tamil Brahmans is that the 1930s and 1940s have been little studied by historians compared with earlier decades when controversy over the age of consent and child marriage was at its height. On the other hand, despite a shortage of directly corroborative evidence, a close connection between marriage customs and standards of girls’ education almost certainly existed, and a rising marriage age, as Raman concludes, did promote improvements in female education.36 Our own ethnographic data, from case studies on individual Vattima women, have allowed us to analyze the relationship between marriage and female education in their community.37 Most elderly Vattima women married as children between the 1930s and 1960s said that schooling for girls did not normally stop until the onset of menstruation, after which they joined their husbands. Hence the amount of education most girls received depended on their age at menarche, although a few were withdrawn from school straight after marriage. Typically, though, girls’ schooling ended in their midteenage years between the fifth and eighth standards. But when it ended was also affected by the proximity of suitable schools; almost all

The Changing Position of  Women / 135

villages had primary schools, but very few had secondary schools, so that village girls could not continue after the fifth standard, unless their parents arranged private tuition at home, like Uma’s did. Traveling to and from a school several miles away in slow bullock carts, the sole means of transport in the past, would have been impractical. (Village boys attending a secondary school usually stayed in its hostel or with relatives living nearby.) But the fundamental reason why nearly all Vattima—and other Brahman—girls had to leave school after marriage or menarche pertained to morality. No mature married girl was permitted to risk her moral reputation through an encounter with a male stranger—or even the rumor of one—outside the safety of her own home. With very few exceptions, therefore, young wives could not be allowed out to go to school; if they were accused of immorality before cohabitation, their husbands’ families might refuse to accept them, and in any event the public reputations of both partners’ families would be seriously damaged. Most conservative, rural Tamil Brahmans probably evaluated girls’ education and its moral perils much like Vattimas in the mid-twentieth century. In urban areas, however, traveling to school was easier and attitudes tended to be more enlightened, so that girls were more likely to stay in school after they came of age. Among the Vattimas, change began to occur after Independence. Kamala, born in 1935, was married after puberty at sixteen because her father, a doctor employed by the Madras government, had to comply with the Sarda Act. Kamala married a twenty-three-year-old engineer and air force officer, and started to live with him in Pune (Poona) at nineteen. By then she had completed a two-year intermediate course at college. She had also been offered a place in a medical college, but did not take it up, because neither Kamala nor her husband ever thought she would go out to work, so training as a doctor seemed pointless. Karpagam was born in 1937 and brought up in Tiruchirappalli; she married Kamala’s brother, who became a company manager, when she was twelve and he was nineteen. Thus Karpagam’s marriage contravened the Sarda Act, but any repercussions would have befallen her parents, not her father-in-law, the doctor. Karpagam’s father was a liberal-minded writer, whose mother apparently had modern views about girls’ education. Before joining her husband in Madras, Karpagam was allowed to complete her Secondary School Leaving Certificate (SSLC). She won admission to a women’s college in the city and her husband encouraged her to study there, but Karpagam explained that she was nervous about her competence in English. She does not regret missing college, as she has been happy as a homemaker and would never have taken a paid job anyway.

136 / Chapter Four

Kamala and Karpagam, like some other Vattima women of their generation, could complete their education because they had fathers or fathersin-law—or adult husbands—who belonged to a small cohort of men with progressive views about female education. These men, who almost all had urban, professional or managerial occupations, had the authority to insist that girls should complete their education before starting to cohabit, even if other relatives, such as the girls’ parents, were opposed; a few men were supported by senior women in their families. Some modern-minded men opposed prepuberty marriage too, but the critical factor in their attitude toward girls’ education was that they disapproved of secluding young wives after menarche. During the sixties and seventies, progressive views began to spread among the Vattimas. That in turn meant that responsibility for a daughter, which her parents used to transfer to her husband’s family when she came of age, now remained with them until she had completed her education—preferably by obtaining the SSLC—and was deemed ready for marriage. Whether consciously aware of it or not, those who favored better girls’ education were simultaneously saying that parents should look after their daughters for longer than they had done when prepuberty marriage was the norm. Growing support for improved female education was ultimately decisive in stopping prepuberty marriage among Vattimas, whereas for other Tamil Brahmans—who had abandoned the custom earlier—it was probably the other way round, inasmuch as a rising marriage age promoted better female education. In all cases, however, the crux of the matter was a girl’s destiny after she was married or started to menstruate; the end of child marriage meant that her own family, not her husband’s, had to decide whether she could stay at school and that change was vital for improving education for all Tamil Brahman girls, with the possible exception of a privileged minority educated at home.

Education, Employment, and Marriage between the 1940s and 1980s In the first two or three decades after Independence, few Vattima girls were well educated and very few women had jobs outside the home. Much the same was true of many other Tamil Brahman girls and women, even though they were marrying after puberty. Yet some change was taking place, even in villages. Thus in Kumbapettai in 1952, no Brahman woman went out to work, but in 1976 one was a clerk, one a midwife, and two were seamstresses; there were also eight college or university students. Furthermore,

The Changing Position of  Women / 137

after the death of a Brahman from the village employed in a government office in Thanjavur, his widow was appointed to his post—even though the dead man’s family raged at this radical departure from orthodoxy.38 In towns and cities, the number of Tamil Brahman women who had completed school or college education was probably growing fairly quickly; some of these women also had paid employment. But unfortunately, there is little evidence about them, apart from some material in Patricia Caplan’s study of upper-middle-class women, many of them Brahmans, in Chennai in 1974–75. Caplan’s informants included well-educated housewives married to men in professional or managerial employment, but most of them, who married shortly after graduation, had never worked and never seri­ ously contemplated doing so. Caplan also had a sample of sixty-six professional women, mostly in their forties or fifties, who included teachers, lawyers, doctors, social workers, and government and bank officials, as well as some working in family businesses, almost all of whom had university degrees. About half these women were married, but the rest were not, which is a strikingly high proportion, given the near universality of marriage— especially for women—in Indian society. Most professional women complained about the difficulties and discrimination they faced, mainly owing to widespread disapproval of their “deviant” decision to work outside the home, the proper place for a morally respectable wife and mother.39 The disapproval described by Caplan has not vanished, but it is less common than it was. Our interviews among middle-class women in Chennai in 2003–5 included a few Brahman graduates who had worked before and after marriage. Two of them were teachers in their fifties who had started work in the 1970s. The older woman, a company manager’s daughter, has been a schoolteacher since she first qualified. She has mainly pursued her career in Chennai, but also taught in north India, where her husband, an IIT graduate and engineer, was working (although she is now separated from him). The younger woman, probably from a modest family background, retired in 2003 after a career in Chennai schools. After qualifying, she worked for several years before marriage and continued afterward, apart from a break when her children were young. But from 1980 until 1992, her husband was away working as a welder in the Gulf. Since neither woman mentioned it, we do not know whether they had to work because they were effectively single mothers, or whether they faced disapproval for working outside the home. At least by the 1980s, though, Tamil Brahmans appear to have regarded teaching—like working in government offices, banks, or insurance companies, as well as the other professional occupations mentioned by Caplan—as generally acceptable types of female employment. All the same,

138 / Chapter Four

the overwhelming majority of women, irrespective of age, always firmly put their families’ interests before their careers and usually stopped work­ ing when they had children, unless they could rely on child care provided by other family members, normally mothers or mothers-in-law. M. N. Srinivas’s article on the changing position of Indian women, published in 1977, helps to put the Tamil Brahmans of that time in a broader context. Srinivas discussed educated, mainly high-caste, urban women, probably depending quite heavily on his personal observations in Delhi and Bangalore. By the 1970s, women in this category were employed in a range of white-collar and professional jobs, such as teachers and social workers, doctors and nurses, and officials and clerks. The “career woman in India is very visible,” said Srinivas, but he exaggerated in claiming that “it is significant that the society at large has quietly accepted women’s assumption of new roles.”40 In reality, as Caplan and others reported, there was plenty of prejudice against working women throughout India. Srinivas also described the growing popularity of education for girls, especially in urban areas, and explained how parents generally wanted their daughters edu­ cated “to give them the utmost advantage in the marriage market,” rather than to obtain jobs, which usually meant that women must not be “over­ educated” because men almost always looked for partners whose education was inferior to their own. On the other hand, “colleges and universities provide respectable waiting places for girls who wish to get married” and among the high-caste, urban middle classes “the possession of a basic degree is now regarded as essential for obtaining a ‘good’ groom.”41

The Development of Companionate Marriage Srinivas’s observations about education and marriage are amply supported by evidence from all over India, where matrimonial considerations have powerfully driven rising standards of girls’ education, especially in uppercaste, middle-class groups. This is certainly the case among Tamil Brahmans. It is therefore logical to interrupt our discussion of female education and employment in order to examine modern change in the marriage system. At least by the late 1970s, almost all educated, employed Tamil Brahman men—including most Eighteen-Village Vattimas, despite their subcaste’s conservatism about marriage—normally wanted wives who had completed the SSLC and many preferred graduates. Young women often married as soon as they finished their education, or even a year or two earlier if studying for a degree, whereas men—who must always be older than their wives—did not usually marry until they had found secure employment.

The Changing Position of  Women / 139

Sons, as explained in chapter 3, have long been encouraged to study science and technology, engineering and medicine—the paths to well-paid, high-status professional and managerial jobs. Daughters, on the other hand, because marriageability normally mattered far more than employment, have usually studied arts and humanities, although teaching qualifications and the vocational commerce degree (BCom) were also popular with some girls and their parents. Medicine is a special case as an esteemed profession that has always been open to women, as well as men. In practice, though, as noted earlier, medicine has become less attractive to all Tamil Brahmans in recent decades owing to the impact of the reservations system. Tamil Brahman men began to prefer educated wives for the same reasons as countless other Indian men: in particular, it was assumed that they would be more congenial partners for the companionate marriages that were emerging as the norm, as well as more competent mothers who could help with their own children’s education. A key feature of modern compa­ nionate marriages is that young men and women are actively involved in selecting their own partners, whereas in traditional arranged marriages, although a young man may be consulted and may be able to reject a poten­tial bride, a young woman often has very little say and no power of veto over a potential groom at all. And even if she has sympathetic parents who do consult her, she and her husband-to-be cannot meet each other before the wedding. For thirty or forty years after prepuberty marriage disappeared, arranged marriages of the traditional kind were still the norm among Tamil Brahmans, even in urban, upper-middle-class families. Indeed, although they are admittedly more old-fashioned than most Tamil Brahmans, arranged marriages of the traditional kind were still common until recently among the Minakshi temple priests in Madurai. Thus one priest, explaining how successfully he had arranged his twenty-four-year-old daughter’s marriage in 2001, told Fuller that she never even met her prospective husband when he came to the brief meeting between the two families, because everything about the match was agreed so quickly.42 That is an extreme case, but, a few decades earlier, many Tamil Brahmans would have arranged their daughters’ marriages similarly and would not have thought the priest’s conduct untoward. The development of companionate marriage is linked to the rising age of marriage, particularly for females. Nowadays, middle-class Tamil Brahman women normally marry in their early twenties, and men in their late twenties or early thirties. Self-evidently but significantly, twenty- or thirtyyear-olds—unlike young children or even adolescents—can make informed decisions about possible spouses. Moreover, they usually insist on being

140 / Chapter Four

fully consulted, so that the selection of marriage partners has now become the joint responsibility of young men and women, together with their parents or other senior kin. There is also a link between companionate marriage and the reduction in completed family size that occurred among middle-class Tamil Brah­ mans in the mid- to late twentieth century. That development helped to undermine the traditional preference for sons over daughters, which has steadily faded away, so that few Brahman parents still worry about it and few have more than two children, even if they are both girls. Within their smaller families, daughters are more valued than they were and, in relation to their general welfare, parents treat girls and boys much more equally today than in the past. Until recently, almost all south Indians, including Tamil Brahmans, had a normative preference for marriages between cross-cousins (mother’s brothers’ and father’s sisters’ children) or between uncle and niece (that is, “elder sister’s daughter’s marriage” in anthropological terminology). In other words, the ideal spouse was a relative within a specific category, either a known consanguine or a “classificatory” relative as defined by the rules of the south Indian, Dravidian kinship system. In practice, in all social groups, the proportion of marriages complying with the normative preference was always fairly small; for example, Gough gave a figure of 12 percent for Brahman unions in Kumbapettai in the 1950s.43 Nonetheless, the preference for close-kin marriage was strongly expressed and virtually universal, until it suddenly began to alter about one generation ago for reasons that Isabelle Clark-Decès explores.44 Today, most south Indians—except for some elderly, conservative people—positively prefer marriages between partners who are unrelated or only very distantly related. Nearly all Tamil Brahmans, like the vast majority of ordinary Indians, continue to marry within their own caste, although marriages across subcaste boundaries are commoner than they were, particularly among the bulk of Aiyars, who tend to be less conservative than Aiyangars. Many Tamil Brahmans, too, regard a union with another south Indian Brahman as acceptable, but usually only if no suitable partner can be found within their own caste. For all marriages today, the attributes of individual men and women matter greatly and certainly more than in the past. Middle-class Tamil Brahmans, in common with other Indians, particularly emphasize education and employment. Thus everyone talking about grown-up children or grandchildren, and their spouses or marriage prospects, invariably itemizes each individual’s qualifications and career. People just take it for granted

The Changing Position of  Women / 141

that any sensible conversation about marriage must also be about education and employment. Other things being equal, the most attractive husbands are of course those with degrees from elite colleges or advanced professional qualifications, as well as high-status, well-paid jobs in medicine, engineering, IT, and so on, either in India or overseas—especially in much-favored America.45 For women, education and employment are ranked more or less identically, although men’s educational qualifications are normally a little higher than their wives’. But many men prefer “home-loving” wives, who will not work outside the home after marriage, so that they can fully devote themselves to domesticity and motherhood—although some skill in classical Carnatic music or dance, Bharatanatyam, is also appreciated. On the other hand, more men than in the past are now happy to have working wives and growing numbers of young women have career ambitions, so that they refuse to marry men who will stop them taking paid employment. Usually, when a marriage is in the offing, possible partners are suggested by relatives, friends, or other contacts, or they are found through Internet sites and other matrimonial agencies, such as those run in some temples by Brahman voluntary associations; newspaper advertisements tend to be a last resort. The young people’s photographs and horoscopes are then exchanged by the respective families. They collect more information and start to discuss a possible marriage if a match seems likely, especially because educational and employment criteria are satisfied, but also because the two young people like the look of each other. The couple is also encouraged to try to get to know each other better through telephone calls, letters, or e-mails, or by meeting in person. If they, and their parents and other close senior kin, are all in favor, an agreement to marry is announced. Weddings are expensive, although not exceptionally so by the criteria common in contemporary India, or indeed in many other countries. All the same, in urban, upper-middle-class Tamil Brahman circles today, the bride’s family can expect to spend around ten lakhs, one million rupees. About two-thirds of this amount goes on the bride’s dowry or strı¯dhanam, consisting of gold sovereigns, silver, and diamond jewelry, as well as articles for her new home. Most of the rest is spent on new clothes and presents for family members, plus the reception. The groom’s family buys new clothes and presents for its own members, as well as the bride’s wedding sari. Payments in cash or kind to the groom’s family are often small or nonexis­ tent, unlike the sizable “groomprice” paid in some communities. Explicit financial negotiations are unusual; instead, the bride’s side makes informal contact through intermediaries to make sure that the groom’s expectations about the dowry and the splendor of the reception will be met, and that all

142 / Chapter Four

parties are content. A young couple keen to marry may urge their parents to be reasonable and avoid arguments about money. Some older people criticize the new modes of consultation because they give naïve youngsters too much say, but almost everyone recognizes that pushing a marriage through against the objections of either a young man or a young woman is no longer possible. In addition, most parents accept that listening to their children carefully is wise, because divorce, albeit very rare, has become a risk. More positively, though, Tamil Brahmans, like many other Indians, now believe that the future couple’s congeniality and happiness are vitally important; that is the primary reason for consulting young people and encouraging potential partners to get to know each other before a marriage is finalized. Young women, however, sometimes express more concern about future mothers-in-law than husbands, believing that the former can cause more trouble. Today, most couples live separately as nuclear families, especially in urban areas, although some become part of the husband’s joint family. But very few women ever regard the relationship with a mother-in-law lightly, however far away she is. In the contemporary marriage system, senior kin continue to matter. Modern companionate marriages are still arranged, and what has altered is that parents and grown-up children together now select the partners, whereas previously children—especially daughters—had little or no say. “Love marriages,” in which young people choose their partners by themselves, remain uncommon, even among well-educated, upper-middle-class Tamil Brahmans, although they are less rare than they were. Most young people accept that their more experienced parents and other senior relatives can probably find better marriage partners than they could do by themselves; Caplan heard the same opinions in the 1970s.46 Thus Revati, an IT professional in her early twenties, said that she and her friends constantly discussed love marriage, which she would ideally prefer because it “seems to be the right kind of marriage.” But she was very unsure that she could find a suitable partner by herself and her parents were already looking for one; she also stressed the importance of family support for any decision about marriage. “Why take unnecessary risks?” said Revati; when asked if any marriage is not a risk, she responded by saying, “Yes, but finding someone on our own is more risky. Why should we do that? We should leave it to our elders.” Revati, however, was also sure that her parents would never force her to marry someone she did not like. Revati is actually a Kannada Brahman, but her opinions are common among young, middle-class Tamil Brahman women as well.

The Changing Position of  Women / 143

Sometimes, especially by sons living overseas, selecting a partner is left almost entirely to parents. Arjun, for example, is a Vattima IT consultant living in the Chicago suburbs with his wife Leela, a financial analyst. Arjun moved to America in 1996 and married Leela in 2002, when he was in his midtwenties and she was a little younger. Arjun’s parents in Chennai examined four marriage proposals for him, including Leela’s, and then met her several times before deciding that she was the right girl. Arjun and Leela probably communicated with each other before the wedding, but they did not meet. Arjun, in fact, wanted his parents to conduct all the important discussions and he relied on them to select a wife with whom he would be happy. Leela was consulted in person by her parents and agreed to the marriage proposal. Companionate marriage, as it has developed over the last generation or two, has enabled young people to make more informed choices about potential partners than their predecessors, even though they often decide with little real knowledge about each other. Men have benefited, but for women the change for the better has been much more radical. As people of all ages recognize, young Tamil Brahman women today have far more control over their own matrimonial affairs than their mothers or grandmothers had, let alone women in earlier generations who were child brides. And although young married women are still subject to many constraints, especially from domineering mothers-in-law, they can rarely be harassed or exploited like girlish wives in the past. Finally in this section, let us reiterate the point that companionate marriage has not undermined endogamy. Indeed, most Tamil Brahmans—like other Indians—assume that caste or subcaste endogamy contributes positively to marital companionship, because partners with the same cultural background are most likely to form a strong bond. Similar views are found everywhere, of course, but in India a specific connection between shared caste culture and good companionship is prominent. Endogamy, therefore, is functionally consistent with modern companionate marriage and not a mere survival from an earlier epoch. Moreover, because partners are selected by matching educational and employment criteria, as well as by judging their potential happiness as a compatible couple, the contemporary marriage system plays a fundamental role in reproducing class as well as caste. Our data show, in particular, that urban, upper-middle-class young people overwhelmingly tend to marry each other, so that a caste-cum-class unit— the Tamil Brahman upper-middle class—is in practice a predominantly endogamous status group.

144 / Chapter Four

Education and Employment since the 1980s The development of companionate marriage and the expectations of potential husbands have been significant factors in encouraging Tamil Brahman girls to enter higher education, as we have seen. But young women’s attitudes toward paid employment have changed as well. This is partly because young women brought up as equal to their brothers and as well educated as them, or nearly so, are less inclined than their mothers were to confine themselves to motherhood and domesticity; instead, they actively want to work and in this ambition they are often supported, or at least not opposed, by their parents. Many older women, though not all, positively encourage their daughters and granddaughters to take up opportunities that they missed. Some older people disapprove of young women’s assertiveness, especially when they drag their feet about getting married, but everyone agrees that real change has occurred. During the 1980s, the attitudes toward female education and employment prevalent in earlier years were changing and they have continued to do so, more fully and rapidly, since economic liberalization began in the early 1990s. Meenakshi, a Tamil Brahman born in 1962, who lives in Chennai, has a BA in English literature and has never been out to work. Meenakshi was talking about her two daughters’ education and said that the decisions made by her and her husband, a bank manager, would have been the same if they had had sons. They did not want their daughters to go into banking, but hoped they would find jobs in the better-paid technology sector. Both daughters studied mathematics, science, and computing; the elder, twenty at the time of the interview in 2004, was then an engineering student who also wanted to do an MBA later. When she finished college in the 1980s, Meenakshi said, most women graduated “for the sake of it,” not for career reasons, but now, like her own elder daughter, “girls know what they’re doing.” Another girl with a similar outlook is the only daughter of Gita, a Vattima woman born in 1970, who was brought up in Chennai, graduated with a BCom in 1990, and married immediately afterward; her husband is an engineering company manager. Gita had wanted to postpone her marriage to study for a master’s degree, but that was not allowed and after her wedding she soon became pregnant. Gita, however, had only wanted to study for longer and, unlike some of her Brahman friends at college, she never desired a paid job. In 2009, Gita’s daughter became the first Vattima girl to enter IIT–Madras and she intends to work after graduating, probably in IT. Like many other IIT students, female or male, she also wants to study for a further degree, probably an MBA, either in India or abroad.

The Changing Position of  Women / 145

Gita’s daughter’s admission to an IIT makes her exceptional, but her attitude to further study and employment is typical of young, upper-middleclass Tamil Brahman women who “know what they’re doing.” People we spoke to in our research, whether or not they were IT professionals themselves, generally agreed that the IT industry has been critical in altering young women’s perceptions. The differences between IT companies and older, “core” engineering ones have encouraged young women to study engineering, or other technological disciplines, with a view to a career in the IT sector. Based on her research in the late 1980s, partly with students at IIT–Madras, Carol Mukhopadhyay commented on the overwhelmingly male environment of engineering departments, as well as the working conditions in manufacturing and construction, where engineers regularly work in dirty factories or on outdoor sites alongside male manual workers. As a result, in contrast to the natural sciences or medicine, for the moral reputation of girls the “most socially dangerous field by far is engineering” and most parents actively steered their daughters away from it, thus further reinforcing male dominance over the profession.47 Not surprisingly—though it hardly differs in the West, despite lack of concern about the moral danger of engineering—there are hardly any women on the managerial staff of Chennai’s engineering manufacturing companies and during our research we actually managed to meet only one, a north Indian women in her midtwenties, who was working as a graduate trainee in Chakra Vehicles’ marketing division. The working environment in the IT sector, especially in the major software companies, is very different from that in manufacturing companies, as we explained in the previous chapter.48 In particular, IT professionals do “clean” work alongside colleagues in city offices, so that it is seen as both fitter and morally safer for women than “socially dangerous,” older types of engineering, despite the late working hours sometimes required. Hence IT has radically modified the aversion toward an engineering career described by Mukhopadhyay, because in the last two decades it has encouraged growing numbers of young women to go to engineering colleges, find jobs in software companies, and compete on equal terms with men. Flat hierarchies define the corporate ideology of IT companies, as mentioned earlier, and in reality, too, there is less inequality, including gender inequality, than in manufacturing firms. How the IT sector treats women is important for evaluating the changing position of educated, professional women, including Tamil Brahmans. Our informants, almost without exception, insisted that there is no gender-based inequality among individuals in the major software companies. Smitha Radhakrishnan reports the

146 / Chapter Four

same findings from her research in Bangalore and Mumbai.49 One salient feature is the virtual consensus that women and men have equal technical skills in software engineering, computer programming, and other related fields, which is consistent with the general assumption in India—unlike the West—that both sexes are equally good at mathematics, science, and technology.50 Mala, a project team leader in ICS, is a non-Brahman who was in her late twenties in 2005. Mala firmly insisted that the company has “no gender gap,” by which she meant that gender is irrelevant for recruitment, ap­ praisal, promotion, salary levels, work allocation, project team composition, or other significant issues at work. Other women shared her view. All staff normally belong to mixed-sex project teams, women are frequently in charge of men, and everyone asked about team composition said that it depends on the skills that are needed. Some people (men as well as women) say that women tend to be better project managers than men, because women are more sensitive to personal problems and social relationships within teams. Yet even women who argue that promotion should take more account of these qualities, so that it depends less on technical attainment, do not see gender bias in ICS as the main issue. Thus Parvati, a team leader mentioned in chapter 3, complained about her insensitive male project manager and compared him unfavorably with his female predecessor, but she made it clear that she was not saying he had been promoted because he was male and was not accusing him of discriminating against women on his team— any more than he, as a Malayali, was discriminating against Tamils. Team performance depends on cooperation and members who do not pull their weight are resented by others. Women who take time off to care for children can irritate fellow team members, for instance, but so can men who spend time in the office chatting about all and sundry to friends. Most grievances, in fact, are about particular individuals who waste time, whereas generalized complaints about either men or women are rare. Team activities—such as lunches, which are fairly frequent—are designed to promote solidarity among all members, but most informal socializing by IT professionals is divided by gender. This tends to mean that bonding between male managers and other men on their teams, who sometimes socialize a lot, can be fairly strong, whereas it is weaker between female managers and other women, and weakest between managers and team members of the opposite sex. These variations may give men advantages—for example, in lobbying male managers for overseas assignments—but their impact on vital issues like appraisal and promotion is not clear-cut. Thus, for instance, male managers may support male friends in appraisals, but they may also

The Changing Position of  Women / 147

deliberately not do so to avoid damaging accusations of favortism. In any case, though, nobody ever cited any specific cases in which they believed that gender bias had unfairly affected somebody’s career. Women’s equality with men is often linked to “empowerment,” which is a term in common parlance among IT professionals.51 Lakshmi, a married Tamil Brahman, was an assistant project manager in ICS in her early thirties when we met her. She said that individuals are all selected “on their strengths only” at ICS and she also talked about her “empowerment” in some detail. Thus Lakshmi explained that she feels empowered when she represents ICS to a customer abroad, because it enables her to cope with pressure—which brings out “one’s inner ability”—and creates a sense of achievement that women can scarcely find outside the IT sector, as she can see from her friends in other lines of work. She is personally respected, too, for what she has achieved in her family and among her relatives. Lakshmi’s empowerment also comes from her “financial freedom,” as she put it; high salaries allow women to enhance their negotiating power in the family domain, particularly by resisting demands for excessive domestic labor from husbands or mothers-in-law. Lakshmi further referred to her “exposure”; she had moved to Chennai to join ICS in her early twenties and said that the exposure in the city—compared with the town in Tamilnadu where she grew up—was “too good to miss.” For all IT professionals, though, the most vital form of exposure comes from working overseas on project assignments. Lakshmi was sent to Britain for three months in 2003. It was “tough,” mainly because she left her young son behind, but the “big exposure” directly contributed to the “high level” of empowerment that she now feels. Women’s equality and empowerment at work are significantly affected, however, by marriage and children. In contrast to the unmarried professional women studied by Caplan in Chennai in the 1970s, we never came across any women IT professionals in their thirties or forties who had remained single to pursue their careers, and all the single women we met assumed that they would marry eventually. That is partly explained by a more relaxed attitude about married, employed women in Chennai compared with three decades ago, but partly also by more conservative pressure to get married than in “Westernized” cities like Bangalore, where Radhakrishnan met a few women IT professionals who had chosen to remain single.52 Most women IT professionals from all communities carry on working when they get married; single women consistently say, too, that they want a husband who will let them choose whether to work after marriage. They also say they want to marry someone who understands the industry’s special working

148 / Chapter Four

conditions, which means that most prefer husbands who are IT professionals as well. In contrast, the majority of male IT professionals are (or hope to be) married to nonworking, “home-loving” wives. At present, there appear to be enough men in the minority to satisfy the wishes of women IT professionals, and the discrepancy in preferences is not a noticeable problem. Marriage itself does not necessarily have much effect on a woman IT professional’s working life, assuming that she stays in her job, although she may have to move if her husband works in a different place. Of course, for any woman, marriage entails a major restructuring of relationships and responsibilities. In some households, wives have to work a “double shift” before and after office hours, and adjustment may be particularly difficult if a wife moves into her husband’s family home, instead of setting up a new conjugal household. On the other hand, IT professionals can usually resist demands from exploitative mothers-in-law better than other married women, partly because they are away at the office for much of the time, but more importantly because their large salaries contribute so much to the household budget that they cannot be trifled with. When women have children, their familial priorities and responsibilities move to a new level, so that childbirth, rather than marriage itself, normally has the greater impact on women IT professionals’ working lives and forces them to make decisions about their family-work “balance.” Very few mothers rely on paid child care; instead, they usually depend heavily on their parents-in-law and often their own parents as well. Anuradha, who appeared in the previous chapter, has been particularly fortunate. Anuradha, her husband, and their young daughter lived with his parents, and her own parents lived in the same apartment block, so that she could always rely on family members for child care. Lakshmi’s circumstances were not as easy. In 2004, she lived in an apartment with her husband, a manager in a data transcription company, and their son, who was five. Lakshmi’s parents lived far away from Chennai, but her husband’s parents were on the next street. In the morning, she took her son to school and after school, in early 2004, her parents-in-law looked after him until she returned home. Later that year, she made the unusual decision to employ a personal tutor for her son, not because he really needed tuition, but so that her parents-in-law were not weighed down with child care, for they also looked after their own daughter’s son. On weekends, Lakshmi went to work only if absolutely necessary, so that she could catch up with housework and spend more time with her son. Lakshmi praised ICS’s flexitime system, which helped her to manage the balance between family and work through “proper planning” and “working effectively.” Yet

The Changing Position of  Women / 149

Lakshmi was anxious about her son and did consider a career break, because she felt that she was not looking after him properly. She and her brother were trying to persuade their parents to move to Chennai so that they could help her, because she disliked being dependent on her parents-in-law, even though they were supportive and understanding. In the future, though, said Lakshmi, when she had met her financial needs and personal ambitions, she intended to find a less demanding job outside the IT sector. Implicitly at least, she acknowledged that her empowerment had its limits. Jayashree eventually accepted that she could not cope with both her children and her job. Jayashree, a married Tamil Brahman, was an assistant project manager at ICS in her midthirties in 2004, and she had a twelveyear-old daughter and a toddler son. Her husband also worked in IT and they lived near his parents, who mainly cared for the little boy, but periodically they visited her husband’s brother in America. Then Jayashree’s own parents, who lived in a different part of Chennai, looked after her son, so that she described herself as a “nomad” moving between their houses. Jayashree found herself heavily burdened by the pressures of combining family life and work, and she dwelled on her guilt about her son. Finally, she resigned from ICS to spend more time with him, although she did contract work at home for the company. The problems faced by Lakshmi and Jayashree are not unusual, of course.53 In ICS and other software companies, women with children often leave their jobs or take career breaks sooner or later. But many IT professionals decide to resign as soon as their children are born, and even some very enthusiastic young women say that they want and expect to become full-time mothers later. All the major software companies recognize the high rate of “attrition” among married women and try to address the problem with “family-friendly” policies. These policies can be helpful to some women IT professionals. Nevertheless, almost all women, however empowered they feel, consistently give their families higher priority than their work and careers; in short, as Radhakrishnan crisply comments, they put “family first, job second.”54

Conclusion During the twentieth century, child marriage ended and the standard of Tamil Brahman girls’ education steadily rose; by its close, girls and boys were almost equally well educated, many young women had degrees, and more and more women were going out to work. These trends have continued to the present day and the IT industry has had a pronounced impact,

150 / Chapter Four

because it has given young women an opportunity to work in a new, prestigious, and even glamorous industry. Women IT professionals belong to diverse castes, but in Brahman eyes, the presence of so many Brahman women, as well as men, provides further evidence that their intelligence and intellect specially qualify them for IT. In the major software companies, where many women describe themselves as “empowered,” the discourse of empowerment is not just wishful thinking. Gender equality in software companies is a social reality that means a lot to women when at work; very importantly, too, male IT professionals learn to treat their female colleagues as equals. Furthermore, women IT professionals have acquired a personal status and autonomy in their own families and social circles that their mothers—and the vast majority of Indian women—have hardly ever enjoyed. Of course, only a minority of young women will ever be employed by the top IT companies—and an even smaller number will rise to become company executives. But such careers are no longer beyond the imagination of girls, or their parents, and nowadays a lot of young women are determined to start a professional career, even if they give it up later when they have children. Women IT professionals—whose numbers have risen rapidly because the software companies recruit thousands of new staff every year—have become role models for countless other young, middle-class women, both Brahman and nonBrahman, who can now see opportunities for themselves that scarcely existed twenty years ago. Yet IT professionals, like almost all working women in India, continue to put family first, job second, as we have seen. Wanting to work outside the home and pursue a career are arguably inconsistent with accepting traditional female roles as wives and mothers; in the case of IT professionals, who emphasize their empowerment and equality with male colleagues, that inconsistency looks more like blatant contradiction, as Radhakrishnan, and Upadhya and Vasavi, contend.55 Radhakrishnan specially looks at how women in a transnational professional class are expected to remain authentically “Indian,” whereas for Tamil Brahmans within India—or even overseas—their Brahman identity is normally at least as salient as their broader Indian one. But her general point holds well: among all women IT professionals, “‘balancing’ the desires and motivations of the individual with the duties and obligations of the family” is strongly emphasized, and women who can strike this “balance” properly fulfill a “dominant model of femininity” that is also expressive of upper-middle-class status.56 None of this means, though, that women working in IT—or indeed in other occupations—who put family responsibilities before their own ca-

The Changing Position of  Women / 151

reers or, more specifically, choose to leave their jobs when they have children, are suffering from false consciousness or shrunken individual agency. Of course, these are complex issues that have been extensively debated—by feminists and others, in India and the West—for many years. Nonetheless, it appears to us, the Tamil Brahman women we met are realistic about the opportunities and constraints that exist for them. Moreover, they are certainly no longer trapped in the old, restrictive patriarchal system, because the position of women in relation to marriage, education, and employment has altered radically over the last hundred years. In the nineteenth century, first of all in Bengal, the emergent urban middle class of colonial India mostly favored improved education for girls, who would then be better wives and mothers, but in Madras not much changed among the Tamil Brahmans who dominated this class until prepuberty marriage came to an end. But when it did end, girls’ educational standards rose and Tamil Brahmans, in common with other Indians, came to see female education as an index of modern, middle-class status. In the past, the Tamil Brahmans’ caste status significantly depended on ritual purity, especially female purity, and their compliance with scriptural orthodoxy, which were both demonstrated by prepuberty marriage. Today, purity and orthodoxy are much less salient. Moreover, in relation to the position of women, the old order has been virtually inverted, so that—oversimplifying to make the point—the Tamil Brahmans’ claim to superiority over other castes, which used to depend on extreme inequality between men and women, now depends on the relative absence of inequality between them. In reality, giving up child marriage brought the Brahmans into line with the great majority of non-Brahmans. But Brahmans, of course, do not see the change in that way, which might imply that they were undergoing “deSanskritization.” Instead, in their eyes, it is an intrinsic part of their transformation into a modern, middle-class caste, in which women as adults have a full say in their own marriage arrangements and are well educated like men. The near parity between the sexes in these respects is crucial to the status of the Tamil Brahman middle class, especially the upper-middle class, whose well-educated members are employed in professional and managerial occupations. Functionally, too, that status is reinforced by the ability of both sexes to take up these occupations, even if relatively few women actually pursue long-term careers. To be sure, gender inequality has not vanished among Tamil Brahmans. Women always have to face the competing pressures of family and career as men do not; in addition, to mention a few salient examples, the domestic division of labor is usually unequal, many women still observe menstrual

152 / Chapter Four

pollution, female sexuality is normally strictly controlled, and women, es­ pecially when single, are subject to much more moral surveillance than men. All the same, it is striking how many ideas about female inferiority that used to matter a great deal have been abandoned by Tamil Brahmans during the last century or so. That suggests, at least in this case, that caste inequality is more deeply rooted than gender inequality, in line with Carol Mukhopadhyay and Susan Seymour’s observation that “elite, upper-status women” from the “formerly most prototypically patrifocal” groups “have achieved the most educationally and . . . appear to be challenging traditional patrifocal structures and ideology.”57 Perhaps unsurprisingly, Tamil Brahman men have in the end more readily ceded status to Brahman women than Brahmans collectively have done to non-Brahmans, so that notions of caste superiority have persisted, while gender inequality has diminished. The superiority that Manu prescribed and Tamil Brahman men once took for granted has gone and, in our experience, all Tamil Brahman women, young and old, insist—like Uma, who opened this chapter—on how much has changed and how much more control they now have over their own lives.

Five

Urban Ways of Life

The vast majority of  Tamil Brahmans now live and work in towns and cities. They usually retain few, if any, links with their ancestral villages, although there are exceptions, including members of the Eighteen-Village Vattima subcaste, who often hold strong views about the relative merits of urban and rural ways of life. Most Tamil Brahmans also know next to nothing about rural life and tend to assume—like other Indian urbanites—that villages are primitive places where no civilized person should live. As we saw in chapter 1, Brahman urban migrants had to face new problems concerning their caste status, ritual purity, and moral conduct by making compromises and adjustments. In general, though, individuals and families adapted to urban life fairly easily and quickly, within a generation or two. It is obvious but important, too, that the Indian middle class is first and foremost an urban class, so that the Tamil Brahmans’ transformation into a middle-class caste has been simultaneously a process of urbanization. The majority of Tamil Brahmans live in Chennai and other towns and cities in Tamilnadu, but a large minority is dispersed across the rest of urban India, while a small but growing number is overseas; population estimates can be found in the appendix. We look at four urban sites in particular: Chennai; Bangalore and Mumbai (Bombay), which share some salient features; and the outer suburbs of American cities. Most of our evidence comes from interviews, rather than direct observation of urban living. We also know more about Chennai than the other places, but Bangalore, Mumbai, and America are discussed, because they are such important settlement areas for Tamil Brahmans and loom so large, in contrasting ways, in their collective imagination. In discussing urban ways of life, Herbert Gans’s critical revision of Louis Wirth’s influential essay on urbanism is helpful. Wirth argued that cities

154 / Chapter Five

are characterized by “secondary” social contacts, which are “impersonal, superficial, transitory, and segmental,” whereas “primary” contacts among kin and neighbors, for instance, are the norm in villages.1 As well as showing that Wirth’s generalizations about urbanites are poorly supported by empirical evidence on American cities, Gans usefully identified a third category of “quasi-primary” relationships among neighbors, which are “more intimate” than secondary contacts but “more guarded” than primary ones; “quasi-primary” relationships are particularly characteristic of suburbs. He also noted that in socially and ethnically homogeneous urban and suburban areas, socialization among neighbors is more frequent and social ties are stronger than in heterogeneous ones.2 In Chennai, where more than 20 percent of all Tamil Brahmans live, middle-class Brahmans are the dominant group in an extensive area of the city, so that a relatively high degree of social homogeneity exists there. That homogeneity is in turn linked to fairly high levels of caste-based solidarity and social control, mainly exercised through networks of primary kin and quasi-primary neighborly relationships. In Bangalore and Mumbai, as well as in American suburbs, on the other hand, Tamil Brahmans—and indeed Brahmans in general—are almost always a minority group in their residential areas, so that they live in a more socially heterogeneous environment with correspondingly less caste-based social control, fewer or weaker primary and quasi-primary relationships, and more individual autonomy. These variations in urban ways of life, as well as the differences between them and rural ways, are particularly important for women, who are normally more subject to social control than men and often very vocal about it as well. Another significant issue for many Tamil Brahmans is the perceived contrast between India and the United States, which promises freedom and opportunity, but also threatens primary family and kinship ties, as well as the continuation of Indian traditions.

Chennai and Its Brahman Suburbs Madras was founded in 1639 as a colonial trading post and, during the next hundred years, it expanded by incorporating nearby villages, as well as the old Hindu temple town of Mylapore to the south. By the mid-eighteenth century, Madras was the most important British city in India, although this position was subsequently ceded to the two other colonial port cities, Bombay and Calcutta. During the colonial period, Madras steadily grew, but in the early twentieth century, according to Dupuis, it still resembled a “big village,” although it had become a “heterogeneous conurbation” by the mid-

Urban Ways of Life / 155

dle of the century—even if it was still not surprising to see “cows pastured under coconut palms” within its boundaries.3 Until relatively recently, the city’s image as a rather old-fashioned, slow, bucolic place lingered on, as the title of a recent anthology, The Unhurried City, suggests.4 Before Andhra Pradesh was formed in 1953, many Telugu speakers, as well as Tamils, regarded Madras as their capital, but since then Telugu numbers have fallen and the city has become predominantly Tamil, notwithstanding the many non-Tamil minorities living there. Chennai’s population in 2011 was 4.7 million, with 8.7 million in the greater metropolitan area.5 In the 1990s and early 2000s, the city’s face visibly changed, as new hotels, office blocks, and the huge, plate-glass com­ plexes erected by the IT companies began to spring up rapidly. Supermarkets, shopping malls with stores full of classy clothes or electronic goods, modern banks and ATMs, coffee bars, pizza parlors, Internet centers, travel agents, car dealers, and numerous other outlets for middle-class consumption opened in Chennai as well. Hence consumer choices and up-to-date facilities that used to be scarce in Chennai compared with Bangalore or Mumbai—India’s most modern and “Westernized” cities—have become widely available. All these developments occurred, middle-class people in Chennai told us, as a result of the demand from highly paid IT professionals and nonresident Indians (NRIs) returning to Chennai to work in the IT sector. Truly, the IT industry has become the emblematic symbol of Chennai’s transformation into a genuinely modern city. In fact, though, all the developments are a cumulative outcome of economic liberalization, foreign investment in and around the city (for example, by Ford, Hyundai, and other multinational companies), rising middle-class prosperity, and the state government’s promotion of Chennai as a thriving metropolis in order to attract investors. But in spite of its new attributes, middle-class people in Chennai like to say that the city has not entirely discarded its past, so that now it has the “right balance” between tradition and modernity. Let us now turn to the Tamil Brahmans in Chennai. One hundred years ago, Brahmans were most heavily concentrated in Georgetown, the old colonial city’s center, and nearby Triplicane (Tiruvallikeni) to its south, although significant numbers lived in Mylapore, which was the most desirable residential area for wealthy, elite Brahmans. In these and other old parts of the city, often close to temples, there were exclusively Brahman agraha¯rams like those in villages.6 Nungambakkam, for example, where Satyamurti from Tindivanam lived, was a Brahman village with an agraha¯ram that was incorporated into the expanding city, although its rural origins were still visible until the 1980s.7 In Georgetown, a large agraha¯ram surrounded the

156 / Chapter Five

eighteenth-century Kachaleeswara temple, dedicated to Shiva as Kaccha¯lı¯s´­ vara.8 By the late 1980s, however, when Fuller’s former research assistant, a Brahman from Madurai, lived in this agraha¯ram with his uncle, several buildings were used as small workshops and hardly any Brahman residents remained; the last ones had gone within a few years. In 2005, we visited a domestic priest and his young family living in a small house in an agraha¯ram adjacent to one of Mylapore’s many temples; a few other houses, some fairly decrepit, were still occupied by Brahmans, but the rest were not and most residents were non-Brahmans. To the best of our knowledge, urban agraha¯rams in Chennai and other cities in Tamilnadu, such as Madurai, have all ceased to be exclusively Brahman like their rural counterparts. When Brahmans first left their villages for Madras, however, few of them moved into agraha¯rams. Instead, they lived in houses on ordinary streets that were not restricted to Brahmans. Wealthy lawyers like Bhashyam Aiyangar built huge mansions in Mylapore, especially on Luz Church Road, at the turn of the last century, but the vast majority of Brahmans in the city lived according to their means in smaller houses or rented rooms within shared accommodation. During the course of the twentieth century, the geographical distribution of Brahmans altered as well, so that not many still live in Georgetown or the north of the city, and numbers have also declined in Triplicane. The heaviest concentration of Brahmans today is in and around Mylapore (still a desirable area, although many poor people also live there), Alwarpet and Teynampet, T. Nagar (Thyagaraya Nagar), which is a large planned suburb first developed in the 1920s, and West Mambalam. These areas form a contiguous belt of inner southern suburbs, all predominantly middle class, to the north of the Adyar River. Brahmans probably make up between one-half and two-thirds of the residents in this belt (see appendix), which is regarded by everyone as the city’s main Brahman area, together with the prosperous suburbs of Adyar and Besant Nagar just south of the river. Around half of all Chennai’s Brahmans live in this southern area, but fairly large numbers are dispersed across the city’s western suburbs and further out to the south as well, where much of the property is cheaper. On the southwestern fringes of the city, Brahmans are also concentrated in the former villages of Nanganallur, Chromepet, and Madipakkam. In Mylapore, T. Nagar, and Chennai’s other older suburbs, there used to be many “garden houses” built in the late nineteenth or early twentieth centuries that stood in their own grounds, but almost all of them have been demolished and replaced by blocks of flats a few stories high, interspersed with shops and offices. The vast majority of residents now live in these flats, which vary considerably in size and standard of accommoda-

Urban Ways of Life / 157

tion; only very wealthy people live in the handful of detached houses that remain. Because more Tamil Brahmans are concentrated in Chennai than anywhere else, it is the one city that many of them can and do regard as their own. Hence, to adapt an argument of Ulf Hannerz’s about “lifestyle” minorities in large cities, even within the socially heterogeneous urban fabric of Chennai, the small Tamil Brahman minority—especially in the southern suburbs—can acquire a “critical mass” as a cohesive, relatively homoge­ neous, subcultural group.9 One significant product of this critical mass is the extensive network of primary and quasi-primary social relationships already mentioned. With the possible exception of some fairly small suburbs in cities such as Madurai or Tiruchirappalli, no residential areas exist in any other urban centers in India, let alone overseas, in which Tamil Brahmans constitute as much as half the population. Almost everywhere, in fact, except in unusual agraha¯rams such as Sripuram and Tippirajapuram, Tamil Brahmans today live alongside much larger numbers of non-Brahmans or people from other regional groups. In southern Chennai, however, Tamil Brahmans—together with smaller numbers of other south Indian Brahmans—can mainly live among their own kind. The Brahman majority in Chennai’s southern suburbs is partly a function of how the housing market is organized, because, as Christine Auclair shows, property developers collaborate with landowners, who are often Brahmans, and then seek out Brahmans as purchasers or tenants.10 In many blocks, such as the one in Mylapore she describes, Brahmans are as many as 80 percent of the residents and in such buildings, “vegetarianism . . . is an implicit condition” for occupancy, and “caste is certainly an important criterion.”11 Actually, vegetarianism is quite often an explicit condition, in the sense that landlords or existing residents may ask prospective purchasers or tenants about their dietary regime. Not all vegetarians in Tamilnadu are Brahmans and open discrimination on caste grounds can provoke serious disputes. Nonetheless, everyone knows that questions about vegetarianism are a coded way of saying that a block of flats is wholly or mainly occupied by Brahmans, who will expect others to respect them and their customs, particularly by not cooking meat or fish. A lot of non-Brahmans find Brahman scrupulousness intolerable. One non-Brahman of our acquaintance, who lived in a block where almost everyone else was Brahman, was delighted to discover fish being sold in sealed plastic packets in a new supermarket nearby, so that at last he could indulge his passion for fish without creating the smells that would have led to quarrels with his neighbors; such petty deception may be quite common. Probably, too, because Brahmans

158 / Chapter Five

now prefer to buy rather than rent, landlords may be finding it harder to get them as tenants. Hence Brahmans cannot always ensure their monopoly in residential buildings. Even so, in many parts of southern Chennai, Brahmans can expect to have a lot of neighbors who are also Brahmans. Suburban blocks of flats are never true agraha¯rams, of course, but they come as close to them as any residential complex in a modern city could. The majority of households in urban areas are based on nuclear families, although some include additional relatives, often the household head’s elderly or widowed parents or parents-in-law, who have sometimes had to leave their rural homes because they could no longer manage without assistance and ready access to medical care. In urban flats and houses, unlike those in agraha¯rams or villages in general, residents tend to keep their front doors closed. Groups of children frequently play outside, but most adults do not wander in and out of each other’s homes unannounced, sit and gossip outside the front door, or stroll on the streets to talk to neighbors. Urban life is more private than rural life, as well as more family-centered, and the contrast is often commented on by people familiar with both settings, who say, for instance, that neighbors don’t know each other in cities as well as in villages. Such comments are mostly accurate, although Indians usually seem better acquainted with their neighbors than most middle-class residents of  Western cities. Circles of friends and acquaintances normally include people from different communities. On the other hand, from childhood into old age, the friendship networks of middle-class Brahmans in Chennai typically contain a lot of other middle-class Brahmans, who may meet together to visit a temple, go shopping, drink coffee, discuss their children’s education, or just chat. For Brahmans in south Chennai, a high proportion of the people they know, as well as other family members and relatives, are likely to live in the same or nearby neighborhoods, sometimes in the same building or on the same street, so that even when meetings are not prearranged by telephone, as so many now are, they often just happen by chance when out on an errand. Today, most Brahmans usually dress like everyone else, so that they are not visibly distinct. But caste can often be identified by speech because Brahman and non-Brahman Tamil dialects vary considerably, rather like social class can be identified by accent in England. Audibly if not visibly, therefore, there is a palpable Brahman presence in many localities, such as the roads in T. Nagar that became familiar to us during research. In Chennai and Tamilnadu as a whole, as Brahmans are well aware, they are a small, rather unpopular minority, and complaints about insults or teasing on the streets are not unusual. But across the city’s southern suburbs, they have

Urban Ways of Life / 159

safety in numbers and less to fear, and they frequent a predominantly Brahman public space of a size and type existing nowhere else.

Social Control in Chennai and Villages Living among one’s own kind brings solidarity and familiarity, but it comes with social control and surveillance as well. Social control is an issue of special concern to women, which we discussed with many Tamil Brahmans, particularly Eighteen-Village Vattimas, although their social networks are unusually close-knit, mainly because they belong to a small, overwhelmingly endogamous subcaste. There is an Eighteen-Village Vattima subcaste association with active memberships in Chennai and Bangalore. Its main objectives are to help needy Vattimas (for example, with educational expenses), to keep a register of members’ names and addresses, and to hold periodic general meetings to discuss matters of common concern. A few dozen people attend the meetings, although some are just there to catch up with friends, and many Vattimas take no interest in the association at all. Even so, Vattimas are more collectively minded than the majority of urban Tamil Brahmans, who generally do not belong to caste or subcaste associations. In various ways, therefore, Vattimas are rather atypical, so that we can generalize only cautiously from them to other Tamil Brahmans. In Chennai, Vattimas were commonly talking about other Vattimas who live nearby and with whom face-to-face contact—at home or in the street, at family gatherings or religious festivals—is quite frequent, so that if there is anything interesting to report, it will not be long before lots of people know about it. That is not true to the same extent in other urban localities, except for small towns like Kumbakonam or Mayiladuthurai, which were the destinations of many early urban migrants and still contain sizable populations of Vattimas and other Tamil Brahmans, whose quasi-primary networks are often extensive and close-knit. Kumbakonam, which is close to several Vattima villages, was described by one Vattima resident as “twoin-one,” a place with the advantages of both a town and a village, although he is an unusual townsman because he is a farmer, who regularly visits his village. Other residents of Kumbakonam, which is a famous pilgrimage center with numerous temples, also emphasized that it has urban facilities, such as good education and “many things to do,” as one woman learning Carnatic music commented while comparing the town with less interesting Tippirajapuram. But Kumbakonam is also generally regarded as a quieter and more traditional place than Chennai, so that it is in a sense halfway between city and village.

160 / Chapter Five

Almost everyone agrees that there is far more gossip in villages—by which they mean agraha¯rams—than in towns and cities, and that pressure to conform is generally stronger in villages. As we saw in chapter 1, agraha¯rams used to be exclusively occupied by Brahmans and women rarely left their confines. Neither is true today, but Brahman women living in agraha¯rams, unless they work or study in a nearby town, still spend almost all their time within them. Inevitably, therefore, they and their neighbors know a lot about each other. Men gossip as well as women, of course, but the pressure to conform falls much more heavily on women and those who live in urban areas are particularly sensitive about it. Thus even though some Vattima men say that they want to go back to their ancestral villages, their wives rarely do. Savita was born in 1970, brought up in Chennai, and now lives in T. Nagar; her husband is a company manager and they have one daughter who is a student. Savita’s husband and his brother, who would both like to retire to their village, still own some land there, which is cultivated by their father. Talking at home in Chennai, Savita’s husband said that life in a village is more peaceful than in the city, people know their neighbors better and give each other support, which is important in old age, and villages are no longer backwaters, because they now have good transport facilities and even Internet connections. Savita countered that rural facilities remain inferior and the electricity supply is exasperatingly unreliable. She also insisted that there is plenty of stress in villages, as well as constant jealousy and gossip, and there is very little to do, except visit the local temples, so that villages are mostly boring. Urbanites, indeed, perennially complain about being bored when they go to villages, although it is worth noting that villagers themselves complain much less. Older people who have spent their lives in villages do not normally find them boring, and they are actually more likely to get bored when they visit their urban children and grandchildren, and have to stay in small flats where they feel confined, unable to converse freely with their neighbors as they can in the agraha¯rams.12 In addition to boredom, Savita mentioned the more stringent observation of menstrual pollution in villages compared with cities, which she and almost all urban Vattima women find irksome, so that they try to avoid any rural visits during their periods. On another occasion, we talked to Savita and her father and brother in their native village. Savita’s father was happy that he returned from Chennai to his village, where he owns land, but her mother was not, so that her parents still spend quite a lot of time in the city; her brother, who used to work in the United States, now lives in Chennai and also thinks he might retire to the village. But Savita insisted that too

Urban Ways of Life / 161

many old customs still prevail in villages, including the notion that women are inferior; she did not mention menstrual pollution then, but gave as an example the fact that women must wait for their food until men have finished, whereas in Chennai everyone eats together. To sum up, Savita flatly declared, only men want to return to villages, whereas women never do. Savita exaggerated, but only slightly. To be sure, a few women are happy to return to rural life. For example, we met an elderly Tamil Brahman couple who had moved into the husband’s ancestral agraha¯ram in a village near Thanjavur, after he retired from his job in Baroda, Gujarat, where they had lived for thirty years. He was the one who wondered whether they had made the right decision, complaining that it was boring in villages, whereas his wife was more content and said that settling down in the village was quite easy, although one had to be careful because there was so much gossiping. Nonetheless, much as Savita said, the majority of women in Chennai and other urban areas, especially those of her own age or younger, certainly do not want to live in villages. Nandini, a Vattima born in 1971, shares Savita’s views. She and her husband, a businessman, live with their daughter in T. Nagar. He would like to return to his village agraha¯ram, partly so that he can look after his elderly parents, who still live there. But Nandini was unhappy about this plan and said that she would stay in Chennai, because their daughter must go to school there. Like Savita, Nandini also complained that there is nothing to do in villages and too many old customs survive; for instance, there are strict rules about when to get up and when to go to bed, and too much fuss about pollution, so that her daughter would be told off if she carelessly touched a tumbler with her mouth when drinking from it. Women, she further insisted, have no independence in villages. Neither Nandini nor Savita specifically mentioned their mothers-in-law, but some wives (or their husbands) do, because for many women village life would mean sharing a house with a mother-in-law or living near her, which could only exacerbate the problem of surveillance. Some Vattimas complain about gossip in Chennai, however much worse it may be in villages, although one woman claimed—unlike almost every­ one else—that gossiping was worse in the city than the villages. More typical is Urmila, who is in her forties and moved to Chennai with her husband when she was twenty. When Urmila was being interviewed in her West Mambalam flat, a Vattima relative and neighbor dropped in to ask about her daughter’s recent engagement. After she had left, Urmila commented: “This is Vattima. See how she comes and inquires. She need not do so, but she does because that is how it is in our community”; Vattimas, she said, are all like one’s “immediate neighbors.” Another Vattima

162 / Chapter Five

woman in her sixties, who has been in Chennai for over forty years and lives with her daughter-in-law in Mylapore, said that they avoided other Vattimas. They wondered why Haripriya wanted to meet them at all and then suggested a neighboring woman as a much better informant, as she is widely known as a “big gossip.” It is, of course, precisely because so many Brahmans in south Chennai live alongside other Brahmans that they are always subject to some scrutiny from neighbors belonging to their own caste, who are linked to them by quasi-primary ties. Surveillance among Vattimas is probably tighter than among most Tamil Brahmans, especially other Aiyars belonging to bigger subcastes whose kinship networks are looser and more dispersed. Certainly, though, there is widespread pressure on all Brahmans to conform to “respectable” norms, notably by avoiding any open consumption of meat, fish, or alcohol. Brahmans also generally assume, with good reason, that other Brahmans, especially relatives and neighbors, are likely to notice and gossip about any unconventional or inappropriate behavior, particularly by women, although in Chennai’s suburbs women do not have to put up with male control of the streets and public space, as they do in some Indian urban neighborhoods.13 Young single women naturally find the pressure to conform most intense and they sometimes complain bitterly that even wearing a T-shirt and jeans—let alone “provocative” behavior such as public display of affection with a boyfriend—can lead to disparaging comments in Chennai, as it probably would not in Bangalore or Mumbai. There is, we must emphasize, significant social control, surveillance, and “moral policing,” particularly of young women, in all castes and communities in Chennai, which is partly a product of the city’s prevailing conservatism. Even so, at least for Tamil Brahmans, social control dependent on networks of primary and quasi-primary ties is one dimension of their critical mass in the city, especially in the southern suburbs, that is very difficult to avoid. A lot of  Tamil Brahmans—not only Vattimas—regard this social control as a real drawback of life in Chennai. But by no means everyone agrees and many others believe that the advantages of living close to other Brahmans count for much more.

Chennai as a “Balanced City” For many middle-class Brahmans, and non-Brahmans as well, Chennai has various desirable features that outweigh its hot climate, chronic water shortages, and other inconveniences. Some people referred to Chennai’s good doctors and hospitals, but the city’s schooling system was most commonly

Urban Ways of Life / 163

mentioned. Many insisted that children can be better educated in Chennai than anywhere else in India or even abroad. Such comments, of course, refer to the city’s best, private, English-medium schools, which uppermiddle-class children attend (see chap. 3). Chennai’s educational assets also include IIT–Madras, the medical colleges, and several leading engineering colleges in and around the city. Another of Chennai’s attractions is its perceived traditionalism, most evident in religion and cultural activity, which coexists with its latter-day modernity. Chennai has numerous churches and mosques, but for the Hindu majority what matters is the vast number of temples, old and new, and large and small, dedicated to all the deities in the pantheon. Almost everyone, therefore, can find a place of worship that appeals to them within a short distance of home. There are, too, many other sites in which, for example, gurus deliver discourses about Hindu scripture or groups of people gather to sing devotional hymns (bhajana). A few people complain about religious insincerity and “show off” in Chennai; in the eyes of the majority, however, the city’s multifarious religious sites and activities (discussed in chap. 6) make a vital contribution to the respect for religious traditions that is widely extolled. When people praise Chennai for its “culture,” they are usually referring not to films or other forms of popular culture, but to “high” culture, especially classical Carnatic music and Bharatanatyam, classical dance, which both have a strongly religious ethos imbued with Sanskritic Hinduism. This is particularly true if the speakers are upper-middle-class Tamil Brahmans, like the company manager resident for many years in Bangalore and Delhi, who claimed that Chennai is India’s “cultural capital.” As we discuss in the next chapter, Brahmans dominate Carnatic music and (to a lesser extent) Bharatanatyam, and their critical mass in the city is a major reason why these elite art forms flourish there. Classical music and dance are appreciated by non-Brahmans as well as Brahmans, but middle-class Brahmans most often single them out. For these Brahmans, Chennai’s prized traditionalism finds expression in an amalgam of “great tradition,” Sanskritic high culture and religiosity, much as Singer saw in Madras in the fifties and sixties.14 The combination of traditionalism and postliberalization modernity ensures that Chennai today has the “right balance” between old and new. Thus middle-class people, particularly Brahmans, who like contemporary Chennai commonly contrast it with either provincial towns that are too backward, or metropolitan cities, especially Bangalore and Mumbai, that are too modern and Westernized. Naturally, not everyone agrees. Some people think Chennai itself is too modern, so that they prefer towns like

164 / Chapter Five

Kumbakonam, or they decry the lack of true culture in Chennai, compared with more traditional Tamil cities such as Madurai. More people, though, particularly those with experience elsewhere in India or overseas, criticize Chennai as still too old-fashioned, narrow-minded, and villagey. Some Mal­ ayalam and Telugu speakers also complain that over time Chennai has become more parochially Tamil and less “cosmopolitan,” a term normally meant to signify a mixture of regional, linguistic, and religious groups. An interesting comment came from one young Tamil Brahman IT professional, who said that he prefers living in Chennai, where he was born and brought up, but thought that this was really because it’s his “comfort zone,” so that he ought to move to more cosmopolitan Bangalore or Mumbai to learn about different kinds of people. Comparisons between Chennai and Bangalore are constantly made, because so many people frequently move between them for family or business reasons, and know both cities well. Quite a lot of Chennai’s residents actually prefer Bangalore, but others actively dislike it, condemning it as a superficial, glitzy, over-priced, “artificial” city typified by its foreign food, bars, and frivolous “party culture.” Not everyone is so vehement, but many people in Chennai, especially upper-middle-class Tamil Brahmans, told us that their “balanced” city has now acquired the virtues of Bangalore and Mumbai without their vices.

Bangalore and Mumbai Bombay, founded as a colonial port city, eventually grew into India’s commercial and business capital. Throughout the modern era, this huge metropolis has drawn in millions of migrants from all over the country in search of a better living and by 2011 greater Mumbai’s population exceeded 18 million. After Independence, in Sunil Khilnani’s words, the city “became permanently lodged in the popular imagination as a totem of modern India itself,” partly owing to the impact of the Bollywood films produced there.15 Bangalore, once the capital of Mysore princely state, developed as a major industrial center in the early twentieth century. After Independence, it became the site for several government-owned corporations in defense, telecommunications, and other advanced technology sectors; it is now the premier center of India’s IT industry. In the first decade of this century, Bangalore was the fastest-growing metropolitan city in the country and greater Bangalore’s population reached 8.5 million in 2011, almost equal to Chennai’s.16 Especially for the “new rich” professional and managerial upper-middle class, writes Khilnani, Bangalore has become “the strongest alternative incarnation of Indian modernity” to Mumbai.17

Urban Ways of Life / 165

Tamil Brahmans first migrated to Bombay in the 1920s. They mainly settled in Matunga, a residential neighborhood north of the city’s historic center, which became known as a south Indian area. The majority of Matunga’s south Indian immigrants between the twenties and forties were Brahmans, especially Tamil Brahmans from the Tirunelveli and Thanjavur regions, and Palghat Aiyars from Kerala. They established themselves as a south Indian, neighborhood-based community defined by caste, native place, language, and class—specifically their status as middle class. Cooperative societies dominated by Brahmans were set up, as well as temples and cultural associations, which all played important roles in community life.18 Mumbai’s principal south Indian temples are still in Matunga, but Tamil Brahmans now prefer to live in Chembur, which is a newer and more spacious residential suburb a few miles further out, although many are also scattered throughout the city’s northern commuter belt. Estimating their total number is extremely difficult, but Mumbai (which they still call “Bombay”) is almost certainly the Tamil Brahmans’ third-largest settlement area, after Chennai and Bangalore. During the colonial period, large numbers of  Tamils from many different communities migrated to Bangalore and settled near its British military cantonment; a significant proportion of them were Brahmans, who worked in the cantonment and in the Mysore state’s bureaucracy. Since Independence, Tamils have continued to move to Bangalore and now make up about onefifth of the city’s population. (Bangalore’s new Kannada name Bengalooru is not used very widely and never by Tamils.) Bangalore contains the largest concentration of  Tamil Brahmans outside Tamilnadu and between 100,000 and 150,000 is a reasonable guess for their population. The majority reside in suburbs scattered throughout the city, although Malleswaram, an old suburb north of the city center, used to be favored by Tamil Brahmans, who are still numerous among the middle-class residents of the flats that have replaced its former garden houses. In Malleswaram and some other localities, many blocks of flats and residential streets contain clusters of neighboring Brahman households, but they are rarely as densely concentrated as in southern Chennai. Parenthetically, it is worth mentioning here that Delhi attracted sizable numbers of  Tamil Brahmans even before Independence and, probably until the 1980s, the majority of Tamils and other south Indians in the capital were Brahmans. Almost all had white-collar or professional jobs, many in the central government’s bureaucracy. Karol Bagh in central Delhi was their main settlement area.19 Since the 1990s, the declining appeal of government employment, counterbalanced by the capital’s huge expansion and

166 / Chapter Five

development as a business center, must have significantly altered the Tamil Brahmans’ social profile in Delhi, but we have no data about it and did not interview people in the city during our research. Most Tamil Brahmans, we are sure, would agree with Khilnani that Mumbai and Bangalore are the two most modern, Westernized metropolitan cities in India. In the early twentieth century, however, when Bangalore was smaller and less important than Madras, Bombay’s economic dynamism gave it unrivaled pulling power. Some migrants went to work in offices and others made money in small businesses, but some rose a lot higher. Nilakantha, a Tamil Brahman retired engineer who was born in 1924, was taken there as a baby by his parents, who had been living in Tiruchirappalli. Nilakantha’s father, a graduate, had a job as a librarian, but he decided to try to improve his lot in Bombay, where he first got a job as a linesman in the city’s telephone company. He worked his way up, had postings in Ahmadabad, Gujarat, and in Karachi, and by 1946 he was the company’s general manager. For his father, said Nilakantha, Bombay was a “land of opportunities” just like America today. Stories of people who made it in Mumbai are common enough, but, as Nilakantha implied, the city no longer has the magnetism that it once had for Tamil Brahmans. Even so, despite the chronic housing and transport problems that everyone complains about, Mumbai is still seen as a good place for work by the Tamil Brahmans, mostly Vattimas, who talked to us there. Mahadevan, a retired company manager, who arrived in 1962, said that there was always more opportunity in Mumbai, his job was challenging, and there was not much “caste feeling,” which negatively affects working life. Shanmugam, a retired engineer who ended his career as a dock manager, moved to Mumbai in 1965 and said that to work there was always a great opportunity; the city has also been the only place in India where someone can get a job with or without education. Younger men, such as Kailasa, who works in a bank, expressed similar views: career prospects are good, levels of professionalism are high, unlike in Chennai, and one can get things done in Mumbai, as one can’t elsewhere. In Bangalore, the men we talked to had less to say about working there, except for a few IT professionals who commented on the advantages of being located in the IT sector’s premier center. Although men generally agree with them, Vattima women are particularly forthright about the good life in Bangalore and Mumbai, which they see as safe cities for women that also offer more freedom and indepen­ dence than Chennai or anywhere elsewhere in Tamilnadu. Vasudha, Mahadevan’s wife, was particularly eloquent. When they married in the 1960s,

Urban Ways of Life / 167

Vasudha was only fourteen, ten years younger than her husband, and she left Tamilnadu to join him in Mumbai two years later. Vasudha (who knew Hindi) used to meet Mahadevan after work and together they would go to the cinema or walk on the famous Chowpatty beach. She recalled how much she enjoyed her young life in the city with her new husband in their conjugal home and how lucky she was. In Tamilnadu, whether in a village or a city, no sixteen-year-old bride—even if she and her husband were living by themselves—could have enjoyed the personal freedom that Vasudha had in Bombay in the sixties. One of Vasudha’s cousins is Karpagam (mentioned in chap. 4), who married at twelve and joined her husband when she was seventeen in 1954. The couple first lived in Chennai and Ernakulam (Kochi’s twin city) in Kerala; they then moved to Bombay in 1962 and finally to Bangalore in 1969. Vasudha and Karpagam both agreed that they were very fortunate, compared with their other female cousins, to have spent most of their adult lives away from Tamilnadu, especially outside the villages. Karpagam compared her attitude to Bangalore with the younger generation’s toward America, which she has visited many times. Bangalore, like America, she commented, offers independence to individuals, especially in comparison with the villages. “I don’t miss the village,” said Karpagam; “Here I’m on my own. I can sleep late, get up late. There, in the village, our next-door neighbor will say: ‘Ah nice . . . sleeping till seven a.m!’” Nandini in Chennai made a similar point and, trivial though it sounds, the right to get up late epitomizes a personal freedom that women value in the cities. Sita is a Vattima, who was born in 1955 and brought up in Calcutta; she studied for a BSc in mathematics, an unusual degree for a woman in her day, at a college in Tiruchirappalli. After graduating in 1976, Sita married and moved to Bangalore, where she found a job as a bank officer. She was encouraged to do so by her husband and his parents, and says that other Brahman women did not criticize her, but that was probably because Bangalore was always much less conservative about working women than Chennai or anywhere else in Tamilnadu, as Sita emphasized. Indeed, compared with Chennai, Bangalore is consistently seen as less conservative and more “cosmopolitan”—an epithet used by several informants to describe its mixed populace—and Sita is very happy that she could settle in Bangalore, although she may retire with her husband to “nice and quiet” Kumbakonam. Anandhi is an accountant in her late twenties, who was born in Mumbai, grew up in Chennai, and moved to Bangalore after marrying an electronics engineer. In Chennai, she had a job in an office dominated by Tamil Brahmans, where she felt secure and did well, but she heard too

168 / Chapter Five

much Brahman gossip. In Bangalore, Anandhi works for a multinational IT company, where she is a “drop in the ocean” in a “multicultural” firm, and enjoys a more challenging “expanded world.” Moreover, like her husband, Anandhi is glad that she knows only one other Vattima family in her part of Bangalore, a city where “nobody knows what you do,” whereas people are always watching or passing comment in Chennai. Almost exactly the same remark was made by some Vattimas in Mumbai. Thus Kailasa, while acknowledging that there are “advantages and disad­ vantages” to having relatives around, described Chennai as really “a Vattima village,” and his wife Shyamala said that people ask too many questions there. Their own housing estate in Chembur—where most neighbors are Tamils or Sindhis—is “like a village,” but unlike a real village, neighbors “don’t poke their noses” into other people’s affairs. In Mumbai, Shyamala said, echoed by her husband and a cousin who was visiting, “we can do whatever we want.” In both Bangalore and Mumbai, some Vattimas said that their friends were Tamil Brahmans, whereas others mentioned people from different communities, both Tamils and non-Tamils. In both cities, inviting all one’s neighbors to festival celebrations appears to be a common custom. Without detailed ethnographic material about social networks, we cannot judge the accuracy of people’s statements about freedom from surveillance in Bangalore and Mumbai, but they are consistently expressed, especially by women, and they are certainly heartfelt. Despite the advantages of life in Bangalore and Mumbai, several people said they hoped to return to Tamilnadu after retiring from work, although Bangalore is itself a popular choice for retirement, especially for Tamil Brahmans who want to return to the south after working in northern India and prefer it to Chennai. One common motivation for retiring to Tamilnadu is financial, because Bangalore and Mumbai are both very expensive cities. To the best of our knowledge, however, plans to go back to Tamilnadu—to Chennai, a town like Kumbakonam, or an ancestral village—are confined to Vattimas who have regularly visited family members there, so that they have kept up links with their home state. Many adult Vattimas in Bangalore and Mumbai fall into this category. But most of their children do not and they have often been to Tamilnadu only on holiday. These children also speak Hindi or English better than Tamil, especially in Mumbai (as well as Delhi). For them, like many other middle-class Tamil Brahmans dispersed across India whose ancestors emigrated a long time ago, Tamilnadu is no longer their homeland in any meaningful sense, even though their caste identity is still salient.

Urban Ways of Life / 169

To sum up: unlike Tamil Brahmans in Chennai who criticize Bangalore and Mumbai, almost all those who actually live in the two cities prefer them, and emphasize the personal opportunities and liberties available. To some extent, social control and surveillance are lighter in Bangalore and Mumbai than Chennai because middle-class morality is more “modern” and relaxed, especially in relation to women. It is equally if not more important, though, that residential areas in Bangalore and Mumbai are more socially heterogeneous—or “cosmopolitan”—so that nobody is surrounded by neighbors from the same caste, and controlled by closely knit networks of primary and quasi-primary social relationships.

The American Suburbs The American dream—the idea that the United States is a land of boundless freedom, prosperity, and opportunities—has been a powerful magnet for generations of immigrants. The majority of South Asians who went to the US after the 1965 Immigration Reform Act were neither poor nor persecuted, however; instead, they were doctors, engineers, scientists, and other highly skilled workers who mostly came from urban, upper-middle-class families. Yet America offered these migrants excellent career prospects and material prosperity, as well as high standards of education for themselves and their children, which were on a scale unattainable at home. From the 1980s onward, South Asians (mostly Indians), together with East Asians (mostly Chinese and Koreans), began to be seen as “model minorities”— ethnic groups whose above-average incomes and overall success testify both to their own commitment to education and hard work, and to the reality of the American ideology of limitless, merit-based, individual achievement.20 Many Tamil Brahmans, talking to us in both India and the US, freely acknowledged how much they have been influenced by images of America as a promised land. Approximately half of all Tamil Brahman overseas emigrants have probably gone to the US, one-quarter to the Middle East, and the rest to Europe, Southeast Asia, Australia, and elsewhere. In all foreign countries, men outnumber women, mainly because men are more likely to go when young and single, as students or for jobs, and many women emigrate only after marriage. In the Middle East, the sex ratio is very uneven, because although it offers good jobs and high salaries, the region is seen as uncongenial for women and even married men may leave their wives and families behind in India. European countries or Australia, for example, are attractive

170 / Chapter Five

destinations for some people. But for the great majority of  Tamil Brahmans, like countless other Indians, America has a unique appeal that eclipses everywhere else. The Tamil Brahmans, mostly Vattimas, we met in America live in the outer suburbs of Chicago, Dallas, Milwaukee, Philadelphia, Pittsburgh, and Syracuse, New York, as well as Edison, New Jersey, a town within New York’s commuter belt that has a sizable South Asian population. There are, of course, many Tamil Brahmans living elsewhere as well, for example on the West Coast, whom we could not visit. We interviewed the members of twelve families, focusing particularly on how people compared life and work in America with India, but time did not permit detailed discussion of important questions about, for example, ethnic minority identity, citizenship, or attitudes toward race. In general, though, our admittedly limited evidence is consistent with that of other scholars, such as Sunil Bhatia, who carried out research among Indians in suburban Connecticut.21 The outer suburbs or “exurbia,” often vilified as “sprawl,” are very different as social and physical spaces from the central areas or even the inner suburbs of most American cities, including the “Little Indias” where ethnic shops and restaurants are concentrated close to areas settled by poorer South Asian immigrants, such as Jackson Heights, New York, or Devon Avenue in north Chicago.22 Obviously, too, the American outer suburbs are radically different from Indian urban neighborhoods. In places like T. Nagar in Chennai or Chembur in Mumbai, which are usually called “suburbs,” people live in houses or flats built close to shops, offices, schools, places of worship, and other public buildings, all connected by noisy, traffic-filled streets and bustling footpaths. All these characteristics make Indian city neighborhoods very unlike the low-density, pedestrian-free, clean, quiet, residential outer suburbs where the majority of middle-class Tamil Brahmans and other South Asians live in the United States. Most Tamil Brahmans we visited live on housing estates in huge modern houses of the kind nicknamed “McMansions” by their critics. Life in the suburbs is overwhelmingly dependent on cars and hardly anyone walks anywhere. There is no public transport, so that all commuting to work has to be done by car. Shops and restaurants are located in malls alongside major highways, and other facilities—as well as friends’ houses—are usually far away. A school bus service normally exists, but otherwise children, old people, or others who cannot drive always have to be given a ride. A substantial proportion of South Asian middle-class immigrants to the United States, like almost all our Tamil Brahman informants, have spent little or no time in inner-city areas, except as students, because they settled

Urban Ways of Life / 171

in the suburbs as soon as possible. Especially after the Second World War, the affordable suburban family home became central to American middleclass ideals; in the suburbs, people “lived the American dream” and the “dominant American way of life came to be rooted in and crafted around the suburbs.”23 Immigrants from South Asia (as well as East Asia) rapidly adopted the same ideal, so that, as Madhulika Khandelwal, who studied Indians in New York City, writes: “Indian immigrants perceived suburban life to be a significant part of their American dream, achievable by, as well as emblematic of, their economic success.”24 Despite the presence of some ethnic minorities, the outer suburbs are predominantly white, and African Americans in particular have often been excluded from them. Our Brahman informants generally reported that their own suburbs, though mainly white, contained South and East Asian minorities, but no or hardly any African Americans. These reports were probably correct, but they were not just neutral statements of fact, because for South Asians, like other Americans, the suburban ideal has a racial dimension, whereby the white suburbs are opposed to black inner-city “ghettoes” characterized as violent.25 For South Asians, this perspective may be reinforced by adopting a white-majority viewpoint, itself partly motivated by fear that whites may identify brown people as black.26 But South Asians are also victims of racism themselves, mainly but not entirely white racism; sometimes they acknowledge, for example, that racial “glass ceilings” at work are fairly common. Yet there is also a countervailing tendency, especially in the professional and managerial, upper-middle class, to deny that white racism really affects South Asians. Thus, people claim, everyone with talent who works hard can succeed in the American “color-blind meritocracy,” as the South Asians’ “model minority” status proves, even though this success often demands an accommodation to American mores, while remaining authentically Indian, that can be full of stressful contradictions, as Bhatia vividly shows.27

Life and Work in America In our conversations with Tamil Brahmans, the advantages and disadvantages of America compared with India became the main topic, and the focus was on work and the family. Shankara, a Vattima in his early thirties in 2006, is an engineering graduate who first obtained a job in the US in 2000 and works as an IT professional in a large American company in the Midwest; Gauri, his wife, is in her twenties and does not have a job. Like many women, Gauri first went to the US after marriage, not to study or

172 / Chapter Five

work. The couple live in a typical outer suburb and had no children when we met them. Shankara summarized a common view very simply when he said that for him the “work environment” is more important, whereas for Gauri it is the “social environment,” meaning primarily the family and kinship network. Shankara said that he went to America mainly to make money, but later decided that the more professional organization of work is equally if not more important. In the US, he said, people complete their work efficiently and then leave the office at the end of the day, whereas in India they are forced to work longer, more unsociable hours. Much the same opinion was expressed by Partha, who lives with his wife, Prithi, and their young son in one of Chicago’s outer suburbs. Partha is an engineering graduate who moved to the US in 1994 and works as a business systems consultant in a bank; he emphasized the more professional working environment in America, noting that time—including leisure and vacation time—is more valued than in India, and that managers do not constantly check up on their subordinates, who have more individual responsibility. In general, men, as well as women with careers, agreed that job opportunities and material rewards are much better in the US, although some may have changed their mind after 2008 when the recession deepened in the West while the Indian economy continued to grow. People also agreed that merit and hard work are more consistently rewarded in America, and gave the impression that racism did not hinder them significantly; in this context, they often commented adversely on Indian caste-based reservation policies. Mythily, however, did refer to a glass ceiling defined by both gender and race. Mythily, who lives in suburban New Jersey, is a university administrator who arrived in the US in 1990. She is an unusually self-reliant woman who largely depends on her own salary because her husband, a scientist, works elsewhere; their two sons live with her. Mythily is proud of her career progress, which she partly attributes to the opportunities for women in America, but she also believes that she will not be promoted much higher in the university because, even though people will not admit it, she will be blocked by the glass ceiling. Arjun and his wife Leela, who were mentioned in chapter 4, live in Chicago’s outer suburbs. Arjun commented that Indians are foreigners without equal rights and discrimination in the workplace can occur, although he also said he had experienced more racism in Europe than America. Discussing racial problems at work, Arjun and Leela specifically said that they tried to “fit in”: for example, they both try to speak with a more American accent at work and Arjun, when talking

Urban Ways of Life / 173

to his male colleagues, takes an interest in baseball (even though he much prefers cricket). Concerning food and drink, some of the Tamil Brahmans we met talked about the topic very pragmatically and made it clear that they were no longer strictly vegetarian. Others were more sensitive and just dropped hints about their relaxed dietary practices in America, while one or two people asked us to say nothing to relatives in India, lest they got upset. In fact, many Tamil Brahman men and even more women do remain vegetarian and teetotal in the US, especially in their own homes. Nevertheless, it is normal to make some compromises—for example, by eating fish or white meat and sometimes drinking alcohol with American colleagues in restaurants—and this is particularly common among men with professional and managerial jobs, whose career progress may be impeded if they appear inflexible at business lunches or office social events. Gauri, talking about the “social environment,” said that she wanted to marry someone who would take her to the United States; she admitted that she had thought America would be like heaven, although she realized it was not soon after arriving. Gauri’s rosy view of America is shared by other young Tamil Brahman women in India, so that single men in the US are valued for “America varan” (“American alliance”) unions.28 Indeed, Shankara’s father in India commented about marriage proposals that “once your son is in the US, people start pestering you.” But despite coming down to earth quickly, Gauri has still been happy in America. She said she missed her family and relatives, but she was emphatic that one primary advantage of living in America is privacy, lack of social constraints, and freedom from the pressures exerted by relatives and other Vattimas, especially a mother-in-law. These pressures, she and Shankara agreed, would be serious even if they lived by themselves in India, especially if they were in Tamilnadu, where people would interrogate Gauri about what kind of food she cooks for her husband and what kind of clothes she wears at home, or even demand to know why they both like Hollywood films as much as Tamil ones. Talking to Shankara and Gauri, it was clear, too, that personal freedom in America was highly valued by them as a young couple, who could experience newly married life without interference from others. Arjun and Leela similarly identified personal freedom as one of the main advantages of America. Leela emphasized this point forcefully; she said she was much less constrained by her relatives and other Vattimas than she could ever be in India, and she had also gained a self-confidence in the US that she lacked before. It is striking that Gauri, Leela, and other women in

174 / Chapter Five

America talked about personal freedom very like women in Bangalore and Mumbai. A minority of women are less enthusiastic about America. When Prithi first arrived, she hated it; isolated in her suburban house while Partha was at work, she was lonely, bored, and homesick. After two years, her son was born and life started to improve, and she later started work as an accoun­ tant. Nonetheless, Prithi struggled to tell us what she liked about America, apart from personal independence, and she did not agree that social pressure from kin is a serious problem in India. Prithi’s initial misery in America was not unusual and we know of similar cases, but she may miss her family more than most women and she certainly cares less about American freedoms. In America, parties at religious festivals and holidays, and weekend buffet dinners, are standard ways of socializing among upper-middle-class Indians living in the outer suburbs. They feature prominently and amusingly in The Namesake, Jhumpa Lahiri’s novel about Bengalis in New England, which brilliantly portrays the pleasures, pains, and perplexities of suburban life in ways that often ring true for Tamil Brahmans as well.29 Tamil Brahmans, like other Indians, frequently meet each other as well when visiting temples or participating in cultural activities, and when taking their children to mother-tongue lessons or classes for music, dance, Kumon mathematics tuition, and countless other pursuits. Almost all these activities require extensive driving and illustrate how far “community” in the suburbs has become detached from locality. Tamil Brahmans do not seem to go to as many dinners as Lahiri’s Bengalis or the Indians in Connecticut studied by Bhatia.30 All the same, some of them engage in plenty of social and cultural activity, especially in places with large Indian populations, such as Edison, New Jersey. Arvind and Nirmala, who migrated to the US in 1994, live in Edison. Arvind, an engineer, works in New York City and commutes each day in a shared vehicle with six Indian friends, most of them south Indian Brahmans; Nirmala is a teacher in a local school. The couple’s two daughters attend several extracurricular classes—for Bharatanatyam and S´lokas (“shlokas,” or scriptural verses praising Hindu deities), as well as music and swimming—and these, together with friends’ birthday parties, usually fill up their weekends. On Sunday evenings, the family spends four hours with a devotional singing group; they often go a nearby temple as well. Arvind and Nirmala and their daughters also have plenty of Indian friends in their neighborhood. Within the US, in Arvind’s view, Edison is “the best place to settle down because of Indian culture.”

Urban Ways of Life / 175

But that is not everyone’s opinion; moreover, some Tamil Brahmans complained about too many social events in their area. One man living in Pittsburgh’s outer suburbs works in Edison, where he stays during the working week, but both he and his wife, like another couple we met in Pittsburgh, emphasized that they would not live in Edison precisely because it has too many Indians. The man who works in Edison said that he and his wife knew about twenty-five Indian families in the Pittsburgh area and socialized with them, especially at festivals; in his opinion, however, the parties have become too big and frequent as the local Indian population has grown. Two women friends living in the outer suburbs of Dallas similarly said that too many Indians now live in the area, so that they had cut down on weekend socializing. Thus people disagree about the ideal concentration of  Tamils, south Indian Brahmans, or other Indians, but most are wary not only about demands on their time, but also about too much scrutiny from others at social events or meeting places such as temples. Plainly, dispersed residence in the outer suburbs strongly shapes the patterns of social interaction among Tamil Brahmans in America, so that they differ considerably from those in high-density, urban neighborhoods in India, whether in Chennai, Bangalore, or Mumbai. Obviously, too, despite some concern about meeting too many other Indians, the amount of social control within American neighborhoods is normally minimal. After all, since the families living nearby are usually white Americans, not South Asians, nobody is likely to comment on a teenage daughter’s flimsy clothing, for example, even if the neighbors sometimes complain that the grass needs cutting. Quasi-primary neighborhood relationships are therefore thin, which is consistent with the fact that the postwar American suburbs described by Gans were more compact and ethnically homogeneous than the newer outer suburbs where Tamil Brahmans live today.31 Finally, because it is so important, we reiterate the point that our informants talked about freedom from oppressive social control and surveillance—and its value for women in particular—in very similar ways in Bangalore, Mumbai, and the American suburbs, which are all consistently contrasted with Chennai or other localities in Tamilnadu, but occasionally with Edison, New Jersey.

Children and Grandparents When children are born in America, their mothers commonly lose much of their personal freedom and their housebound confinement may be worsened by reluctance to use child care centers or paid child carers. Often,

176 / Chapter Five

Indian couples bring their parents to the United States for up to six months at a time (the maximum permitted by visa regulations), so that they can be present at childbirth and help to look after young children. In some families this system works well, at least for a few years, but in others it does not. One difficulty is that although some elderly parents are enthusiastic about visiting America, most are not, mainly because they have to spend so much time within the four walls of their children’s homes without enough company and enough to do. As Shankara’s father said, what’s wrong with America is that there is no “timepass” recreation to ward off boredom. El­ derly men, wherever they live in India, often describe America as “house arrest” or a “luxury jail,” because they loathe isolation in the suburbs, which they cannot escape, because they cannot drive. They find that there is often nowhere to walk, nobody to talk to apart from their own children and grandchildren (who may be at work or school during the week), and very little to do except watch Indian films on television. Women, who can occupy themselves with small children and cooking, usually find the situation more tolerable, but they also become bored eventually, especially when children grow up and need less attention. Weekend outings to temples and socializing with other Indians, as well as holiday visits to tourist sites, usually find favor. But from Monday to Friday, boredom is a curse for many el­ derly Tamil Brahmans visiting their families in the American suburbs. Some continue to make regular visits anyway, but others, especially men, stop going after a while, particularly when they can no longer tolerate the long flights or the winter cold in the northern states. If they do not travel to the US, old people must wait for their children’s and grandchildren’s visits to India, which typically last for two or three rushed weeks every two or three years. Almost everyone maintains regular contact with family members via telephone and webcam calls, and e-mail messages. But despite the periodic visits and frequent communication, people worry that overseas emigration will eventually undermine family and kinship ties. The Tamil Brahmans we met in the US were all first-generation immigrants and most were fairly young, so that even their oldest children were still students, but they all recognized that boys and girls brought up in America normally have weaker links with their grandparents and other relatives than their cousins in India. They also realistically accepted that when their children grew up, they would probably not marry other Tamil Brahmans, so that caste-based kinship networks would become still looser. Almost inevitably, Tamil Brahman children raised in the US will start to identify themselves as Desis or “hyphenated” South Asian Americans. As Shalini Shankar explains, Desi teenage subculture has its own distinctive

Urban Ways of Life / 177

features shaped by class, caste, and other factors, but the “model minority” stereotype consistently places enormous pressure on young South Asians, who manage it in diverse ways.32 Even so, when they become adults, young people largely uphold their parents’ values in relation to the family and marriage, because they are invested in remaining members of their own community.33 Among second-generation South Asians, transnational family ties generally remain important as well.34 Quite possibly, therefore, our Tamil Brahman informants were overly fearful about the risks to family and kinship, and the problems of socializing children into Indian culture and values. Nevertheless, concern about those risks and problems is pervasive, and they are the primary reason why transnational migration is perceived as radically different from migration within India. Keeping in touch with family members in Bangalore or Mumbai is not thought to be difficult and both cities, albeit “Westernized,” are still in India, so that they endanger “traditional” culture far less than distant America does.35 Hence for Tamil Brahman first-generation migrants to the US, as well as for their family members and relatives in India, the potential damage to kinship, together with concern about children’s socialization, is the greatest penalty paid for pursuing the American dream.

Staying in America or Returning to India Talking about the advantages and disadvantages of America naturally leads to debate about whether it is better to stay or go back to India. In this context, the two main issues are caring for elderly parents and bringing up children, although the availability of good jobs in India, especially for men, is also important. Citizenship and immigration status matter too, because most Indians want to secure the legal right to work and live in the US before leaving for India, in case they change their mind and decide to return to the US after all. Return migration is commoner than it was, although men who decided to quit America and go home, like Krishna and Ratnam, the IT professionals described in chapter 3, are still fairly unusual. Return migration is facilitated by the IT industry, however, because overseas projects and the global market for software engineers enable IT professionals to move backward and forward between India and foreign countries relatively easily. Sometimes they do so repeatedly. Rajesh, who has a doctorate from IIT–Kanpur, Uttar Pradesh, is a manager at a major IT company; he married Usha, who has a doctorate from an American university, in 2000. At that time, Rajesh was working in Chennai and Usha was still a student. Rajesh then went to work

178 / Chapter Five

on a project in the US for six months, returned to Chennai, went back to the US for another six months, and again returned to Chennai. In 2002–3, he worked in Canada for six months and when he finished, Usha had completed her PhD and they both returned to Chennai, where Usha got a job as a mathematics researcher and gave birth to a son. In late 2005, Rajesh obtained a transfer within his firm to North Carolina, so the family moved there, but after six months Rajesh was moved to New Jersey. His mother then came to stay for six months and Usha appreciated her help with her child, but she did not find it easy to live with her mother-in-law. In 2007, Usha obtained a job in an IT company in New Jersey and her own parents came to visit, but they actually spent much of their time in North America with Usha’s sister and her baby in Canada. By this time, for various family and work reasons, Rajesh and Usha had decided to return to India in early 2008, but in the end they did not; once again, they postponed a final decision about where to live. Rajesh and Usha have moved back and forth more than most people, but their dilemma about where to settle down is a common one and Tamil Brahmans in America told us that they constantly discussed returning to India. Almost certainly, though, once they have children and settled careers, they will—like most of the Indians in Connecticut studied by Bhatia—”postpone the idea of going back home to India, although never completely.”36 One forthright exception to the usual attitude toward return migration was Anupama, who went with her husband, an IT professional, to the Netherlands in 1985–89 and settled in the US in 1990. When her friend Janaki, who arrived in the US in 1996, commented that she and her husband were debating the matter, Anupama declared that after a few more years they would stop, as most Indians eventually do. Anupama was probably right. In any case, it is certainly true that many more people talk about returning to India—and continually postpone the decision—than actually go back. Caring for elderly parents is a key issue, especially for those with no siblings left in India, and it is one major reason why some people go home. Many Tamil Brahmans say that Indian cities are now full of lonely old people whose sons and daughters send them dollars, but only ever come on short visits, and there is a widespread feeling that overseas migration, unlike moving around India, damages the properly reciprocal relationship between parents and children. In reality, of course, grown-up children in Mumbai, say, cannot easily look after elderly parents in Tamilnadu. But the problem is not regarded as so intractable, whereas sons and daughters overseas—however well-meaning and financially generous they are—can never provide the personal care for aged parents that is part of an Indian’s

Urban Ways of Life / 179

traditional duty. On the other hand, most elderly parents we have met in India pragmatically accept that they cannot expect their children to return home to look after them, whatever their good intentions. A few go into old people’s homes, but most rely for help on relatives, friends, neighbors, or self-help groups. Concerning children, parents often say they would prefer to raise them in India, so that they absorb their own culture and values, and are not corrupted by Western immorality and materialism. This opinion is held most strongly about daughters, whose moral character is more at risk than sons’. Grandparents visiting America can teach young children some of their traditions, but one vital means of imparting Indian—or Brahman—culture and values is to take children to temples and religious festivals, send them for religious instruction, and enroll them in mother-tongue classes. For girls, Carnatic music and Bharatanatyam classes are popular or even imperative. As children grow older, as Bhatia observes, it also becomes “important to keep a sense of home life through the maintenance of Indian traditions,” even though the “meaning of India and Indian culture” gradually changes over time.37 All the same, the birth of children tends to postpone any firm decision about returning to India, which is unlikely to be made at all unless parents act when their children are still young. Realistic parents who eventually accept that they are not returning, said Anupama, learn how to “bring up the kids in America,” which “is more of what to do than what not to do.” In other words, it should not be all don’ts about dating, sex, alcohol, and drugs. As far as we can tell, most Tamil Brahman parents try hard to raise their children as members of a model minority who combine “traditional” values with high levels of educational achievement—although they do disagree about how hard to push their children to succeed at school—and they normally want them to study in America’s top colleges and universities, which is easier to achieve if the family stays in the country. But it is also crucial, as we have implied, that when children get older—especially when they become teenagers in high school—they are likely to identify themselves positively as South Asian Americans and will strongly resist any proposal to return permanently to India. That was undoubtedly recognized by Anupama, whose daughters were twenty-one and seventeen when she spoke, whereas Janaki’s daughters were only seven and four. Thus it is actually very rare for any Tamil Brahman (or other Indian) family with older children to return to live in India, irrespective of any other reasons for going home. Lastly, let us note that attitudes about Indian culture and “Americanization” are not uniform. For example, several Tamil Brahman girls in their

180 / Chapter Five

teens or early twenties were accomplished Bharatanatyam dancers, who had had a public debut (aran·getram) and performed in public. But the mother of one of them freely admitted that her student-aged daughter, with whom she was very close, ate meat and drank alcohol like her father, although she also wanted to go and work in India, at least for a few years, to experience the country for herself. In Edison, Arvind was confident that his fourteenyear-old daughter, who was due to make her public Bharatanatyam debut in Chennai, would always be interested in Indian culture. The girl agreed that her father’s confidence was well founded, but also declared she had always known that she was both Indian and American, had as many white friends as Indian ones, was already thinking about her university education in America, and was sure she wanted to stay in the country. She may of course change when she gets older, but it is very likely that, in common with other girls who have learned to dance, as well as their brothers and sisters, she will eventually become a South Asian American belonging to the Tamil Brahman diaspora settled in the United States.

Conclusion Middle-class Tamil Brahmans, especially in the professional and managerial upper-middle class, almost all live and work in urban areas. Unlike many urban migrants in India, they are also thoroughgoing urbanites, not “urban villagers” who have brought their rural ways with them to the towns and cities. In shaping the Tamil Brahmans’ urban ways of life, it is important that urban and suburban neighborhoods—unlike exclusively Brahman agraha¯rams—are socially heterogeneous, so that contact with people from diverse castes, communities, and ethnic groups, many of them strangers, is the norm. Neighborhoods, however, tend to be homogenous in relation to social class. But towns and cities are not all the same and the Brahmans’ way of life varies considerably from small Tamilnadu towns like Kumbakonam to the city of Chennai, and from Chennai to Bangalore and Mumbai; in the outer suburbs of American cities, it is different again. Brahmans are generally aware of these variations and often have strong views about the quality of life in different urban areas, partly derived from personal experience and partly through reports from family members, relatives, friends, and colleagues. In this chapter, one central theme has been the contrast between Chennai—the city with the largest concentration of Tamil Brahmans—which is seen to combine tradition and modernity, and Bangalore and Mumbai, which are inimitably modern. In defining this contrast, one

Urban Ways of Life / 181

crucial variable is the level of caste-based social control and surveillance, which—though not as extreme as in villages—is considerably stronger in Chennai, especially in its Brahman-dominated southern suburbs, than in more socially heterogeneous or “cosmopolitan” Bangalore and Mumbai. In suburban America, where South Asians have highly dispersed social networks and live as an ethnic minority alongside white Americans, social control and surveillance are still weaker. Women are invariably more concerned than men about social pressure because they are more subject to it. The majority of women—especially those who are actually familiar with rural life—are adamant that they do not want to live in restrictive, “gossipy” villages; many of them are content in Chennai, but many also dislike the surveillance prevalent in its Brahman neighborhoods; most of those living in Bangalore and Mumbai, as well as in the American suburbs, value the greater autonomy and freedom that they find in these places. To put it another way, because gender inequality is reinforced by social control, the significant improvement in the position of Tamil Brahman women over the last century or so, which we discussed in the previous chapter, is also directly linked to urbanization. In some respects, overseas migration is merely an extension of the urban migration and circulation within India that Tamil Brahmans have undertaken since the nineteenth century. Nonetheless, migration to the United States has a unique quality owing to the potency of the American dream. However quickly the dreamers come down to earth after arrival and however realistic many people are, only America is ever thought of as a land of boundless freedom and opportunities. In America, both women and men, but especially women, can enjoy a personal autonomy that is hard to find in India. Yet life there also endangers family and kinship ties, as well as children’s socialization into Indian culture and values. These are regarded as serious problems by many Tamil Brahmans in America, as well as their relatives in India—as indeed they are by many South Asians belonging to transnational groups. Return migration has become commoner, but most emigrants, irrespective of caste or community, are still unlikely to go back to India. Moreover, their children, when they grow up, will probably identify themselves as second-generation South Asian Americans. How, in the case of Tamil Brahmans, that new identity will affect their old caste identity—or whether, perhaps, it will undermine it entirely—are questions on which we can only speculate, so that they should be left to future researchers.

Six

Religion, Music, and Dance

In January 2012, on a Sunday morning so that people could easily attend, a special ritual was held in the Ratnagirishwara (Ratnagirı¯s´vara [Shiva]) temple in Besant Nagar, Chennai. Most devotees at this temple, which was founded in 1968, are upper-middle-class Tamil Brahmans living in the vicinity. The ritual, which lasted several hours and had been held in previous years, was a homa, in which oblations were made to the deities by dropping fruits, grains, and other items into a sacrificial fire, while a dozen Brahman pandits chanted Sanskrit verses from the Vedas, the most ancient and authoritative sacred texts of Hinduism. The oblations were offered in particular to Hayagrı¯va and Medha¯ Dakshinamurti (Daks·ina¯mu¯ rti), who are respectively Vishnu and Shiva as gods of learning and wisdom. According to publicity distributed in the temple beforehand, the ritual was designed to benefit students approaching their end-of-year examinations, to whom the following plea was made: “Since these homams are exclusively arranged for the benefit of school/college going students and those pursuing professional higher studies, we would request students to participate in large numbers and receive the Grace of Sri Hayagrivar and Sri Medha Dakshinamurthy.” Supplicating the deities before important examinations is common practice among Hindus. Most students, however, would probably say a prayer or make a small offering to their own favorite god or goddess—or to Sarasvatı¯, preeminently the goddess of learning and music—rather than participate in a homa for forms of Vishnu and Shiva who are rarely worshipped. Homas are standard in south Indian temples, but they are Vedic rituals that are distinctively Brahmanical and Sanskritic, and only a temple run by and for Brahmans would organize such an event for students, even if the publicity did emphasize that “all are welcome to participate.” It is striking, too, that this elaborate ritual was specially organized for students, including those

184 / Chapter Six

studying for professional examinations, because it so obviously reflects the preoccupation with examination success typical among the many uppermiddle-class Brahman families resident in Besant Nagar. In the Hayagriva and Dakshinamurti homa, the connection between socioeconomic goals and ritual is unusually clear. Nonetheless, middle-class Tamil Brahmans predictably have a general predisposition—in Weber’s phrase, an “elective affinity”—for particular forms or styles of Hinduism that are broadly consistent with their social status and economic interests. Moreover, now as in the past, religious belief and practice are personally and collectively important for the vast majority of them, so that they are key constitutive elements of Tamil Brahmanhood. In this chapter, which does not pretend to present a detailed ethnography of Brahman religion, we focus on features that are particularly salient for understanding the Tamil Brahmans’ place in the modern world.

Brahmans and Brahmanism in Tamilnadu According to classical, orthodox tradition, the Brahmans, the highest of the four varn ·as, have the sole right and duty to teach, study, and recite the Vedas. The Brahmans are also the custodians and interpreters of the dharmas´a¯stra, the corpus of religious and legal texts, including The Laws of Manu, which were composed over many centuries to set out the Hindu dharma—religious law and duty, but in its extended sense the sociocosmic moral order in its entirety.1 The dharmas´a¯stra, as the work of Brahman lawgivers, presumes a worldview in which Brahmans alone have the power of sacred knowledge. As well as constituting a class of religious literati, Brahmans serve as priests, in both temples and homes. Religion has always been crucial to the Tamil Brahmans’ relationship with non-Brahmans. From the outset, therefore, the non-Brahman movement—especially the “Self-Respect” movement founded in the 1920s by E. V. Ramasamy—did more than challenge the Brahmans’ preponderant share of government jobs and near-monopoly of political power. It also rejected the despised appellation of Shudra—the lowest varn · a to which south Indian non-Brahmans traditionally belong—because it implied slavery and degraded birth, and consistently attacked “Brahmanism,” the discourse and ideology justifying the Brahmans’ supremacy as guardians of the Vedic, Sanskritic tradition deemed superior to its Tamil counterpart.2 The ancient belief that south Indian Brahmans are Aryans, who originally came from the northern heartland of Vedic religion and culture, was reshaped in the late nineteenth century by Western, “scientific” theories

Religion, Music, and Dance / 185

of racial evolutionism, following the earlier philological discovery that the Dravidian language family, which includes Tamil, is distinct from the Indo-European family, whose ancestral language was originally identified as Sanskrit. According to protagonists of the non-Brahman Dravidian movement, the Aryan Brahmans from the north had successfully introduced caste hierarchy and suppressed the egalitarian social system of the indigenous people, who belonged to a separate Dravidian race. At one and the same time, the Brahmans had superimposed Vedic Hinduism authorized by Sanskrit scripture onto the autochthonous religion of the Dravidians based upon the cult of local deities, although “neo-Shaivist” Dravidian intellectuals also portrayed their religion as a form of pristine worship of Shiva. Furthermore, Sanskrit’s preeminence in Brahmanical religion and culture had relegated Tamil to the inferior status of a local vernacular, thereby extinguishing the classical Tamil traditions of language and literature, which were in reality older than their Sanskrit counterparts.3 The development of Dravidianist ideas alongside the rise of the nonBrahman political movement led to a series of “turn-of-the-century culture wars,” as Sumathi Ramaswamy aptly calls them.4 Even though antiBrahmanism has declined as a political force since the late 1970s, these culture wars—for example, about Tamil’s “classical” status vis-à-vis Sanskrit—have continued until the present day.5 They have also followed fundamentally the same course for a century or more, always positing a sharp dichotomy between hierarchical Brahmanism, Sanskrit, and Sanskritic religion and culture, on the one hand, and putatively egalitarian anti-Brahmanism, the Tamil language, and Tamil or Dravidian, nonSanskritic religion and culture, on the other.

Sanskritic Hinduism and the Great Tradition in Tamilnadu Irrespective of its politicized character in modern Tamilnadu, some sort of distinction between religious beliefs and practices that are Brahmanical, Sanskritic, and (in theory) sanctioned by scriptural texts, as opposed to ones that are non-Brahmanical, non-Sanskritic, unsanctioned, and therefore “folk” or “popular,” has always been salient for Hindus themselves everywhere in India. In one guise or another, the analogous idea of higher and lower “levels” has also consistently made its appearance in scholarly analyses of Hinduism, as well as of other “world religions.” In his sociology of religion, Weber powerfully argued that different status groups with dif­ ferent social, economic, and political ideas and interests are predisposed toward—or have an elective affinity for—different forms of religion, whether

186 / Chapter Six

scriptural or popular, ethical or prophetic, intellectual or emotional, and so on.6 More obviously influential in the anthropology of India, however, is Srinivas’s distinction, first formulated in the 1950s, between “Sanskritic Hinduism,” which is pan-Indian and most prevalent among Brahmans, and “non-Sanskritic Hinduism,” which is local or regional and mainly lowercaste.7 In addition, around the same time, Robert Redfield developed a broader, cross-cultural concept of the “great” and “little traditions” within a civilization, which he and Singer discussed in relation to the “cultural role of cities,” the main sites for transmuting folk cultures into great traditions because they are home to the “literati,” the traditional intellectuals.8 Singer then employed the same concept in his research on Madras.9 Srinivas’s and Redfield’s concepts, which have been extensively discussed and criticized, are not identical; nevertheless, they have commonly been merged, so that the great tradition is equated with Sanskritic religion and culture, and the little tradition with its non-Sanskritic counterparts.

Brahman Priests In Tamilnadu, the distinction between the Brahmanical, Sanskritic great tradition and the non-Brahmanical, Tamil little tradition is institutionalized most plainly in the region’s ubiquitous temples, but it is also apparent in domestic worship and rites of passage, and actually pervades the entire religious domain. The priesthood, too, is correspondingly segmented by caste. Before turning to temples and rituals, we briefly describe the Brahman priests’ main roles and principal subdivisions. First and foremost, Brahman priests are divisible into temple priests or arcakas (“archaka”), and domestic priests, usually known in Tamilnadu as sa¯stris, rather than purohitas. Temple priests are of course responsible for conducting the worship (pu¯ ja¯), for the gods and goddesses in temples— both the “public worship,” which comprises daily worship and periodic festivals, and “private worship” on behalf of individuals making offerings to the deities. In Shiva’s temples, almost all priests belong to the endogamous Adishaiva, Shivacharya (s´iva¯ca¯rya), or Gurukkal (gurukkal·) subcaste, which is conventionally ranked below the other Smartas like a fifth Aiyar subcaste, although the Adishaivas’ true status as Brahmans was formerly questioned. Around 3 percent of Tamil Brahmans belong to the Adishaiva subcaste.10 In the Thanjavur region, some temple priests belong to the Choliyar (coliyar) ¯ subcaste. In Shiva’s preeminent temple in Tamilnadu, the Nataraja temple in Chidambaram, however, the priests belong to the small, exclusive subcaste of Dikshitars (dı¯ks·itar), who claim superiority over Adishaivas.11 The

Religion, Music, and Dance / 187

priests in Vishnu’s temples are not members of a separate Aiyangar subcaste, equivalent to the Adishaivas. In general, temple priests in older temples have hereditary rights of office, whereas in newer ones they are paid employees. Priests working in large, wealthy temples that attract a lot of welloff, middle-class devotees—such as the Minakshi temple in Madurai—can earn a good living nowadays.12 Most temple priests, however, are less fortunate and those serving in village temples are often impoverished. The priests in non-Sanskritic temples are nearly all non-Brahmans, who are known as pu¯ca¯ris. Sa¯stris, whether Smarta or Sri Vaishnava, do not belong to distinct subcastes. Domestic priests, who take responsibility for both auspicious and inauspicious rituals, officiate at weddings, funerals, and several other rites of passage; they also perform pu¯ ja¯s, especially at festivals, for individuals and families, as well as exceptional rituals conducted, for instance, to inaugurate and bless a new venture. Domestic priests work in their clients’ homes, public buildings, wedding halls, premises reserved for funeral rituals, and other sites. At mortuary rituals, the dreadful task of absorbing the sins of the deceased by eating the special food offered to its spirit, which occurs on the final eleventh and twelfth days of the mourning period, is normally discharged by either family members or very poor Brahmans who need the money. In Chennai, Smarta and Sri Vaishnava domestic priests normally serve only Brahman clients belonging to their own section, and a separate, lower-status group of Brahman domestic priests, the Panchangakkarars (pañca¯n·gakka¯ra), serve non-Brahman families. The same system probably exists more widely in Tamilnadu, as information on Panchangakkarars in the Kaveri delta region, where they are Telugu Brahmans, indicates.13 Many domestic priests, like temple priests, are poor. When K. Subramaniam studied them in the 1960s, almost all the men she met in Madras and Mayiladuthurai were struggling to secure clients and to earn enough to support themselves and their families.14 In those days, many domestic priests still had hereditary links with client families, but these have progressively disappeared, especially in urban areas. Hence priests now offer their services in an open market, even though many people usually call upon the same priest whenever they need one. In this market, men with a high reputation for ritual skill and professionalism—and a good business sense—can make a lot of money, especially from affluent, upper-middle-class clients in Chennai and other cities, where there is a shortage of priests. Modern, successful priests typically charge a fixed fee for a ritual and regularly employ other poorer priests, who may be younger or less enterprising, to assist them.

188 / Chapter Six

The priesthood has always been a demeaning occupation, so that all Brahmans working as priests—not just Adishaivas—are conventionally regarded as inferior by other Brahmans. Furthermore, priests are mostly unpopular with the general public and are maligned, often unfairly, as incompetent and avaricious. Some priests, as just mentioned, make a good living today and some now work in temples overseas, where they may earn plenty of money and settle down with their families. But they are the minority and a large number of priests, owing to their low status and poverty, are unhappy with their lot and want their sons to take up secular employment instead. Many have done so and the Brahman middle class now includes priests’ descendants.

Temples, Domestic Worship, and Rites of Passage Throughout India, Shiva and Vishnu are preeminent as the two great gods (deva) of Hinduism. In Brahman villages in the Kaveri delta, as we saw in chapter 1, there is normally a Vishnu temple inside the agraha¯ram and a Shiva temple at its edge. In Tambraparni valley villages, the layout is more variable. Thus in Mel Ceval, the main Vishnu temple stands at the end of the old agraha¯ram and another smaller one is by the newer agraha¯ram, but the big Shiva temple is well outside the old agraha¯ram, whereas in Yanaimangalam, there is a Shiva temple at the end of the agraha¯ram and the Vishnu temple is at the edge of the settlement.15 In all temples dedicated to Shiva or Vishnu, the priests are Brahmans, Sanskrit is the primary ritual language, and only vegetarian food is offered in worship. The goddess Devi, who is Shiva’s wife Pa¯rvatı¯ and Vishnu’s wife Lakshmi (Laks ·mı¯), is worshipped in variant forms in both gods’ temples, but she is more a prominent, partially independent deity in Shaiva temples. In addition, even though Ka¯·liyamman or a cognate goddess serving as a village’s tutelary deity is rep¯ resented in her own temple as unmarried, she is typically identified as a potential or actual consort of Shiva as well. In every village, the village goddess’s temple, as well as temples and small shrines for other little “village deities” (gra¯madevata¯), are located in and around the main, non-Brahman settlement area. In the Dalits’ colony, there are separate village deities’ temples. For almost all village deities, the priests are non-Brahman pu¯ca¯ris and Tamil is the ritual language; for some village deities, vegetarian offerings during pu¯ja¯ are periodically complemented by animal sacrifice (bali).16 Village temples for Shiva and Vishnu are (or were) controlled by Brahman trustees, although many of the larger ones—like almost all the state’s major temples in both urban and rural areas—are now administered by the Tamil-

Religion, Music, and Dance / 189

nadu government’s Hindu Religious and Charitable Endowments Department. Shiva’s and Vishnu’s temples were generally closed to Untouchables until 1947, when the Madras Temple Entry Authorization Act was passed. In Brahman villages in the past, non-Brahman access to Vishnu temples, though not Shiva temples, was commonly restricted as well. Village deities’ temples have usually been open to all castes, although few Brahmans go to them. Only Dalits normally worship in their own colonies’ temples. In reality, the correspondence between types of temple and caste divisions in Brahman villages was more complex than our brief sketch implies, as Béteille’s study of temples in Sripuram shows.17 Moreover, as Diane Mines demonstrates for Yanaimangalam, the relationships among and between deities and villagers are neither fixed nor uncontested. Hence there are “multilateral discourses” engaging divine and human beings, which generate arguments about the village as a social and physical space, and about “alternative formulations of possible social worlds”—although Brahmans, significantly, were too weak in Yanaimangalam in the 1980s to impose their ideas.18 Finally—and a point of much wider significance—we must emphasize that the conceptual distinctions made in the academic study of Hinduism between the Sanskritic and non-Sanskritic, or great and little traditions, have been criticized as seriously misleading, mainly because they imply that the religion is actually split into separate strata. It is easy to show that this is false: one example among countless others is that the village goddess, a little deity, may be seen as both distinct from and identical with Shiva’s consort, a great deity. This fluid, contextual logic of the “one and many” is fundamental to the polytheistic Hindu pantheon, so that in reality no clear-cut distinction can ever be drawn between great Sanskritic and little non-Sanskritic deities. Broadly speaking, the same applies to every facet of the distinction between the two levels. Even in the case of village temples, therefore, the bifurcation into Brahmanical, Sanskritic, great-tradition religion, and its non-Brahmanical, Tamil, little-tradition counterpart, is fuzzier than it appears to be, notwithstanding the general assumption of almost everybody, Brahmans and non-Brahmans, that two separate religious domains exist within south Indian Hinduism.19 In towns and cities, like villages, the same distinction between two categories of temples—subject to the same qualifications—can be drawn. The Brahmanical, Sanskritic temples dedicated to Shiva, Vishnu, and Devi, and to gods closely allied with them—notably Murugan or Subrahman ·ya, ¯ Shiva’s younger son—include the vast, spectacular temples originally built by the Chola and Nayaka kings in Kanchipuram, Madurai, Srirangam (near Tiruchirappalli), Thanjavur, and many other places, as well as later

190 / Chapter Six

constructions like the principal temples in Chennai. In urban areas, there are also countless temples for village deities (pace their name). Brahmanical temples in towns and cities were closed to Untouchables before templeentry legislation was passed. They have rarely been controlled by Brahmans alone, however, unless they are located inside agraha¯rams. In the past, Brahman devotees sometimes had privileged access to the inner shrines, but otherwise all urban temples—including Vishnu’s—were and are normally accessible to both Brahmans and non-Brahmans equally. Hence the distinction between Brahmanical and non-Brahmanical temples—and their deities, priests, and styles of worship—has always been less well correlated with caste divisions in urban centers than in Brahman villages. In both rural and urban areas, many Hindu rituals are carried out in homes, rather than temples. In most homes, worship is performed regularly before the domestic shrine, which contains small images or pictures of gods, goddesses, and gurus favored by family members, as well as photographs of deceased ancestors. Worship may be carried out daily or weekly, often by a woman but sometimes by the male household head; it may be more or less elaborate according to how much time is available or whether it is a special day, such as the festival of a popular god or goddess. As well as regular worship, numerous domestic festivals are celebrated, typically with special rituals and food offerings. Hindus of all castes conduct regular worship and celebrate festivals at home, although some festivals are confined to particular communities, and there is some variation in ritual practice between Brahmans and nonBrahmans. For instance, Brahmans have more elaborate methods of purification prior to a ritual than non-Brahmans, so that they can attain a special state of purity (mat·i) higher than the ordinary purity (cuttam) otherwise required. Some (though not all) Brahmans recite Sanskrit mantras (sacred formulae), whereas non-Brahmans never do, and Brahmans—upholding the orthodox position—sometimes insist that worship without mantras is not really worship at all. There are other differences, too, but in most respects domestic ritual is similar in Brahman and non-Brahman homes.20 The rites of passage observed by Brahmans and non-Brahmans, as well as how they celebrate them, also differ somewhat. Weddings, for example, vary in detail from one community to another, but one difference often highlighted is that for Brahmans the union is sealed by the Vedic ritual of the “seven steps” (saptapadı¯), when the couple walks round the sacred fire, whereas for non-Brahmans the crucial moment is tying the bride’s wedding pendant or ta¯li round her neck, a distinctively south Indian rite. Brahman and non-Brahman funerals differ considerably, and traditional rules decree

Religion, Music, and Dance / 191

that death pollution and mourning last longer for non-Brahmans (sixteen to thirty days) than Brahmans (twelve days), although nowadays many people in all castes shorten these periods. One ritual reserved for Brahmans is the upanayana, when a boy or young man is invested with his sacred thread and becomes “twice-born” and fully Brahman. Ideally, the upanayana is done before puberty, the practice in more orthodox families, but it is often delayed until just before marriage. “Twice-born” Brahman men should undertake various regular rituals specific to their status, notably the earlymorning sandhya¯, and some of them comply or try to do so. Our outline of differences between Brahmanical, Sanskritic and nonBrahmanical, non-Sanskritic deities, priests, temples, and rituals is far from exhaustive, but it serves to illustrate the dichotomy running through the religious domain, which Brahmans and non-Brahmans alike generally take for granted. In fact, because this religious dichotomy is empirically fuzzy in many contexts, it is primarily ideological, but now as in the past, it plays a vital role in sustaining the broader social and cultural differences between Brahmans and non-Brahmans. For the Brahman minority, it also serves to reinforce their own collective identity and sense of superiority as privileged custodians of the Sanskritic great tradition.

Hinduism in Chennai When Singer conducted research on the great tradition in Madras city between 1954 and 1964, the majority of his informants were Smarta Brahmans whom he described as “leading representatives of the Great Tradition of Sanskritic Hinduism.”21 Singer, who did a lot of his research in Mylapore, was initially guided by V. Raghavan, an eminent Smarta professor of Sanskrit and long-term secretary of the Madras Music Academy, the preeminent Carnatic music association discussed later in this chapter. Raghavan unequivocally portrayed Sanskritic Hinduism and culture as the core of the great tradition “integrating” the diversity of India.22 Later, Singer partly distanced himself from Raghavan, “whose version of the Great Tradition . . . represented only one of several interpretations,” and he “stopped looking for the Great Tradition and gave up the effort to select the orthodox version of Sanskritic Hinduism.”23 In the end, Singer presented a “first approximation” to the “structure” of Sanskritic Hinduism in Madras as a constellation of “sacred centers” (homes, temples, monasteries, pilgrimage sites), “sacred specialists” (priests, monks, Sanskrit pandits, astrologers, and others), and “sacred rites” (domestic and temple).24 He then amplified this outline, before extending his survey into the cultural sphere of music and dance in

192 / Chapter Six

particular, while also emphasizing the importance of bhakti, devotionalism, in urban religion.25 Singer’s ethnographic analysis is rather miscellaneous, although consis­ tent with his conclusion that no single great tradition of Sanskritic Hinduism existed in Madras. It is open to various criticisms: for instance, “Singer’s privileging of Smarta voices” so that he reproduced their caste and class biases, his neglect of temple Hinduism, and his overstated claim that devotionalism is a response to modern urban life.26 On the other hand, as Mary Hancock implies, Singer’s account can still be read as a description of a polythetic set of beliefs, practices, and sensibilities constituting the Brahmanical, Sanskritic great tradition as normatively understood by uppermiddle-class Brahmans in Madras in the fifties and sixties. That understanding, to a great extent, is an outcome of the Brahmans’ critical mass in the city, especially in its southern suburbs, as noted in the previous chapter. Much of what Singer described remains in place, even though much has changed as well, especially since the late 1980s, following the rise of Hindu nationalism and the spread of its Hindutva (“Hinduness”) ideology, and the efflorescence of temple building and renovation that has been central to Hindu revivalism in Tamilnadu.27 Another notable development, common throughout India, is the growing appeal of gurus, “god-men,” and other charismatic saintly figures. These changes have been explored in Chennai by Hancock, who focused on women and domestic ritual; Joanne Waghorne, who investigated temple Hinduism; and John Harriss, who restudied Singer’s “industrial leaders” in 2000.28 In our research, we collected copious information about religion, although we did not do detailed fieldwork on the topic. Our material is best about middle-class people in Chennai, both Brahman and non-Brahman. Consistently with Waghorne’s findings, but not Singer’s, “visiting temples” was the commonest response to questions about customary religious belief and practice. The majority of people from all castes said that they go to temples regularly, with only a handful saying they never do. Men and women with jobs typically visit temples on the weekend, although a few go more or less often, whereas women without paid employment, especially if they are not looking after small children, commonly go on one or more days each week, sometimes on the days favored for worshipping particular deities. Virtually all the temples mentioned to us—except for a couple visited by non-Brahmans—are Brahmanical, Sanskritic ones and most people naturally visit those near their homes. In T. Nagar, for example, the most popular are the Agastya temple in which Shiva is the presiding deity; the Venkateshwara ·kat es´vara [Vishnu]) temple run by the Tirumala Tirupati Devasthanam, (Ven ·

Religion, Music, and Dance / 193

which controls the god’s main temple—the richest and most popular in India—at Tirupati in Andhra Pradesh; and the Sringeri Sharada Pitha (S´a¯rada¯ Pı¯·t ha) temple owned by the Sringeri monastery (“mutt,” mat·ha). In Besant Nagar, the Varasiddhi Vina¯yaka (Ganesha) temple is popular and the Ratnagirishwara (Shiva) temple is specially favored by Brahmans. These five temples, like many others in Chennai, have all been built relatively recently in line with the city’s expansion. Local trusts founded the Agastya temple in 1943, the Ratnagirishwara temple in 1968, and the Varasiddhi Vinayaka temple in 1979, and each was constructed in stages. The Venkateshwara and Sringeri temples opened in 1975 and 1995, respectively. The newer temples differ in structure from the older, traditional ones, such as the great temple dedicated to Shiva as Kapa¯lı¯s´ vara, which was built in Mylapore in its present form around 1800, although a temple has existed on the site for much longer. The Kapalishwara temple, like others in Chennai dating from the same period, resembles much older, larger temples of Shiva in Tamilnadu, especially in the arrangement of different deities’ shrines. Thus in old Shaiva temples, only one or two minor images of Vishnu are usually found, while in Vaishnava temples there are none of Shiva. Modern temples, by contrast, all contain prominent images of both great gods, as well as other subsidiary Shaiva and Vaishnava deities, and their layout is usually different. The Agastya temple—like some others dedicated to both Shiva and Vishnu—actually employs both Shaiva and Vaishnava priests. With some rare exceptions, such as the ancient Suchindram temple in southern Tamilnadu, temples where both great gods are prominently worshipped are modern urban innovations. They reflect the mutual endeavor of both Aiyars and Aiyangars—and in some cases nonBrahmans as well—whose “sectarian” divisions no longer matter much in the modern, urban, middle-class environment. Hindus go to new temples, like old ones, to have a “sight” or “vision” (dars´ana) of the deities, to praise and pray to them, to make offerings to them, and to receive their grace in the form of prasa¯da, which may be sacred ash from Shiva, holy water from Vishnu, vermilion powder from the goddess, or other items such as food. Many people also attend temples during the regular morning and evening worship conducted by the priests, and visit on special festival days. But most new temples also have the space to host a range of other activities and events; indeed, in both the Venkateshwara and Sringeri temples, the main architectural feature is a large, spacious hall, with all the deities’ images positioned at one end. Temple space is used for public meetings, religious discourses, philanthropic projects (such as free clinics for the poor), Carnatic music concerts, and classes to learn s´ lokas,

194 / Chapter Six

devotional hymns, or Vedic texts, as well as elaborate rituals like the homas for students in the Rathnagirishwara temple described above. Activities of this kind do occur in old temples as well, but they are commoner in new ones, and they mirror the contemporary social, cultural, and civic concerns of the urban middle class alongside their religious ones.29 And because Brahmans are the majority of middle-class residents in much of southern Chennai, their concerns are naturally at the fore. Temples of the older traditional type, especially large ones, tend to be noisy, bustling places full of activity, in which music is played, bells are rung, mantras are chanted, children play, and priests and devotees continually walk around and talk to each other. When they are crowded, modern temples can be noisy, too, but in general their atmosphere is more sedate. In the Venkateshwara temple, devotees stand quietly in line to pass before his image in orderly fashion, but the Sringeri temple displays a greater contrast with the older style. Like another new temple described by Waghorne, it is exceptionally clean and its spaciousness ensures that it is not “cramped” like old temples, a comment she heard frequently from middle-class Hindus.30 At the end of the Sringeri temple’s hall, there are marble images of the god·kara), the eighth-century dess Sharada, Vinayaka, and Adi Shankara (A¯di S´an Kerala Brahman philosopher of advaita, “monism,” from whom the Sringeri ·kara¯ca¯rya) heading the monastery claim spiritual deShankaracharyas (S´an scent. Worship is regularly performed before these images by the Brahman priests, but it is done quietly, so as not to disturb the peace in the hall, where devotees sometimes chant s´lokas or recite Vedic texts, but otherwise sit in meditation, respecting the notice requesting silence. One Brahman woman in T. Nagar said that she spends a few minutes meditating there after her daily walk in a park, while another prefers the Sringeri temple for its peacefulness, although she regularly visits other temples as well. Not everyone in T. Nagar chooses to worship in the Sringeri temple, of course; nonetheless, its cleanliness, spaciousness, and serenity undoubtedly have a strong appeal to middle-class Brahmans with a predilection for calm and restraint in the expression of religious devotion. Apart from visiting temples, several people answering questions about religion mentioned their daily performance of domestic worship and celebration of periodic festivals, although all this is routine in most Hindu homes. Some women, both Brahman and non-Brahman, reported regular fasting, usually weekly or monthly, which requires avoidance of rice-based meals, but may not mean total abstinence from all food. One Brahman woman in T. Nagar, who said that she prays a lot at home, fasts every Tuesday; another, who regularly worships in a local Subrahmanya temple,

Religion, Music, and Dance / 195

also fasts every Tuesday and during the Tamil month of ka¯rttikai (November– December), which are the god’s special weekday and month. Several Brahman men said they carry out the sandhya¯ ritual at home, more or less fully, either daily or sometimes only on the weekend, and others claimed they were trying to start doing it regularly. One or two men flatly declared that they do not perform the sandhya¯ because they “don’t believe in it” and others calling themselves “not very religious” implied that it was because they ignored this daily ritual. Somewhat surprisingly, only two Brahman women told us they belonged to bhajana, or devotional hymn-singing groups, which does not adequately reflect their known popularity among Tamil Brahmans, now as in the past. Many Brahmans also attend classes for learning s´lokas (one of which is described below) and other devotional texts, such as, for example, the Lalita¯ Sahasrana¯ma in praise of the goddess Lalita in Sanskrit, and the Divya Prabandham, the corpus of devotional poems for Vishnu composed by medieval saints, in Tamil.31 Among Brahmans, only the woman who meditates daily in the Sringeri temple said she had a personal guru, although several people go to hear discourses by notable gurus and religious leaders, which frequently take place in Chennai and are advertised daily in the Hindu newspaper. Among the most important of these figures are the Shankaracharyas, who are discussed shortly. One Tamil Brahman guru with a strong appeal to businessmen and educated professionals in Chennai is Swami Dayananda, who combines “traditional” Sanskritic Hinduism with teaching attuned to modern, uppermiddle-class concerns.32 The immensely popular “god man” Sathya Sai Baba also attracted many Tamil Brahman disciples, who grieved deeply when he died in 2011.33 But many other gurus have substantial followings as well, although some people pointedly told us that they do not follow them. One or two informants said that now, or when younger, they have not been religious at all. They are exceptions, however, and for the vast majority of Tamil Brahmans living in Chennai, as well as elsewhere, religion—specifically Sanskritic Hinduism—plays a very important part in their lives. Not far from the Rathnagirishwara temple in Besant Nagar are the premises of the Satya Narayana Theertham (Satya Na¯ra¯yan ·a [Vishnu], tı¯rtha [sacred place, crossing point]), which can aptly conclude this section. At the Theertham, which opened in 2001, Brahmans can carry out their funerary rituals, after the body has been burned in the nearby electric crematorium or elsewhere. One of the Brahmans who founded the Theertham, which is run by a trust controlled by Sri Vaishnavas, told us that it was easy to raise money for the buildings after first announcing the plan in the Varasiddhi Vinayaka temple. Brahmans were “suffering,” he said, because they could

196 / Chapter Six

not fulfill their filial obligations to deceased parents or other relatives; since most people live in flats or small houses in Chennai, they have no space to perform death rituals, which should last thirteen days in all (although the period is often shortened). Most people employ their own domestic priests to take charge of the rituals, but they are usually assisted by priests and cooks employed by the Theertham. Some of these men are Aiyangars and others Aiyars; in death, if no longer in modern life, this division remains important, so that priests and cooks can work only for families belonging to their own division. Brahmans who need assistance are helped by the Theertham; some of them are men living overseas who have to fly home in a hurry to conduct their parents’ funerals and may have next to no idea about what to do. Although obviously different from the Ratnagirishwara temple organizing homas for students, the Satya Narayana Theertham also exemplifies the combination of Brahmanical orthodoxy and civic responsibility that characterizes the religion of Tamil Brahmans, especially those in the upper-middle class, in contemporary Chennai.

A Sanskrit S´ loka Class in a Chennai Temple Because Tamilnadu’s Brahmanical, Sanskritic temples are such prominent manifestations of Brahmanism, they and their priests have been the targets of the non-Brahman movement and radical Dravidianists since the days of E. V. Ramasamy. The Sanskrit language itself is important in this context and is one of “Sanskritic” Hinduism’s key features, as Srinivas’s coinage was meant to indicate. In the Tamil country, Sanskrit has always been linked to Brahmans, although nowadays priests are more likely to know some Sanskrit than ordinary, higher-status Brahmans, who have no reason to learn it.34 One telling story about Brahman attitudes to language is quoted by Sumathi Ramaswamy. When U. V. Swaminatha Aiyar, who became a distinguished scholar of Tamil, was a young man in the 1870s, an acquaintance asked his father what he did and was told that he read Tamil. The acquaintance incredulously asked why he could not study English or Sanskrit: “If he studies English, he would benefit in this world. The study of Sanskrit will prepare him for the other world. Studying Tamil will bring him neither benefit.”35 Such Brahman disdain for “vernacular” Tamil has long outraged the Dravidianists, and a polar opposition between Sanskrit and Tamil has been at the heart of the culture wars described by Ramaswamy. The use of Sanskrit by Brahmans is therefore never a neutral act. Something of the Brahmans’ outlook on the centrality of Sanskrit in Hinduism can be illustrated with a case study of a s´loka class. Savitri is an

Religion, Music, and Dance / 197

upper-middle-class Smarta Tamil Brahman woman who lives in T. Nagar; she was in her early forties when interviewed in 2004. Savitri’s husband used to work in the armed services and is now in business; they have two sons, both at school at the time of the interview. Savitri’s father was a businessman in Bangalore, where she was brought up and her brothers and sisters still live. Savitri is the teacher at a s´loka class held in the evenings at the Sringeri temple, which in 2004 was attended by about thirty women from different castes, all described by Savitri as “middle-class.” In 1997, the temple had started s´loka classes for children, taught by a school Sanskrit teacher who is a Telugu Brahman. In 2004, twenty-nine boys and girls in the fourth to seventh standards, mostly aged between nine and twelve and mostly Brahmans, were in the class. But in 1997, the class had included Savitri’s sons and when the temple managers discovered that she was knowledgeable in the s´lokas, which her father had taught her, they asked her to teach a new class for adult women, which began a year later. The curriculum of s´lokas for both the women’s and children’s classes is set by the Sringeri Shankara­ charya. The oral teaching method follows the age-old Hindu system, whereby the teacher successively utters the stanzas that the students then repeat and progressively memorize. Savitri emphasized the importance of pronouncing the s´lokas accurately, which students find difficult because many Sanskrit phonemes do not exist in Tamil. Sanskrit, said Savitri, is supposed to be the oldest language, although she also mentioned the claim that Tamil is older; she was clear, though, that Sanskrit is “the language of the gods,” which is the orthodox, traditional belief of most Tamil Brahmans, except for the Sri Vaishnava Tengalai minority, which gives primacy to Tamil. In her classes, Savitri explains the meaning of the verses, which praise a wide range of gods and goddesses, because she believes that reciting with understanding is very different from just reciting. Once a week and at festivals, the women in the class worship Sharadambal in the temple by chanting verses in her praise; the children do so once a month. According to Savitri, s´lokas matter not only for worship, but also because “we are losing our culture because of Western influence” and s´lokas “impart a lot of culture along with knowledge.” The children’s teacher reported that many of their parents shared the same opinion. Savitri is an accomplished musician, too, who used to perform Carnatic music on the radio in Bangalore, and later taught music and hymn singing in Chennai. In Chennai, she said, people are extremely knowledgeable about Carnatic music, as they are not in Bangalore. Savitri took it for granted that “music and Sanskrit go along together.” Most of her students are not musical, but she teaches them to sing hymns and chant the s´lokas with the right musical rhythm,

198 / Chapter Six

and sometimes uses popular film music to help them learn. In addition to the women’s and children’s s´loka classes, the temple also runs Vedic classes for men, in which they learn to recite Vedic texts forming part of the worship in Sanskritic temples. For Savitri, who was particularly articulate on the topic, Sanskritic Hinduism and Carnatic music are the core of middle-class Brahman religion and culture, but she also felt that they are endangered by Westernization. Although not shared by everyone, this is a common view. Thus one of her students, a retired Brahman schoolteacher, said that one reason for joining the s´loka class was her feeling that Brahmans are losing interest in their own traditions; some non-Brahmans, by contrast, are actually becoming more interested and that is why they join the class. One non-Brahman did indeed say that she joined Savitri’s class because she felt badly informed when visiting the Sringeri temple and other temples in T. Nagar, and she was now thinking about learning Sanskrit as well. She has become a vegetarian, too, and although her husband supports her, other relatives tease her for following rules meant only for Brahmans. Political controversy over the language of temple rituals—particularly the arcana (anglicized as “archana”), when an offering is made to a deity while reciting its names—has been erupting in Tamilnadu ever since the 1920s, with repeated demands that arcanas should be done in Tamil, not Sanskrit.36 In the 1970s, when the DMK government was pursuing a vigorously anti-Brahman religious policy, it passed legislation designed to abolish the hereditary temple priesthood; its Hindu Religious and Charitable Endowments Department also instructed temples to introduce Tamil arcanas. At the time, M. Karunanidhi, the chief minister, is reported to have declared that if the gods in south India cannot tolerate Tamil arcanas, they should move to the north. The government’s religious policy later became more moderate, under both DMK and ADMK regimes, but that has scarcely modified the polarization between Sanskrit and Tamil in the temples.37 Savitri’s s´loka class, therefore, is not just a pious initiative to teach women sacred verses; it is also an active assertion by a Tamil Brahman that the deities should be praised in Sanskrit and they will be, whatever non-Brahman politicians say. Implicitly at least, it is also an assertion that the priests in Brahmanical, Sanskritic temples must continue to be Brahmans. And since everyone in Tamilnadu with any interest in temples knows about the controversies over language and the priesthood, Savitri’s students, including the non-Brahmans, undoubtedly understand that learning Sanskrit s´lokas is about more than just the right way to praise the gods and goddesses.

Religion, Music, and Dance / 199

The Shankaracharyas For Savitri, it matters a lot that the Sringeri Shankacharya sets the curriculum of the s´loka class and is ultimately its true guru. The heads of the monasteries in Kanchipuram in Tamilnadu and Sringeri in Karnataka are rival claimants to be Adi Shankara’s true spiritual descendant in south India. The majority of Tamil Smartas recognize Kanchipuram’s primacy, but a significant minority has always looked to Sringeri, which has an active presence in Chennai and elsewhere in Tamilnadu. For most Sri Vaishnavas, the supreme monk is the head or Jı¯r of Ahobila monastery in Andhra Pradesh, a Vadagalai institution, although he has never been as influential in Tamilnadu as the Shankaracharyas. The Jı¯r’s line of spiritual succession ultimately descends from Ra¯ma¯nuja, the eleventh-century Tamil Brahman philosopher of vis´is··t a¯dvaita, “qualified monism.” Throughout most of the twentieth century, the Kanchipuram monastery’s power and influence steadily grew under the leadership of Chandrasekharendra Saraswati, a Kannada Brahman born in Tamilnadu, who was installed as its head in 1908 and died in 1994 in his hundredth year. Chandrasekharendra patently manifested the vaidika (“Vedic”) ideal of ascetic renunciation combined with religious learning, and he commanded extraordinary veneration in south India, especially among Tamil Brahmans; he became dignified as the “supreme teacher” or Paramacharya (parama¯ca¯rya) and was widely regarded as divine. Singer met Chandrasekharendra and was impressed by him in the 1950s. As Singer’s summary of their conversation shows, Chandrasekharendra held extremely conservative views, which he frequently expounded in his discourses, and he was virtually the only prominent, public figure who consistently defended Brahmans, Brahmanical Hinduism, and Brahmanism in general against Dravidianists, nonBrahman politicians, and progressives of all stripes.38 Whether or not they shared his conservatism, Chandrasekharendra was admired and even revered by the leading Brahman businessmen in Chennai who talked to Singer and, half a century later, to Harriss.39 He was also revered by a broader group of upper-middle-class Brahmans, especially Aiyars, who made up the monastery’s core support, helped with its financial and legal affairs, and enabled it to interact with the government and other leading institutions.40 It owes much to Chandrasekharendra’s authority that even the DMK and ADMK governments of Tamilnadu have regularly taken heed of Kanchipuram’s views when formulating policy on religious matters, especially temples. Chandrasekharendra and his successors have consistently

200 / Chapter Six

emphasized the value of Sanskrit, and have also encouraged the building of new temples and the renovation of old ones, especially Shaiva shrines. Adishaiva temple priests have appreciated the Shankaracharyas’ support, while also fearing their influence over temple affairs, for there is a history of rivalry between the priests and monks.41 In 1954, Jayendra Saraswati was installed as the junior Shankaracharya and Chandrasekharendra’s heir. Chandrasekharendra largely retired from the monastery’s management in the 1970s and Jayendra then expanded its activities considerably, made the monastery richer, and attracted more support from non-Brahmans as well as Brahmans. By the 1990s, Jayendra was also involved in Hindu nationalist politics.42 After the Paramacharya’s death, criticism of his successor began to grow, allegations of corruption in the monastery were made, and Jayendra acquired some powerful enemies. In 2004, he was dramatically arrested and charged as a conspirator in a temple manager’s murder; in 2013, Jayendra, along with his junior Vija­ yendra Saraswati, was finally acquitted, but what this affair has done to the Kanchipuram monastery’s reputation in the long run is impossible to say.43 Tamil Brahmans who have been loyal to Kanchipuram—including many Eighteen-Village Vattimas who talk about the Shankaracharyas’ special connection with their community—often draw a veil over recent events and avoid any talk about Jayendra’s arrest. Instead, they reiterate their reverence for the late Paramacharya, whose reputation is undimmed. Sringeri attracts followers for essentially the same reasons as its rival. Among Tamil Brahmans, Sringeri could never match Kanchipuram while Chandrasekharendra was its active head, but after Jayendra expanded his monastery’s appeal to non-Brahmans, Sringeri’s more conservative outlook enhanced its position in some Brahman circles. Glenn Yocum argues that the Sringeri monastery, its head, its interpretation of monistic philosophy and Brahmanical traditions, and the ritual to install the Shankaracharya (which he observed in 1989), together “constitute a symbolic ensemble, a system of discourse, that legitimates and perpetuates a set of power relations in which Brahmins are superordinate and the non-Brahmin rest are subordinate.”44 Such an assertion of Brahman superiority is attractive not only to old-fashioned traditionalists, but also to many modern, urban middle-class Brahmans. In the final analysis, leaving aside their influence with government, the Shankaracharyas—both Sringeri and Kanchipuram until its recent catastrophe—were and are important to Tamil Brahmans because they are the preeminent, Brahman ascetic monks and renouncers. As the most Brahmanical of all Brahmans, they are also vital figures in constructing, per­ petuating, and defending Brahmanical, Sanskritic Hinduism in south India.

Religion, Music, and Dance / 201

Carnatic Music For Savitri, to repeat, “music and Sanskrit go along together.” Her view is shared by many Tamil Brahmans, so that for them the Brahmanical, Sanskritic great tradition embraces not only religion, but also classical music and dance, which are the quintessential expressions of elite, high culture in Tamilnadu. Brahmans play a dominant role in south Indian Carnatic (Karnatic) music and, though probably to a lesser extent, in south Indian dance, Bharatanatyam (bharatana¯t·yam, “Indian dance”). For both Carnatic musicians and Bharatanatyam dancers, Chennai is the city where everyone with ambition wants to appear on stage. “Traditional” Carnatic music and Bharatanatyam, which are invariably qualified as “classical,” were largely created in the first half of the twentieth century.45 In the early modern period, Carnatic music mainly depended on royal and lordly patronage, and was mostly performed in courts and temples. It particularly thrived in the Maratha kingdom of Tanjore where three great composers—Tyagaraja (1767–1847), Muthuswami Dikshitar (1725– 1834) and Shyama Sastri (1762–1827)—flourished in the late eighteenth and early nineteenth centuries. These three men, who were later canonized as the “trinity” of classical Carnatic music, refined the composition of devotional songs, kriti (kr ·ti) or kı¯rtana, which they set to a range of melodies or ra¯gas. Most songs were in Telugu or Sanskrit, though some were in Tamil, but even the Telugu and Tamil were strongly inflected by Sanskrit. Tyagaraja and Shyama Sastri were Telugu Brahmans, and Muthuswami Dikshitar a Tamil Brahman; almost all courtly musicians were also Brahmans. The style of music played in temple rituals was deemed inferior and temple musicians belonged to non-Brahman castes. During the nineteenth century, Carnatic music still prospered in the courts, especially in the Mysore and Travancore kingdoms, but it declined in Tanjore, which the British annexed in 1855. At the turn of the twentieth century, Carnatic music moved decisively to Madras city, which has been its principal center ever since. There its main patrons were members of the new urban middle class, especially Brahmans, and they included elite figures, such as V. Bhashyam Iyengar and S. Subramania Iyer, the high court judges who have already appeared several times in this book. According to Lakshmi Subramanian, the Brahman elite in particular attached a lot of “symbolic significance” to listening to and appreciating music, which contributed to building their “sense of community, with exclusivist overtones.”46 At one and the same time, traditional music had to be reorganized as part of a cultural reconstruction process that was developing “explicitly nationalist overtones.”47 Hence there was a close

202 / Chapter Six

parallel between musical reform, and wider social and political reform. At a Congress meeting in Madras in 1927, it was decided to set up the Madras Music Academy, which was founded a year later. The academy’s opening ceremony was performed by Sir C. P. Ramaswamy Aiyar, another leading lawyer and member of the Mylapore political group in the 1920s, and incidentally the last dewan of Travancore from 1936 to 1947; the academy’s founding president was U. Rama Rao, a Saraswat Brahman from Karnataka also prominent in Madras politics.48 Music was organized in the city by newly established sabha¯s, or associations, and performed in concert settings, mainly in public halls, though sometimes in temples. Even in new secular spaces, however, the music retained its predominantly religious ethos; indeed, “music was sacred and therefore had to be divested of all accretions of sensuality and corrupt usage.”49 The Madras Music Academy brought together leading patrons and other music lovers, and its main objectives were promoting Carnatic music and encouraging the study of its theory and practice. The academy established an annual festival in the Tamil month of ma¯rkali, mid-December to ¯ mid-January, which became the highlight of the year in the Carnatic music calendar. During the festival, as it has done since 1929, the academy stages a conference and series of concerts, which close with a convocation. The academy’s festival program has steadily expanded over the years; in the 2012–13 season, it included eighty-five concerts in eighteen days, which were immediately followed by a seven-day Bharatanatyam dance festival with twentysix performances.50 In addition, elsewhere in Chennai, more than 2,500 performances were staged by numerous sabha¯s, so that the city’s music and dance festival is now one of the largest events of its kind in the world. The inaugurator of the academy’s conference and the convocation president are either renowned musicians or distinguished figures from other walks of life. The list of names for both honors since 1929 is a roll call of the great and the good, including numerous maharajas, politicians, bureaucrats, judges, and businessmen, as well as some scholars and scientists. A significant minority of these people have been Tamil Brahmans: for instance, the physicist and Nobel laureate C. V. Raman and the eminent administrator V. T. Krishnamachari (both mentioned in chap. 2), who inaugurated the conferences in 1933 and 1961, respectively. In 2010, another Tamil Brahman scientist, Venkatraman Ramakrishnan, who won the Nobel Prize for chemistry in 2009, presided over the convocation; in 2011, N. Murali, a member of the family owning the Hindu newspaper group was reelected president of the academy. High-profile Tamil Brahmans are not the only

Religion, Music, and Dance / 203

people with prominent public roles in the academy, but they are very well represented, as they have been since its earliest days.51 The academy in its formative years played the key role in “defining” and “consolidating” Carnatic music, a process with close parallels in Hindustani music.52 As Amanda Weidman explains in detail, the process of defining Carnatic music, which provoked dispute among leading protagonists, “depended on the selection, from a number of heterogeneous musical traditions, of particular sounds, performance conventions, and repertoire that would come to be identified with the indigenous ‘classical’ music tradition of South India.” Carnatic music had to be “traditionally” Indian, but it also had to be “classical,” and hence modeled on Western classical music “with its notation, composers, compositions, conservatories, and concerts.”53 This modeling had many different aspects. For example, the fiddle, a European instrument brought to south India to play dancing and marching tunes, was redefined as the classical violin and then became a key instrument in Carnatic music. One man associated with the academy was C. Subrahmanya Ayyar—C. V. Raman’s brother and S. Chandrasekhara’s father—who was an accomplished amateur violinist and author of works on Carnatic music and the violin; Raman also did some research on the acoustics of musical instruments, including the violin.54 Although the distinguished scientists in his family make it very exceptional, Subrahmanya Ayyar’s own biography illustrates well how Carnatic music’s development was linked to the rise of the new Tamil Brahman professional class in Madras. As we saw in chapter 2, Subrahmanya, like Raman, joined the government service as an accountant in 1907. On the whole, Subrahmanya Ayyar led a more conventional life than his brother as a prosperous, upper-middleclass Tamil Brahman. In 1902, when he was seventeen, he married an eleven-year-old girl, and the couple had ten children. In 1918, he set up his family home in Mylapore, although his career as a government accountant took him all over India. In 1940, Subrahmanya Ayyar retired and was free to spend time on music.55 In Carnatic music, pride of place is given to the voice and singers are preeminent in musical ensembles. Probably the most far-reaching of all modern developments in Carnatic music was that “the voice—the vocal nature of Indian music and its ties to the oral tradition—came to stand for [the] essential difference” between it and Western classical music.56 Moreover, this was a particular kind of voice which, among other features, depended on modern ideas of the self and interiority, and pertained to notions of female virtue and middle-class ideals.57 Women were encouraged to learn music

204 / Chapter Six

and especially to sing, in private and eventually in public. For many Brahmans in the urban elite and wider middle class, Carnatic music—as well as Bharatanatyam—was successfully classicized by its “transformation . . . into ‘arts’ fit for upper-caste, middle-class ‘family women’” in the period between the 1920s and 1940s.58 Classical music, as Weidman puts it, “was seen as the soundtrack for the modern marriage and the modern home.”59 The woman who most perfectly personified the music’s transformation was M. S. Subbulakshmi (1916–2004), almost universally lauded as India’s greatest female vocalist, whose voice was celebrated for its captivating “naturalness.”60 But that natural quality was perceived as gendered, so that by the 1940s, “the connection between women and music had ascended to the level of an assumption.”61 Subbulakshmi was a devada¯si’s daughter from Madurai, who married a prominent Brahman in Madras. Devada¯sis, the dancers whose art was transformed into Bharatanatyam, have been consistently denigrated, as we shall see shortly. Subbulakshmi’s biography therefore perfectly captures the progress of both Carnatic music and its female artists from a disreputable past to a modern, classical present that epitomizes middle-class Brahman social respectability and cultural refinement. The modern development of Carnatic music was of course concurrent with the rise of the non-Brahman Dravidian movement, which was one factor encouraging Brahmans to create a new cultural sphere in Madras. At the same time, the Dravidianists were challenging the claim that Carnatic music—with its repertoire of mostly Telugu and Sanskrit songs—could genuinely represent the culture of south India, particularly Tamilnadu.62 Thus music, like language and religion, became a politicized arena for competition between Brahmans and non-Brahmans, and some Brahmans took a firm stand on it. One of them was T. T. Krishnamachari, a judge’s son, who was a successful businessman, leading Congressman, and central government minister after Independence; he was also one of the Madras Music Academy’s founders. In 1941, he published an article stating that the Tamil Icai (“music”) movement, which had been launched in 1929, was more about politics than music, and those who favored Tamil songs knew nothing about musical art and aesthetics; indeed, Krishnamachari patronizingly wrote, “the tendency of the common man is to destroy anything that he cannot understand.”63 But the Tamil Icai movement did not undermine Carnatic music, let alone destroy it; rather, it gave rise to a distinction that grew sharper over time between “classical” music and “devotional” music, in which Tamil songs sung by non-Brahmans predominate, so that language and caste became salient for defining the two musical forms.64 Music, too, was never crucial for the cultural identity of non-Brahmans in the early

Religion, Music, and Dance / 205

twentieth century, whereas it was for Brahmans; to some extent, perhaps, as Subramanian suggests, music became still more vital to Brahmans when their political power faded away after Independence.65

Bharatanatyam The developments of Bharatanatyam and Carnatic music were closely linked. Devada¯sis, literally “slaves of god,” used to dance as part of the ritual in south Indian temples and have often been described in the literature as “temple dancers.” Davesh Soneji’s detailed study has shown, however, that many of them were courtesans and secular dance artists, so that a narrow focus on temple-based devada¯sis is misleading.66 In the Maratha court at Tanjore, for example, an elaborate, aesthetically innovative, dancing repertoire was developed and in the late nineteenth century, when many Tanjore dancers moved to Madras, courtly dance forms became the basis for salon dances (or “nautches”) in the city.67 Soneji convincingly portrays the rich artistic life of south Indian devada¯sis, past and present, which has been obscured by their denigration as prostitutes whose dancing was little more than crude sexual display. The long campaign to reform the “disreputable” devada¯sis culminated in the Madras Devadasis (Prevention of Dedication) Act of 1947. For our discussion of Bharatanatyam and Tamil Brahmans, however, the distortion of the devada¯sis’ story is actually more salient. Devada¯sis were all non-Brahmans and, before the 1920s, the only women performing Carnatic music in public also belonged to the devada¯si community. To create both classical music and dance, and to make them accessible to and respectable for middle-class women, especially Brahmans, it was necessary “to divest the devadasi of her artistic heritage,” in Subramanian’s words.68 This was achieved in the twenties and thirties, with the support of the Madras Music Academy, by reforming and reinventing the devada¯sis’ “immoral” arts as Bhar­atanatyam. Two Tamil Brahmans, E. Krishna Iyer (a lawyer) and Rukmini Devi Arundale, played influential roles in this process. In 1932, challenging the campaign against devada¯sis and their art, Krishna Iyer defended the aesthetic value of dancing and successfully tabled a resolution in support of dance performance and patronage at the academy. Thereafter the academy sponsored dance performances and by the late 1930s, according to Janet O’Shea, “adolescent Brahman girls were appearing on the academy’s stage. This foray of high-caste, middle-class young women into performance and its support by the Music Academy . . . affirmed the respectability of bharata natyam.”69

206 / Chapter Six

Rukmini Devi (1904–86) gave her debut performance in Madras in 1935. One year later, she founded the Kalakshetra Foundation, now located near Besant Nagar, which became the leading center for teaching Bharatanatyam. Rukmini Devi was trained by older devada¯sis and their “dance masters,” but she strove to eliminate erotic display from dancing, which was reinterpreted as a devotional art form rooted in ancient Sanskrit tradition. Partly because she was herself a respectable, middle-class Brahman, Rukmini Devi successfully convinced a wider public that Bharatanatyam was a classical, spiritual, legitimate style of dance that all honorable women could practice.70 Reinvented Bharatanatyam was not entirely Brahmanical and Sanskritic, however. Tanjore Balasaraswati (1918–84), for instance, was a leading dancer who did not repudiate her devada¯si origins. She gave her first performance at the music academy in 1933, forged a longstanding relationship with it, and had the firm support of key figures like V. Raghavan. Yet, writes O’Shea, “Balasaraswati’s perspective did not fit easily with the largely elite nationalism of the Music Academy.”71 Balasaraswati defended the devada¯sis’ artistic integrity and traced Bharatanatyam’s origins to their community heritage, rather than Sanskrit texts, and she upheld the erotic, devotional style that Rukmini Devi sought to eradicate.72 Balasaraswati’s partner from the 1930s was R. K. Shanmugham Chetty, a powerful non-Brahman businessman and politician, who was also a founder of the Tamil Icai Sangam (Tamil Music Association); despite her connection with the academy, Balasaraswati shared his interest in the Tamil musical tradition.73 Among Bharatanatyam dancers and enthusiasts, Balasaraswati is still a revered figure. In India during the forties, however, she succumbed to Rukmini Devi’s growing influence, so that she was very publicly relegated to second place at a major festival in Bombay in 1945. After the 1947 Devadasis Act, moreover, Balasaraswati’s opportunities to perform, as well as her audience and income, went into decline.74 In developing the style of Bharatanatyam now mainly performed in Tamilnadu, therefore, Rukmini Devi prevailed. “Brahman taste,” Soneji argues, has configured Bharatanatyam through an aesthetic of refined control that contrasts with the looser bodily style, unstructured improvisation, and more explicit eroticism of devada¯si dance.75 Rukmini Devi was crucial, too, in ensuring that, by the late 1940s, dance, like music, “had not only ceased to threaten feminine propriety, it had come to confirm it.”76 The new dispensation took time to prevail, but by the 1950s, Brahman women were more often performing in public as dancers, as well as singers, than they had ever done before. By then, too, as chapter 4 showed, prepuberty marriage had largely disappeared and growing numbers of Brahman girls were in school

Religion, Music, and Dance / 207

or college. For more and more middle-class Brahman men, the ideal bride would be well educated, but she would also be able to sing or dance.

Music, Dance, and Contemporary Brahmans Data on the caste of professional concert musicians in Chennai show that between 1928–29 and 1976–77 the proportion of Brahmans remained constant at around 65 percent and the proportion of women at around 20 percent. About three-quarters of the preeminent vocalists were Brahmans; among instrumentalists, almost all veena (Indian lute) players and most violinists were Brahmans. The majority of women were vocalists; the remainder mainly played the veena or sometimes the violin. Among amateur musicians, women made up about 40 percent of the total.77 More recent data are unavailable and would now be almost impossible to gather, although well over 20 percent of professionals must be women today. Whatever the true figures, it is widely assumed in Chennai—by both the general public and knowledgeable experts—that Brahmans overwhelmingly dominate Carnatic music, especially its upper echelons, as well as Bharatanatyam to a slightly lesser extent. The late R. Ramachandran, for instance, founded Hamsadhwani, one of Chennai’s major Carnatic music associations, which organizes an annual winter festival for nonresident Indian musicians; speaking out of a long experience, Ramachandran regretfully told us in 2003 that he could not involve any non-Brahmans in Hamsadhwani, because they just assumed it had nothing to do with them. The Brahman musicians who perform in Chennai, it should be noted, include a sizable minority of Kannada and Telugu Brahmans, alongside the Tamil Brahmans; many of the non-Brahmans, notably Kerala Nayars, come from outside Tamilnadu. Irrespective of caste background, however, who is invited to perform on a public stage always has a lot to do with money (artists’ fees and expenses, or lack of them, and sponsors’ resources), patronage networks, and competition among sabha¯s, all of which attracts severe condemnation from some critics, who also deplore the impact on the quality of the winter festival.78 Carnatic musicians and Bharatanatyam dancers with established artistic reputations enjoy high prestige in Tamilnadu and those who can claim a link with them like to do so. One of the most famous Carnatic vocalists of the last century was Semmangudi Srinivasa Iyer (1908–2003), who was a nephew of Tirukodikaval Krishna Iyer (1857–1913), a leading early violinist. Since Semmangudi belonged to the Eighteen-Village Vattima subcaste, Vattimas constantly mention his name when listing their community’s achievements. One man was proud to present us with a family tree

208 / Chapter Six

showing how he is related to Semmangudi and several other people also mentioned their family connection to these Vattima musicians. Among the Tamil Brahmans we have met in Chennai and elsewhere, a considerable number of women said that they or their daughters have learned to sing, play an instrument (usually the veena), or dance. In our research, we met only one woman who is a professional musician—a vocalist in her twenties now pursuing her career in Chennai—although Savitri used to sing on the radio in Bangalore, and some others have achieved a high enough standard in music or dance to perform in public. Many, of course, have not become so proficient, but it is more or less taken for granted in middle-class Tamil Brahman families that girls should learn Carnatic music or Bharatanatyam, not least to improve their marriage prospects—just as middle-class Maharashtrian girls should learn Hindustani classical music or Bengali girls Rabindranath Tagore’s songs. For a middle-class girl in Mumbai, therefore, as Janaki Bakhle experienced as a teenager, “music and dance were somewhat like finishing school,” because they were “important markers of modern refinement that aided one’s chances in the marriage circuit.”79 And a well-bred, middle-class Bengali girl, who should ideally have some skills in fine arts and dance, is customarily asked to sing one of Tagore’s songs when a prospective bridegroom first sees her.80 On a smaller scale and without affecting marriage prospects, Western ideas about middle-class girls and classical ballet are not so different from those found in India.81 The first public performance as a Carnatic musician or Bharatanatyam dancer is a major rite of passage, on which parents often spend a lot of money for their daughters. This is especially so for a dancer’s debut, the lavish aran˙getram ceremony that usually presents a girl “to an audience consisting of friends, family, neighbors, and business associates of her parents.”82 The cost is extremely high for Brahmans living in America, who choose to hold their daughters’ debuts in Chennai. In America, too, the religious dimension of Carnatic music and Bharatanatyam tends to be very strongly emphasized, and both arts have become markers of Indian ethnic minority identity, which has raised some concern about the parental and community pressure exerted on girls to learn them.83 In spite of the cultural value accorded to music and dance by Tamil Brahmans, many individuals, predictably enough, are not particularly keen on them. We did not conduct a detailed survey of leisure interests, but when people were questioned, they mentioned visiting temples—or watching sport or soap operas on television—more often than listening to Carnatic music, or attending concerts or dance performances. Even among middleclass Brahmans, Tamil “devotional” and film music may actually be more

Religion, Music, and Dance / 209

popular than classical music. Nonetheless, numerous Brahmans did express a liking for Carnatic music and said they sometimes attended concerts, especially during the winter season, and some are serious enthusiasts, as Caplan also found in the 1970s.84 Plainly, the high prices and long queues for admission to many concerts in Chennai’s annual festival are one indication that there is a large, general audience for the music, which includes both Brahmans and non-Brahmans; the press, particularly the Hindu, also gives extensive coverage to the festival. Television broadcasts of music and dancing, which are more frequent now than in the past, are widely watched as well. Non-Brahman Carnatic musicians periodically appear on TV shows, even though they are sparsely represented among professional concert performers. But irrespective of how many people from different castes actually play music or dance—or take an enthusiastic interest in them—the received wisdom that Brahman artists are the most numerous and distinguished en­ ables all Tamil Brahmans both to claim Carnatic music and Bharatanatyam as their own esteemed cultural preserve, and to interpret their “virtuosity in the arts . . . as a sign of moral and cultural eminence.”85

Conclusion The Brahmans’ preeminent role as custodians of Sanskritic Hinduism has been significant for their caste status from ancient times to the present day. During the twentieth century, Carnatic music and Bharatanatyam, which both have a strongly religious ethos, were largely fashioned into their current forms by Brahmans, who dominate the two arts. One outcome is that religion, music, and dance are all vital components of the Brahmanical, Sanskritic great tradition as conceptualized by contemporary Tamil Brahmans. Most top Carnatic vocalists and many musicians, as well as Bharatanatyam dancers, are women today, as we have seen, and middle-class Brahman girls are also expected to learn music or dance. Until the mid-twentieth century, however, it was shameful for women and girls to perform in public, so that attitudes have changed completely. Sumathi Ramaswamy, describing her Tamil Brahman childhood in Delhi, says that she began Carnatic music classes when she was seven, just like “so many young girls who grow up in post-Independence India in bourgeois families burdened with the task of preserving ‘Indian tradition’ even while aspiring to be ‘modern’ and ‘Westernized.’”86 Religious devotion is desirable in both daughters and sons, but daughters are expected to take far more responsibility for preserving cultural traditions. For Tamil Brahmans—like members of some other high-caste groups—an improvement in the position of women has been accompanied

210 / Chapter Six

by a growing expectation that girls should learn classical music or dance, so that their investment in “traditional,” Sanskritic high culture actually becomes, like their good secular education, a sign of the whole caste’s urban, middle-class modernity. Social class, as well as caste, is indeed critical in this context, because classical music and dance—like Brahmanical, Sanskritic religion itself in many respects—are status markers of the middle class, particularly the upper-middle class. Hence, as Waghorne puts it, Brahman practices can be a “key measure of classy behavior.”87 Middle-class non-Brahmans are often attracted to Brahmanical religion—like several students in Savitri’s ´sloka class—and sometimes, too, to classical music and dance. In effect, these non-Brahmans are “Sanskritizing” themselves, but as part of a process of upward mobility within the class system, rather than the caste system. Moreover, in contemporary Tamilnadu, middle-class non-Brahmans can dissociate Brahmanical, Sanskritic religion and culture from the Brahman caste, so that they can hold them in esteem without acknowledging the social superiority of Brahmans. Respect for the great tradition, in other words, need not be coupled with deferential respect for Brahmans. Yet for most Tamil Brahmans, the two kinds of respect ought to go hand in hand. And in their eyes, Sanskritizing non-Brahmans, however they may themselves look at it, are actually endorsing the Brahmans’ superiority as expressed by both their ancient religious prerogatives and their modern preeminence in Carnatic music and Bharatanatyam. Hence the great tradition of Sanskritic Hinduism, and classical music and dance, counts for a great deal in sustaining the Tamil Brahmans’ collective identity and selfesteem—in Chennai, elsewhere in India, or overseas as part of a South Asian ethnic minority. At the same time, however, it also plays a key role in culturally constituting the Tamil Brahman upper-middle class as a distinctive status group within the wider, contemporary society.

Seven

Tamil Brahmans as a Middle-Class Caste

We have sought to describe and explain the Tamil Brahmans’ transformation from a mainly rural, traditional caste elite into a modern, urban middle class. Tamil Brahmans are well represented in the upper-middle class as educated professionals and managers, and we have mainly focused on them, although many Brahmans also have subordinate, white-collar jobs that place them in the lower-middle class. In general, however, all Tamil Brahmans today regard themselves as urbanites and members of the middle class, so that their caste and class statuses are intertwined and Tamil Brahmanhood is congruent with middle classness. As our discussion of education and employment particularly has shown, Tamil Brahmans joined the urban middle class in colonial Madras earlier and faster than members of other castes and communities; moreover, their disproportionately large share of professional and managerial employment, particularly in the postcolonial period, has primarily been maintained through a process of upper-middle-class reproduction, rather than the mere persistence of old caste privileges. During the twentieth century, child marriage ended and girls became better educated, so that the position of women steadily improved, which is—in their own eyes—a key sign of the Tamil Brahmans’ modern, middle-class status. Yet it is also important for Tamil Brahmans that they uphold ancestral religious and cultural values as custodians of the great tradition of Sanskritic Hinduism, and classical, south Indian music and dance. The caste’s relationship with this great tradition is highly salient for defining the ideal style of life of the Tamil Brahman, upper-middle-class status group. Plainly, Tamil Brahmans can be aptly described as both modern and traditional. But that well-worn description—which would equally apply to countless other social groups in India and the rest of the world—explains little by itself. We therefore look at the Tamil Brahmans in a comparative

212 / Chapter Seven

context to see what is distinctive about both their modernity and their middle-class status.

Colonial Bengal, Bombay, and Madras The histories of the three provinces of Bengal, Bombay, and Madras in the nineteenth century differed in ways that matter for understanding the Tamil Brahmans. In Bengal, the new middle class emergent under British rule styled itself the bhadralok, the gentry or “respectable people,” and its principal constituents were three Bengali high castes, Brahmans, Baidyas, and Kayasthas. Status differences among these three castes, and between them and the lower castes, were important; for example, like Tamil Brahmans, the Bengali high castes practiced prepuberty marriage and prohibited widow remarriage, which were primary markers of their superiority to other castes. Yet in Bengali society overall, caste divisions were less rigid and weighty than in the south, even before the distinction between Brahmans and nonBrahmans in Madras hardened during the colonial period. Moreover, for the bhadralok, a prestigious, refined culture based on education, literary and artistic skills, and mastery of the Bengali language, counted for more than caste status itself for their social dominance in Bengal. Calcutta, the capital of both British India (until 1911) and Bengal, was a large and prosperous commercial city, with a total population (including its suburbs) of around 790,000 in 1881. In that year, Brahmans, Baidyas, and Kayasthas made up 12.8 percent of Bengal’s total Hindu population of 17 million, but in the Calcutta municipality more than one-quarter of the Hindu population belonged to the bhadralok, so that the city “was undoubtedly the centre of Bengali bhadralok civilisation.”1 By the late nineteenth century, the bhadralok’s income from land had declined substantially and increasingly depended instead on jobs in the colonial system as administrators, lawyers, teachers, and clerks, and on other forms of professional, managerial, and white-collar employment. Brahmans, Baidyas, and Kayasthas were better educated than the rest of the population and they dominated higher and professional education; in 1883–84, for example, 85 percent of college students in Bengal belonged to the bhadralok castes and their literacy, in both Bengali and En­ glish, was well ahead of other communities.2 Broadly speaking, the bhadralok was therefore dominant in Bengal much like the Brahmans in Madras, but it was not made up of only one caste group and formed a substantially larger minority of the general population. In Maharashtra, the heartland of the old Bombay Presidency, Brahmans were the socially superior, dominant caste in many regions, but non-Brahman

Tamil Brahmans as a Middle-Class Caste / 213

hostility to them had a long history. In the whole province in 1881, Brahmans were 5.5 percent of the population; the two largest Maharashtrian Brahman castes were the Deshasthas, followed by the Chitpavans, who had risen to power under the precolonial Peshwa rulers and displaced the Deshasthas from their primary position. In 1881, Bombay, the province’s capital city, had a population of around 773,000 and was rapidly developing as a major industrial and commercial center. Especially in commerce, Gujaratis were the leading group, although the tiny Parsi community was highly successful and better educated even than the Brahmans. During the second half of the nineteenth century, Chitpavans in particular, but also Deshashtas and other Brahmans, migrated to Bombay and entered its educational institutions in large numbers. They then found employment in administration, law, teaching, and professions such as medicine and engineering, as well as in managerial positions and lower-level, white-collar jobs. In higher education, Brahmans were preponderant; in 1882, in the Brahman-dominated city of Poona, 93 percent of Hindu students in the elite Deccan College were Brahmans, although in Elphinstone College in Bombay the proportion was only 46 percent—far higher than their share in the general population, but low compared with the student body in Poona or, of course, in Madras.3 Furthermore, mainly because several different social groups competed for primacy in Bombay, the Chitpavans and Deshasthas “could never monopolise classroom and college as the bhadralok could in Calcutta.”4 In education and employment in Bombay and its capital city, Maharashtra’s Brahmans were in the lead, but not as decisively as their peers in Madras, and no homogenous group dominated Bombay city like the bhadralok in Calcutta. In many respects, though, the three provinces differed most in their contrasting paths of social, religious, and political reform, and their intellectual controversies. Rammohan Roy was the first great figure in the Bengali Re­ naissance, which marked the beginnings of Indian reform and nationalism in response to the impact of the West. Roy founded the monotheistic religious reform movement known as the Brahmo Samaj in 1828, partly in response to Christian criticism of Hindu polytheism, and the Brahmos became extremely influential among the bhadralok during the rest of the century, although they were strongly opposed by orthodox conservatives. There were also many other new religious developments; particularly inspirational for the bhadralok in the late nineteenth century was the mystic Ramakrishna Paramahamsa, whose principal follower, Swami Vivekananda, founded the Ramakrishna Mission in 1897. Religious reform was closely aligned with social reform—for example, in relation to child marriage and widow remarriage—

214 / Chapter Seven

and religious conservatives were usually socially conservative as well. Political nationalism was also gathering strength among the bhadralok, so that, even in the 1870s, a “militant and defiant mood was certainly one face of Bengali nationalism.”5 Running through and across all these social, religious, and political currents were vigorous intellectual debates about Bengali, Hindu, and Indian society and culture, the revitalization of their traditions and the evaluation of European modernity, the importance of science and rationality, the significance of reason in constituting the modern subject and the individual in society, and many other related issues. All these debates, which in Bengal sometimes displayed an obsessive and extreme reaction to the West compared with other regions of India, were stimulated both by intellectual excitement and, as Tapan Raychaudhuri explains, by an anxiety to understand Europe, to compare its civilization with India’s, and to decide which of its features should be accepted and which rejected.6 In Bengal, too, as Dipesh Chakrabarty demonstrates, the nineteenth-century discussion of widow remarriage generated intensive reflection on women’s suffering and the human capacity for compassion. Much of this took place through the new medium of the novel and it progressively led to an understanding of the modern subject that fused European ideas with, in particular, Bengali idealizations of kinship and extended patriarchal families, so that the “Bengali modern self was not quite the bourgeois modern self of Europe.”7 In Maharashtra in the mid-nineteenth century, two of the earliest reform movements, which were led by Gopal Hari Deshmukh and Jotirao Phule, attacked the iniquities of caste and Brahman dominance. Thus antiBrahmanism was not new in Maharashtra, but later, in 1920, a NonBrahman Association resembling Madras’s Justice Party was established. An influential religious and social reform movement, the Prarthana Samaj, was founded in 1867; although modeled after the Brahmo Samaj, it emphasized its connection with Maharashtrian Hindu traditions and eschewed the Bengali movement’s hostility to conservative orthodoxy. One of the Prarthana Samaj’s members was Mahadev Govind Ranade, a Chitpavan Brahman, who was a leading moderate nationalist and reformer involved in establishing Congress in 1885. Two other key figures at the turn of the century, also Chitpavans, were Gopal Krishna Gokhale, the liberal and Congress “moderate,” and Bal Gangadhar Tilak, the “extremist” advocate of radical Hindu, Maharashtrian nationalism. In Maharashtra, like Bengal, especially among educated Brahmans, there was also a significant intellectual ferment, as D. D. Karve, son of Dhondo Keshar Karve, a prominent Chitpavan reformer, explains. The “new Brahmans,” as Karve calls them, were attracted by West-

Tamil Brahmans as a Middle-Class Caste / 215

ern ideas of liberty and equality, and women’s emancipation, for example, but “they did not entirely abandon their heritage,” as the “tremendous popularity of Shivaji as a symbol of resistance to foreign rule in Maharashtra” demonstrates.8 This outline merely scratches the surface of Bengali and Maharashtrian reformism, nationalism, and intellectual controversy, but it serves to highlight the contrast with the more limited developments in the south. The Hindu Literary Society, which had both Brahman and non-Brahman members, was active in Madras city in the 1830s and 1840s. In 1852, a branch of the Bengali-led British Indian Association was set up, but later it was independently reorganized as the Madras Native Association, which was dominated by non-Brahman merchants and sought to represent Indian interests to the government, although it lasted only ten years. In both the society and the association, in addition to criticism of Christian missionary activity, there was some lively, but short-lived, discussion about social reform and the impact of colonialism.9 In the later nineteenth century, Madras’s new print media provided space for debates about child marriage, female education, and the position of women, as well as about other issues such as the “modernization” of Tamil language and literature; the Tamil novels emerging in this period played a vital role in these debates. The notable Telugu Brahman reformer, Kandukuri Viresalingam, was also active in Madras in the late nineteenth century. Sascha Ebeling justifiably rejects the notion that the nineteenth century was a stagnant, “dark” period in Tamil society and culture.10 Hence it would be an overstatement to say that there was no battle of ideas in south India. In some areas, however, such as Vizianagaram in northern Andhra, Velcheru Narayana Rao tells us that Viresalingam’s reformism had little sway, colonialism’s impact was relatively weak, and the distinctive forms of south India’s precolonial modernity retained much of their vigor.11 Moreover, in Madras and other urban centers, even the new middle class—which, according to Narayana Rao, “forgot its immediate past in favor of colonial modernity, which led it to devalue tradition”—did not engage in debates as intensive as those in Bengal or Bombay.12 Thus Heimsath exaggerated only slightly in saying that: “No Madrasi formulated the kind of intellectual and spiritual responses to Western thought which were occupying the attention of certain Bengalis, Gujaratis, and Maharashtrians.”13 A lot changed after 1900, but before then there were few signs in Madras of the vibrant reformism and nationalism emergent in colonial Bengal and Bombay; little of the innovatory literature, intellectual ferment, and elite anxiety that had characterized the life of the bhadralok, and some “new Brahmans” in Maharashtra,

216 / Chapter Seven

for many years; and next to none of the anti-Brahmanism that had been part of Maharashtrian social protest for a long time. As we have also seen, the small Tamil Brahman minority became dominant in Madras to an extent unparalleled either by Bengali Brahmans within Calcutta’s larger bhadralok group, or by Maharashtrian Brahmans within the more heterogeneous middle class of Maharashtra, especially in Bombay city. Indeed, the Tamil Brahmans—once they had displaced the Deshastha Brahmans in the south in the second half of the nineteenth century—had a more or less clear run into Madras’s middle class until the non-Brahman movement got under way, unhampered by competition from rival groups (except Telugu Brahmans), reform movements, or subversive new ideas. It is, incidentally, very unclear why the division between Smartas and Sri Vaishnavas faded in importance and did not impede Tamil Brahman progress, whereas (to speak counterfactually) it might have become a new focus for status rivalry within the middle class. The comparison between Tamil and Telugu Brahmans is instructive, although there is hardly any relevant literature on the latter, so that our discussion owes much to the knowledge of Narayana Rao.14 Telugu Brahmans, as explained in chapter 1, are divided into two major sections: the smaller, “orthodox” Vaidika and the larger, “secular” Niyogi. Until fairly recently, the vast majority of Telugu Brahmans taking up modern education and employment have always been Niyogis, who regard the Vaidikas as inferior. In the past in coastal Andhra, as well as in the Tamil country where many were settled, Niyogis were mostly landlords, although they were also well represented among literate service people. In the nineteenth century, many rural Niyogis migrated to towns and cities and, in the Madras Presidency around 1900, Telugu Brahmans were only slightly behind Tamil Brahmans in the educational and employment fields. Madras was the capital city for Telugus, as well as Tamils, but Tamil Brahmans probably benefited from their greater concentration in the Tamil districts forming Madras’s extensive hinterland to the south and west. Compared with Tamil Brahmans, fewer Telugu Brahmans went into medicine and engineering, and even in administration and law they tended to lose out, especially in Madras city, to the more numerous, better organized Tamil Brahmans, who favored their own in patronage networks. In addition, Tamil Brahmans actually occupied many senior government positions in the Andhra country and increasingly saw themselves as superior to the Telugus. It is also salient that because Tamil Brahmans are not split into orthodox and secular sections, those who belong to priests’ families, or are just conservatively religious, can take up modern education and employment with-

Tamil Brahmans as a Middle-Class Caste / 217

out blatantly transgressing caste traditions, in contrast to Telugu Vaidikas, who sometimes bitterly opposed any deviation from the Brahman dharma, like the Sanskrit pandit cited in chapter 1.15 Furthermore, the orthodoxsecular division may partly explain why Telugu Brahmans tend to have a weaker sense of their own identity outside Andhra than Tamil Brahmans do outside Tamilnadu. But probably more significant is the course of the non-Brahman movement in coastal Andhra, which was strong in the early twentieth century, when many leading figures in the Justice Party were Tel­ ugus. By the 1940s, however, the movement was losing impetus in Andhra, partly because it was becoming a Dravidian movement completely focused on Tamil and Tamilnadu. That in turn meant that Telugu Brahman identity was not continually reinforced by insistent, hortative claims that they were alien Aryans from the north.16 After Independence and the formation of Andhra Pradesh in 1956— partly the outcome of Telugu resentment about Tamil dominance—the Telugu Brahmans’ position actually worsened because coastal Andhra had no large city of its own. Telugu Brahman urban migrants therefore had to leave their own region, but the closest major cities apart from Madras were Bangalore, where Kannada and Tamil Brahmans were numerous and powerful, and Hyderabad, Andhra’s capital, which was dominated by Muslims. Those who went to these cities often prospered, but Telugu Brahmans appear to have been less successful than Tamil Brahmans, in the long run, in entering the urban, professional and managerial, upper-middle class.

The Brahmans’ Response to Modernity Modernity, of course, has multiple aspects, referents, and meanings. Not all of them are necessarily relevant to the experience of Tamil Brahmans during the colonial and postcolonial periods. But those which plainly are include English education, especially at higher and professional levels, which differed in both content and pedagogy from traditional Indian education; the colonial government’s bureaucratic and legal systems, which in theory, though often not in practice, were organized by new principles of rationality and equity; Western science and technology as systems of knowledge and fields of practical activity, notably in medicine, engineering, and (in the long run) information technology; new ideas about the status of women and the low castes, often inspired by European and Christian ideas, which challenged orthodox, hierarchical norms; new concepts of the self, the individual, and the citizen; the development of nationalism and the Indian nation-state, as well as representative and democratic political systems that

218 / Chapter Seven

undermine elite minorities; the expansion of capitalism and the market economy, in both the agrarian and new industrial sectors, and the growth of the modern city; and—last but not least in this list—social, cultural, and religious reform, which commonly takes the guise of revivalist traditionalism, but is nevertheless modern in many respects. At the turn of the twentieth century, the temperature of debate about the impact of colonialism and modernity started to rise in Madras. M. S. S. Pandian discusses this change in relation to Tamil Brahmans who “had now to reconstitute themselves . . . by responding to the dual demands made on their selfhood by colonialism, i.e. to be historically authentic and modern at once.”17 For example, M. K. Acharya, addressing a meeting of Brahmans in Madras in 1921, deplored the community’s decline and wanted Brahmans “to give up their present pursuits and go back to the performance of those duties prescribed by Vedas and Upanishads.” The meeting also passed a resolution in favor of persuading Brahman youths “to seek employment only in such professions as will not interfere with their proper performance of Brahmana-Dharma,” which echoed nineteenth-century Bengali complaints that the time regimes of salaried work severely disrupted the daily rituals ordained by the dharma for upper-caste men.18 Acharya symbolized the “anxiety-ridden” figure of the English-educated Brahman and exemplified the dilemma of trying to be both traditional and modern in everyday life in the colonial milieu. One strategy to resolve this dilemma, argues Pandian, was “to present extant Brahminical practices as always already modern,” mainly by claiming that they were really scientific because, for instance, ritual pollution is about maintaining hygiene.19 A second and more potent strategy was religious reform, whereby a newly revitalized Brahmanical, Sanskritic Hinduism could be represented as “the essence of the Indian nation.”20 In both these strategies, the Theosophical Society, which located its headquarters in Madras in 1883, had a major role in promoting the idea that a purified Brahmanical Hinduism was both “scientific” and the authentic expression of Indian spirituality in opposition to Western materialism. Annie Besant, who became the society’s president in 1907, was active in the Indian nationalist movement and also close to leading Brahman figures in the powerful Mylapore group. In Theosophy, Brahmanical Hinduism and Indian nationalism formed a potent nexus, which helped to create the Tamil Brahman as a “colonial hybrid,” in Pandian’s phrase, who “metonymically equated his culture with that of the nation and simultaneously embraced the protocols of colonial modernity” to access power and material resources.21 Consequently, religious reform, including the claim that Hinduism is scientific, tended to reinforce the distinction between Brah-

Tamil Brahmans as a Middle-Class Caste / 219

mans and non-Brahmans in Tamilnadu, although it closely resembled its earlier counterparts in Bengal, Maharashtra, and elsewhere in north India in many other respects. Clearly, too, much of the anxiety felt by Acharya and other Brahmans was fueled by a desire to maintain their superiority over other castes, as exemplified by Acharya’s vehement defense of prepuberty marriage during the Sarda Act debate in 1929 (see chap. 4). On the other hand, even by the standards of his day, Acharya was probably unusually conservative, for trying to persuade ordinary Brahman young men to give up “present pursuits” and return to Vedic duties in 1921 was obviously a lost cause. In fact, despite the worries of men like Acharya, the majority of Tamil Brahmans seem not to have been plagued by anxiety about their place in the world in the early twentieth century. V. Geetha and S. V. Rajadurai are hostile to what they decry as Brahmanism, but they are most likely right to say that the “Madras elites,” overwhelmingly Brahman, were “self-assured in their historic (and spiritual) patriotic mission and complacent in their sterling public virtues,” which is why they came to “mistrust and treat with lofty contempt the Non-Brahmin Manifesto [of 1916] which was openly critical and scornful of that mission.”22 It is consistent with that self-assurance and complacency that, as Washbrook argues, “a striking feature of modern Brahmin culture in the South is how little modernity continued to be resented after it had been accommodated—in sharp contrast, say, to the angst of the Kolkata [Calcutta] bhadralok.”23 Evidence about ordinary Brahmans’ states of mind is admittedly thin, but the available information about how, for example, Brahmans adapted to urban life, and took up the new professions of medicine and engineering, supports Washbrook’s argument. Compared with the new middle classes of Bengal and Bombay, including of course their Brahman sections, Tamil Brahmans belatedly and perhaps rather placidly embraced modernity—or at least its desirable and advantageous aspects—but they did so relatively easily and without perceiving any serious damage to their religious beliefs and practices, and their traditions as a whole. In south India, Tamil Brahmans contrast most extremely with the Nambudiri Brahmans of Kerala, who turned their back on all modern change, so that most of them “withdrew haughtily to their estates, and refused to allow their sons and daughters any form of modern education,” a situation that did not change until the mid-twentieth century.24 Among Telugu Brahmans, as already implied, Niyogis generally adjusted to modern change much like Tamil Brahmans, whereas Vaidikas resisted it. In Mysore princely state in the early twentieth century—where Mysorean Kannada Brahmans challenged

220 / Chapter Seven

the immigrant Tamil Brahmans’ dominance of government posts—disputes arose between “progressive” and “orthodox” Brahmans over the reform of old customs, notably child marriage, although the “progressive” outlook mainly won out in the end, as T. S. Ramesh Bairy shows.25 Today, Brahmans in Karnataka, as in Tamilnadu, are an overwhelmingly urban population strongly represented in the occupational middle class.26 Yet, Ramesh Bairy argues, the forces of modernity have been consistently problematic for Kannada Brahmans, and they have become increasingly equivocal about their own caste identity. Hence the Brahman, in his or her own eyes, is a “self under siege” facing a “crisis of legitimacy.” Brahmans also resent being “branded” as Brahmans, whether they adhere to orthodox customs or not, within a social and political discourse dominated by Karnataka’s non-Brahman movement.27 Ramesh Bairy provides many quotations from his informants, mostly expressing regret or complaint about the Brahmans’ condition, to support his argument. Almost everything they say is familiar to us from conversations with Tamil Brahmans. The overall tone is different, however, for reasons that are unclear, and in our experience middle-class Tamil Brahmans hardly ever suffer from the crisis of identity described by Ramesh Bairy. In Tamilnadu after Independence, as explained in chapter 6, Chandrasekharendra Saraswati, the Shankaracharya of Kanchipuram dignified as the Paramacharya, persistently defended traditional Brahman values and warned against deviation from them. Nonetheless, the Shankaracharya was widely supported and revered by urban middle-class Brahmans, who combined their own inherited Brahmanical beliefs and practices with a modern outlook. When Singer wrote about Madras’s leading industrialists, including several Brahmans, in the 1960s, he explained their ability to combine the old and new through the concept of “compartmentalization,” defined as the conceptual and physical separation of traditional and modern spheres to reduce conflict between them, as symbolized, for example, by wearing different clothes at home and in the office.28 In his restudy of industrialists in Chennai nearly fifty years later, however, Harriss found that all his infor­ mants (one of whom had read Singer) rejected the notion that they engaged in compartmentalization.29 That does not prove that compartmentalization, as described by Singer, did not exist. Nevertheless, Srinivas, in discussing similar evidence, labelled it “contextual variation”; this formulation, which does not posit any clear-cut separation between traditional and modern, is more convincing than Singer’s, even though Srinivas too easily assumed that modifying behavior according to context would always normally lead to more complete secularization.30

Tamil Brahmans as a Middle-Class Caste / 221

Alternatively, we might regard contextual variation as a kind of codeswitching, which is displayed by easy movement across the often fuzzy boundary between traditional and modern practices and attitudes. In our experience, contextual variation or code-switching is characteristic of contemporary Tamil Brahmans. Certainly, there is concern about disappearing tradition, for example, in relation to those who have gone to America. Nonetheless, the majority of Tamil Brahmans—whether resident in Tamilnadu, or elsewhere in India or overseas—see themselves as both fully modern and authentically traditional. On the other hand, Tamil Brahmans in general have not become more fervently or ideologically traditionalist as well, unlike the Minakshi temple Brahman priests, whose interests are bound up with preserving hereditary rights and their Adishaiva subcaste’s religious legitimacy.31 Indian modernity, to paraphrase Gyan Prakash, is always accompanied by an “aura of dislocation and disorientation,” for modernity and tradition do not negate each other, but enable each other’s reformulation.32 As a general proposition, this is surely right. Yet dislocation and disorientation, as we understand them, are actually remarkably absent from the Tamil Brahmans’ certain sense of themselves as both modern and traditional. Their confident combination of modern ways and Brahmanical tradition may not make Tamil Brahmans unique, but it does contribute significantly to their own sense of personal and collective identity today.

The Tamil Brahman Middle Class in Comparative Perspective In the introduction, we explained that a neo-Weberian definition of “social class” would guide our discussion of the Tamil Brahmans. A key feature of Weber’s analysis, as further explicated by Giddens, is that economic classes, which group people in similar occupations together, become social classes through a process of structuration. In this process, classes tend to become status groups sharing a common style of life, which is seen by their members as distinctive; social mobility between classes is also relatively restricted. In modern capitalist societies, the middle class, which has continually expanded, has divided into a professional and managerial, upper-middle class, and a white-collar, lower-middle class. To a significant extent, the upper- and lower-middle classes have become separate social classes. This separation is evident in the Indian middle class, not least among Tamil Brahmans, who identify themselves as a middle-class caste, but also recognize upper and lower strata within it. Hence the making of the Tamil Brahman middle class has also been the making of a division within it; this book, as we have emphasized, is mainly about the Tamil Brahman

222 / Chapter Seven

upper-middle class. The occupational basis of this class, primarily professional and managerial employment normally requiring higher educational qualifications, was examined in chapters 2 and 3. Chapter 3 concluded by discussing the process of upper-middle-class reproduction. We have also examined, mostly in chapters 1 and 5, urban migration, urbanization, and urban ways of life, which are important features of a social class that is always also overwhelmingly urban. Crucial to the Tamil Brahman middle class, particularly the upper-middle class, is the decline in gender inequality and the improved position of women, especially in education, as well as the development of contemporary companionate marriage, which were all discussed in chapter 4. Chapter 6 looked at the importance of the Sanskritic great tradition for the Tamil Brahman upper-middle class in particular. In sum, the making of the Tamil Brahman middle class, and its internal division, can be understood as a process with a series of analytically distinguishable economic, social, and cultural components that are shaped by employment, education, gender, urbanism, religion, and “high” culture, to list only some of the main factors. In the introduction, we also briefly described the Indian middle class and its diversity, noted its connection with modernity, and pointed out that consumerism is not invariably important, as it is not, for example, to middleclass Tamil Brahmans valuing self-discipline and moderation. We drew attention, too, to their systematic tendency to equate traditional Brahman and modern middle-class values. This can be illustrated by the use of language, especially among upper-middle-class Tamil Brahmans, who quite often talk about “educated people” to refer to the “middle class,” but sometimes do so in order to refer to themselves without explicitly naming their own caste. Conversely, they may talk about “uneducated people” to refer to those who are poor, low-caste, or both. In fact, the Christians in Madras studied by Caplan sometimes spoke in this way as well (see the introduction), and such usages and euphemisms, whereby caste and class become figures of speech for each other, are common in modern India.33 But there can be more to it than euphemism. Thus a notable development in present-day Tamilnadu is rising standards of education, especially for girls, in non-Brahman peasant castes that have not previously attached much weight to education. When this is discussed, Brahmans—and indeed some non-Brahmans too—may say that the lower castes are becoming like Brahmans. Whether this comment is reasonable or not, it is significant that, in Brahman eyes, educational improvement simultaneously implies that lower-caste people, as part of their upward mobility into the middle class, are adopting superior Brahman values, much as if they were Sanskritizing themselves by

Tamil Brahmans as a Middle-Class Caste / 223

becoming vegetarian or learning ´slokas. Sanskritization, broadly conceived, can hence be a route to upward class mobility, rather than or in addition to upward caste mobility.34 By the same token, although non-Brahmans would disagree, middle-class values are regarded by Brahmans as paradigmatically their own, and that in turn is consistent with how the Tamil Brahman middle class—especially its upper stratum—has isomorphically developed as both a social class and a status group in Weberian terms. The caste-class isomorphism found in the Tamil Brahman upper-middle class is an unusual feature. Indeed, one of the closest parallels may be the Parsi community, especially the majority group living in Mumbai, which is described by Tanya Luhrmann as “urbanized, middle-class, Englishspeaking, urbane.”35 But the Parsis, who are of course famed for their business prowess and are not a true caste anyway, are exceptional in many respects, so that they do not really help us to assess how unusual Tamil Brahmans are. In fact, there is relatively little up-to-date information about closely comparable groups and most research on the middle class has not focused on specific castes. All the same, some tentative conclusions are possible, mainly in relation to other Brahman and non-Brahman castes that are or were landowning groups. We begin with Brahmans. Telugu and Kannada Brahmans have been mentioned already; like Tamil Brahmans, they now predominantly belong to the urban middle class and have left the land, but there appear to be some significant differences, including the Telugus’ lesser representation in the educated, professional upper-middle class, and the Kannadas’ apparent ambivalence about their own caste identity. Among Kannada Brahmans, the Havik (Havakya) subcaste is unusual because many of its members still live on the land in villages in coastal Karnataka, although growing numbers are migrating to urban areas for education and employment.36 One group of Kannada Brahmans, the Saraswat Brahmans from Kanara, had become increasingly urbanized and dependent upon urban employment by the 1930s, with Bombay as the main center of settlement, but no data exist to extend Conlon’s historical study closer to the present day.37 In Bengal, Brahmans are an overwhelmingly urban, middle-class group, but their social status and collective identity are mainly subsumed within the larger bhadralok, as we have seen, so that their caste per se matters less than it does for Tamil Brahmans. Among Maharashtrian Brahmans, the limited evidence suggests that the Chitpavans closely resemble Tamil Brahmans, particularly in their pursuit of education and employment both before and after Independence.38 On the other hand, because Chitpavans are one of several Maharashtrian Brahman groups—including their historic

224 / Chapter Seven

rivals, the Deshasthas—they are not the dominant Brahmans of their region like the Tamil Brahmans are in Tamilnadu. Another difference, which is linked to how the non-Brahman movement developed in Maharashtra, is that many Chitpavans have been core supporters of right-wing Hindu nationalism, strongly motivated by resentment over the loss of their old privileges, whereas Tamil Brahmans have typically reacted to their political eclipse by withdrawing from active politics. The populous Kanya-Kubja caste of Uttar Pradesh is one north Indian Brahman group on which there is useful material. As R. S. Khare shows, Kanya-Kubjas invested in English education and moved into urban, middleclass employment as bureaucrats, lawyers, doctors, engineers, and so on, as well as lower-level, white-collar staff, from the late nineteenth century; they also have a pattern of extensive migration comparable to that of  Tamil Brah­ mans.39 Although Khare discusses rural Kanya-Kubjas only briefly, his material indicates continuing connections with villages and the land.40 One such village is Karimpur, where the dominant landowners are Kanya-Kubjas; despite many changes between 1925 and 1998, including improved education and some urban employment, Karimpur’s Brahmans have not emigrated to urban centers on a large scale.41 Compared with Tamil Brahmans (or Kannada and Telugu Brahmans), Kanya-Kubjas are therefore less thoroughly urbanized, even though many of them have also joined the urban middle class, in both its upper and lower strata. The Kashmiri Brahmans or Pandits are very unusual, because one group of them served in the Mughal Empire’s bureaucracy, migrated to Muslim cities across north India, and adopted much of their Islamic court culture. Another Pandit group later served the Sikh rulers of Punjab in Lahore. In the colonial period, especially in Punjab, the Pandits became well educated and strongly represented in administration, law, and similar fields.42 MostPandits, however, always lived in the Kashmir Valley, mainly as rural landowners, although some became urban migrants.43 In the 1990s, threatened by Muslim insurgents, the vast majority of the valley’s Pandits left and went to camps in Jammu, or moved to Delhi and other centers. Historically, the Kashmiri Brahmans were exceptional because, unlike Tamil or other Brahmans, an entire branch of them assimilated to elite Islamic culture and, according to Henny Sender, “surrendered much of their tradition in the process, to an extent almost unimaginable for other migrant groups.”44 The closest parallels to them are the Kayasthas, a loosely structured caste of literate service people in north India, who also became Islamicized when serving Muslim rulers. One group of Kayasthas migrated south to Hyderabad and their history was studied by Karen Leonard, who found that their fortunes

Tamil Brahmans as a Middle-Class Caste / 225

during the colonial and postcolonial periods varied, although some became educated and found professional and white-collar employment, between the 1920s and 1960s.45 The ethnographic and historical evidence on different Brahman groups is far from comprehensive. Nevertheless, it mostly confirms the conventional wisdom about their early and extensive entry into the urban middle class, both the professional and managerial upper-middle class, and the white-collar, lower-middle class. There are Brahmans who remain on the land, like some Haviks and Kanya-Kubjas, but they are probably exceptional today. Turning now to non-Brahmans, in south India the high-ranking, landowning Nayars are interesting, partly because in Kerala it is they, not the Nambudiri Brahmans, who migrated to urban areas for modern education and employment. Nayars had a caste tradition of education and literacy and, in the early twentieth century, their male literacy rates were the third highest in the Madras Presidency, after the Tamil and Telugu Brahmans.46 Nayars went to Madras in greater numbers from British Malabar than Cochin or Travancore, whose rulers developed their own educational and administrative systems. Particularly in Travancore, though, Tamil Brahmans dominated the bureaucracy and the resulting Nayar resentment was one factor encouraging their entry into modern education, although competition with prospering Syrian Christians also played a role.47 In the long run, a large proportion of Nayars became educated professionals, managers, and whitecollar workers, and now belong to the urban middle class. Yet many still live in the countryside; thus in the early 1990s, 24 percent of households in a southern Kerala village were Nayar and among employed Nayar men, about a quarter worked as government servants, a quarter in agriculture, and a quarter in trade, while most of the rest had emigrated to other places in India or the Gulf.48 Particularly valuable for our comparative discussion are the nonBrahman, Gounder industrialists in the textile town of Tiruppur in western Tamilnadu, who are described by Geert De Neve.49 The Gounders are a large caste of peasant landholders in the region, and nearly all of Tiruppur’s successful entrepreneurs come from a rural background. Senior men take pride in their village roots, capacity for manual labor, simple style of life, and “traditional” values; they also readily acknowledge their negligible education. But they now know that things must change, because Gounders need to catch up with educated, high-caste communities and have to find new ways of being “respectable.” The second generation of industrialists, the senior men’s sons, have almost all been educated; by the 1980s they

226 / Chapter Seven

were usually graduates, by the 1990s postgraduate degrees were the norm, and by the 2000s study abroad was common. After a few years’ delay, girls from the industrialists’ families started to be better educated as well, so that most now have degrees, although none has studied abroad. Disapproval of women who go out to work is still widespread, but several young, educated women work in their own families’ businesses. Most marriages are still arranged between Gounders within the local area and kinship ties matter a great deal. De Neve shows that the Gounder industrialists are engaged in constructing a new middle-class identity for themselves, which involves a major departure from the old ways, because it depends on younger men and women becoming highly educated urbanites, as well as business people. Yet the development of the Gounder urban middle class remains incomplete—a work in progress, so to speak—compared with its Tamil Brahman equivalent. The same conclusion probably applies to the vast majority of mainly agrarian, non-Brahman communities in contemporary Tamilnadu, even though we lack the evidence to demonstrate this. In much of coastal Andhra, Kammas and Reddis are the dominant landowning castes. Carol Upadhya studied Kammas in the 1980s, when they were benefiting from the green revolution; many had become rich farmers, but others were businessmen and a minority were educated professionals. A lot of Kammas were migrating to towns, but there was considerable circulation between urban and rural areas, and the new rich, rural elite was also acquiring urban, middle-class aspirations and styles of life. English-medium education was becoming more desirable, the children of the rich were gaining college degrees, and graduate sons often sought urban employment or started businesses, whereas education for daughters was mainly about enhancing their marriage prospects. Kammas, therefore, were progressively joining the urban middle class in the 1980s, but many of them stayed in the countryside, creating a kind of urban lifestyle there. In recent years, investment in education and urbanization have continued, numerous men are now professionals and entrepreneurs in the IT sector, and the community is well known for its wealth, although many Kammas still retain close links with the land and rural society, unlike Tamil and other Brahmans.50 Rather similar to the Kamma case is a final example from north India. A group of rich Jat landowners in Bijnor, Uttar Pradesh, were studied by Roger and Patricia Jeffery, and Craig Jeffrey in the early 2000s. Their data show that rich Jats are trying to diversify out of agriculture by setting up businesses and by investing in education for their sons, who can then look for paid employment. Some young men do secure professional, managerial, or white-collar jobs and emigrate to urban centers, but many others fail to

Tamil Brahmans as a Middle-Class Caste / 227

do so and prolong their college education, partly to avoid facing unemployment or returning to work on the family farm. In raising the local Jats’ social status, suitable marriage alliances are also important, and the ideal brides are well-educated girls with no aspiration to work outside the home. The outcome for rich Jats whose strategies are successful is that they can join “the broad church of the Indian middle classes” by cultivating a modern, urban style of life within their rural homes; “they have actively acquired the accoutrements of urban and urbane middle-class life, even whilst constantly trying to keep the taint of the rustic and the rural at bay.”51 Gounders, Kammas and Jats—the three groups for which there is goodquality, directly comparable evidence—vary in many ways, of course, but they are all non-Brahman, dominant landowning castes, which are becoming urban and middle-class to different degrees. For each of these groups, unlike the majority of Brahmans wedded to salaried employment, business is equally or more important. Roughly speaking, the Tiruppur Gounders are now mainly urban, the Kammas straddle the urban-rural divide, and the rich Jats remain predominantly rural. In all three cases, though, the development of fully fledged, urban middle-class groups is only partial, mainly because these non-Brahman communities have not turned their back on land and village life like the Brahmans. That may change in the future, but at present the evidence points to two firm conclusions. First, Brahmans (together with unusual service castes, such as Kayasthas) have been more completely transformed than any non-Brahman agrarian castes into modern, urban, middle-class groups. Second, although it is impossible to say whether they are unique, Tamil Brahmans are extremely unusual in how fully they have been transformed into an urban middle-class caste, so that they now constitute a social class-cum-status group, internally divided into upper and lower strata, which is itself structured by an isomorphism between Tamil Brahmanhood and middle classness.

Conclusion In his study of Russian Jews, Yuri Slezkine shows that after they migrated to towns and cities, this small minority of the population became disproportionately well represented in education and professional employment, as well as banking and business. Russian Jews also became increasingly secularized and gave up many religious traditions.52 The outcome in the early twentieth century was that the “Jews were becoming modern faster and better than . . . anybody else in Russia.”53 More generally, Slezkine argues: “Modernity was about everyone becoming a service nomad: mobile,

228 / Chapter Seven

clever, articulate, occupationally flexible, and good at being a stranger,” so that the Jews “came to represent . . . modernity everywhere.”54 Slezkine’s contrast between modern service nomads and the traditional, dominant settled population is overdrawn and his bold assertion about Jewish modernity is obviously controversial. Nonetheless, the comparison between Jews and Tamil Brahmans is good to think with. This has nothing to do with the false claim by some Brahmans that they have been persecuted in Tamilnadu like Jews in Europe. Rather, it is because in many respects apart from their religiousness, the Brahmans fit Slezkine’s description well, not least in their own telling. Furthermore, as urban, educated professionals and managers— especially but not only in the IT sector—many of them have become service nomads in the global economy, who can be plausibly held up as the ideal workers in the globalized “knowledge” or “informational” economy that has developed at the core of the neo-liberal, capitalist world system. How Tamil Brahmans became representatives of modernity has been a central theme of this book. As we have seen, the Tamil Brahmans’ partial separation from village society and detached attitude to the land initially encouraged them to move to towns and cities, whose educational and economic opportunities were the main attraction. In the long run, the caste’s traditional “vocational ethic” of learning and education has clearly been a flexible asset in enabling and encouraging Brahmans, who once belonged to a landed elite, to diversify into a range of occupations in a modern economy that is “rational” in Weber’s terms. Yet there is also an important sense in which Tamil Brahmans were not just marginal in the villages, where they lived in their exclusive agraha¯rams, for they were and are marginal in Tamil society in its totality. Tamil Brahmans, to borrow an insight from Sudipta Kaviraj’s analysis of linguistic segmentation, belong like all Brahmans to a pan-Indian “self-recognizing elite,” which has a “superior caste-for-itself character,” unlike the lower castes, who are more “deeply segmented” regionally and linguistically.55 Their myth of origin always portrayed Tamil Brahmans as people who came from India’s northern heartland and, even though they are Tamil speakers, they have customarily regarded “vernacular” Tamil as inferior to Sanskrit, the classical language of all-Indian Hindu civilization; they were also quickly willing to learn English, the national language of Indian modernity. The non-Brahman, Dravidian movement, of course, has also insisted that Brahmans are Aryan outsiders, which has further reinforced their marginality vis-à-vis the predominantly non-Brahman, Tamil society and civilization. This marginality, coupled with their experience of migration and circulation within and beyond Tamilnadu since the

Tamil Brahmans as a Middle-Class Caste / 229

nineteenth century, has encouraged Tamil Brahmans to look outward to the rest of India and overseas as places in which to live and work. In addition to their educational and occupational credentials, the Tamil Brahmans’ expansive outlook benefited them as the Indian national economy developed, especially after Independence; it has benefited them even more under conditions of economic liberalization and globalization, so that it has helped to strengthen their position within the all-Indian and transnational upper-middle class. All in all, therefore, postliberalization economic developments in India are tending to reinforce, rather than weaken, the urban, middle-class Tamil Brahmans’ privileged status, as well as their distinctive identity and subjectivity, which owe a lot to how Brahmans have been portrayed as separate, marginal outsiders, by both themselves and the nonBrahman majority. Tamil Brahmanhood has not become an anachronism at the turn of the twenty-first century; on the contrary, being Tamil Brahman continues to matter to an extent that is highly unusual, if not unique, within the contemporary Indian urban middle class, especially the upper-middle class. “Tam Brams,” now stock characters in this class, both at home and abroad, are unlikely to fade from view for many years to come.

Appendix

Tamil Brahman Demographics

Tamilnadu and South India The 1931 Census of India is the last one containing full data about caste; after Independence only data about Scheduled Castes and Tribes have been collected. It is therefore still necessary to estimate caste populations by extrapolating (with increasing unreliability) from the 1931 census. In 2012, however, a socioeconomic and caste census was carried out, in which people were asked to identify their own caste. When this book’s proofs were being corrected in early 2014, no results from this census had been published. Nor was it clear how much information would eventually be issued, but it presumably will include the size of the Brahman population in Tamilnadu and other states, perhaps broken down by subdivisions (Tamil Brahmans, Telugu Brahmans, etc.) and geographically by districts, urban and rural areas, and so on. If and when the census figures appear, some estimates below (whether or not they turn out to be reasonably accurate) will become redundant. The Madras Presidency districts of Madras, Chingleput, North Arcot, South Arcot, Salem, Coimbatore, Tanjore, Trichinopoly, Madura, Ramnad, Tinnevelly, and Nilgiris, together with the princely state of Pudukkottai, more or less correspond to present-day Tamilnadu. In 1931, the total population of this region was 22.85 million, of which 563,685 (2.5 percent) were Brahmans; 442,491 or 78 percent of them (1.9 percent of the total population) were Tamil Brahmans and 11 percent were Telugu Brahmans, the largest minority.1 P. Radhakrishnan, using data from the censuses of 1891 to 1921, also estimated the Brahman population of  Tamilnadu as 2.5 percent of the total.2 The 2011 census recorded the population of  Tamilnadu at 72.14 million.3 Assuming that Tamilnadu’s Brahman population has grown at the same rate as the general population, its estimated size in 2011 was 1.78 million, of which 1.39 million were Tamil Brahmans. In fact, that assumption is likely to be wrong. Middle-class Brahman parents have usually had only one or two children

232 / Appendix

since the 1980s, so that the caste has probably undergone the “demographic transition” more completely than most of Tamilnadu’s population; there has also been a relatively high rate of emigration from the state. Hence the actual Brahman population is almost certainly lower than the above estimate. Many Tamil Brahmans in south India live outside Tamilnadu, especially in Kerala (formerly British Malabar, and Travancore and Cochin princely states), coastal Andhra Pradesh, and southern Karnataka (formerly Mysore princely state). Some of them moved from the Tamil country during the colonial period or more recently, but others—such as Palghat (Palakkad) Brahmans in Kerala—have been settled outside it for centuries and are in practice almost entirely separated from Tamil Brahmans within Tamilnadu, just like many Telugu Brahmans in Tamilnadu are separated from their community in Andhra Pradesh. In 1931, there were 52,230 Tamil Brahmans living within the Madras Presidency, but outside the Tamilnadu region (mostly in Malabar and coastal Andhra), 39,985 in Travancore, and 21,754 in Cochin.4 In the 1931 census of Mysore, there are no data on the Tamil Brahman population. Tamil Brahmans were concentrated in Bangalore and Mysore, however; in the two cities, there were 111,338 Tamil speakers and 57,313 Brahmans, and it is a reasonable guess that around 30,000 of them were Tamil Brahmans.5 The estimated population of Tamil Brahmans in the Kerala, coastal Andhra, and Mysore regions in 1931 is therefore 144,000; if it increased at the same overall rate as in Tamilnadu, the maximum total in 2011 was approximately 455,000. Some Tamil Brahmans who lived in south India outside Tamilnadu have probably moved back since the formation of linguistic states in the 1950s, thus partially counterbalancing overall emigration from Tamilnadu. By 1931, many Tamil Brahmans had left south India for cities further north, notably Bombay. In Bombay, Tamil Brahmans could easily have represented as many as half the 10,000 Tamil speakers then living in the city.6 But despite migration northward, Tamil Brahman numbers outside the south in 1931 were still relatively low and may be ignored here. To conclude, the estimated total Tamil Brahman population was approximately 586,000 in 1931—three-quarters in the Tamilnadu region and onequarter outside—and it grew to a maximum size of 1.85 million by 2011, by when the proportion outside Tamilnadu had risen considerably.

Tamilnadu Villages The 1871 Census of Madras published tables for each district showing the population by caste of individual villages, listed by taluks (district subdivisions).7 All the Eighteen-Village Vattima villages can be found in the tables,

Tamil Brahman Demographics / 233 Table 1  The Brahman population in Eighteen-Village Vattima villages 1871 Village Anandatandavapuram Arasavanangadu Konerirajapuram Kundalur Mandai Maratturai Molaiyur Mudikondan Palur (Mel Palaiyur) (Pandaravadai-)Mappadugai Puliyur Semmangudi Sengalipuram Tattattimulai-Anaikkuppam Tediyur Tippirajapuram Tuttukudi Vishnupuram All villages

Population

2005–8

Percentage of Households Population population remaining

324 147 422 203 182 201 226 589 124 88 7 134 418 93 343 403 128 340

40 20 40 6 16 4 12 55 11 8 23 2 54 4 20 82 8 32

114 57 114 17 45 11 34 156 31 23 65 6 153 11 57 233 23 91

4,372

437

1,241

35 39 27 8 25 5 15 26 25 26 Increased 4 37 12 17 58 18 27 28

Source (1871 data): Census of Madras: Statement of Population in 1871, Tanjore District. The small village of Sittanvalur, which could not be located in the statement, is omitted.

except the small village of Sittanvalur, which was probably included within a nearby larger village. The other Brahman villages studied by modern researchers and mentioned in this book can also be found, except for Sripuram (Tillaisthanam) and Mel Ceval. Tillaisthanam appears as a Ma­ratha royal estate (“mokasa”), but the entry does not cover the entire settlement; Mel Ceval’s absence is probably an error, since all the villages linked to it are listed and it also appears in the 1881 tables (which do not enumerate castes). The tables are summed for each taluk and, added together, the totals should obviously show the number of villagers, including Brahmans, living in each district. In fact, though, some totals are slightly higher than those given for district populations in the main census report, whereas others are not low enough to account for the urban population.8 Hence errors, even though they are relatively small, must have been made in tabulation or computation. Nonetheless, if we assume that the figures for individual villages are roughly correct, estimates of the decline in the rural Brahman population are possible. In table 1, the population and household numbers for Tippirajapu­ram were collected in a survey. For other Vattima villages, the estimates of Brahman household numbers in 2005–8 were obtained from knowledgeable,

221 414 412 125 341* 116 249

Dusi Gangaikondan Kumbapettai Manjapalayam Sripuram Thyagasamuthiram Yanaimangalam

33 (1959) 97 (1960) 36 (1952) 30 (1963) 92 (1962) 13 (1958) 60*(1950)

Households

1950s/60s

112 388* 178 128 341 52 240*

Population 16 (1983) 18 (2008) 6 (2005) 19 (1971) 65 (2011) 2 (2007) 11 (1990)

Households 74 72* 24* 76* 210 8* 47

Population

Most Recent Date

33 17 6 61 62 7 19

Percentage of population remaining

Note: * indicates estimated figures. Sources: Census of Madras: Statement of Population in 1871, North Arcot District (Dusi); Tanjore District (Kumbapettai, Sripuram, Thyagasamuthiram); Tinnevelly District (Gangaikondan, Yanaimangalam); Guhan and Bharathan 1984: 16, table 4 (Dusi); Harriss, Jeyaranjan and Nagaraj 2012: 42–43, table 4 (Gangaikondan); Gough 1981: 156–57, table 8.1 (Kumbapettai); Mencher 1970: 205, table 2 (including 1871 population); 1978: 294–95, table B.2 (figures differ slightly from the 1970 table) (Manjapalayam); Béteille 1996: 26, table 1 (Sripuram); Sivertsen 1963: 28, table 1 (Thyagasamuthiram); D. P. Mines 2005: 7, 9, table 1.1 (Yanaimangalam).

Population

Village

1871

Table 2  The Brahman population in selected Tamilnadu villages

Tamil Brahman Demographics / 235

longstanding residents. The population figures for 2005–8 are estimated from the household numbers on the basis that the mean Brahman household size in Tippirajapuram, which is 2.84, holds good for all the Vattima villages. Table 1 shows that the “percentage of population remaining”—that is, the figure for 2005–8 as a percentage of that for 1871—varies considerably, but the weighted mean calculated for all villages is 28 percent and the median value is 25 percent. Puliyur is excluded as anomalous because its Brahman population has risen, but this is almost certainly an error, possibly because in 1871 the boundaries of the village, which is very close to Nagappattinam town, differed from those today. In table 2, figures for the 1950s and 1960s, when research was carried out in all but one of the villages, are included. Several figures are estimates; population estimates are worked out using a mean household size of 4.0, which should be fairly accurate for the mid-twentieth century, but may be too high for recent years. The Yanaimangalam household figure for the 1950s is Mines’s estimate. The Sripuram 1871 population total is not actually an estimate, but merely reflects the fact that the population was almost certainly no lower than in 1962. Given different dates for the most recent data, the entries in table 2 are not properly comparable, although we note that the median “population remaining” is 19 percent. Owing to the discrepancies between the 1871 census figures for individual villages and districts already referred to, the population of village Brahmans, as opposed to the small minority in towns, cannot be worked out. Hence for our calculations, the total Brahman population of 440,476 in 1871 in the districts more or less corresponding to the state today (excluding Madras city) is assumed to be rural. On that basis, in the light of the “percentage population remaining” values in tables 1 and 2, we estimate that the Brahman population in Tamilnadu’s villages today is, at most, around 20 to 25 percent of the 1871 total, which equals approximately 100,000; approximately 78,000—rounded to 80,000—of them are Tamil Brahmans. That means that no more than around 5.6 percent of Brahmans in Tamilnadu are now rural, compared with the overwhelming majority of them in the late nineteenth century, and that around 4.3 percent of all Tamil Brahmans live in villages in their home state.

Chennai From the nineteenth century onward, Brahmans started to migrate in increasing numbers to Madras and other southern towns. The census reports give caste population figures for each district, but not urban-rural

236 / Appendix

distributions, so that the net Brahman movement from rural to urban areas cannot be calculated. Migration to Madras city can be calculated, however. In 1871, the city’s total population was 397,552, of which 11,783 (3.0 percent) were Brahmans; the census for the city noted this “very insignificant proportion,” which scarcely differed from the proportion in the province overall.9 In later decades, however, migration accelerated and by 1931, when the city’s population was 647,230, its Brahman population had grown fivefold to 58,761 (9.1 percent), so that Madras, mainly through immigration, accounted for a sizable fraction of the natural growth in Brahman numbers. In 1931, Tamil Brahmans were 70 percent of the city’s Brahmans and Telugu Brahmans were 12 percent.10 Telugus have been leaving the city since Independence, however, and the Tamil Brahman proportion has probably risen to the same level as in the state as a whole, around 78 percent. Around 1971, Weinstein did an interview-based survey of 2,000 households in Madras, which involved a complex classification of the city into five “social areas.” Weinstein reports that the Brahman percentage of the population in the five areas was 12.0, 12.2, 12.9, 14.6, and 18.0; the last and highest figure significantly applied to the area with the highest rate of professional and managerial employment.11 There are no data to weight these five figures, but their crude mean equals 13.9 percent, equivalent to approximately 343,000 Brahmans in the city’s total population of 2.47 million in 1971. That figure amounts to about one-third of  Tamilnadu’s Brahman population, which would have been approximately 1 million, 2.5 percent of the state’s population of 41 million.12 Weinstein’s survey is the only source that permits an estimate of Madras’s Brahman population during this period. In 1911, Tamil Brahmans in Madras were most heavily concentrated in central Georgetown (the old colonial city’s center), south Triplicane, and south Mylapore; the Telugu Brahmans’ distribution across the city differed slightly.13 By the late twentieth century, the residential pattern had altered. Thus in Chennai today (see chap. 5), the southern suburbs of Mylapore, Alwarpet and Teynampet, T. Nagar, West Mambalam, Adyar, and Besant Nagar form the main Brahman areas. Fairly large numbers of Brahmans are also dispersed across the city’s western and outer southern suburbs, but not many still live in Georgetown and their numbers have also declined in Triplicane. On the southwest outskirts of the city, Brahmans are also concentrated in the former villages of Nanganallur, Chromepet, and Madi­pakkam. In 2005, we commissioned a survey of Chennai ward number 126, which covers most of T. Nagar. The survey estimated the ward’s caste composition by gathering information from several key informants, normally long-term residents, in each street or apartment complex; it recorded a total of 4,765 households, of which 2,255

Tamil Brahman Demographics / 237

(47.3 percent) were Brahman and 94 (2.0 percent) were Scheduled Caste. The 2001 census for the ward enumerated 4,887 households and a total population of 21,521, of which 2.4 percent were Scheduled Caste.14 The closeness of the survey and census household totals—and the very small Scheduled Caste percentages—suggest that some confidence can be placed in the survey results. Assuming that a household size of 4.4 is the norm throughout T. Nagar, its Brahman population in 2001 can be estimated at 9,922. Eighteen wards correspond more or less exactly to Mylapore, Alwarpet and Teynampet, T. Nagar, West Mambalam, Adyar, and Besant Nagar, and their total population in 2001 was 440,931.15 Hence the area’s estimated Brahman population would be 208,560, if their representation were the same as in T. Nagar. In the early 1990s, however, Auclair carried out surveys of residential buildings in Adyar, Mylapore, and T. Nagar, covering nearly 300 households in each suburb; she found that Brahmans made up 62.6, 74.5, and 64.5 percent of the Hindu population (the overwhelming majority) in Adyar, Mylapore, and T. Nagar, respectively.16 Our survey of T. Nagar probably produced more accurate results than Auclair’s; nonetheless, her figures suggest that ours may be on the low side. Extrapolating from Auclair’s results suggests that around 300,000 residents in the southern suburbs are Brahmans, but the best guess may be a figure between our two estimates of 250,000. Continuing emigration from the countryside means that total Brahman numbers in Chennai must have risen from approximately 343,000 in 1971 and have probably reached at least 500,000 in the early 2000s, with roughly half resident in the inner southern suburbs. Our rough estimate for Chennai’s Tamil Brahman population is therefore 390,000, 28 percent of all those living in Tamil­ nadu and 21 percent of the total population.

Towns and Cities in Tamilnadu According to Guilmoto, useful data on caste populations in Tamilnadu’s towns and cities—apart from his own material from Tiruvannamalai—are extremely sparse.17 Hence we can only resort to simple arithmetic. If the Brahman population estimates for Chennai (500,000) and villages (100,000) are subtracted from Tamilnadu’s (1.78 million), 1.18 million Brahmans (920,000 Tamil Brahmans) live in urban centers other than Chennai.

The Rest of India and Overseas The number of Tamil Brahmans living in India outside Tamilnadu is even harder to estimate than the number within it. In the Kerala, coastal Andhra,

238 / Appendix

and Mysore regions, as explained above, the total Tamil Brahman population may have risen to a maximum of 455,000 by 2011. But there are no data that could be used to adjust this total for the effect of migration flows in and out of Tamilnadu, or other southern states, since the 1930s. Neither are there any data to estimate the net outflow to north India or overseas. Most probably, the proportion of Tamil Brahmans living outside Tamilnadu is now higher than one-quarter, the estimate for 1931, but by how much is impossible to work out. Migration to Bangalore has been particularly vigorous, both from Tamilnadu and elsewhere, and the circumstantial evidence is consistent with the popular assumption that Bangalore is the Tamil Brahmans’s single largest settlement after Chennai; Mumbai is almost certainly the city in third place and Delhi may be fourth. Inbanathan’s informants guessed that Delhi’s Tamil Brahman population in the 1980s was around 27,000, or half the total number of Tamils in the capital.18 Tamil Brahmans sometimes say that their numbers in Delhi have fallen, because fewer now work for the central government, although the city’s enormous expansion must have drawn in others in alternative occupations. In truth, though, estimating Tamil Brahman numbers in these three cities is almost impossible, although our best guess is that at least 100,000–150,000 live in Bangalore, upward of 50,000 in Mumbai, and still at least 25,000 in Delhi, giving a total of around 200,000. Sizable numbers also live in many other urban areas throughout India. Matrimonial services provide an indirect source of information. Table 3 is based on data about residence for 6,000 prospective grooms on lists issued in 2003–4 by the Mahalingapuram temple in Chennai, which ran a service exclusively for Tamil Brahmans. Table 4 is a simple analysis of the “partner profiles” of Tamil Brahman Iyer and Iyengar men and women, for one day in December 2011, which appeared on the website of the online service Jeevansathi.com, which a researcher can access very easily.19 Information about occupations shows that the Mahalingapuram temple’s clients include a larger proportion of lower-middle class men than Jeevansathi’s, who nearly all have professional and managerial jobs. It cannot be assumed that matrimonial service clients are typical members of the caste. Nonetheless, both tables provide indirect evidence that Bangalore, and secondly Mumbai, are the Tamil Brahmans’ leading population centers after Chennai, and that Delhi (or Hyderabad) comes next. Also interesting is the information about people outside India, although it is highly unlikely that as many as 15 percent of all young Tamil Brahmans now live abroad, since matrimonial services are probably more heavily used by or on behalf of overseas residents. On the other hand, the figures showing

Table 3  Residence of Tamil Brahman men on Mahalingapuram Temple lists, 2003–4 Percentage of sample population

Location

Number of men

Chennai Tamilnadu (except Chennai) Bangalore South India (except Bangalore) Mumbai North India (except Mumbai) All India

3,239 771 431 236 182 314 5,173

54.0 12.9 7.2 3.9 3.0 5.2 86.2

449 198 96 59 25 827

7.5 3.3 1.6 1.0 0.4 13.8

6,000

100.0

North America Middle East Asia and Australia Europe Africa All overseas All locations

Table 4  Residence of  Tamil Brahman men and women on Jeevansathi website, December 2011 Men Percentage of sample population

Women

Number

Percentage of sample population

Location

Number

Chennai Bangalore Hyderabad Mumbai Delhi Rest of India All India

620 370 88 289 98 374 1,839

28.5 17.0 4.0 13.3 4.5 17.2 84.5

163 143 37 143 36 97 619

22.3 19.6 5.1 19.6 4.9 13.3 84.8

159 9 48 27 25 70 338

7.3 0.4 2.2 1.2 1.1 3.2 15.5

72 13 5 6 7 8 111

9.9 1.8 0.7 0.8 1.0 1.1 15.3

2,177

100.0

730

100.1

United States Canada United Arab Emirates United Kingdom Australia Rest of world All overseas All locations

240 / Appendix Table 5  Estimated Tamil Brahman population in different locations in 2011 Location

Population

%

Chennai Rest of urban Tamilnadu Tamilnadu villages All Tamilnadu

390,000 920,000 80,000 1,390,000

21.1 49.7 4.3 75.1

200,000 160,000 100,000

10.8 8.6 5.4

1,850,000

99.9

Bangalore, Mumbai, Delhi Rest of India Overseas Total population

that the United States accounts for half the emigrants, with the Middle East in second place for males, are consistent with ethnographic and anecdotal evidence about Tamil Brahman emigration. We tried to make an estimate of Tamil Brahman numbers in the US, from US census data on Tamil speakers and the “Asian Indian” population, but it depended on too many dubious assumptions. A plausible guess might be 50,000 Tamil Brahmans in the US, and another 50,000 in the rest of the world.

Summary Conclusion Table 5 summarizes the Tamil Brahmans’ demographics set out above. As we have seen, all these population estimates are really guesstimates, but some are rather firmer than others; the “rest of India” category is a merely residual one. But despite all these caveats, we submit that these figures do give a reasonable indication of the contemporary size and geographical distribution of the Tamil Brahman population.

Subcastes The 1891 census was the only one to record subcaste populations. The majority of Tamil Brahmans are either Smartas (Aiyars) or Sri Vaishnavas (Aiyangars). Smartas are divided into four principal subcastes: Vadama, Brahacharanam, Ashtasahasram, and Vattima. Today the priestly Adishaiva subcaste is normally regarded as Smarta as well; the Choliya subcaste of the Tanjore region is also reputedly included among Smartas.20 The Sri Vaishnavas are divided into Vadagalai and Tengalai sections. The 1891 census also

Tamil Brahman Demographics / 241 Table 6  Tamil Brahman subcastes %

Subcaste

Population

Vadama Brahacharanam Ashtasahasram Vattima Choliya Adishaiva/Gurukkal

110,166 46,269 13,918 9,756 9,118 8,961

37.8 15.9 4.8 3.3 3.1 3.1

All Smartas

198,188

67.9

Vadagalai Tengalai

54,034 39,527

18.5 13.5

All Sri Vaishnavas

93,561

32.1

291,749

100.0

All

Source: Census of India 1891, vol. 14, Madras, Tables, British Territory, tables 16 and 16-A, pp. 311–13, 361–64.

listed several other small Tamil Brahman subcastes, which are ignored here. Table 6, based on figures for the Madras Presidency’s Tamil districts (excluding North Arcot and Pudukottai), shows that two-thirds of  Tamil Brahmans are Smartas—more than half being Vadamas—and one-third are Sri Vaishnavas; the relative size of the two sections is probably much the same today.

Glossary

This selective glossary only supplies meanings strictly relevant to this book; it also excludes most names and terms occurring only at one or two places in the text. For the transliteration system, see the “Note on Transliteration”; listed in parentheses are words whose spelling is modified when diacritics are removed.

a¯ca¯rya:  priest; guru; teacher A¯di Dra¯vid ·a:  “original, first Dravidian”; Tamil Untouchable or Dalit castes ¯ Adis´aiva (Adishaiva):  “first Shaiva”; Shaiva temple priests’ subcaste agraha¯ram:  Brahmans’ street or quarter Aiyan˙ kar (Aiyangar): Sri Vaishnava Tamil Brahman (and honorific title) Aiyar: Smarta Tamil Brahman (and honorific title) aran˙getram:  Bharatanatyam debut performance bhajana:  devotional hymn bhakti:  devotion Bharatana¯t·yam (Bharatanatyam):  “Indian dance”; classical south Indian dance Bra¯hman · as · a:  Brahman, first of four varn ceri (cheri):  Untouchable or Dalit quarter, hamlet cuttam:  ritually pure devada¯si:  traditional or temple dancer Devı¯:  the goddess of Hinduism dharma:  religious law; moral code dharmas´ a¯stra:  corpus of religious law Gan ·es´a (Ganesha): Shiva’s elder son; god of beginnings and obstacles gra¯madevata¯:  village deities gra¯mam (kira¯mam):  village homa:  ritual of oblations into fire ja¯ti:  caste

244 / Glossary

kuladevam:  family deity laukika:  secular; worldly mat·i:  high ritual purity mira¯sida¯r:  hereditary landlord Nat·ara¯ja: Shiva as “lord of the dance” Niyogi:  “employed”; secular section of  Telugu Brahmans pa¯rppa¯n:  Tamil term for Brahman, usually derogatory ¯ pu¯ja¯:  worship sandhya¯:  Brahman men’s daily, morning ritual S´an˙ kara¯ca¯rya (Shankaracharya):  head of Kanchipuram and Sringeri monasteries sa¯stri:  Brahman domestic priest S´iva (Shiva):  one of two great gods of Hinduism ´sloka (shloka):  verse praising deity S´u¯dra (Shudra):  servant, fourth of four varn · as; Tamil non-Brahman, usually derogatory Sma¯rta:  larger of two main sections of Tamil Brahmans; Aiyar S´rı¯ Vais·n ·ava (Sri [Shri] Vaishnava):  smaller of two main sections of Tamil Brahmans; Aiyangar

suman˙galı¯:  auspicious married woman whose husband is alive ta¯li:  wife’s marriage emblem upanayana:  sacred-thread ceremony; male Brahman initiation ritual u¯r:  village vaidika:  “Vedic”; Brahmanically orthodox; orthodox section of Telugu Brahmans varn · a:  four “classes” of traditional Hindu society Vis·n ·u (Vishnu):  one of two great gods of Hinduism

Notes

Introduction

1. 2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Dupuis 1960: 50–58. Wildavsky 2011. On food in Chennai, see P. Caplan 2008. Fuller 1992: 88, 93–95. See especially Arooran 1980; Baker 1976; M. R. Barnett 1976; Béteille 1969; Geetha and Rajadurai 1998; Hardgrave 1993; Irschick 1969; 1986; Pandian 2007; S. Ramaswamy 1997; Saraswathi 1974; N. Subramanian 1999; Suntharalingam 1974; Washbrook 1976; 1989. Irschick 1969: 358–67; Saraswathi 1974: 199–209. Pandian 2007: 234–35. Price 1989: 162. Dirks 1996: 291–92. Pandian 2007: 236. Irschick 1969: 362. M. R. Barnett 1976: 22. Sattanathan 2007: 52. See also Baker 1976: 28; Geetha and Rajadurai 1998: 61–62. On Ramasamy, see Dirks 2001: chap. 12; Geetha and Rajadurai 1998: chaps. 6–13; Pandian 2007: chap. 6. N. Subramanian 1999: 83. Ibid.: 120, 153. Béteille 1996: 50–52. On Navaratri displays in Chennai, see Hancock 1999: 246–54. P. Viswanathan 2009. Mosse 2012: 245–46. Baviskar and Ray 2011; Donner and De Neve 2011; Misra 1961. Béteille 2003: 81. Donner and De Neve 2011: 3. Ibid.: 7. Ibid.: 13. Baviskar and Ray 2011: 2–3. See Fuller and Narasimhan 2007: 141–44.

246 / Notes to Pages 16–32 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48.

Donner and De Neve 2011: 8–10. Dickey 2012: 570. Ibid.: 574–77. Weber 1978: 541–56. Ortner 1998: 11. Weber 1978: 926–40. Ibid.: 305. Giddens 1973: 105. Ibid.: 42–49, 105–12, 177–88. See also Crompton 2008: 33–35, 41–42; Hall 1997: 19–25. See Crompton 2008: chap. 4. Béteille 1991: 5–9. Fuller and Narasimhan 2007: 121–22. L. Caplan 1987: 11–13, 66–76, 117–18, and passim. Dickey 2012: 61–62, 73–74, and passim. Deshpande 2003: 108, 116. Vaid 2012. Upadhya 2007: 1864. Srinivas 1966: 147–58; see also Srinivas 2002: chaps. 33–36. N. Kumar 1992: 17, and introduction passim. Madan 1975: 151. Chap t e r On e

1. 2. 3. 4. 5. 6.

7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

Census of Madras, 1871, vol. 2, table 8, p. 36. Extract from the “Report of Mr Wallace on the Settlement of  Tanjore for Fasly 1214, 1 May 1805” (Bayley and Hudleston 1862: 94–98), p. 95. Washbrook 1975: 24. Washbrook 1976: 87. See Ludden 1985: 34–41. Extract from “Mr Place’s Final Report on the Jagir, 6 June 1799” (Bayley and Hudleston 1862: 38–70), p. 41; Extract from the “Report of Mr S. Lushington on Tinnevelly, 29 December 1800” (Bayley and Hudleston 1862: 77–85), p. 78. Wilson 1855: 17. Knipe 1997. Hemingway 1915a: 43, 53. Anantha Murthy 1976; see also Nanjundayya and Ananthakrishna Iyer 1928: 318, 320, 408, 507. Ananthakrishna Iyer 1912: 291, 316; Subbarama Aiyar 1918: 123; Kolenda 2003. See, e.g., Hemingway 1915b: 80; Pate 1917: 101. See Krishnama Acharya 1918; Lokanathan 1918; Natarajan 1940; Veeraraghavan 1940. Dirks 2001: 70–73. Geetha and Rajadurai 1998: 57–60; Dirks 2001: 240–41. See Gough 1956; 1960; 1969 (1955); 1981; 1989. “Kumbapettai” is a pseudonym, but we know its real name. See Béteille 1996 (1965). Béteille has agreed (personal communication) that Sripuram’s real name, Tillaisthanam, can now be revealed. Gough 1981: 155–57. Ibid.: 191.

Notes to Pages 32–47 / 247 20. Ibid.: 199–200. 21. Census of Madras. Statement of Population of 1871 in Each Village of the Tanjore District, pp. 130–31. 22. Gough 1981: 201–2. 23. Béteille 1996: 24, 26, 73. 24. Ibid.: 90, 92. 25. Ibid.: 111–18, 131. 26. Gough 1981: 213–17. 27. Ibid.: 238. 28. Ibid.: 223. 29. Ibid.: 159. 30. Ibid.: 251, 294. 31. Ibid.: 169. 32. Gough 1960: 38. 33. Béteille 1996: 35. 34. Gough 1981: 294–95. 35. Ibid.: 159, 251, 294. 36. Ibid.: 169. 37. Ibid.: 249. 38. Ibid.: 301. 39. Béteille 1996: 29. 40. Ibid.: 277–86 (appendix 2). 41. Ibid.: 39, 152. 42. Weber 1948: chap. 7; 1978: 926–40. 43. See Gough 1981; 1989. 44. Béteille 1996: 79–91, 195. 45. Gough 1989: 240–45, 325. 46. Personal communication from Rama Kausalya using data collected by Lakshmi Narayanan. 47. Census of Madras, Statement of Population in 1871 in Each Village of the Tanjore District, pp. 164–65. 48. Sivertsen 1963: 28, 33, 66. 49. See Subrahmanyam 1975. 50. Census of Madras, Statement of Population in 1871 in Each Village of the Tanjore District, pp. 178–79. Information on the number of houses in the most important Vattima agraha¯rams appears on unpublished plans drawn by and used with permission from Pierre Pichard, a member of a research team at the École Française d’Extrême Orient, Pondicherry, in the mid-1990s. 51. Hemingway 1933: 242. 52. Census of Madras, Statement of Population in 1871 in Each Village of the Tanjore District, pp. 238–39, 186–87. 53. Ibid., pp. 160–61. 54. Reiniche (1981) gives a detailed description of such houses in the Tirunelveli region. 55. Narasimhan 2011: 252–53. 56. Mosse 2012: 245; Narasimhan 2011. 57. Fuller and Narasimhan 2010: 231–33. 58. Gough 1981: 319–32. 59. Béteille 1996: 24–25.

248 / Notes to Pages 47–62 60. 61. 62. 63. 64. 65. 66.

67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99.

Ibid.: 29. Ibid.: 52–53. Gough 1981: 151, 353. Béteille 1996: 25. Daniel 1984: 69–70, chap. 2 passim. Dupuis 1960: 340–45; Mencher 1972: 44–46. “Manjapalayam” and “MM” are pseudonyms for the village Joan Mencher studied. D. P. Mines 2005: 10–16; Reiniche 1978: 140–45; 1979: 5–8, 25, 27. “Yanaimangalam” is a pseudonym, but Diane Mines has told us the village’s real name and location. Daniel 1984: 67. Ibid.: 86–89. Gorringe 2005: chap. 5. Béteille 1996: 115. Ludden 1985: 91. Gough 1981: 238; Béteille 1996: 283. Ludden 1999: 144; see also Irschick 1994: 196–202. Natarajan 1940: 61. Ludden 1985: 90. Béteille 1996: 117; Ludden 1999: 186. Beck 1972: 9. Ludden 1999: 144. Gough 1989: 277–78; Yanagisawa 1996: 253–55. Ludden 1999: 187. Gough 1956: 832–33. Gough 1981: 201–2, 264–65; 1989: 262–63, 272–73. Béteille 1996: 114–15. M. R. Barnett 1976: 25. Gough 1981: 294. S. A. Barnett 1975: 166. Daniel 1984: 61–62. Nishimura 1998: 29–31. Harriss 1982: 111. Reiniche 1996: 138. De Neve 2006: 29–30. Fuller and Narasimhan 2010: 231. Washbrook 2010: 604–5. Sarma 2007: 42. Hancock 1999: 106; cf. Narasimhan 2011: 246–47. Narasimhan 2011: 259. Venkatachalapathy 2006: chap. 1. Conlon 1977: 174. Rudner 1994: chap. 4. Chap t e r Tw o

1. 2. 3. 4.

Neild-Basu 1984: 9–14. B. Raman 2012: 32–38. Frykenberg 1965: 233. Ibid.: 77; cf. Suntharalingam 1974: 18.

Notes to Pages 62–73 / 249 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

28. 29. 30. 31. 32. 33. 34.

35. 36. 37. 38. 39. 40. 41. 42.

Wagoner 2003: 795–97. Raghavaiyangar 1893: 337. Suntharalingam 1974: 113–14. Ibid.: 71–72, 125. Census of Madras, 1871, vol. 1, p. 136; Cox 1881: 265. Row 1883: 171. Suntharalingam 1974: 78. Washbrook 2010: 611. Suntharalingam 1974: 58, 66. Ibid.: 114. Ibid.: 65–79, 151–54. Irschick 1969: 16–19; Suntharalingam 1974: 112–15. Irschick 1969: 14. Paul 1991: 8, 10. Washbrook 1976: 284. Price 1989: 162–77. Lingat 1973: 207–23. S. Bayly 1999: chap. 2. C. A. Bayly 1988: 153–54. B. Raman 2012: 76, chaps. 3 and 4 passim. Seal 1968: 124–35, 129. B. Raman 2012: 16–18, 193–94, chaps. 5 and 6 passim. See annual Report on Public Instruction in the Madras Presidency. In the reports from 1916/17 to 1921/22 and 1923/24, consistent student statistics appeared in vol. 1, General Tables, table 3(a); in 1922/23 and 1924/25 to 1938/39, they appeared in vol. 2, part 2, A-Subsidiary Tables for Indians, table 1(b). Report on Public Instruction in the Madras Presidency, 1901/2, vol. 1, table 3; 1916/17, vol. 1, table 3a; 1938/39, vol. 1, table 3a. Irschick 1986: 71–73, 258–59. See annual Report on Public Instruction in the Bombay Presidency. In the reports from 1923/24 to 1942/43, consistent student statistics appeared in appendix 1. Arnold 2000: 64; see also D. Kumar 2006: 114, 133–36. Report on Public Instruction in the Madras Presidency, 1854/55, p. 42; 1858/59, pp. 77– 78; Report on Public Instruction in the Bombay Presidency, 1857/58, p. 70. Report on Public Instruction in the Madras Presidency, 1883/84, p. 57; 1887/88, p. 63; 1895/96, pp. 40–41. Report on Public Instruction in the Madras Presidency, 1883/84, p. 59; 1901/2, p. 20; 1902/3, p. 14; 1904/5, p. 12 (Brahmans preponderant); 1887/88, p. 66 (lack of artisan classes); 1897/98, p. 50 (lack of non-Brahmans). D. Kumar 2006: 186. Fuller and Narasimhan 2008c: 180, n. 5; cf. Irschick 1986: 69, citing different data. Application records are kept in the Institutions of Civil Engineers, Mechanical Engi­ neers, and Electrical Engineers in London, whose librarians and archivists we thank. See Visvesvaraya’s autobiography (1951); Heitzman 2004: 36–39, 60–62, 222–24, 278–79. Weber 1978: 436; 1967: 57, 111–17. On Ramanujan, see especially Kanigel 1991. Kanigel 1991: 36; Nandy 1995: 106–7. Babu 2007.

250 / Notes to Pages 73–94 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65.

66. 67. 68.

69. 70. 71.

Nandy 1995: 96–103. Ibid.: 138–39. Ibid.: 136. Wali 1991: 34–35, 40–46. Indian Scientists: Biographical Sketches (1929): 196. Chakrabarty 2000: 254. Sur 2011: 29. Indian Scientists: Biographical Sketches (1929): 218. Sur 2011: 141. Ibid.: 147–55. Ibid.: 161–64, 168–72. Wali 1991: chaps. 3, 8, and epilogue. Ibid.: 233–34. Arnold 2000: 167. On Bose, see Arnold 2000: 174–76; Nandy: 1995: 46–87; Prakash 1999: 80–81, 229–30. Prakash 1999: 77; see also van der Veer 2001: 74–77. Prakash 1999: 80. Ibid.: 104–5. Ibid.: 84. Some of our information about Seshia Sastri’s life derives from Kamesvara Aiyar (1902). Dirks 1987: 389–90; Waghorne 1994: chap. 2. Fuller and Narasimhan 2008c: 176–77. See: http://www.vangalheritage.com (Vangal Family Tree, Main Vaishnava Families) (accessed July 15, 2013). We thank V. L. Vijayaraghavan, compiler of the website, for his assistance. As the website has grown, so the amount of information about individuals on it has been reduced; our discussion partly relies on information available when the website was first accessed in September 2003. Washbrook 1976: 235. Gough 1981: 201–2. Gough’s fieldnotes are preserved in the Kathleen Gough Fonds in the University of British Columbia Archives, Vancouver; see http://www.library.ubc.ca/archives /u_arch/gough.pdf (accessed July 15, 2013). The three examples are taken from the material on Brahman houses with serial numbers 3 (box 14–6), 15 (box 14–7), and 27 (box 14–9). We thank Sanjeev Routray for collecting the material from the archives for us. Nair 1978: 41–42. Kumar, Heath, and Heath 2002: 2985. Giddens 1973: 107. Chap t e r Th r e e

1. 2. 3. 4. 5. 6.

Ramadorai 2011: 1–11, 19, 30–39, 102, 141–42. Report of the Tamil Nadu Second Backward Classes Commission, 1985 (chaired by J. A. Ambasankar), “Chairman’s recommendations,” table D, pp. 76–86. See P. Radhakrishnan 1996. See Chuyen 2004: 170–74; Hancock 1999: 39–42. Kidder 1974: 17. Misra 1961: 332–35.

Notes to Pages 94–117 / 251 7.

8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

18. 19. 20. 21. 22.

23. 24. 25. 26. 27. 28.

29. 30. 31. 32.

For the estimated 1970 and 1997 figures, see World Bank, India: Scientific and Technical Manpower Development in India (2000), fig. 1.2, p. 46; for the 2006 figures, see Gereffi et al. 2008: 16. These two publications also discuss the debatable accuracy and international comparability of data on Indian engineers issued by the All India Council for Technical Education. Anandakrishnan 2005: 352; Venkatesan 2005; report in Hindu, August 26, 2010, http://www.thehindu.com/news/national/article594331.ece (accessed July 15, 2013). Singh 1976: 61, 158; Inbanathan 1997: 63, 107, n. 1. M. R. Barnett 1976: 266–67. Harriss 2003a. Baker 1984: 284–85, 316. See Singer 1972: ch. 8; Harriss 2003a. Berna 1960: 70. Harriss 2003a: 331–33. Fuller and Narasimhan 2007: 122, 125–26. Data from National Council of Applied Economic Research, reported in Deccan Herald, August 2, 2010, see http://www.deccanherald.com/content/85347/rich-indian -households-outnumber-low.html (accessed July 15, 2013). The average exchange rates of the rupee against the dollar and pound in 2010 were (in round figures): 1 USD = 45 INR; 1 GBP = 71 INR. Taking purchasing power parity into account, a middle-class city dweller in 2010, using the same amount of money, could probably buy at least three times as much in India as in the US or UK. L. Caplan 1987: 92. See, e.g., Baviskar and Ray 2011: 1–11; Deshpande 2003: 144–48. Parry 2003: 221–23. Harriss 2003b: 765. These data are issued by NASSCOM (the IT companies’ trade association). The basis for the list of “top twenty players in IT services” is not explicit, although data formerly issued by NASSCOM show that the top companies are also those with the largest export revenues; Cognizant Technology Solutions is excluded from this list because its corporate headquarters is in the US. See http://www.nasscom.in/industry -rankings (accessed July 15, 2013). See http://www.tcs.com/about/corp_facts/Pages/default.aspx (accessed July 15, 2013). See http://www.cognizant.com/RecentHighlights/Corporate_Fact_Sheet.pdf (accessed July 15, 2013). http://www.careers.tcs.com/CareersDesign/Jsps/FaqEntrylevel.jsp (accessed July 15, 2013). See Fuller and Narasimhan 2006; Krishna and Brihmadesam 2006; Upadhya 2007. Fuller and Narasimhan 2008b: 190. A report about IT professionals’ age formerly on the NASSCOM website was unavailable in 2012. See: http://www.cognizant.com/aboutus/employee-development (accessed July 15, 2013); http://www.infosys.com/sustainability/diversity/pages/index.aspx (accessed July 15, 2013). For more information about TCS’s early years, see Ramadorai (2011: chap. 2). See Fuller and Narasimhan (2007: 127–33; 2008b: 196–206; 2008c: 189–90) for further cases. D’Mello and Sahay 2008; Mukherjee 2008; Upadhya and Vasavi 2006. Upadhya 2007: 1864.

252 / Notes to Pages 118–131 33. 34. 35. 36. 37. 38. 39.

Krishna and Brihmadesam 2006; Upadhya 2007. Béteille 1991. See Bourdieu 1986; cf. Béteille 1991: 15–17. Fuller and Narasimhan 2006: 260–62. Crompton 2008: 112–16, 127–31, 134–35. Kumar, Heath, and Heath 2002: 2986–87. Béteille 1991: 25. Chap t e r F o u r

1. 2. 3. 4. 5. 6. 7. 8.

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

30.

Laws of Manu, 5.148 (1991: 115). N. Kumar 2007: 131. Census of India 1891, vol. 13, Madras, report, p. 180. Nagam Aiya 1906: 298–302. S. A. Raman 1996: 84–86, 103–7. Census of India 1931, vol. 14, Madras, pt. 1, pp. 271–72. Kane 1974: 445. For a small selection from the historical literature on these topics, see Chatterjee 1993; Forbes 1996; N. Kumar 2007: chaps. 5–7; Sarkar 2001; Seth 2007: chap. 5; Sinha 2006: chap. 4. For Madras in particular, see S. A. Raman 1996; Sreenivas 2008; Suntharalingam 1974: chap. 7. Seth 2007: 142. Ibid.: 135–36; cf. Chatterjee 1993: 6, chaps. 6–7 passim. Ebeling 2010: 19–21, 29, 184–200, 238; Rajam Aiyar 1998 (afterword by Blackburn, pp. 162–63). Suntharalingam 1974: 326. Heimsath 1964: 111–12. Sreenivas 2008: 69–70, 92–93. Mayo 1927: pt. 1 passim, 135–38. Sinha 2006: 154. Ibid.: 171–72, 181–82. Sreenivas 2008: 75. M. K. Acharya and M. S. Sesha Ayyangar spoke repeatedly and lengthily against the bill in September 1929; see Legislative Assembly Debates (Official Report), vol. 4, pp. 240, 262, 887; vol. 5, pp. 1056, 1130, 1284 (September 1929). On Acharya, see Pandian 2007: 92. See Sreenivas 2008: chap. 3. Ibid.: 77; for non-Brahman views on marriage, see Hodges 2005. On Subbalakshmi, see Felton 1966; Forbes 1996: 57–60. See N. Kumar 2007: 148–50; S. A. Raman 1996: 107–10. See Felton 1966: 152–64; Forbes 1996: 59–60; Sinha 2006: 174. Felton 1966: 131–32. S. A. Raman 1996: 189. Sinha 2006: 152. Census of India 1931, vol. 14, Madras, pt. 1, p. 340. Census of India 1911, vol. 12, Madras, pt. 2, table 14, p. 124; Census of India 1921, vol. 13, Madras, pt. 2, table 14, p. 126; Census of India 1931, vol. 14, Madras, pt. 2, table 7, pp. 98–99. Because the ambiguous age cohorts in the earlier censuses differ from the clear ones in 1931, direct comparison between their figures is impossible. Fuller 2003: 73; Gough 1956: 841.

Notes to Pages 131–164 / 253 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57.

Sinha 2006: 194, 152. Gulati 2005: 108–14. Sivaraman 2006: 128, and passim. Fuller and Narasimhan 2008a: 738–40. S. A. Raman 1996: 251. Ibid.: 253. Fuller and Narasimhan 2013: 64–78. Gough 1989: 283. P. Caplan 1985: 32, 84–99. M. N. Srinivas 1977: 232. Ibid.: 233. Fuller 2003: 75. Gough 1956: 844. Clark-Decès 2011. Kalpagam 2005. P. Caplan 1985: 42. Mukhopadhyay 1994: 114–16. See also Fuller and Narasimhan 2008b. S. Radhakrishnan 2011: 101–6. Mukhopadhyay 2004: 476, 481; see also Fuller and Narasimhan 2008b: 195–96. See also S. Radhakrishnan 2011: 78–86. Ibid.: 127–34. For further case studies, see Fuller and Narasimhan 2008b: 203–7. S. Radhakrishnan 2011: 149. Ibid.: 83–84; Upadhya and Vasavi 2008: 34–35. S. Radhakrishnan 2011: 147–48, 50–51. Mukhopadhyay and Seymour 1994: 25. Chap t e r F iv e

1. 2. 3. 4. 5.

6.

7. 8. 9. 10. 11. 12. 13. 14. 15.

Wirth 1938: 8, 12. Gans 1975: 196–97, 199. Dupuis 1960: 432. Lakshmi 2004. See http://www.census.tn.nic.in/census2011data/ppt2/Urban_Agglomerations _Tamil_Nadu.pdf; http://www.census.tn.nic.in/index.php?ppt2.php (table 1) (accessed July 15, 2013). On agraha¯rams in present-day Chennai, see reports in Hindu, October 9, 2011, and April 30, 2012, http://www.thehindu.com/todays-paper/tp-features/tp-property plus/article2519286.ece, http://www.thehindu.com/todays-paper/tp-national /article3368467.ece (accessed July 15, 2013). Dupuis 1960: 427–28; M. Mines 1994: 116–18. M. Mines 1994: 71–73. Hannerz 1996: 133. Auclair 1998: 245–46. Ibid.: 239, 241. See Fuller 2011. Donner 2006: 153–54. See Singer 1972: pt. 2. Khilnani 1997: 136.

254 / Notes to Pages 164–189 16. See http://www.censusindia.gov.in/2011-prov-results/paper2/data_files/india2 /Million_Plus_UAs_Cities_2011.pdf (accessed July 15, 2013). 17. Khilnani 1997: 148. 18. Rao 2007: 186–96. 19. Inbanathan 1997: chap. 4; Singh 1976. 20. See Bhatia 2007: 19–20, 184–97; Khandelwal 2002: 5, 94; Purkayastha 2005: 1, 89–90, 93–94; Rajagopal 2001: 248–49; Rangaswamy 2000: 93–94; cf. Kibria (2002: 157–58) on Chinese and Korean Americans. 21. Bhatia 2007. 22. Khandelwal 2002: 25–30; Shukla 2003: 106–18; Rangaswamy 2000: 88–91, 273–82. 23. Beauregard 2006: 142. 24. Khandelwal 2002: 18; see also Bhatia 2007: 96–97. 25. Beauregard 2006: 15–17. 26. Prashad 2000: 93–95; Rajagopal 2001: 248–53. 27. Bhatia 2007: 193–95, chaps. 4–6. 28. Kalpagam 2005. 29. Lahiri 2004: chaps. 3–4, and passim. 30. Bhatia 2007: 50–60. 31. Gans 1975: 197–200. 32. Shankar 2008: 5–9, and chap. 6. 33. Ibid.: chap. 7. 34. Purkayastha 2005: chap. 3. 35. See also S. Radhakrishnan 2011: 168. 36. Bhatia 2007: 222. 37. Ibid.: 222. Chap t e r Si x

1. 2.

3.

4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

See Laws of Manu, passim, and Doniger’s introduction (1991: xvi–lxi). See especially M. R. Barnett 1976: chaps. 2–3; Geetha and Rajadurai 1998: chaps. 1–3, 8–9; Irschick 1969: chap. 8; Pandian 2007: chaps. 4–6; N. Subramanian 1999: chap. 2. In addition to works cited in n. 2, see Arooran 1980; Dirks 2001: 134–48, 257–75; Hellman-Rajanayagam 1995; S. Ramaswamy 1997: chaps. 1–2; Trautmann 1997: 2–15, 37–41, 143–64, 172–78, 190–94; 2006: 186–87, 202–8. S. Ramaswamy 1997: 28. See Arooran 1980: chap. 4; S. Ramaswamy 1997: chap. 2, 199–204; Venkatachala- pathy 2009. Weber 1978: chap. 6, pts. v–vii; cf. Weber 1948: chap. 11. M. N. Srinivas 1952: chap. 7. Redfield and Singer 1954. Singer 1972: pt. 2. See appendix. On Adishaivas, see Fuller 1984: chaps. 2–4; 2003: chap. 2. Thurston and Rangachari 1909: 338–42; Younger 1995: chap. 1. Fuller 2003: chap. 2. Béteille 1996: 62–63; Gough 1981: 205–6, 255–56; Subramaniam 1974: 12, 32, 42. Subramaniam 1974: 19–34. D. P. Mines 2005: 126–27; Reiniche: 1979: 21–28, figs. 1, 2. On categories of deities, see Fuller 1992: chap. 2. Béteille 1996: appendix 1.

Notes to Pages 189–204 / 255 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43.

44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56.

D. P. Mines 2005: 145–46. Fuller 1992: 24–28, 54–56. On domestic ritual, see Hancock 1999: chaps. 3–5; Logan 1980. Singer 1972: 70. Raghavan 1956. Singer 1972: 82, italics original. Ibid.: 84–86. Ibid.: chaps. 4–6. See Hancock 1999: 64–67; Fuller and Harriss 2005: 218; Waghorne 2004: 10, 127; Fuller 1992: 160–61. Fuller 2003: chap. 5. Hancock 1999; Waghorne 2004; Harriss 2003a; Fuller and Harriss 2005. Waghorne 2004: 236–38. Ibid.: 25. See Hancock 1999: 93, 199–201; Narayanan 2007: 186–88; Singer 1972: chap. 6. Fuller and Harriss 2005. On Sai Baba, see Babb 1987: pt. 3; T. Srinivas 2010. Fuller 1999: 43–47. S. Ramaswamy 1997: 208. Ibid.: 137–44. Fuller 2003: 114–16. Singer 1972: 86–89. Ibid.: 341–42; Harriss 2003a: 353–55. Cenkner 1983: 143; Hancock 1995: 909–10; 1999: 124, 205–16; Mines and Gouri- shankar 1990: 777–78. Fuller 2003: 126–30. Ibid.: 131, 136. Jayendra’s case and its repercussions were widely reported; see, e.g., reports in Frontline, December 3, 2004, pp. 4–12 (T. S. Subramanian 2004; Venkatesan 2004; S. Viswanathan 2004); for the acquitted, Hindu, November 28, 2013: http://www. thehuidu.com/news/national/tamil-nadu/kanchi-seers-acquitted-um-sankaraman -case/article539743.ece?ref=relatedNews (accessed January 21, 2014). Yocum 1992: 85. See O’Shea 2007; L. Subramanian 2006; Weidman 2006; also Narayanan 2007; Peterson and Soneji 2008. L. Subramanian 2006: 42–43. Ibid.: 74. See ‘About Music Academy: History’ on the Academy’s website: http://www.music academymadras.in/home.php (accessed July 15, 2013). L. Subramanian 2006: 17. See ‘Events’ on the Academy’s website: http://www.musicacademymadras.in/events. php (accessed July 15, 2013). See ‘About Music Academy: Annual Conferences’ on the Academy’s website: http:// www.musicacademymadras.in/home.php (accessed July 15, 2013). L. Subramanian 2006: chaps. 2, 3; cf. Bakhle 2005. Weidman 2006: 5. Ibid.: 40–49. Wali 1991: 34–35, 40–47. Weidman 2006: 5.

256 / Notes to Pages 204–218 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87.

Ibid.: 6. Ibid.: 115. Ibid.: 136. L. Subramanian 2006: 136–37; Weidman 2006: 112–15, 123–28. Weidman 2006: 128. Peterson and Soneji 2008: 14–17. Quoted in Weidman 2006: 184. Weidman 2006: 178–91. L. Subramanian 2006: 166. Soneji 2012: 3–12. Ibid.: chap. 1. L. Subramanian 2006: 132. O’Shea 2007: 37. Ibid.: 38. Ibid.: 90. Ibid.: 90–91, 121–22. Knight 2010: 111–12; see also Weidman 2006: 151, 168. Knight 2010: 132–33, 147. Soneji 2012: 25. O’Shea 2007: 108. L’Armand and L’Armand 1983: 427–33. See, e.g., Menon 2012. Bakhle 2005: viii. Mukulika Banerjee, personal communication. Gaston 1992: 150–51. O’Shea 2007: 154. Gaston 1992: 160; O’Shea 2007: 154–55; L. Subramanian 2006: 176; Weidman 2006: 22–23. P. Caplan 1985: 81. Soneji 2012: 223. S. Ramaswamy 1997: xix. Waghorne 2004: 145. Chap t e r S e v e n

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Seal 1968: 49. Ibid.: 60–62. Ibid.: 89. Ibid.: 83. Raychaudhuri 1988: 19. Ibid.: 11. Chakrabarty 2000: 141, and chap. 5 passim. Karve 1963: 3–4. See Suntharalingam 1974: 32–57. Ebeling 2010: chap. 1 and passim. Narayana Rao 2007. Ibid.: 161–62. Heimsath 1964: 112. In conversation with Fuller in 2009.

Notes to Pages 218–232 / 257 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55.

Sarma 2007: 42. U. Ramaswamy 1978: 291, 298–99. Pandian 2007: 26. Ibid.: 35; Chakrabarty 2000: 218–24. Pandian 2007: 37. Ibid.: 60. Ibid.: 62. Geetha and Rajadurai 1998: 6. Washbrook 2010: 604. Mencher 1966: 189–90. Ramesh Bairy 2009: 226–43. Ibid.: 86–89. Ibid.: 242, 252, 243, 256, and chap. 6 passim. Singer 1972: 320–21. Harriss 2003a: 349–51. M. N. Srinivas 1966: 123. Fuller 2003: 5–6, 77–79, 112–13, 166–67. Prakash 1999: 234. L. Caplan 1987: 12; Deshpande 2003: 146; Dickey 2012: 572; Frøystad 2006: 162– 64, 170–71; Fuller 1996: 16–17. Cf. Fuller 1999: 36; Kapadia 1995: 11, 47. Luhrmann 1996: 38, chap. 1 passim. Ramesh Bairy 2009: 88–90; Ullrich 2010. Conlon 1977. See Patterson 1970. Khare 1970: 4–5, 116–28, and passim. Ibid.: 18, 125, 136. Wadley 1994: 77–105; 2000: 320–28. Sender 1988: introduction, chaps. 3–5, 16, and passim. Madan 1965. Sender 1988: 34. Leonard 1978: chap. 13. Irschick 1969: 16–17. R. Jeffrey 1976: chap. 6. Osella and Osella 2000: 28, 57–58. De Neve 2011. Upadhya 1997; Xiang Biao 2007: 31–32, 34. Jeffery, Jeffery and Jeffrey 2011: 163, italics original, and passim; C. Jeffrey 2010: 63–71, chap. 3. Slezkine 2004: chap. 2. Ibid.: 153. Ibid.: 30, 39. Kaviraj 2010: 96–97. A pp e n d i x

1. 2.

Census of India 1931, vol. 14, Madras, pt. 2, table 17, pp. 307–8; provincial table 2, pp. 332–41. P. Radhakrishnan 1989: 507.

258 / Notes to Pages 232–240 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

http://www.census.tn.nic.in/state_ppt.php (accessed July 15, 2013). Census of India 1931, vol. 14, Madras, pt. 2, table 17, pp. 306–7; vol. 28, Travancore, pt. 2, table 17, pp. 154–55; vol. 21, Cochin, pt. 2, table 17, p. lxxii. Census of India 1931, vol. 25, Mysore, pt. 2, table 15, pp. 204–5, table 17, pp. 230–32. Census of India 1931, vol. 9, Cities of the Bombay Presidency, pt. 1, statement 42, p. 37. See Census of Madras: Statement of Population in 1871 for each district (full details in bibliography). Cf. Census of Madras Presidency 1871, vol. 2, supplementary table 8, p. 36. Census of the Town of Madras, 1871, p. 83. Census of Madras Presidency 1871, vol. 2, supplementary table 8, p. 36; Census of India 1931, vol. 14, Madras, pt. 2, table 17, pp. 307–8; provincial table 2, pp. 332–41. Weinstein 1974: 44, table 11. http://www.archive.org/details/TamilNaduPrimaryCensusAbstract1971 (accessed July 15, 2013): Tamil Nadu 1971 Census Handbook, union table A-1. Census of India 1911, The City of Madras: Statistical Tables, table 7, pp. xxxii–xxxiv. http://census2001.tn.nic.in/pca2001.aspx (accessed July 15, 2013): ward 126. http://census2001.tn.nic.in/pca2001.aspx (accessed July 15, 2013): wards 96, 113, 114, 115, 116, 124, 126, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152. Auclair 1998: 239. Guilmoto, Reiniche, and Pichard 1990: 32, 36. Inbanathan 1997: 63, 107. http://www.jeevansathi.com/ (accessed July 15, 2013). Thurston and Rangachari 1909: 340–42.

Bibliography

Anandakrishnan, A. 2005. “Higher Education.” In Tamil Nadu State Development Report, 337–84. Chennai: Government of  Tamil Nadu. Ananthakrishna Iyer, L. K. 1912. The Cochin Tribes and Castes. Vol. 2. Madras: Higgin­ botham for Government of Cochin. Anantha Murthy, U. R. 1976 [1965]. Samskara: A Rite for a Dead Man. Translated by A. K. Ramanujan. Delhi: Oxford University Press. Arnold, David. 2000. Science, Technology and Medicine in Colonial India. Cambridge: Cam­ bridge University Press. Arooran, K. Nambi. 1980. Tamil Renaissance and Dravidian Nationalism, 1905–1944. Ma­ durai: Koodal. Auclair, Christine. 1998. Ville à vendre: Voie libérale et privatisation du secteur de l’habitat à Chennai (Inde). Pondicherry: Institut Français de Pondichéry. Babb, Lawrence A. 1987. Redemptive Encounters: Three Modern Styles in the Hindu Tradition. Delhi: Oxford University Press. Babu, D. Senthil. 2007. “Memory and Mathematics in the Tamil Tinnai Schools of South India in the Eighteenth and Nineteenth Centuries.” International Journal for the History of Mathematics Education 2 (1): 15–37. Baker, Christopher J. 1976. The Politics of South India, 1920–1937. Cambridge: Cambridge University Press. ——— —. 1984. An Indian Rural Economy, 1880–1955: The Tamilnad Countryside. Delhi: Ox­ ford University Press. Bakhle, Janaki. 2005. Two Men and Music: Nationalism in the Making of an Indian Classical Tradition. New York: Oxford University Press. Barnett, Marguerite Ross. 1976. The Politics of Cultural Nationalism in South India. Prince­ ton, NJ: Princeton University Press. Barnett, Stephen A. 1975. “Approaches to Changes in Caste Ideology in South India.” In Essays on South India, edited by Burton Stein, 149–80. Honolulu: University Press of Hawaii. Baviskar, Amita, and Raka Ray. 2011. “Introduction.” In Elite and Everyman: The Cultural Politics of the Indian Middle Classes, edited by Amita Baviskar and Raka Ray, 1–23. New Delhi: Routledge. Bayley, W. H., and W. Hudleston, eds. 1862. Papers on Mirasi Right Selected from the Records of Government. Madras: Pharoah and Co.

260 / Bibliography Bayly, C. A. 1988. Indian Society and the Making of the British Empire. Cambridge: Cam­ bridge University Press. Bayly, Susan. 1999. Caste, Society and Politics in India from the Eighteenth Century to the Modern Age. Cambridge: Cambridge University Press. Beauregard, Robert A. 2006. When America Became Suburban. Minneapolis: University of Minnesota Press. Beck, Brenda E. F. 1972. Peasant Society in Kon˙ku: A Study of Right and Left Subcastes in South India. Vancouver: University of British Columbia Press. Berna, James J. 1960. Industrial Entrepreneurship in Madras State. Bombay: Asia. Béteille, André. 1969. “Caste and Politics in Tamilnad.” In Castes: Old and New, 152–96. London: Asia. ——— —. 1991. “The Reproduction of Inequality: Occupation, Caste and Family.” Contributions to Indian Sociology (n.s.) 25 (1): 3–28. ——— —. 1996 [1965]. Caste, Class, and Power: Changing Patterns of Stratification in a Tanjore Village. 2nd ed. Delhi: Oxford University Press. ——— —. 2003. “The Social Character of the Indian Middle Class.” In Middle Class Values in India and Western Europe, edited by Imtiaz Ahmad and Helmut Reifeld, 73–85. New Delhi: Social Science Press. Bhatia, Sunil. 2007. American Karma: Race, Culture, and Identity in the Indian Diaspora. New York: New York University Press. Bourdieu, Pierre. 1986. “The Forms of Capital.” In Handbook of Theory and Research for the Sociology of Education, edited by John G. Richardson, 241–58. Westport, CT: Green­ wood Press. Caplan, Lionel. 1987. Class and Culture in Urban India: Fundamentalism in a Christian Community. Oxford: Clarendon Press. Caplan, Patricia. 1985. Class and Gender in India: Women and Their Organizations in a South Indian City. London: Tavistock. ——— —. 2008. “Crossing the Veg/Non-Veg Divide: Commensality and Sociality among the Middle Classes in Madras/Chennai.” South Asia: Journal of South Asian Studies 31 (1): 118–42. Cenkner, William. 1983. A Tradition of Teachers: S´an˙kara and the Jagadgurus Today. Delhi: Motilal Banarsidass. Census of India 1891, vol. 13, Madras, Report; vol. 14, Madras, Tables. Madras, 1893. Census of India 1911, vol. 12, Madras, pt. 2, Imperial and Provincial Tables. Madras, 1912. Census of India 1911, The City of Madras: Statistical Tables. Madras, 1912. Census of India 1921, vol. 13, Madras, pt. 2, Imperial and Provincial Tables. Madras, 1922. Census of India 1931, vol. 9, Cities of the Bombay Presidency, pt. 1, Report. Bombay, 1933. Census of India 1931, vol. 14, Madras, pts. 1, Report; 2, Imperial and Provincial Tables. Madras, 1932. Census of India 1931, vol. 21, Cochin, pt. 2, Tables. Ernakulam, 1933. Census of India 1931, vol. 25, Mysore, pt. 2, Tables. Bangalore, 1932. Census of India 1931, vol. 28, Travancore, pt. 2, Tables. Trivandrum, 1932. Census of India 1971, Tamil Nadu, Primary Census Abstract. http://www.archive.org/details /TamilNaduPrimaryCensusAbstract1971 (accessed July 15, 2013). Census of India 2001, Tamil Nadu, Primary Census Abstract. http://census2001.tn.nic.in /pca2001.aspx (accessed July 15, 2013). Census of India 2011, India, Provisional Population Totals, Urban Agglomerations, Cities Having Population 1 Million and Above. http://www.censusindia.gov.in/2011-prov-results

Bibliography / 261 /paper2/data_files/india2/Million_Plus_UAs_Cities_2011.pdf (accessed July 15, 2013). Census of India 2011, Tamil Nadu, Provisional Population Tables: Urban Agglomerations and Cities. http://www.census.tn.nic.in/census2011data/ppt2/Urban_Agglomerations _Tamil_Nadu.pdf (accessed July 15, 2013). Census of India 2011, Tamil Nadu, Provisional Population Total, PPT2, Urban Agglomeration—Tamil Nadu. http://www.census.tn.nic.in/index.php?ppt2.php (accessed July 15, 2013). Census of India 2011, Tamil Nadu, Provisional Population Total—Tamil Nadu State. http:// www.census.tn.nic.in/state_ppt.php (accessed July 15, 2013). Census of India 2011, Tamil Nadu, Rural-Urban Distribution of Population (Provisional Popula­ tion Tables). http://www.census.tn.nic.in/census2011data/Presentation_rural_urban _2011.pdf (accessed July 15, 2013). Census of Madras Presidency, 1871, vol. 1, Report; vol. 2, Supplementary Tables. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the Chingleput District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the Coimbatore District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the Madura District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the North Arcot District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the Salem District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the South Arcot District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the Tanjore District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the Tinnevelly District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of Madras. Statement of Population in 1871 in Each Village of the Trichinopoly District, Arranged According to Area, Caste and Occupation. Madras, 1874. Census of the Town of Madras, 1871. Madras, 1873. Chakrabarty, Dipesh. 2000. Provincializing Europe: Postcolonial Thought and Historical Difference. Princeton, NJ: Princeton University Press. Chatterjee, Partha. 1993. The Nation and Its Fragments: Colonial and Postcolonial Histories. Princeton, NJ: Princeton University Press. Chuyen, Gilles. 2004. Who Is a Brahmin? The Politics of Identity in India. New Delhi: Manohar. Clark-Decès, Isabelle. 2011. “The Decline of Dravidian Kinship in Local Perspectives.” In A Companion to the Anthropology of India, edited by Isabelle Clark-Decès, 517–35. Oxford: Wiley-Blackwell. Conlon, Frank F. 1977. A Caste in a Changing World: The Chitrapur Saraswat Brahmans, 1700–1935. Berkeley: University of California Press. Cox, A. F. 1881. A Manual of the North Arcot District in the Presidency of Madras. Madras: E. Keys. Crompton, Rosemary. 2008. Class and Stratification. 3rd ed. Cambridge: Polity Press. Daniel, E. Valentine. 1984. Fluid Signs: Being a Person the Tamil Way. Berkeley: University of California Press.

262 / Bibliography De Neve, Geert. 2006. “Economic Liberalisation, Class Restructuring and Social Space in Provincial South India.” In The Meaning of the Local: Politics of Place in Urban India, edited by Geert De Neve and Henrike Donner, 21–43. London: Routledge. ——— —. 2011. “‘Keeping It in the Family’: Work, Education and Gender Hierarchies among Tiruppur’s Industrial Capitalists.” In Being Middle-Class in India: A Way of Life, edited by Henrike Donner, 73–99. London: Routledge. Deshpande, Satish. 2003. Contemporary India: A Sociological View. New Delhi: Viking. Dickey, Sara. 2012. “The Pleasures and Anxieties of Being in the Middle: Emerging Middle-Class Identities in Urban South India.” Modern Asian Studies 46 (3): 559–99. Dirks, Nicholas B. 1987. The Hollow Crown: Ethnohistory of an Indian Kingdom. Cambridge: Cambridge University Press. ——— —. 1996. “Recasting Tamil Society: The Politics of Caste and Race in Contemporary Southern India.” In Caste Today, edited by C. J. Fuller, 263–95. Delhi: Oxford Univer­ sity Press. ——— —. 2001. Castes of Mind: Colonialism and the Making of Modern India. Princeton, NJ: Princeton University Press. D’Mello, Marisa, and Sundeep Sahay. 2008. “Betwixt and Between? Exploring Mobilities in a Global Workplace in India.” In In an Outpost of the Global Economy: Work and Workers in India’s Information Technology Industry, edited by Carol Upadhya and A. R. Vasavi, 76–100. New Delhi: Routledge. Donner, Henrike. 2006. “The Politics of Gender, Class and Community in a Central Cal­ cutta Neighbourhood.” In The Meaning of the Local: Politics of Place in Urban India, edited by Geert De Neve and Henrike Donner, 141–58. London: Routledge. Donner, Henrike, and Geert De Neve. 2011. “Introduction.” In Being Middle-Class in India: A Way of Life, edited by Henrike Donner, 1–22. London: Routledge. Dupuis, Jacques. 1960. Madras et le nord du Coromandel: Étude des conditions de la vie indienne dans un cadre géographique. Paris: Librairie d’Amérique et d’Orient, AdrienMaisonneuve. Ebeling, Sascha. 2010. Colonizing the Realm of Words: The Transformation of Tamil Literature in Nineteenth-Century South India. Albany, NY: State University of New York Press. Felton, Monica. 1966. A Child Widow’s Story. London: Gollancz. Forbes, Geraldine H. 1996. Women in Modern India. Cambridge: Cambridge University Press. Frøystad, Kathinka. 2006. Blended Boundaries: Caste, Class, and Shifting Faces of “Hinduness” in a North Indian City. Delhi: Oxford University Press. Frykenberg, Robert E. 1965. Guntur District, 1788–1848: A History of Local Influence and Central Authority in South India. Oxford: Clarendon Press. Fuller, C. J. 1984. Servants of the Goddess: The Priests of a South Indian Temple. Cambridge: Cambridge University Press. ——— —. 1992. The Camphor Flame: Popular Hinduism and Society in India. Princeton, NJ: Princeton University Press. ——— —. 1996. “Introduction: Caste Today.” In Caste Today, edited by C. J. Fuller, 1–31. Delhi: Oxford University Press. ——— —. 1999. “The Brahmins and Brahminical Values in Modern Tamil Nadu.” In Institutions and Inequalities: Essays in Honour of André Béteille, edited by Ramachandra Guha and Jonathan P. Parry, 30–55. Delhi: Oxford University Press. ——— —. 2003. The Renewal of the Priesthood: Modernity and Traditionalism in a South Indian Temple. Princeton, NJ: Princeton University Press. ——— —. 2011. “Timepass and Boredom in Modern India (review article).” Anthropology

Bibliography / 263 of This Century 1 (May 2011). http://aotcpress.com/articles/timepass-boredom/ (ac­ cessed July 15, 2013). Fuller, C. J., and John Harriss. 2005. “Globalizing Hinduism: A ‘Traditional’ Guru and Modern Businessmen in Chennai.” In Globalizing India: Perspectives from Below, edited by Jackie Assayag and C. J. Fuller, 211–36. London: Anthem Press. Fuller, C. J., and Haripriya Narasimhan. 2006. “Engineering Colleges, ‘Exposure’ and In­ formation Technology Professionals in Tamil Nadu.” Economic and Political Weekly 41 (3) (January 21–27): 258–62. ——— —. 2007. “Information Technology Professionals and the New-Rich Middle Class in Chennai (Madras).” Modern Asian Studies 41 (1): 121–50. —. 2008a. “Companionate Marriage in India: The Changing Marriage System in a ——— Middle-Class Brahman Subcaste.” Journal of the Royal Anthropological Institute (n.s.) 14 (4): 736–54. ——— —. 2008b. “Empowerment and Constraint: Women, Work and The Family in Chen­ nai’s Software Industry.” In In an Outpost of the Global Economy: Work and Workers in India’s Information Technology Industry, edited by Carol Upadhya and A. R. Vasavi, 190–210. New Delhi: Routledge. ——— —. 2008c. “From Landlords to Software Engineers: Migration and Urbaniza­ tion among Tamil Brahmans.” Comparative Studies in Society and History 50 (1): 170–96. ——— —. 2010. “The Agraharam: The Transformation of Social Space and Brahman Status in Tamilnadu during the Colonial and Postcolonial Periods.” In Ritual, Caste, and Religion in Colonial South India, edited by Michael Bergunder, Heiko Frese, and Ulrike Schröder, 219–37. Halle, Germany: Verlag der Franckeschen Stiftungen. ——— —. 2013. “Marriage, Education, and Employment among Tamil Brahman Women in South India, 1891–2010.” Modern Asian Studies 47 (1): 53–84. Gans, Herbert J. 1975 [1962]. “Urbanism and Suburbanism as Ways of Life: A ReEvaluation of Definitions.” In City Ways, edited by John Friedl and Noel J. Chrisman, 189–206. New York: Crowell. Gaston, Anne-Marie. 1992. “Dance and the Hindu Woman: Bharatana¯t.yam ReRitualized.” In Roles and Rituals for Hindu Women, edited by Julia Leslie, 149–71. Delhi: Motilal Banarsidass. Geetha, V., and S. V. Rajadurai. 1998. Towards a Non-Brahmin Millennium: From Iyothee Thass to Periyar. Calcutta: Samya. Gereffi, Gary, et al. 2008. “Getting the Numbers Right: International Engineering Educa­ tion in the United States, China, and India.” Journal of Engineering Education 97 (1): 13–25. Giddens, Anthony. 1973. The Class Structure of the Advanced Societies. London: Hutchinson. Gorringe, Hugo. 2005. Untouchable Citizens: Dalit Movements and Democratisation in Tamil Nadu. New Delhi: Sage. Gough, E. Kathleen. 1956. “Brahman Kinship in a Tamil Village.” American Anthropologist 58 (5): 826–53. ——— —. 1960. “Caste in a Tanjore Village.” In Aspects of Caste in South India, Ceylon and North-west Pakistan, edited by E. R. Leach, 11–60. Cambridge: Cambridge University Press. ——— —. 1969 [1955]. “The Social Structure of a Tanjore Village.” In Village India, edited by McKim Marriott, 36–52. Chicago: The University of Chicago Press. ——— —. 1981. Rural Society in Southeast India. Cambridge: Cambridge University Press.

264 / Bibliography ——— —. 1989. Rural Change in Southeast India, 1950s to 1980s. Delhi: Oxford University Press. Guhan, S., and K. Bharathan. 1984. Dusi: A Resurvey (MIDS Working Paper 52). Madras: Madras Institute of Development Studies. Guilmoto, Christophe, Marie-Louise Reiniche, and Pierre Pichard. 1990. Tiruvannamalai: Un lieu saint sivaïte du sud de l’Inde. Vol. 5, La ville. Paris: École Française d’Extrême Orient. Gulati, Leela. 2005. “The Tyranny of Tradition.” In A Space of Her Own: Personal Narratives of Twelve Women, edited by Leela Gulati and Jasodhara Bagchi, 105–24. New Delhi: Sage. Hall, John R. 1997. “The Reworking of Class Analysis.” In Reworking Class, edited by John R. Hall, 1–37. Ithaca, NY: Cornell University Press. Hancock, Mary. 1995. “Hindu Culture for an Indian Nation: Gender, Politics, and Elite Identity in Urban South India.” American Ethnologist 22 (4): 907–26. ——— —. 1999. Womanhood in the Making: Domestic Ritual and Public Culture in Urban South India. Boulder, CO: Westview. Hannerz, Ulf. 1996. Transnational Connections: Culture, People, Places. London: Routledge. Hardgrave, Robert L., Jr. 1993 [1965]. “The Dravidian Movement.” In Essays in the Political Sociology of South India, 1–80. New Delhi: Manohar. Harriss, John. 1982. Capitalism and Peasant Farming: Agrarian Structure and Ideology in Northern Tamil Nadu. Bombay: Oxford University Press. ——— —. 2003a. “The Great Tradition Globalizes: Reflections on Two Studies of ‘ The In­ dustrial Leaders’ of Madras.” Modern Asian Studies 37 (2): 327–62. ——— —. 2003b. “‘Widening the Radius of Trust’: Ethnographic Explorations of  Trust and Indian Business.” Journal of the Royal Anthropological Institute (n.s.) 9 (4): 755–73. Harriss, John, J. Jeyaranjan, and K. Nagaraj. 2012. “Rural Urbanism in Tamil Nadu: Notes on a ‘Slater Village’; Gangaikondan, 1916–2012.” Review of Agrarian Studies 2 (2). http://ras.org.in/rural_urbanism_in_tamil_nadu (accessed July 15, 2013). Heimsath, Charles H. 1964. Indian Nationalism and Hindu Social Reform. Princeton, NJ: Princeton University Press. Heitzman, James. 2004. Network City: Planning the Information Society in Bangalore. Delhi: Oxford University Press. Hellman-Rajanayagam, Dagmar. 1995. “Is There a Tamil ‘Race’?” In The Concept of Race in South Asia, edited by Peter Robb, 109–45. Delhi: Oxford University Press. Hemingway, F. R. 1915a. Godavari (Madras District Gazetteers). Vol. 1. Madras: Govern­ ment Press. ——— —. 1915b. Tanjore (Madras District Gazetteers). Vol. 1. Madras: Government Press. ——— —. 1933. Tanjore (Madras District Gazetteers). Vol. 2, incl. suppl. Madras: Govern­ ment Press. Hodges, Sarah. 2005. “Revolutionary Family Life and the Self Respect Movement in Tamil South India, 1926–49.” Contributions to Indian Sociology (n.s.) 39 (2): 251–77. Inbanathan, Anand. 1997. Migration and Adaptation: Tamils in Delhi. Delhi: Kalinga. Indian Scientists: Biographical Sketches. 1929. Madras: G. A. Natesan. Irschick, Eugene F. 1969. Politics and Social Conflict in South India: The Non-Brahman Movement and Tamil Separatism, 1916–1929. Berkeley: University of California Press. ——— —. 1986. Tamil Revivalism in the 1930s. Madras: Cre-A. ——— —. 1994. Dialogue and History: Constructing South India, 1795–1895. Berkeley: Uni­ versity of California Press.

Bibliography / 265 Jeffery, Roger, Patricia Jeffery, and Craig Jeffrey. 2011. “Are Rich Rural Jats Middle-Class?” In Elite and Everyman: The Cultural Politics of the Indian Middle Classes, edited by Amita Baviskar and Raka Ray, 140–63. New Delhi: Routledge. Jeffrey, Craig. 2010. Timepass: Youth, Class, and the Politics of Waiting in India. Stanford, CA: Stanford University Press, 2010. Jeffrey, Robin. 1976. The Decline of Nayar Dominance: Society and Politics in Travancore, 1847–1908. London: Chatto and Windus for Sussex University Press. Kalpagam, U. 2005. “‘America Varan’ Marriages among Tamil Brahmans: Preferences, Strategies and Outcomes.” Indian Journal of Gender Studies 12 (2–3): 189–215. Kamesvara Aiyar, B. V. 1902. Sir A. Sashiah Sastri, K.C.S.I., an Indian Statesman: A Biographical Sketch. Madras: Srinivasa, Varadachari and Co. Kane, P. V. 1974 [1941]. History of Dharmas´a¯stra. Vol. 2. Poona: Bhandarkar Oriental Re­ search Institute. Kanigel, Robert. 1991. The Man Who Knew Infinity: A Life of the Genius Ramanujan. New York: Washington Square Press. Kapadia, Karin. 1995. Siva and Her Sisters: Gender, Caste, and Class in Rural South India. Boulder, CO: Westview. Karve, D. D., ed. 1963. The New Brahmans: Five Maharashtrian Families. Berkeley: Univer­ sity of California Press. Kaviraj, Sudipta. 2010 [1994]. “On the Structure of Nationalist Discourse.” In The Imaginary Institution of India: Politics and Ideas, 85–126. New York: Columbia University Press. Khandelwal, Madhulika S. 2002. Becoming American, Being Indian: An Immigrant Community in New York City. Ithaca, NY: Cornell University Press. Khare, R. S. 1970. The Changing Brahmans: Associations and Elites among the Kanya-Kubjas of North India. Chicago: The University of Chicago Press. Khilnani, Sunil. 1997. The Idea of India. London: Penguin. Kibria, Nazli. 2002. Becoming Asian American: Second-Generation Chinese and Korean American Identities. Baltimore, MD: Johns Hopkins University Press. Kidder, Robert L. 1974. “Formal Litigation and Professional Insecurity: Legal Entrepre­ neurship in South India.” Law and Society Review 9 (1): 11–38. Knight, Douglas M., Jr. 2010. Balasaraswati: Her Art and Life. Middletown, CT: Wesleyan University Press. Knipe, David M. 1997. “Becoming a Veda in the Godavari Delta.” In India and Beyond: Aspects of Literature, Meaning, Ritual and Thought (Essays in Honour of Frits Staal), edited by Dick van der Meij, 306–32. London: Kegan Paul. Kolenda, Pauline M. 2003. “Memories of a Brahman Agraharam in Travancore.” In Caste, Marriage and Inequality: Essays on North and South India, 125–64. Jaipur: Rawat. Krishna, Anirudh, and Vijay Brihmadesam. 2006. “What Does It Take to Become a Soft­ ware Professional?” Economic and Political Weekly 41 (30) (July 29): 3307–14. Krishnama Acharya, P. 1918. “Dusi, North Arcot District.” In Some South Indian Villages, edited by Gilbert Slater, 84–94. London: Oxford University Press. Kumar, Deepak. 2006 [1995]. Science and the Raj: A Study of British India. Delhi: Oxford University Press. Kumar, Nita. 1992. Friends, Brothers, and Informants: Fieldwork Memoirs of Banaras. Berke­ ley: University of California Press. ——— —. 2007. The Politics of Gender, Community, and Modernity: Essays on Education in India. Delhi: Oxford University Press.

266 / Bibliography Kumar, Sanjay, Anthony Heath, and Oliver Heath. 2002. “Determinants of Social Mobil­ ity in India.” Economic and Political Weekly 37 (29) (July 20): 2983–87. Lahiri, Jhumpa. 2004. The Namesake. London: Harper Perennial. Lakshmi, C. S., ed. 2004. The Unhurried City: Writings on Chennai. New Delhi: Penguin. L’Armand, Kathleen, and Adrian L’Armand. 1983. “One Hundred Years of Music in Ma­ dras: A Case Study in Secondary Urbanization.” Ethnomusicology 27 (3): 411–38. The Laws of Manu. 1991. Translated by Wendy Doniger and Brian K. Smith. London: Penguin. Legislative Assembly Debates (Official Report), vols. 4–5 (September 1929). Simla: Govern­ ment of India, 1929. Leonard, Karen I. 1978. Social History of an Indian Caste: The Kayasths of Hyderabad. Berke­ ley: University of California Press. Lingat, Robert. 1973 [1967]. The Classical Law of India. Berkeley: University of California Press. Logan, Penelope. 1980. “Domestic Worship and the Festival Cycle in the South Indian City of Madurai.” PhD thesis, University of Manchester. Lokanathan, P. S. 1918. “Gangaikondan, Tinnevelly District.” In Some South Indian Villages, edited by Gilbert Slater, 53–76. London: Oxford University Press. Ludden, David. 1985. Peasant History in South India. Princeton, NJ: Princeton University Press. ——— —. 1999. An Agrarian History of South Asia. Cambridge: Cambridge University Press. Luhrmann, T. M. 1996. The Good Parsi: The Fate of a Colonial Elite in a Postcolonial Society. Delhi: Oxford University Press. Madan, T. N. 1965. Family and Kinship: A Study of the Pandits of Rural Kashmir. Bombay: Asia. ——— —. 1975. “On Living Intimately with Strangers.” In Encounter and Experience: Personal Accounts of Fieldwork, edited by André Béteille and T. N. Madan, 131–56. Delhi: Vikas. Mayo, Katherine. 1927. Mother India. London: Jonathan Cape. Mencher, Joan P. 1966. “Namboodiri Brahmins: An Analysis of a Traditional Elite in Ker­ ala.” Journal of Asian and African Studies 1 (3): 183–96. ——— —. 1970. “A Tamil Village: Changing Socioeconomic Structure in Madras State.” In Change and Continuity in India’s Villages, edited by K. Ishwaran, 197–218. New York: Columbia University Press. ——— —. 1972. “Continuity and Change in an Ex-Untouchable Community of South In­ dia.” In The Untouchables in Contemporary India, edited by J. Michael Mahar, 37–56. Tucson: University of Arizona Press. ——— —. 1978. Agriculture and Social Structure in Tamil Nad: Past Origins, Present Transformations and Future Prospects. New Delhi: Allied. Menon, Sadanand. 2012. “Chennai’s Annual ‘Mad Cow’ Syndrome.” Business Standard, January 7, 2012. http://www.business-standard.com/india/news/sadanand-menon -chennais-annual-mad-cow-syndrome/460956/ (accessed July 15, 2013). Mines, Diane P. 2005. Fierce Gods: Inequality, Ritual, and the Politics of Dignity in a South Indian Village. Bloomington: Indiana University Press. Mines, Mattison. 1994. Public Faces, Private Voices: Community and Individuality in South India. Berkeley: University of California Press. Mines, Mattison, and Vijayalakshmi Gourishankar. 1990. “Leadership and Individuality in South Asia: The Case of the South Indian Big-Man.” Journal of Asian Studies 49 (4): 761–86.

Bibliography / 267 Misra, B. B. 1961. The Indian Middle Classes: Their Growth in Modern Times. London: Oxford University Press. Mosse, David. 2012. The Saint in the Banyan Tree: Christianity and Caste Society in India. Berkeley: University of California Press. Mukherjee, Sanjukta. 2008. “Producing the Knowledge Professional: Gendered Geogra­ phies of Alienation in India’s New High-Tech Workplace.” In In an Outpost of the Global Economy: Work and Workers in India’s Information Technology Industry, edited by Carol Upadhya and A. R. Vasavi, 50–75. New Delhi: Routledge. Mukhopadhyay, Carol C. 1994. “Family Structure and Indian Women’s Participation in Science and Engineering.” In Women, Education, and Family Structure in India, edited by Carol C. Mukhopadhyay and Susan Seymour, 103–32. Boulder, CO: Westview. ——— —. 2004. “A Feminist Cognitive Anthropology: The Case of Women and Mathemat­ ics.” Ethos 32 (4): 458–92. Mukhopadhyay, Carol C., and Susan Seymour. 1994. “Introduction and Theoretical Over­ view.” In Women, Education, and Family Structure in India, edited by Carol C. Mukho­ padhyay and Susan Seymour, 1–33. Boulder, CO: Westview. Nagam Aiya, V. 1906. The Travancore State Manual. Vol. 2. Trivandrum: Travancore Gov­ ernment Press. Nair, K. S. 1978. Ethnicity and Urbanization: A Case Study of the Ethnic Identity of South Indian Migrants in Poona. Delhi: Ajanta. Nandy, Ashis. 1995. Alternative Sciences: Creativity and Authenticity in Two Indian Scientists. Delhi: Oxford University Press. Nanjundayya, H. V., and L. K. Ananthakrishna Iyer. 1928. The Mysore Tribes and Castes. Vol. 2. Mysore: Mysore University. Narasimhan, Haripriya. 2011. “Adjusting Distances: Menstrual Pollution among Tamil Brahmins.” Contributions to Indian Sociology (n.s.) 45 (2): 243–68. Narayanan, Vasudha. 2007. “Performing Arts, Re-Forming Rituals: Women and Social Change in South India.” In Women’s Lives, Women’s Rituals in the Hindu Tradition, edited by T. Pintchman, 177–98. New York: Oxford University Press. Narayana Rao, Velcheru. 2007. “The Play in Context: A Second Look at Apparao’s Kanyasulkam.” In Girls for Sale, Kanyasulkam: A Play from Colonial India, by Gurajada Ap­ parao and translated by V. Narayana Rao, 159–89. Bloomington: Indiana University Press. Natarajan, B. 1940. “Gangaikondan.” In Some South Indian Villages: A Resurvey, edited by P. J. Thomas and K. C. Ramakrishnan, 55–116. Madras: University of Madras Press. Neild-Basu, Susan. 1984. “The Dubashes of Madras.” Modern Asian Studies 18 (1): 1–31. Nishimura, Yuko. 1998. Gender, Kinship and Property Rights: Nagarattar Womanhood in South India. Delhi: Oxford University Press. Ortner, Sherry. 1998. “Identities: The Hidden Life of Class.” Journal of Anthropological Research 54 (1): 1–17. Osella, Filippo, and Caroline Osella. 2000. Social Mobility in Kerala: Modernity and Identity in Conflict. London: Pluto Press. O’Shea, Janet. 2007. At Home in the World: Bharata Natyam on the Global Stage. Middle­ town, CT: Wesleyan University Press. Pandian, M. S. S. 2007. Brahmin and Non-Brahmin: Genealogies of the Tamil Political Present. Delhi: Permanent Black. Parry, Jonathan P. 2003. “Nehru’s Dream and the Village ‘Waiting Room’: Long-Distance Labour Migrants to a Central Indian Steel Town.” Contributions to Indian Sociology (n.s.) 37 (1–2): 217–49.

268 / Bibliography Pate, H. R. 1917. Tinnevelly (Madras District Gazetteers). Vol. 1. Madras: Government Press. Patterson, Maureen L. P. 1970. “Changing Patterns of Occupation among Chitpavan Brah­ mans.” Indian Economic and Social History Review 7 (3): 375–96. Paul, John J. 1991. The Legal Profession in Colonial South India. Bombay: Oxford University Press. Peterson, Indira V., and Davesh Soneji. 2008. “Introduction.” In Performing Pasts: Reinventing the Arts in Modern South India, edited by Indira V. Peterson and Davesh Soneji, 1–40. Delhi: Oxford University Press. Prakash, Gyan. 1999. Another Reason: Science and the Imagination of Modern India. Prince­ ton, NJ: Princeton University Press. Prashad, Vijay. 2000. The Karma of Brown Folk. Minneapolis: University of Minnesota Press. Price, Pamela G. 1989. “Ideology and Ethnicity under British Imperial Rule: ‘Brahmans,’ Lawyers and Kin-Caste Rules in Madras Presidency.” Modern Asian Studies 23 (1): 151–77. Purkayastha, Bandana. 2005. Negotiating Ethnicity: Second-Generation South Asian Americans Traverse a Transnational World. New Brunswick, NJ: Rutgers University Press. Radhakrishnan, P. 1989. “Tamil Nadu Backward Classes.” Bulletin, Madras Development Seminar Series 19 (10): 500–12. ——— —. 1996. “Backward Class Movements in Tamil Nadu.” In Caste: Its Twentieth Century Avatar, edited by M. N. Srinivas, 110–34. New Delhi: Viking. Radhakrishnan, Smitha. 2011. Appropriately Indian: Gender and Culture in a New Transnational Class. Durham, NC: Duke University Press. Raghavaiyangar, S. Srinivasa. 1893. Memorandum on the Progress of the Madras Presidency during the Last Forty Years of British Administration. 2nd ed. Madras: Government Press. Raghavan, V. 1956. “Variety and Integration in the Pattern of Indian Culture.” Far Eastern Quarterly 15 (4): 497–505. Rajagopal, Arvind. 2001. Politics After Television: Religious Nationalism and the Reshaping of the Indian Public. Cambridge: Cambridge University Press. Rajam Aiyar, B. R. 1998 [1893–95]. The Fatal Rumour: A Nineteenth-Century Indian Novel. Translated by Stuart Blackburn. Delhi: Oxford University Press. Ramadorai, S. 2011. The TCS Story . . . and Beyond. New Delhi: Penguin. Raman, Bhavani. 2012. Document Raj: Writing and Scribes in Early Colonial South India. Chicago: The University of Chicago Press. Raman, Sita Anantha. 1996. Getting Girls to School: Social Reform in the Tamil Districts, 1870–1930. Calcutta: Stree. Ramaswamy, Sumathi. 1997. Passions of the Tongue: Language Devotion in Tamil India, 1891–1970. Berkeley: University of California Press. Ramaswamy, Uma. 1978. “The Belief System of the Non-Brahmin Movement in India: The Andhra Case.” Asian Survey 18 (3): 290–300. Ramesh Bairy, T. S. 2009. Being Brahmin, Being Modern: Exploring the Lives of Caste Today. New Delhi: Routledge. Rangaswamy, Padma. 2000. Namasté America: Indian Immigrants in an American Metropolis. University Park, PA: Pennsylvania State University Press. Rao, Nikhil. 2007. “‘South Indians are Like That Only’: Communal Identity in Late Colo­ nial Bombay.” In From the Colonial to the Postcolonial: India and Pakistan in Transition, edited by Dipesh Chakrabarty, Rochona Majumdar, and Andrew Sartori, 182–200. Delhi: Oxford University Press.

Bibliography / 269 Raychaudhuri, Tapan. 1988. Europe Reconsidered: Perceptions of the West in Nineteenth Century Bengal. Delhi: Oxford University Press. Redfield, Robert, and Milton B. Singer. 1954. “The Cultural Role of Cities.” Economic Development and Cultural Change 3 (1): 53–73. Reiniche, Marie-Louise. 1978. “Statut, fonctions et droits: Relations agraires au Tamilnad.” L’Homme 18 (1–2): 135–66. ——— —. 1979. Les dieux et les hommes: Étude des cultes d’un village du Tirunelveli, Inde du sud. Paris: Mouton. ——— —. 1981. “La maison au Tirunelveli (Inde du sud).” Bulletin de l’École Française d’Extrême Orient 70: 21–57. —. 1996. “The Urban Dynamics of Caste: A Case Study from Tamilnadu.” In Caste ——— Today, edited by C. J. Fuller, 124–49. Delhi: Oxford University Press. Report of the Tamil Nadu Second Backward Classes Commission (chaired by J. A. Ambasankar). Madras: Government of Tamil Nadu, 1985. Report on Public Instruction in the Bombay Presidency (annual), 1857/58–1942/43. Bombay: Government of Bombay. Report on Public Instruction in the Madras Presidency (annual), 1854/55–1939/40. Madras: Government of Madras. Row, T. Venkasami. 1883. A Manual of the District of Tanjore in the Madras Presidency. Madras: Lawrence Asylum Press. Rudner, David W. 1994. Caste and Capitalism in Colonial India: the Nattukottai Chettiars. Berkeley: University of California Press. Saraswathi, S. 1974. Minorities in Madras State: Group Interests in Modern Politics. Delhi: Impex India. Sarkar, Tanika. 2001. Hindu Wife, Hindu Nation: Community, Religion and Cultural Nationalism. Delhi: Permanent Black. Sarma, Rani Siva Sankara. 2007 [2002]. The Last Brahmin: Life and Reflections of a ModernDay Sanskrit Pandit. Translated by D. Venkat Rao. Delhi: Permanent Black. Sattanathan, A. N. 2007. Plain Speaking: A Sudra’s Story. Delhi: Permanent Black. Seal, Anil. 1968. The Emergence of Indian Nationalism: Competition and Collaboration in the Later Nineteenth Century. Cambridge: Cambridge University Press. Sender, Henny. 1988. The Kashmiri Pandits: A Study of Cultural Choice in North India. Delhi: Oxford University Press. Seth, Sanjay. 2007. Subject Lessons: The Western Education of Colonial India. Durham, NC: Duke University Press. Shankar, Shalini. 2008. Desi Land: Teen Culture, Class, and Success in Silicon Valley. Dur­ ham, NC: Duke University Press. Shukla, Sandhya. 2003. India Abroad: Diasporic Cultures of Postwar America and England. Princeton, NJ: Princeton University Press. Singer, Milton. 1972. When a Great Tradition Modernizes. London: Pall Mall Press. Singh, Andrea M. 1976. Neighbourhood and Social Networks in Urban India. Delhi: Marwah. Sinha, Mrinalini. 2006. Specters of Mother India: The Global Restructuring of an Empire. Dur­ ham, NC: Duke University Press. Sivaraman, Mythily. 2006. Fragments of a Life: A Family Archive. New Delhi: Zubaan. Sivertsen, Dagfinn. 1963. When Caste Barriers Fall: A Study of Social and Economic Change in a South Indian Village. Oslo: Universitets Forlaget; George Allen and Unwin. Slezkine, Yuri. 2004. The Jewish Century. Princeton, NJ: Princeton University Press. Soneji, Davesh. 2012. Unfinished Gestures: Devada¯sı¯s, Memory, and Modernity in South India. Chicago: The University of Chicago Press.

270 / Bibliography Sreenivas, Mythili. 2008. Wives, Widows, and Concubines: The Conjugal Family Ideal in Colonial India. Bloomington: Indiana University Press. Srinivas, M. N. 1952. Religion and Society among the Coorgs of South India. Oxford: Claren­ don Press. ——— —. 1966. Social Change in Modern India. Berkeley: University of California Press. ——— —. 1977. “The Changing Position of Indian Women.” Man (n.s.) 12 (2): 221–38. ——— —. 2002. Collected Essays. Delhi: Oxford University Press. Srinivas, Tulasi. 2010. Winged Faith: Rethinking Globalization and Religious Pluralism through the Sathya Sai Movement. New York: Columbia University Press. Subbarama Aiyar, S. 1918. “Watakanchery, Cochin State.” In Some South Indian Villages, edited by Gilbert Slater, 123–45. London: Oxford University Press. Subrahmanyam, Y. Subhashini. 1975. Social Change in Village India: An Andhra Case Study. New Delhi: Prithvi Raj. Subramaniam, K. 1974. Brahmin Priest of Tamil Nadu. New Delhi: Wiley Eastern. Subramanian, Lakshmi. 2006. From the Tanjore Court to the Madras Music Academy: A Social History of Music in South India. Delhi: Oxford University Press. Subramanian, Narendra. 1999. Ethnicity and Populist Mobilization: Political Parties, Citizens and Democracy in South India. Delhi: Oxford University Press. Subramanian, T. S. 2004. “The Arrest of Jayendra Saraswati,” “Legal Opinion,” “A Planned Murder.” Frontline, November 20–December 3, 2004, pp. 4–8. http://front lineonnet.com/f12124/stories/20041203006100400.htm, http://frontlineonnet.com /f12124/stories/20041203007500600.htm, http://frontlineonnet.com/f12124 /stories/20041203006400700.htm (accessed July 15, 2013). Suntharalingam, R. 1974. Politics and Nationalist Awakening in South India, 1852–1891. Tucson: University of Arizona Press. Sur, Abha. 2011. Dispersed Radiance: Caste, Gender, and Modern Science in India. New Delhi: Navayana. Thurston, Edgar, and K. Rangachari. 1909. Castes and Tribes of Southern India. Vol. 1. Madras: Government Press. Trautmann, Thomas R. 1997. Aryans and British India. Berkeley: University of California Press. ——— —. 2006. Languages and Nations: The Dravidian Proof in Colonial Madras. Berkeley: University of California Press. Ullrich, Helen E. 2010. “Marital Intimacy: A Four-Generation Study of Change among Havik Brahmins.” In Histories of Intimacy and Situated Ethnography, edited by Karen I. Leonard, Gayatri Reddy, and Ann G. Gold, 105–34. New Delhi: Manohar. Upadhya, Carol. 1997. “Social and Cultural Strategies of Class Formation in Coastal Andhra Pradesh.” Contributions to Indian Sociology (n.s.) 31 (2): 243–68. —. 2007. “Employment, Exclusion and ‘Merit’ in the Indian IT Industry.” Economic ——— and Political Weekly 42 (20) (May 19): 1863–8. Upadhya, Carol, and A. R. Vasavi. 2006. Work, Culture, and Sociality in the Indian IT Industry: A Sociological Study (Final Report to Indo-Dutch Programme for Alternatives to Develop­ ment). http://www.wbitsa.org/documents/idpadfinalreport.pdf (accessed July 15, 2013). ——— —. 2008. “Outposts of the Global Information Economy: Work and Workers in In­ dia’s Outsourcing Industry.” In In an Outpost of the Global Economy: Work and Workers in India’s Information Technology Industry, edited by Carol Upadhya and A. R. Vasavi, 9–49. New Delhi: Routledge. Vaid, Divya. 2012. “The Caste-Class Association in India: An Empirical Analysis.” Asian Survey 52 (2): 395–422.

Bibliography / 271 van der Veer, Peter. 2001. Imperial Encounters: Religion and Modernity in India and Britain. Princeton, NJ: Princeton University Press. Veeraraghavan, A. K. 1940. “Dusi.” In Some South Indian Villages: A Resurvey, edited by P. J. Thomas and K. C. Ramakrishnan, 182–213. Madras: University of Madras Press. Venkatachalapathy, A. R. 2006. In Those Days There Was No Coffee: Writings in Cultural History. New Delhi: Yoda. ——— —. 2009. “The ‘Classical’ Language Issue.” Economic and Political Weekly 44 (2) (Janu­ ary 10): 13–15. Venkatesan, V. 2004. “A Failed Mediator.” Frontline, November 20–December 3, 2004, pp. 8–11. http://frontlineonnet.com/f12124/stories/20041203007901100.htm (ac­ cessed July 15, 2013). ——— —. 2005. “Education: Cry for Corrective Action.” Frontline, August 27–September 9, 2005, pp. 28–37. http://frontlineonnet.com/f12218/stories/20050909004403000 .htm (accessed July 15, 2013). Visvesvaraya, M. 1951. Memoirs of my Working Life. Bangalore: self-published. Viswanathan, Padma. 2009. The Toss of a Lemon. Boston: Mariner. Viswanathan, S. 2004. “A Controversial Career.” Frontline, November 20–December 3, 2004, pp. 11–12. http://frontlineonnet.com/f12124/stories/20041203006300800 .htm (accessed July 15, 2013). Wadley, Susan S. 1994. Struggling with Destiny in Karimpur, 1925–1984. Berkeley: Univer­ sity of California Press. ——— —. 2000. “The Village in 1998.” In Behind Mud Walls: Seventy-Five Years in a North Indian Village, by William H. Wiser, Charlotte V. Wiser, and Susan S. Wadley, 319–38. Berkeley: University of California Press. Waghorne, Joanne P. 1994. The Raja’s Magic Clothes: Re-Visioning Kingship and Divinity in England’s India. University Park, PA: Pennsylvania State University Press. ——— —. 2004. Diaspora of the Gods: Modern Hindu Temples in an Urban Middle-Class World. New York: Oxford University Press. Wagoner, Phillip B. 2003. “Pre-Colonial Intellectuals and the Production of Colonial Knowledge.” Comparative Studies in Society and History 45 (4): 783–814. Wali, Kameshwar C. 1991. Chandra: A Biography of S. Chandrasekhar. Chicago: The Uni­ versity of Chicago Press. Washbrook, David A. 1975. “Political Change in a Stable Society: Tanjore District 1880 to 1920.” In South India: Political Institutions and Political Change 1880–1940, edited by C. J. Baker and D. A. Washbrook, 20–68. Delhi: Macmillan. ——— —. 1976. The Emergence of Provincial Politics: The Madras Presidency, 1870–1920. Cam­ bridge: Cambridge University Press. ——— —. 1989. “Caste, Class and Dominance in Modern Tamil Nadu: Non-Brahmanism, Dravidianism and Tamil Nationalism.” In Dominance and State Power in Modern India: Decline of a Social Order, edited by Francine R. Frankel and M. S. A. Rao, vol. 1, pp. 204–64. Delhi: Oxford University Press. ——— —. 2010. “The Maratha Brahmin Model in South India.” Indian Economic and Social History Review 47 (4): 597–615. Weber, Max. 1948. From Max Weber: Essays in Sociology, edited by H. H. Gerth and C. Wright Mills. London: Routledge and Kegan Paul. ——— —. 1967 [1917]. The Religion of India: The Sociology of Hinduism and Buddhism. New York: Free Press. ——— —. 1978 [1920]. Economy and Society. Berkeley: University of California Press.

272 / Bibliography Weidman, Amanda J. 2006. Singing the Classical, Voicing the Modern: The Postcolonial Politics of Music in South India. Durham, NC: Duke University Press. Weinstein, Jay A. 1974. Madras: An Analysis of Urban Ecological Structure in India. Beverly Hills, CA: Sage. Wildavsky, Ben. 2011. “Think Again: Education.” Foreign Policy (March/April 2011). http://www.foreignpolicy.com/articles/2011/02/22/think_again_education (accessed July 15, 2013). Wilson, H. H. 1855. A Glossary of Judicial and Revenue Terms. London: W. H. Allen. Wirth, Louis. 1938. “Urbanism as a Way of Life.” American Journal of Sociology 44 (1): 1–24. World Bank. 2000. India: Scientific and Technical Manpower Development in India (Report no. 20416-IN, August 30, 2000). http://go.worldbank.org/R5APUDZ000 (accessed July 15, 2013). Xiang Biao. 2007. Global “Body Shopping”: An Indian Labor System in the Information Technology Industry. Princeton, NJ: Princeton University Press. Yanagisawa, Haruka. 1996. A Century of Change: Caste and Irrigated Lands in Tamilnadu 1860s-1970s. New Delhi: Manohar. Yocum, Glenn. 1992. “The Coronation of a Guru: Charisma, Politics, and Philosophy in Contemporary India.” In A Sacred Thread: Modern Transmission of Hindu Traditions in India and Abroad, edited by Raymond Brady Williams, 68–91. Chambersburg, PA: Anima. Younger, Paul. 1995. The Home of Dancing S´ivan: The Traditions of the Hindu Temple in ¯ Citamparam. New York: Oxford University Press.

Index

Acharya, M. K., 129, 218, 219 Adi Dravidas, 7, 30, 33, 34, 35, 36, 37–38, 46, 47. See also Dalits administration. See government and administration agraha¯rams, 1, 25, 31, 49, 54, 59, 160, 161, 188, 190, 228; in Eighteen-Village Vattima villages, 39–40; in Kumbapettai, 32, 33–36, 38; meaning of term, 30–31, 48–49, 53; in Mel Ceval, 48; in Sripuram, 32–33, 36–37, 38, 46–47; in Thyagasamuthiram, 39; in Tippirajapuram, 40–46, 47; in urban areas, 30, 155–57; in Yanaimangalam, 48 All-India Anna Dravida Munnetra Kazhagam (AIADMK), 10–11, 90, 198, 199 American dream, 26, 169, 177 American suburbs. See United States: suburbs in Anandatandavapuram, 39, 40, 233 Andhra Pradesh, 31, 155, 217, 232 anti-Brahmanism, 11–13, 14, 185; in Maharashtra, 214, 216 Arundale, Rukmini Devi, 205–6 Aryans, Brahmans identified as, 11, 56, 184–85, 217, 228 asceticism, 8, 17, 199 Balasaraswati, Tanjore, 206 Bangalore, 13, 26, 69, 75, 93, 138, 153, 154, 168, 217, 238; development of, 164–65; and IT, 23, 107–8, 110, 111, 117, 118, 146, 147; modernity of, 163–64, 166, 167, 180–81; population of, 164 banks, 97–98, 100, 101, 137

Barnett, Marguerite Ross, 11, 54 Bengal, 127, 128, 151, 212, 213–15 Bengali Brahmans, 212, 216, 223 Béteille, André, 13, 32, 36–38, 46–47, 50, 189; on caste and class, 37–38; on middle class, 16, 119, 121 bhadralok, 212–14, 215–16, 219, 223 bhajanas, 163, 174, 194, 195 Bharatanatyam, 27, 141, 163, 174, 179, 180, 202, 207, 208, 209–10; development of, 201, 204, 205–7 Bhashyam Iyengar, V., 65, 66, 80–81, 128, 156, 201 Bombay Presidency, 27, 68, 212–15 Calcutta, 68, 69, 74, 154, 212 Caplan, Lionel, 20–21, 100, 222 Caplan, Patricia, 137, 138, 142, 147, 209 Carnatic music, 27, 141, 163, 179, 193, 197, 207, 208; development of, 201–5; and Sanskritic Hinduism, 198, 201, 209–10 caste: and agrarian class, 34, 37–38; under colonialism, 31, 56; “privatization” of, 15, 44; in Tamilnadu, 7, 13–15; “traditionalism,” 5, 71; and urbanization, 54–55, 58; and varn · a, 6–7, 55, 184 ceris (Dalit colonies), 30, 32, 33, 35, 40, 48, 49, 188, 189 Chakra Vehicles, 101, 103, 104–6, 109, 114, 116, 119, 145 Chandrasekhar, Subrahmanyan, 73, 75–77 Chennai (Madras), 20, 26, 55, 58, 120, 154, 168, 215, 218, 235–37; growth of, 154–55; housing in, 156–58; and IT, 22,

274 / Index Chennai (Madras) (cont.) 108, 109, 110; migration to, 52, 59; modernity and tradition in, 155, 162–64, 180; population of, 155; southern suburbs of, 58, 156–59, 162, 181, 192 Chennai neighborhoods and suburbs: Adyar, 156, 236–37; Alwarpet and Teynampet, 156, 236–37; Besant Nagar, 156, 183–84, 193, 195, 206, 236–37; Georgetown, 155, 156; Mylapore, 81, 154, 155, 156, 191, 193, 236–37; Nun­ gambakkam, 1–2, 155; T. Nagar, 156, 158, 170, 192, 236–37; Triplicane, 155, 156, 236 Chettiyars, 17, 45, 97 Child Marriage Restraint Act. See Sarda Act Chitpavan Brahmans, 213, 214–15, 223–24 Christians, 7, 20–21, 40, 100, 222 classes and status groups, 17–18, 221–23 Cochin, 31, 225, 232 Cognizant Technology Solutions, 107, 108, 111 companies: engineering and manufacturing, 98, 99–107, 115, 119; public-sector, 100–103; Tamil Brahman-owned, 97–98 company managers, 3, 18, 20, 21, 81, 85, 97–98, 118, 211, 228; in IT, 107–9, 112, 114; in manufacturing, 99–100, 103–7 Congress, 9, 10, 65, 90, 129, 202, 214 cooks, Brahman, 8, 20, 58, 196 Dalits, 6–7, 9, 11, 14–15, 41, 42, 49. See also Adi Dravidas Delhi, 13, 96, 108, 138, 165–66, 238 De Neve, Geert, 15–16, 225–26 Deshastha Brahmans, 5, 59, 61–63, 65, 213, 216, 224 devada¯sis, 204, 205–6 dewans, 77, 81, 85; of Baroda, 79; of Mysore, 70; of Pudukkottai, 64, 77; of Travancore, 63, 64, 77, 202 dharma, 2, 56–57, 184, 217, 218 dharmas´a¯stra, 129, 184 Dickey, Sara, 16–17, 21 doctors, 20, 93–94, 137, 138, 169; Tamil Brahman, 71, 98 Donner, Henrike, 15–16 Dravida Kazhagam (DK), 10, 12 Dravida Munnetra Kazhagam (DMK), 10– 11, 90, 198, 199

Dravidian movement. See non-Brahman movement Dusi, 31, 47, 234 economic liberalization, 4, 97, 99, 103, 107, 109, 144, 155, 229; and middle class, 15–16, 19, 20, 21, 27, 101, 102 education: Brahman tradition of, 15, 61, 71, 118; and child marriage, 134–36, 215; English-language, 5, 62–63, 64, 66, 81, 92, 119, 217; female, 92, 124–26, 127, 130, 137–38, 144–45, 151, 211, 222; higher and professional, 63–65, 67–69, 71, 93–96; male, 91–92; and marriage, 138–39, 141; school, 91–93, 119, 163; in villages, 36, 135 Eighteen-Village Vattimas, 21, 23, 200, 207–8, 232–33; in Bangalore and Mumbai, 166–68; and banking, 97; in Chennai, 159–62; and child marriage, 133–36; and land ownership, 53–54; position of women among, 43–44, 57–58, 124–26, 132; subcaste association of, 159; in United States, 171–74; urban migration of, 40, 45–46; villages of, 39–40, 47, 232–35 (see also Tippirajapuram) engineering: industry, 72, 73, 94, 139, 145, 217; profession, 5, 26, 65, 67, 71, 81, 93, 145, 219 engineering colleges, 94–96; Brahman students in, 67, 95; in Chennai (Madras), 64, 67–69, 95, 163; and IT, 108 engineers, 20, 94; Tamil Brahman, 69–72, 98; women, 104, 145 “exposure”, 113, 119–20, 147 funerals, 190–91, 195–96 Gangaikondan, 31, 50, 234 gender inequality, 26, 123–24, 145, 150– 52, 181, 222 goddess (Devi), 8, 188, 189, 193; village, 33, 188, 189 Gough, Kathleen, 32–34, 36, 37, 46–47, 50, 51, 81–83, 131, 140 Gounders, 51, 225–26, 227 government and administration: colonial, 5, 26, 61–63, 64, 65–67, 71, 81, 84–85; officials in, 9, 19–20; post-Independence, 91, 96–97, 100

Index / 275 government and public sector, attitudes toward, 102–3, 116–17 Guindy Engineering College, 64, 95 gurus and “god-men,” 163, 192, 195 Harriss, John, 23, 97, 98, 192, 199, 220 houses, structure of, 43–44 Indian Computer Services (ICS), 109, 112–15, 146–49 Indian Institutes of  Technology (IIT), 93, 94, 95, 107, 112, 144, 145, 163 information technology (IT) industry, 4, 95–96, 107–8; Brahmans in, 15, 22–23, 111, 117–22, 150; companies in, 99, 105, 106, 107–12, 115–17; “flat hierarchies” in, 109, 145–46; offices of, 109, 145; recruitment into, 96, 108, 118, 119, 121, 145 IT professionals, 20, 99, 105, 108–17, 118, 155, 177; women, 108, 109, 110, 114, 145–50. See also software engineers Jats, 226–27 Justice Party, 10, 11, 129, 217 Kallars, 38, 54 Kalyanapuram, 38, 42 Kammas, 226, 227 Kanakkuppillais, 61–62 Kannada Brahmans, 5, 57, 59, 207, 217, 219–20, 223 Kanya-Kubja Brahmans, 224 Karnataka, 31, 220, 223, 232 Kashmiri Pandits, 25, 224 Kaveri delta, river, 21, 29, 30, 32, 39, 40, 47, 187, 188 Kayasthas, 212, 224–25, 227 Kerala, 31, 225, 232 kings, 30, 66, 81, 189 Konerirajapuram, 39–40, 78, 233 Krishnamachari, V. T., 79, 81, 202 Kumbakonam, 39, 52, 159, 164, 167, 168, 180; college in, 64 Kumbapettai, 32–37, 38, 50, 52, 53, 54, 81–83, 131, 136, 140, 234 land, Brahmans’ attitude to, 49–52, 53–54, 224, 225, 228 law, colonial system of, 5, 26, 56, 65–67, 71, 81

law college (Madras), 64, 67–68; Brahman students in, 67 lawyers, 20, 64, 65, 80–81, 84, 93, 98, 137, 156, 212–23 literacy: female, 124; male, 65, 225 Ludden, David, 50, 51, 52 Madhava Rao, T., 63, 64, 77 Madras (city). See Chennai Madras High Court, 9, 64, 65, 80 Madras High School, 63, 64, 77 Madras Music Academy, 74, 191, 202–4, 205–6 Madras Presidency, 9, 61–62, 66, 127, 151, 215–16, 231 Madras, University of, 9, 64, 65, 67–68, 129 Madurai, 16, 21, 30, 156, 157, 164, 189 Malabar, 31, 225, 232 marriage: arranged, 139; and class status, 21, 143; close-kin, 140; companionate, 139–43, 222; endogamous, 6, 140, 143; and women’s employment, 137, 147–49 marriage, child/prepuberty, 26, 123–24, 134–36, 206, 211; and social reform, 126–31; women’s experience of, 131–34 mathematics, 73, 95, 117–18, 146 Mayiladuthurai, 97, 159, 187 medical colleges, 93, 95; in Bombay, 68; Brahman students in, 67–68; in Madras, 64, 67–68, 163 medicine: Brahman attitudes to, 68, 71–72; profession of, 5, 26, 65, 68, 71–72, 81, 93, 94, 95, 139, 217, 219; Western, 76, 217 Mel Ceval, 48, 188, 233 middle class: in colonial period, 151, 212– 13, 215–16, 217, 219; consumerism among, 16–17, 222; incomes of, 19, 20, 99–100; Indian, 15–17, 18–19, 221–22; and non-Brahmans, 225–27; and nonTamil Brahmans, 223–25; Tamil categories of, 20–21. See also Tamil Brahman middle class migration: to Bangalore, 165, 238; to Delhi, 165–66, 238; within India, 69, 83, 84, 97, 177, 181, 232; to Madras, 236; to Mumbai, 165, 238; overseas, 26, 116, 169–70, 177, 178, 181; return, 111, 177–79, 181; and sale of land, 52–54; to United States, 169–70; urban, 26, 32,

276 / Index migration (cont.) 33, 38, 39, 40, 45, 46, 59, 83, 84, 85, 222 mira¯sida¯rs, 29–30, 32, 33, 34, 49–52, 53 modernity, 217–21, 227–28; and middle class, 16, 211–12, 222; precolonial, 215; and tradition, 17, 27, 210, 211, 218–19, 220–21 Mudaliars, 55 Mudikondan, 39, 40, 78, 233 Mumbai (Bombay), 13, 23, 26, 59, 69, 108, 213, 238; modernity of, 163–64, 166, 180; population of, 164 musicians: Brahman, 201, 207–8; nonBrahman, 201, 207, 209 Muslims, 7, 40, 43 Muthusamy Iyer, T., 64 Mylapore group, 65, 76, 202, 218 Mysore, 70, 164, 165, 201, 219, 232 Nagarattars, 55, 59, 98 Nambudiri Brahmans, 219 Narayana Rao, Velcheru, 215, 216 nationalism and nationalist movements: in Bengal, 213–15; Hindu, 192, 200, 224; Indian, 9, 76, 127, 128, 217; in Madras, 128, 129, 201, 206, 218; in Maharashtra, 214–15 Nayars, 65, 207, 225 Non-Brahman Manifesto, 9–10, 23, 26, 219 non-Brahman movement, 9–13, 27, 51, 65, 129, 184, 185, 196, 204, 228; in Andhra, 217; in Karnataka, 220; in Maharashtra, 214 non-Brahmans, 6–8, 13–15, 17, 54–55, 120, 201, 204; and relationship with Brahmans, 33–38, 46–47, 50–51, 128, 151, 218–19 “old” and “new” economies, 100–103, 109, 114, 116 Padaiyachis, 12 Palar valley, river, 30, 31, 47 Pallars, 32, 33, 34 Pandian, M. S. S., 11, 218 Paraiyars, 40 Parsis, 213, 223 pollution, ritual, 35, 58, 161, 218; death, 191; and medicine, 5, 68, 71; menstrual, 35–36, 43, 44, 57–58, 151–52, 160

Prakash, Gyan, 76–77, 221 priests: Brahman, 6, 13, 20, 85, 184, 186– 88, 196, 198; domestic, 187, 196; nonBrahman, 8, 34, 188; temple, 3, 8, 33– 34, 186–88, 190, 193, 194, 198, 200 “primary” and “quasi-primary” relationships, 154, 162, 169, 175 purity and pollution rules, 34, 35, 41, 43–44, 57, 58, 123 purity, ritual, 6, 8–9, 43, 54, 57, 151, 190 Radhakrishnan, Smitha, 145, 147, 149, 150 Raghavan, V., 191, 206 Ramadorai, S., 89–90, 111, 116 Raman, C. V., 73–77, 202, 203 Raman, Sita Anantha, 130, 134 Ramanujan, Srinivasa, 72–73, 75, 117 Ramasamy, E. V., 12, 184, 196 Ramaswamy Aiyar, C. P., 202 Ramaswamy, Sumathi, 185, 196, 209 research projects of authors, 21–25 reservations system: post-Independence, 90– 91, 93, 94, 95, 100; pre-Independence, 10, 68, 69, 90 retirement: to Tamilnadu, 167–68; to villages, 40, 52, 160–61 sacred threads, 12, 13, 57, 191 sandhya¯, 57, 191, 195 Sanskrit: and Brahmans, 47, 61, 196, 190, 198; and Carnatic music, 201, 204; and English, 196; and Shankaracharyas, 200; and Tamil, 6, 47, 185, 196, 197, 228; in temple worship, 188, 198 Sanskritic culture and tradition, 11, 27, 47, 163, 184–85, 201, 209–10, 211, 222 Sanskritic Hinduism, 8, 163, 183, 192, 195, 196, 198, 200, 209–10, 211, 218; anthropological concepts of, 24, 185–86, 189, 191–92 Sanskritization, 24, 210, 222–23 Sarda Act, 126, 128–31, 134, 135, 219 Satyamurti Aiyar, 1–3, 19, 30, 54, 83, 155 schoolteachers, 12, 20, 66, 83; women, 129, 138 science and technology, 15, 73, 95, 139, 146, 217; and religion, 74, 76–77, 218 Self-Respect movement, 12, 184 Sengalipuram, 39, 233 Seshia Sastri, A., 64, 77–78, 81

Index / 277 Shankaracharyas: Kanchipuram, 199–200, 220; Sringeri, 194, 197, 199, 200 Shiva, 6, 8, 183, 185, 188, 189, 193 Shudras, 6, 10, 12, 55, 184 Singer, Milton, 98, 163, 186, 191–92, 199, 220 ´sloka classes, 174, 193, 195, 196–98 social reform, 213–15; and child marriage, 126–31 software engineers, 5, 20, 96, 110, 111, 117, 118, 177. See also IT professionals Srinivas, M. N., 24, 138, 186, 196, 220 Srinivasa Iyer, Semmangudi, 207–8 Sripuram, 32–33, 36–38, 42, 46–47, 50, 53, 157, 189, 233, 234 Subbalakshmi, R. S. (“Sister”), 129–30, 131, 133 Subbulakshmi, M. S., 204 Subrahmanya Ayyar, C., 74, 75, 203 Subramania Iyer, G., 128 Subramania Iyer, S., 65, 66, 74, 76, 128, 201 Tambraparni valley, river, 30, 31, 48, 188 Tamil. See Sanskrit: and Tamil Tamil Brahmanhood, 15, 17, 27, 44, 184, 211, 227, 229 Tamil Brahman middle class: 3, 17, 19–20, 26, 84–87, 119–22, 211, 221–23, 229; and class mobility, 20, 26, 84, 85–87, 118–19, 121, 221; in colonial period, 123–24; as “isomorphic” status group, 17, 27, 223, 227; and music and dance, 201–2, 207, 208, 210; and religion, 184, 193–94, 200, 210; and status of women, 126, 150–52, 222; and urbanization, 153 Tamil Brahmans: claim to superiority of, 9, 12, 15, 26, 34, 36, 128, 151–52, 191, 200, 210, 219; dress of, 12–13, 158; family deities of, 56; “family names” of, 55–56; and languages, 102, 168; population of, 5–6, 29, 231–41; speech of, 13, 158; statistical data on, 22–23; “surnames” of, 6, 14 Tamil Brahman sections and subcastes: Adishaiva, 131, 186–87, 200, 221, 240; Choliyar, 186, 240; Dikshitar, 187; Palghat, 31, 165; Panchangakkarar, 187; Smarta (Aiyar), 6, 8, 32, 187, 192, 199, 216, 240; Sri Vaishnava (Aiyangar), 6,

8, 32, 38, 187, 195, 197, 199, 216, 240. See also Eighteen-Village Vattimas Tamil Icai (“music”), 204, 206 Tanjore, Maratha kingdom of, 62, 63, 201, 205 Tata Consultancy Services (TCS), 89–90, 96, 107, 108, 111–12, 116 Tata group companies, 69, 94, 103, 104, 105, 111 Tediyur, 40, 43 Telugu Brahmans, 5, 65, 207, 216–17, 223, 225, 231, 232, 236; Niyogi, 31, 56, 59, 61–63, 65, 216, 219; Vaidika, 31, 56, 216–17, 219 temples: in Chennai, 163, 183, 190, 192– 94; in Mumbai, 165; in towns and cities, 189–90; in the United States, 174, 175, 176; in villages, 32–35, 37, 39–40, 41, 45, 56, 188–89 tenants, non-Brahman, 32, 33, 34, 37–38, 49, 52 Theosophy, 74, 76, 218 Thiruvenkatachari, C. R., 80–81, 133 Thyagasamuthiram, 39, 234 Tippirajapuram, 23, 25, 39, 40–46, 47, 49, 52, 157, 233 Tiruchirappalli, 29, 52, 157 Travancore, 31, 201, 225, 232 TVS group companies, 98, 101, 103, 105, 106 United States: children in, 175–77, 179–80; (grand)parents’ visits to, 176, 179; “model minorities” in, 169, 171, 177, 179; racism in, 171, 172; social life in, 173–75, 176; suburbs in, 23, 26, 170–71; working in, 172–73 Untouchables. See Adi Dravidas; Dalits Upadhya, Carol, 117, 150, 226 upanayana, 191 urbanization, 26, 52, 54–59, 153, 222 Vaigai valley, river, 30 Vanniyars, 1, 55 vegetarianism, 7–9, 17, 36, 43, 50, 58, 157, 173; and deities, 8–9, 188 Vellalars, 7, 50, 51, 55, 61–62 villages, 25, 26, 232–35; “Brahman,” 30, 47–48; non-Brahman area of, 30, 32–33, 40, 47–49, 188; Tamil terms for, 47–49

278 / Index Vishnu, 6, 8, 183, 188, 189, 193, 195 Visvesvaraya, Mokshagundam, 70 Washbrook, David, 30, 56, 219 Weber, Max, 17–18, 37, 71, 184, 185, 221, 228 weddings, 190; costs of, 141–42 widows, 123, 129–30, 131–32, 133; child, 126, 131–32; remarriage of, 127–28, 212, 213, 214

women: in agraha¯ram, 35–36; and Bharatanatyam, 205, 206–7, 208, 209–10; and Carnatic music, 203–4, 206–7, 208, 209–10; independence of, 161, 166–68, 173–74, 175, 181; and social control, 26, 123, 154, 159–62, 169, 175, 181. See also gender inequality worship: domestic, 190; temple, 186, 188 Yanaimangalam, 48, 188, 189, 234