Patronage as Politics in South Asia 9781107056084, 2014002313

362 137 12MB

English Pages [488] Year 2014

Report DMCA / Copyright


Table of contents :
List of Illustrations
A scandal in Jodhpur
What happened to patronage?
Patronage in South Asianist scholarship
Patronage close-up
The moral logic and the conflict within
Patronage as politics
Idols in Westminster
The Idea of Patronage in South Asia
1 The political economy of patronage, preeminence and the state in Chennai
Clients and patrons: a vignette about social mechanics
Research among castes of the left-hand and right-hand moieties
The public presentation of big-men: Traveling the street
Patronage and Tamil generosity
Locating the roles of the institutional big-man
Arenas of preeminence in the 19th and 20th centuries: Temples and government honours
The institutional big-man at the end of the 20th century
Temples and preeminence among left-hand castes
2 The temporal and the spiritual, and the so-called patron–client relation in the governance of Inner Asia and Tibet
3 Remnants of patronage and the making of Tamil Valaiyar pasts
Ambivalent clients
Fleeing and settling
Indexical density
What happened here?
Your time will come
4 Patronage and state-making in early modern empires in India and Britain
Patron–client relations in South Asian social analysis
Maratha power: limiting patrimony through patronage
The East India Company state
Markets and patronage
Democracy as Patronage
5 The paradox of patronage and the people’s sovereignty
Patrons and politics
Patronage in electoral law
The state and illegitimate influence
Negotiation and the morality of exchange
Patronage, elections and reputation
Concluding case
6 India’s demotic democracy and its ‘depravities’ in the ethnographic longue durée
Know Wiser?
Feast in the time of elections
The law of the fishes
Hierarchy, representation and the depravity of competition
7 ‘Vote banking’ as politics in Mumbai
Aziz Nagar
8 Political fixers in India’s patronage democracy
Social workers and party workers
Clientelistic exchanges
The BJP will help you
Changing channels of patronage
9 Patronage and autonomy in India’s deepening democracy
Changing relations of patronage in Balapalle
Moral elements and politicians-elect
A hero falters
A new expression of autonomous thinking
10 Police and legal patronage in northern India
Patrons and clients in the police
How little kingpins keep their little kingdoms in order
11 Patronage politics in post-independence India
Why does India have high levels of patronage politics?
The timing of mass political mobilisation vs. the professionalisation of bureaucracy
Ethnic heterogeneity
The effects of political competition
Prospects for change
Prospects and Disappointments
12 Kingship without kings in northern India
Patronage democracy, representation and divine kinship
Elected representatives, beneficent donors or violent protectors?
Divine genealogies, caste and the collective nature of patronage
Divine kinship and being extraordinary, or deifying ‘the people’
Fears, violence and strategic caste-based voting
13 The political bully in Bangladesh
In the beginning
Party affinities
Reciprocity, ‘contacts’ and loyalties
Action and violence
Politics as morality
Free-range clients in the underworld
14 The dark side of patronage in the Pakistani Punjab
Changing patterns of dominance
Popular politics and the dominance of landlords
Why elections did not improve political accountability
15 Patronage and printing innovation in 15th-century Tibet
Books as relics and merit-makers
A witness to the Tibetan printing revolution
A devout innovator in pursuit of legitimacy and power
The princely patron of print
The lama who oversaw the printing process
The history of Gung Thang
Printing for change in Tibet
In the 21st century
An ending thought
16 The (im)morality of mediation and patronage in south India and the Gulf
State, migration and the politics of mediation
Connectedness across the Indian Ocean
Expanding one’s networks
Modalities of connectedness and mediation
Morality and legitimacy in the informal life
Recommend Papers

Patronage as Politics in South Asia
 9781107056084, 2014002313

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview


Edited by

Anastasia Piliavsky

Cambridge House, 4381/4 Ansari Road, Daryaganj, Delhi 110002, India Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning and research at the highest international levels of excellence. Information on this title: © Cambridge University Press 2014 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2014 Printed in India A catalogue record for this publication is available from the British Library Library of Congress Cataloguing-in-Publication data Patronage as Politics in South Asia / edited by Anastasia Piliavsky.   pages cm Includes bibliographical references and index. Summary: “The volume shows that in South Asia patronage is no feudal residue or retrograde political pressure, but a vital political force behind South Asia’s burgeoning democratic cultures.”—Provided by publisher. isbn 978-1-107-05608-4 (hardback) 1. Patronage, Political—South Asia. 2. Social structure—South Asia. 3. Political sociology. I. Piliavsky, Anastasia, 1981– jq98.A91P38 2014 306.20954—dc23 2014002313 isbn 978-1-107-05608-4  Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

For my father

That which you call corruption I call influence. John Mortlock (1755–1816) British banker, woollen draper, Member of Parliament, and thirteen times Mayor and ‘Master of the Town of Cambridge’


List of Illustrations ix Foreword by  John Dunn xi Acknowledgments xv Introduction 1 Anastasia Piliavsky

The Idea of Patronage in South Asia 1. The political economy of patronage, preeminence and the state in Chennai Mattison Mines 2. The temporal and the spiritual, and the so-called patron–client relation in the governance of Inner Asia and Tibet D. Seyfort Ruegg 3. Remnants of patronage and the making of Tamil Valaiyar pasts Diane Mines 4. Patronage and state-making in early modern empires in India and Britain Sumit Guha





Democracy as Patronage 5. The paradox of patronage and the people’s sovereignty David Gilmartin 6. India’s demotic democracy and its ‘depravities’ in the ethnographic longue durée Anastasia Piliavsky 7. ‘Vote banking’ as politics in Mumbai Lisa Björkman


154 176

viii  Contents

8. Political fixers in India’s patronage democracy Ward Berenschot


9. Patronage and autonomy in India’s deepening democracy 217 Pamela Price, with Dusi Srinivas 10. Police and legal patronage in northern India Beatrice Jauregui


11. Patronage politics in post-independence India Steven I. Wilkinson


Prospects and Disappointments 12. Kingship without kings in northern India Lucia Michelutti


13. The political bully in Bangladesh Arild Engelsen Ruud


14. The dark side of patronage in the Pakistani Punjab Nicolas Martin


15. Patronage and printing innovation in 15th-century Tibet Hildegard Diemberger


16. The (im)morality of mediation and patronage in south India and the Gulf Filippo Osella


Contributors 395 Bibliography 399 Index 447

List of Illustrations

1 Rajasthan’s Chief Minister, Vasundhara Raje, depicted as goddess Annapurna. 2 1.1 The author and his assistant V. Gopalakrishnan interviewing Kaikkoolar naaḍu headmen in Tirumalasai in 1985. 40 3.1 Kumutam image of the Muttaraiyar statue in Trichy. 98 3.2 Signboard announcing a village branch of the Perumpituku Muttaraiyar Youth Association. 100 4.1 ‘Dun-Shaw’, a cartoon by Gillivray showing a kilted but crowned Dundas as a ‘Bashaw’, or oriental despot, captioned as follows: ‘One foot in Leadenhall Street and the other in the Province of Bengal’. 105 7.1 Kamble with social workers. 181 7.2 Water pipes in Aziz Nagar. 187 9.1 Map of the main regions of Andhra Pradesh prior to its bifurcation in 2013. 219 11.1 Percentage of total misallocated or lost food in the 2003–2004 Targeted Public Distribution System. 272 11.2 The policy mix in India at different levels of competition and economic development.  275 15.1 Map of southwestern Tibet. 347 15.2 The 1407 print of the sBas gdon gsal ba.351 15.3 Blockprinting at the Sera monastery in Tibet. 351 15.4 The Shel dkar (Shekar) fortress, the capital of La stod lho (Southern Lato). 355 15.5 A statue of Bo dong Phyogs las rnam rgyal. 358 15.6 A depiction of Chos kyi sgron ma. 360 15.7 A monk at the Paltsek Research Institute in Lhasa. 363

Foreword John Dunn

Every human society conceives and seeks to apprehend its political life and character in two sharply discrepant ways. It does so through its conceptions of what is valuable and how what it deems valuable can be realised through action, and it does so through its assessment of the impact of power on how and how far these values are realised in the material consequences they have for its human denizens. Neither of these ways can be a surrogate for the other, and they never articulate stably or clearly with one another. But every society faces a standing temptation, intellectual as much as political, to reduce or subordinate one to the other. The modern social sciences privilege assessments of material outcomes; and as soon as they attempt to shape themselves any other way their current renderings rapidly dissolve into complete incoherence. Indigenist politicians and populist intellectuals, by contrast, defend the primacy of their society’s own conceptions of itself and the goods it aspires or hopes to secure. In the vocabulary and political consciousness of the modern West, patronage is a residual element in the acceptable structuring of social, political and economic outcomes, thrust back increasingly into aesthetic domains or the normatively contingent practical privileges open to occupants of particular roles. The official normative ordering of western societies today is systematically egalitarian. Within them, every departure from it is increasingly stigmatised and must be rationalised, if it is to be recuperated at all, through contributions it makes to enhancing equality in some other domain. The result is a society deeply uneasy at its own reality, in endless denial of itself as a structure of power, incapable of recognising its own political substance, or of thinking clearly about the ways in which it is in fact led, or fails to be led, and about the causal relations between political and legal subjection within it and the economic destinies across the ranks of its notionally equal citizens.

xii  Foreword

Patronage as Politics in South Asia presents a very different conception of society, distributed across the Indian subcontinent over a time span far lengthier than the epoch in which the normative order of the modern West took its present shape and established its queasy hegemony. It is an integrally hierarchical order of value, location and responsibility that gives much of their sense to the ways in which its inhabitants live out their lives and structure, experience and judge interactions with one another. (The uncaptured residue is more a matter for novels or poems than for would-be systematic social theory, let alone social science.) Because it covers such vast stretches of time and space and such heterogeneous practices, it makes no attempt to present the order it discloses as a systematic ideology or social totality, as Louis Dumont’s Homo Hierarchicus for example appeared to, at least to some of its readers. What it does show from a startling range of angles is how comprehensively the modern social sciences have disabled themselves to take in the reality of most societies in the world, and even of those that generated them, incorporating these misapprehensions ever more deeply into their faltering efforts at selfunderstanding. It shows how impotent any economistic rendering is to capture the political or social reality of most of the humanly inhabited world, and how impertinent and external a picture it offers even of the structuring of economic outcomes in much of that world. Patronage as Politics in South Asia holds deep lessons for too wide a range of potential readers to list them all. But several points are worth insisting on from the outset. In the first place, it shows how indispensable the focus on social order is to capture the lived political reality of any human society: a task beyond the reach of any paradigm of rational choice. It shows the profoundly crippling cognitive impact of the egalitarian protocol for political, economic and social practices on the comprehension of societies not merely with quite different normative visions but even of the very societies from which that protocol first emerged. It shows the extreme plasticity and the permanent potential efficacy of patronage in settings from political and religious construction in classical Tibet to the exigencies of survival in the slums of Mumbai and the structuring of labour migration from Kerala to the Gulf. It shows that they still shape and inspire potent communities in the subcontinent’s great cities. It shows that they frame and dynamise the intense political life of far the largest democracy in human history, and fit as effortlessly into a political system articulated through

Foreword xiii

legal equality as they did into classical Tibet, Hindu Kingdoms or the Mogul empire. It shows that there is nothing superannuated about the causal presence and weight of such practices across the world, and that they travel as readily beyond the sub-continent as they do within it and are adopted indefatigably precisely because they hold far more utility for those who choose them over competing bureaucratic structures or transnational meliorist facilities. It shows how far even the most penetrating and sensitive of social anthropology has been handicapped by its misgivings and discomfiture at patronage’s frank embrace of inequality, in a setting where even the cultivated and self-conscious heirs of ancient Rome find the idea that friendship can be lop-sided (as most friendship rather evidently is) and be so without jeopardising its reality as friendship, too disconcerting to endorse openly and explicitly. This book shows how easy it remains to take in and comprehend how a society is led and follows or refuses to follow, or how it chooses to act politically through its own structuring of power, by viewing patronage as a huge array of practices incessantly at work, doing good or ill to all they affect. It shows how India’s huge and frenziedly animated democracy could learn to understand itself just the way it is and suggests how far it already does, quietly, on its own and within the relative privacy of its own languages, even if it is still as nonplussed as its western counterparts at combining that quiet self-understanding with the official normative rubric of India’s own Constitution. Very strikingly, too, it suggests a political verdict to which the rest of the world needs to attend every bit as much as India itself. India’s democracy, it intimates, however its material consequences come out in a metric of relative subjugation (where every existing human society comes out pretty mercilessly), is a profoundly democratic outcome, achieved through means that are in essence clearly democratic. As a structuring of material inequality, much of it is still remarkably grim; and that grimness is as unmistakably a product of power as it has ever been. (What, do you imagine, has generated the present class structure of the United States of America: Democracy’s finest and oldest fruit?) Insofar as the material outcome of India’s democracy is recognised, sustained, modified or enforced through public law and the coercive capacities of the state, that law and those coercive capacities act in the end, not just at the alleged behest and on the alleged behalf of all India’s legally equal citizens, but causally and as appropriately as almost

xiv  Foreword

anywhere else in the world, at the direction of India’s own demos, acting at regular intervals for and as itself, at its own iteratively sovereign moments. That just is democracy, and every bit as much so as any specifiable passage in the history of the United States, the United Kingdom, France, Switzerland, classical Athens or any other presumptively democratic exemplar. What democracy means is just that everywhere that has it judges, chooses and acts as and for itself. That is how India judges, chooses and acts. More precariously and hence less resolutely, it is also the way that Pakistan and Bangladesh judge, choose and act on the more intermittent occasions when their armed forces let them.


This volume started at a colloquium held in October 2011 at King’s College, Cambridge, which made it both possible and enjoyable for us all. Chris Fuller, Caroline Humphrey, James Laidlaw, Sian Lazar, Norbert Peabody, Kim A. Wagner and David Washbrook joined in the discussion and made it much more instructive and fun. In different ways and at different stages John Dunn, John Kelly, Jenny Malpass, Yelena and Alexander Piliavsky, Piers Vitebsky, Rob Wallach, Indra and Suresh Chattrapal, Ramesh and Kalla Kanjar, and Arvind and Shweta Singh helped to bring this volume about. David Watson at the Cartographic Unit of the Department of Geography in Cambridge drew the beautiful maps. At Cambridge University Press Suvadip Bhattacharjee, Ranjini Majumdar and Debjani Mazumder have been especially accommodating, and Qudsiya Ahmed displayed a rare combination of patience and enthusiasm in seeing this volume through to print. Editorial work on the volume was supported by research grants from the European and British Research Councils, and by a Research Fellowship at King’s—as fine an intellectual setting as I could wish for. A.P. Jaipur, Rajasthan October 2013

Introduction Anastasia Piliavsky

A scandal in Jodhpur


n the summer of 2007 a curious incident occurred in the city of Jodhpur in the north Indian state of Rajasthan.1 The wife of a Member of the State Legislative Assembly lodged a complaint with the police, accusing a local temple priest of ‘hurting the religious sentiments of the people’. The cause of offence was a poster designed by the priest which depicted Rajasthan’s then Chief Minister, Vasundhara Raje, as the bread-giving goddess Annapurna. On the poster the crowned Miss Raje appeared mounted on a lotus throne, from which she showered an assembly of parliamentarians, legislators and ministers gathered below with golden coins and rays of light (see Figure 1). Either side of her were a pair of guardian lions and two cabinet ministers, portrayed as the ancient Hindu gods Kuber and Indra. Floating just above were the leaders of Raje’s Bharatiya Janata Party (BJP), depicted as Brahma, Vishnu and Shiva, the gods of the Hindu trinity. The incident sent ripples of amusement through the Anglophone press, but in the landscape of South Asian politics this was neither a singular occurrence nor a laughing matter. A host of leading politicians—including President of the Congress Party Sonia Gandhi, head of the People’s Party in Uttar Pradesh (UP) Mayawati, the Chief Minister of Bengal Mamata Banerjee and the former Tamil Chief Minister Jayalalitha—have all received colourful popular veneration. 1

The ideas in this volume and introductory essay have been long in the making. I owe a great debt to Paul Dresch, David Gellner and Jonathan Norton, who in their different ways were there for me when I was first working them out as a doctoral student. I am grateful to Naor Ben-Yehoyada, Lucia Michelutti, Mattison Mines and Pamela Price for their comments, and especially to Piers Vitebsky and John Dunn for braving multiple drafts and for all our conversations. As ever, my greatest debt is to my many hosts and interlocutors in India, and especially to Suresh and Indra Chhattrapal, Ramesh and Kalla Kanjar, and Arvind and Shweta Singh for their tireless generosity, patience and help.

2  Anastasia Piliavsky

Figure 1 Vasundhara Raje depicted as goddess Annapurna. Courtesy of Pandit Hemant Bohra. To most readers of this book, a picture of voters prostrated before a politician (like the one that appears on the cover of this book) will look not just ridiculous or bizarre, it will appear positively obscene: a perversion of every political virtue. This image projects inequality in place of equality, personal bonds where detached judgement should be and subjugation where the free will of individual citizens ought to reign. It makes a mockery of the very idea of representative democracy, a government by people elected to act on behalf of the citizens, not to rule over them as kings or gods. The shock of this vision reverberates far beyond politics. It shakes the foundations of what liberal cosmopolitans believe human beings to be and the society which they live in to be really made of—innately equal and independently judging individuals—the beliefs enshrined in the modern political theology of democracy and the universal ritual of each adult citizen casting a single vote in the isolation of a polling booth.2


On the development of this ritual, see Crook and Crook (2007), Crook (2011) and Gilmartin (2012).

Introduction 3

One look at South Asian popular politics topples all this. In cities and villages across the subcontinent voters adorn political leaders with crowns, garlands, turbans and swords, recite for them praise verses, and fall at their feet in adulation. Time and again, newspapers seem to confirm just how dissolute South Asian politics really is. Votes are cast not by autonomous citizens concerned with their countries’ long-term general good, but by interest groups, or ‘vote banks’, which elect one of their own to provide for them. Parties and politicians do not convince their electors with ideological platforms. They buy votes with short-term benefits. Endemic poverty and governmental dysfunction obliterate free, reasoned and responsible judgement and drive people to exchange their votes for bureaucratic favours, clutches of cash and bottles of hooch. Politicians run the show like royal sovereigns, often at the expense of law. Rates of corruption registered by Transparency International in the region hover at Sub-Saharan levels ( Nepotism, political backwardness and decadence fill news reports, which often attribute this political bedlam on the subcontinent to the prevalence of ‘patronage’ or ‘clientelism’, the common glosses for ‘corruption’. It is seldom clear what exactly these words describe, but what they indicate is never obscure: a perverse and backward political practice, which prevails only where modern states fail. Yet South Asia is not a site of state failure. Far from it. Since the retreat of colonial powers from the subcontinent, it has been one of political modernity’s busiest laboratories, and it has turned out some very striking results. While in many parts of Europe electoral participation has flagged, in most of South Asia it has been steadily on the rise.3 The region is now home to the world’s most populous democracy and one of its most vigorous. The sheer scale of India’s general elections is mindboggling: a population of more than a billion, eager to cast their votes, makes this the single-largest social event in the world. India’s general elections regularly involve over 60% of voters, often substantially exceeding American presidential elections, and local elections often rise to 100% turnout. In 2008 the Hindu Kingdom of Nepal became a Democratic Republic, electing its first (Communist) government with more than 74% of its electors’ votes. In Pakistan, in the spring of 2013 voter participation rose from 40% (in 2008) to a spirited 60%, with people voting in face of great threat. 3

For country-by-country voter turnout data, see the database of the International Institute for Democracy and Electoral Assistance at

4  Anastasia Piliavsky

Yet  these political achievements have not stamped out patronage, which continues to thrive not despite democratic statehood, but alongside and indeed often through it. Clearly, we face not a feudal residue, but a current political form vital in its own right. In this volume we do not treat ‘patronage’ as a term of art and we offer no definitions. Nor do we see it as an unchanging, timeless ‘phenomenon’, a transactional arrangement with a fixed and predetermined content. Rather, we see it as a living moral idiom that carries much of the life of South Asian politics, and society at large. Here ‘patronage’ is an imperfect gloss for a widespread moral formulation which helps us escape the gridlock of liberal political heuristics and see the local actors’ own normative imagination. Collectively, we try to think our way into the region’s own political sense and into the ways in which political communities and attachments, modes of leadership and ways of following shape the hopes and disappointments which political engagement necessarily brings. At the centre of our analysis are the relations which constitute South Asian politics. By working out the ways in which South Asian citizens relate, and think they ought to relate, to one another, we address some central conundrums of political modernity in the region: how people live and think about democracy and the state, what they make of ‘political representation’ and how to understand why the region appears at once so politically engaged and in many ways successful, and so drastically ‘corrupt’ and ‘criminalised’. We hope that those who do not know much about South Asia and those who know a great deal will find this book equally thought-provoking. We also hope that our readers find the accounts collected here, and their implications, helpful for thinking not only about South Asia, but also about politics in other places, including those which they call their home.

What happened to patronage? In the social sciences patronage has had its day. Whilst debates on nationhood, sovereignty, governance, democracy and the state nowadays fill the journals, patronage barely figures. If three decades back academics as distinguished as Ernest Gellner, Shmuel Eisenstadt and James Scott published thick volumes on the subject,4 today few dare to put the words ‘patron’ or ‘patronage’ on the cover of 4

For helpful overviews of this literature, see Scott (1977b), Eisenstadt and Roniger (1981) and Roniger (1981; 2004).

Introduction 5

their books. This avoidance is in part a reaction to the concept’s career  in earlier decades (mainly in the 1960s–1970s) when its analysis, even as the brightest minds applied themselves to it, generated a literature that was overwhelmingly dull. Endless case studies and descriptions of patronage observed around the world yielded little comparative framework or analytical continuity to sustain debate.5 We agreed in 1960 that ‘patronage’ had something to do with asymmetry of status and power, that it involved reciprocity, and that it relied on particular, intimate, face-to-face relations (Powell 1960). More than five decades on we have not moved far beyond this basic picture, but we no longer care to explore it further. The problem runs deeper than the concept’s dispiriting recent career. It issues from deep-seated prejudices and from two beliefs about patronage to which we hold fast, if often unwittingly. We believe that we already know what patronage is, and we believe that it must be a bad thing. At a bare minimum, we think of patronage as a relation between two unequal persons, one of whom holds the upper hand. Patrons are wealthier, politically more potent or otherwise privileged, and they control what others need or want, making their clients at best dependent and at worst oppressed. Call up some images which the phrase ‘political patronage’ brings to mind: a Sicilian peasant kissing the hand of a Mafioso, Vladimir Putin appointing governors to the Russian provinces, pharmaceutical lobbyists at work in DC. The slide show can go on with countless snapshots of political deformity, each capturing a disfigurement of what modern politics should be: equal, disinterested and impersonal. Over the decades, deeply ingrained moral aversion to patronage has frustrated the task of understanding it; description slips inadvertently into evaluation, which in turn poses as analysis. Time and again patronage appears as the cause or symptom of political infirmity. Time and again it is portrayed as a retrograde, oppressive or at best ancillary, institution destined to vanish the moment modern, democratic states take proper hold of the world.6 Time and again analysts forecast its disappearance, time and again lamenting its refusal to go away.7 5

For a sense of the sprawling vastness of this literature, see Eisenstadt and Roniger’s (1980) summative article, in which footnotes listing case studies occupy several full pages of text. 6 For example, Geertz (1960), Eisenstadt (1973, 60), Blok (1974), Boissevain (1974, 147–148) and Eisenstadt and Lemarchand (1981). 7 Gellner’s prediction is typical: ‘where power is effectively centralized … patronage is correspondingly less common’ (1977, 4).

6  Anastasia Piliavsky

Patronage first came to the fore of the social sciences in the 1960s and 1970s, mainly in Mediterranean and Latin American peasant studies. These writings had a broadly Marxist overtone and they dismissed patronage as a sentimentalisation of class inequality (e.g., Leeds 1964; Galjart 1965; Alavi 1973). For them, patronage was a ‘myth’ or the ‘ideology’ of the elites, endorsed by those social analysts who dared to present it as anything but power struggle.8 The language of kinship, friendship and sympathy, they claimed, concealed behind it the brutal mechanisms of dependence and exploitation.9 Sociologist Anthony Hall, for example, wrote that however one may approach patronage, ‘the important fact is the inherently coercive nature of patron-clientage’ (1977, 511, emphasis in original). One could view this struggle from different angles and ask: How do the powerful access, create and misuse their subjects? What degree of control do they have? How do the powerless get access to resources, resist oppression or negotiate for themselves better deals? Yet whichever way analysts turned their viewfinders, the scene in focus remained a site of oppression, submission, resistance and domination. A second academic camp developed a more forgiving view. They suggested that patronage was not everywhere plainly bad news. Patron-client relations, they insisted, did not necessarily propagate inequality or sent modern politics back into feudal darkness, but often achieved the obverse: social mobility and political participation.10 Ties of patronage might assist the poor to wrest resources from the elites or help immigrants access state services. Some reported that patronage and electoral politics often went hand in hand; patronage promoted electoral participation, which in its turn generated fresh patronage bonds. Patron–client relations formed the backbone of ‘traditional’ politics and were the main political tool of tribals, peasants and the urban poor. As modern politics spread into far corners of society and the globe, patronage became a link between 8

For example, Silverman (1977), Lemarchand and Legg (1972), Flinn (1974), Scott (1975) and several essays in Gellner and Waterbury (1977), especially those by Silverman, Gilsenan, Weingrod and Scott. 9 Characteristically, Flinn (1974), Scott (1976), Schmidt et al. (1977) and Malloy (1977); and slightly more recently Bodeman (1990), Knoke (1990), Breman (1993) and Fox (1994). One representative study in this mode appears in this volume (Martin, Chapter 14). 10 On clientelism as the vehicle for peasant movements, see for instance, Landsberger (1969) and Scott (1977a; 1985). On the symbiosis of patronage and democracy, see Gwyn (1962), Chambers et al. (1967), Powell (1971) and Fagen and Tuohy (1972).

Introduction 7

governments and ‘social peripheries’ culturally unfit for or otherwise excluded from direct engagement with the state.11 Because patronage connected bureaucracies to traditional politics, cities to villages and governments to citizens through patron-politicians and brokerbureaucrats, it could paradoxically modernise ‘developing’ states (e.g., Schmidt 1974). Between the late 1970s and the early 1990s, studies of political patronage focused almost exclusively on the ‘developing world’ and its political ‘systems in transition’ (Martz 1997, 14). Insofar as patronage ‘at all interested political scientists, it did so principally as intermediary between the centre and the periphery’ (Dogan and Pleassy 1984, 76; also Eisenstadt 1973, 60; Boissevain 1974, 147–148). If in anthropology patronage has lost its currency, in political science it is still ready money. Political scientists have long agreed the meaning of ‘clientelism’, as they usually term patronage. For them it is an exchange of goods and services for political support. This system of political barter, a ‘distributive’ politics of ‘take there, give here’ (Graham 1990), turns elections into auctions and political choices into calculations of profit.12 Politicians use goods and favours to buy their positions and their electors employ such means as they have at their disposal to wrest resources back from politicians and the state. ‘Machine politics’ could grease the wheels of electoral systems. It might boost political party membership and electoral participation, and even occasionally benefit the poor (e.g., Banfield 1969). But in doing so it depletes democracy of its point and meaning. This politics of purposeful mutual exploitation substitutes moral, responsible judgement and the concern for greater good with the pursuit of selfish, short-term advantage—what political scientist Edward Banfield once called The Moral Basis of a Backward Society (1958). Clientelism is the politics of dire poverty, which turns democracy into a spectacle of desperation and greed. In the past two decades political scientists have slowly recognised that patronage has not slunk away in the face of modernity and democratisation, as their predecessors had hoped, but instead carries on stridently in fully developed, rich countries like Austria, 11

For example, Geertz (1960), Leeds (1964), Wolf (1966), Powell (1970), Lemarchand and Legg (1972) and Boissevain (1974). 12 This approach is ubiquitous. Representative recent examples include Rose-Ackerman (1999), Chandra (2004), Kitschelt and Wilkinson (2007) and Stokes et al. (2012). See also essays by Wilkinson (Chapter 11) and Martin (Chapter 14) in this volume.

8  Anastasia Piliavsky

Japan and the United States.13 These striking revelations make continued analytical resistance to patronage appear increasingly strange. And yet the resistance persists. Workshops and conference panels on clientelism organised around the world at the millennium’s turn, for instance, reached long-familiar conclusions: patronage was a regressive force inimical to democracy, civil society and the market’s free flow (Roniger 2004, 372). Most recent studies are still framed by old questions: Why doesn’t patronage disappear? When may it do so? What might bring about its demise (see Wilkinson, Chapter 11 in this volume)? Over the years, whatever role patronage has played in the literature—as a means of exploitation, a vestige of feudalism, a governmental pathology, a politics of the poor or an ancillary institution—it has never been a site of positive value in its own right. There is no better expression of this attitude than Ernest Gellner’s definition of patronage as ‘not bureaucracy’, ‘not kinship’, ‘not feudalism’, ‘not the market’, ‘not the state’ (1977). But what is it? More than two decades earlier anthropologists had shown that relations between ‘patrons’ and ‘clients’ could be sympathetic and even intimate, based on mutual esteem and affection, like the bonds between kith and kin. In southern Spain Julian Pitt-Rivers (1954) wrote that patronage formed the backbone of politics and social relations, as they were lived and understood. Relations between patrons and clients were best seen as a kind of friendship. Other anthropologists showed that patronage was woven tightly into the fabric of many societies, where people did not see the institution itself as problematic, however dissatisfied they might feel with particular persons involved.14 But this literature left little legacy.15 Campbell’s (1964) superb, intimate study of patron–client relations in rural Greece, for instance, is hardly ever cited today. Resistance to patronage proved so strong that even as fine an anthropologist as Pitt-Rivers found it difficult to take his informants seriously. In his analysis he demurred: if patronage was a kind of friendship, he said, the friendship was necessarily ‘lop-sided’. 13

Roniger and Gunes-Ayata (1994), Briquet and Sawicki (1998), Piattoni (2001) and Kitschelt and Wilkinson (2007). 14 For example, Mintz and Wolf (1950), Pitt-Rivers (1954), Campbell (1964), Foster (1967), Weingrod (1969) and Powell (1970). 15 Exceptions include Martz (1997), Auyero (1999; 2001), Mitchell (2002) and Lazar (2004).

Introduction 9

He wrote that ‘while friendship is in the first place a free association between equals, it becomes in a relationship of economic inequality the foundation of the system of patronage … used to cloak a purely venal arrangement, a rich man using his money to attain his ends’ (1954, 140, emphasis mine). Patronage, in other words, was spoiled friendship. Where relations were unequal, they could not be right. Inequality was inherently oppressive. No one could sanely choose to be someone else’s client, and those who did must have acted under duress.

Patronage in South Asianist scholarship South Asianists have been of a different mind, at least until recently. Once upon a time in South Asianist scholarship patronage was the hub of key debates on status and power, personhood, sociality, economic exchange and polity. Studies of kingship, the history of colonial relations, ethnographies of village exchange and ‘big-men’ in urban centres, and the more recent work on fixers have all focused centrally on patronage. For a long time for historians patronage was the centre of debates about kingship, which cast much light on lasting local conventions of ruler–subject bonds. Historians showed that for as far back as history stretched on the subcontinent, kingly rule relied on the distribution of gifts.16 The king’s defining duty was to provide for (and protect) his subjects. The act of giving linked rulers to their subjects (and their subjects in turn to their subjects), creating chains of gift and receipt that defined kingly realms. Historians showed that giving and receiving were not just economic transactions, but socially and politically constitutive acts which authorised kingly rule and defined political communities. Giving was an expression of generosity, the cornerstone-value of South Asia’s political life. And as such it had to be performed. Whether or not the kings had the wealth to give, they staged their capacity to do so with extreme flamboyance. Royal courts were dazzling spectacles of munificence where kings often gave away much more than they could afford at extravagant feasts and gifting ceremonies. Kingly realms constantly stretched and shrunk with the subjects’ ever-fickle loyalties. As Sumit Guha shows (Chapter 4 in this volume), in 18th-century Maratha polities kingly 16

For example, Dirks (1979; 1987), Stein (1980) and Peabody (1991; 2003).

10  Anastasia Piliavsky

rule was sustained by constant personal engagement, and collapsed when the regime of favours ossified into a regimented system of formal entitlements. This structure of loyalty was replicated on every level—from grand maharajas to ‘little kings’, vassals and vassals of vassals, all the way down to village landlords, producing the series of concentric and identically structured sovereignties which Stanley Tambiah termed ‘galactic polity’ (1977). In this tight structure of mutuality donor-kings depended on the gifts’ recipients no less than recipients depended on kings. Recipients of royal gifts provided not just tax money or manpower in times of war. They—paradigmatically the Brahmin donees—were the source of kingly authority.17 Another camp of historians, the Cambridge School, produced meticulous accounts of local political economies and how these were embedded in networks of patron–client ties, mostly in British India.18 By the middle of the 19th century, India’s ruling class were severely disenfranchised, and colonial administrators took over as the premier patrons. In this political ecology, Indian merchants and bankers emerged as the new class of indigenous patrons who, Chris Bayly argued, came to steer a great deal of politics and much else that happened in India at the time. Many merchants patronised political, and often progressive, activities: new religious movements (including the more ‘Protestant’ Hindu sects), Hindu revivalist campaigns (like the cow-protection movement), new cultural organisations, literary and political publications and the founding of the new Congress Party (Bayly 1973; 1983). Bayly argued that in high colonial India patronage had two sides—a moral and an instrumental—and that these were perennially at odds. All patrons engaged in two types of relations: the essentially commercial vakil patronage meant to maximise profit, and the moral or dharmic patronage of religious and community institutions which enhanced patrons’ standing (1973, 367ff). He suggested that amidst vast shifts in political and economic infrastructures new figures of 17

The paradigmatic donees were the Brahmans, the ultimate providers of kingly authority. To get a sense of debates which once raged around problems of kingly sovereignty and Brahmanical legitimation, see Peabody, Fuller and Mayer (1991). On the classic model of the donor–donee relationship between rulers and their spiritual preceptors, see Ruegg (Chapter 2). 18 Some representative essays can be found in Gallagher, Johnson and Seal (1973). See also Low (1968), Broomfield (1968), Seal (1968) and Bayly (1983).

Introduction 11

pre-eminence appealed to old ideals to legitimate their role as patrons, and not mere financial sponsors. To succeed socially, every bit as much as financially, bankers and dealers had to put on the old kingly shows of largesse. Conflict was built into the system. Generosity required wealth and acquiring wealth corrupted the image of an ideal patron, putting the two modes of patronage inexorably at odds. In this volume David Gilmartin (Chapter 5) shows that Bayly put his finger on something far bigger and more important than he himself may have realised—a clue to a fundamental aspect of patronage—to which I return below. In the era of ‘classic’ ethnography (1930s–1980s) South Asia’s anthropologists took patronage perhaps even more seriously than the historians. At the time they focused chiefly on something known as the jajmani, a system of inter-caste village exchange.19 In the ideal jajmani model each village revolved around a land owning patronfamily (the jajman). Each service-caste (the kamin) performed a unique economic and ritual role—priests conducted rituals, sweepers swept floors, barbers shaved beards—and each caste, in return, received payments and gifts, along with a share of the village harvest.20 This was what anthropologists termed ‘total exchange’ (after Mauss 2002 [1925])—at once economic, political, ritual, and moral— which constituted multi-dimensional social bonds. Over the decades, ethnographic descriptions of jajmani exchange solidified this general scheme into a closed, rigid and internally integrated village ‘system’. At the height of the village ethnography era, the subcontinent often seemed a universe of village-galaxies, each a microcosm of South Asia’s social life as a whole. Patronage was the heart of society, encompassing its hierarchical principles and the ways in which communities, political or otherwise, were composed and related to one another. Like all theoretical perfections, the jajmani system eventually collapsed under the weight of observation. Anthropologists noticed, for instance, that in various south Indian types of village exchange (known by names like baluta, mirasi, paniwallu, padiyal or kaniaci), the 19

Classic accounts include Wiser (1936), Kolenda (1967), Mandelbaum (1970, 159–180) and Dumont (1980, 98ff). 20 Such arrangements were observed by Hindus (Wiser 1936) as well as Muslims, including Pathans (Barth 1959; Marriott 1960; Guha 2004).

12  Anastasia Piliavsky

identities of patrons and clients, the types of payments and quality of relations between them varied a great deal across time and space. There was, they maintained, no single ‘jajmani system’ (Orenstein 1962; 1965; Fukuzawa 1972; Reiniche 1977). By the late 1980s the jajmani system which seemed to bind villages into time capsule ‘village republics’ fell to pieces (Fuller 1989). With it the era of village ethnography too came an end.21 Since then anthropologists turned to other matters: urban anthropology, development, media studies, bureaucracy, middle classes and so on; and for today’s anthropologists the word ‘jajmani’ remains a relic of quaint, positivist ethnography. The time had certainly come to vacate the creaking jajmani edifice. But the baby was thrown out with the bathwater: along with the system anthropologists abandoned the many lessons about relational principles which jajmani studies contained and which still hold vital clues to some of the knottiest puzzles of political life on the subcontinent (Karanth 1987). One thing the jajmani studies have taught us is that in South Asia patronage was never a purely economic or ‘top-down’ power relation. Crucially, it was a relation of status difference. Status asymmetry in this relation, they taught us, was expressed in the language of ‘services’ and ‘gifts’—terms which defined the normative principles organising the relationship and identities of those involved. The donor-servant relation was a profoundly mutual and socially constitutive bond in which all participants were defined relative one another. One was never simply a drummer, barber or priest, but a drummer, barber or priest for this or that patron. Servants, in turn, maintained their patrons’ ritual purity, authorising their pre-eminence and the authority through which they ruled. Both ritually and economically, servants turned landholders into patrons. In their turn, patrons passed down their identities to their servants together with the payments and gifts. Gifts were not just remuneration, they were receptacles of what South Asianists refer to as ‘bio-moral substance’, or a kind of total sociophysical identity.22 This substance was carried most effectively in food and drink, the honoured and paradigmatic gifts in South Asia. As in 21

For more on the rise and fall of jajmani studies, see Piliavsky in this volume (Chapter 6, 156–161). 22 The theory of substantive contingency was at the forefront of the work of Chicago transactionalists led by McKim Marriott and Ronald Inden, who argued that in India exchange was a substantively constitutive process, in which gifts (most importantly food) carried the givers’ nature (e.g., Marriott and Inden 1973; 1977; also Parry 1986; Raheja 1988a; 1988b).

Introduction 13

the kingly realms, all exchanges were put on display, transforming transactions into statements of the relation’s moral content and the qualities of the parties involved. Gifts expressed patronal generosity and clients’ devotion. Later studies of south Indian ‘big-men’ (periyars) put this wisdom in motion, showing how the old values and relational principles worked in contemporary urban life.23 In his work on wealthy entrepreneurs in Chennai, Mattison Mines showed that the relational values present in jajmani exchange were not confined to the rural backwaters, but also shaped institutional development and political life in big, modern cities (see Chapter 1). The old values of munificence and the practices of their display provided a moral frame for urbane big-men’s pursuits and achievements, shaping their public selves and their contributions to public life. Big-men built their careers on grand feasts, ostentatious sponsorship of temples and schools, and other displays of munificence akin to the erstwhile rajas’ or the village jajmans’. Mines showed that the durable, widespread set of ideals and rhetorical tropes was as much the framework for individual action as for the vagaries of subcontinental politics in all its guises. Big-men created new institutions and organisations to benefit their constituents and expand their followings. They were the agents of history, and the identities of institutions (like temples, cooperatives or schools) and communities (like castes or caste associations) were, and were locally seen as, expressions of big-men’s aspirations and their ability to make things happen. Caste as such was not a stable ‘social structure’, but a mobile form of organisation constantly shaped and reshaped by the political and economic pursuits of big-men operating through old moral idioms (Mines 1984; 1994). Yet just like jajmans, big-men dropped off the menu of ethnographers’ interests. With them patronage generally vanished too. Post-modern critiques of empirical inquiry and the ‘literary turn’ shook anthropology’s ethnographic nerve. Research shifted away from villages towards subjects without any determinate location: globalisation, neoliberalism, citizenship, civil society, development, post-colonialism, mass mediation, middle class shopping habits and


Pioneering studies include Mines and Gourishankar (1990) and M. Mines (1994). For examples of more recent work, see Hansen (2005) and D. Mines (2005).

14  Anastasia Piliavsky

so on.24 Today, with South Asianist anthropologists devoting increasing energy to corruption and the state, continued apathy towards patronage appears increasingly ill-timed. While in international news patronage remains a hot item, in the indexes of books published by South Asianist anthropologists the words ‘patronage’, ‘patron’ and ‘client’ hardly figure.25 The subject retains currency in political science, but most of this literature is more attuned to the established sense of clientelism in the discipline than to its South Asian practices and norms. Early studies of post-independence politics showed how patronage worked in electoral politics, political parties and factions. But they failed to explain what patronage meant for its participants and why it remained such a pervasive and central part of local political life.26 Nowadays, just as before, when political scientists mention patronage in South Asia, one can be certain they are writing about political failure: vote banking, party corruption, inequality, ethnic party politics, the capture of governance by elites, failures of distribution and accountability or the overall ‘crisis of governability’.27 Even if by now patronage has been accepted as a central practical feature of politics in the region, ‘the supposition of a tension between patronage and democratic values’, as Gilmartin notes, persists in writings on ‘civil society’ in India, which often embody a normative emphasis on the tension between patronage (arising from ‘tradition’) and the more abstract, individual-oriented values, including openness, equity, and efficiency, commonly associated with democratic (and capitalist) forms of rule (Chapter 5, 128–129).

By binding people into particularistic top-down loyalties and ties of direct exchange, patronage appears to undercut the free choice which ought to guide electoral process; by making voter-clients 24

A striking recent illustration is A Companion to the Anthropology of India (ClarkDecès 2011), which includes nearly 30 chapters, not a single one of which is about village affairs or is even based on research conducted substantially in rural areas. 25 Important recent contributions to South Asian political anthropology (including Fuller and Béneï 2001; Chatterjee 2004; Gupta and Sharma 2006; Sengupta and Corbridge 2010) hardly comment on patronage. When they do mention it, they do so with little explanation, as if we already know what it is (e.g., Brass 1994; Spencer 2007, 85–86; Chakrabarty 2008; Madsen et al. 2011). 26 For example, Bailey (1963), Brass (1965), Rudolph and Rudolph (1967), Weiner (1967) and Fox (1969). 27 Weiner (1967), Kohli (1990), Chandra (2004) and Bardhan and Mookherjee (2012).

Introduction 15

dependent on politician-patrons, it seems to invert the very idea of democracy as the people’s rule. Political scientists Kitschelt and Wilkinson, for instance, argue that clientelism replaces political accountability with a transaction: ‘the direct exchange of a citizen’s vote in return for direct payments or continuing access to employment, goods and services’ (2007, 2). This is a desperate system: voters require continuous proof that politicians will deliver on their promises, and politicians have to construct elaborate (and often unsustainably costly) structures of surveillance and enforcement to ensure that voters whom they had paid off honour the deal (Kitschelt and Wilkinson 2007). Or consider Kanchan Chandra’s (2004) picture of Indian ‘patronage democracy’ in what is now the most widely cited study of political patronage on the subcontinent. India’s citizens do not care for policies or ideologies, only for the goods and benefits they may personally receive. They know that politicians only give to their own—that ‘their’ politicians provide for them and that others provide for others. And they vote accordingly. Political choice is thus an ongoing calculation of loss and profit on both sides: ‘Faced with a choice between parties’, writes Chandra, ‘an individual voter in a patronage democracy should formulate preferences across parties by counting heads, preferring the party that represents elites from her “own” ethnic category to the greatest degree’ (ibid., 13).28 Voters are so caught up in these calculations that they fashion their social identities with no other aim but to maximise gain: ‘Regardless of the good they seek’, contends Chandra, Indian voters are ‘instrumental actors who invest in an identity because it offers them the best available means by which to obtain desired benefits, and not because such identification is valuable in itself’ (ibid., 11). In their turn, politicians scramble to buy most votes and to secure for themselves positions from which they can earn even more (and gain ‘psychic goods’ like prestige) (ibid., 12). The poorer are the voters, the cheaper their votes. The electoral process is thus a mechanism of hand-to-mouth barter fuelled by the desperation of poverty and 28

The use of the term ‘ethnic’ to describe South Asian communities may surprise readers more familiar with descriptions of South Asia as a ‘caste society’. Chandra’s use of the term follows the widely agreed belief that since India’s independence, hierarchically arranged, interdependent castes have transformed into independent groups like ethnic communities vying for economic and political resources, what has been termed the ‘ethnicisation’ of caste (see p. 30 below; also Michelutti, Chapter 12 in this volume)

16  Anastasia Piliavsky

steered by self-serving choice. A billion profiteering voters wielding the abacus of rational choice is an arresting picture. But is it really like this? What happens when we move closer in?

Patronage close-up Pamela Price (Chapter 9) takes us to a village in the former Indian state of Andhra Pradesh (now in Telangana), where villagers tell her how they make electoral decisions and why indeed they vote at all. Focusing on one of south India’s biggest political heroes, the state’s former (and now late) Chief Minister Y.S. Rajasekhara Reddy, better known as YSR, she reflects on the sources of his popularity. YSR’s widespread esteem did not come easy. In the hot summer of 2003 he trekked 1500 kilometres on foot across Andhra wearing simple farmer’s dress and speaking to villagers as he went along. ‘I am not bothered about my comfort’, he was quoted as saying, ‘I am ready to sacrifice my life for the wellbeing of the poor’ (Chapter 9, 221). YSR painted himself in moral hues and he appealed to what people looked for in a politician, and what they valued more generally when they cast their votes. In 2012, just months after being re-elected as Chief Minister, YSR’s helicopter crashed. Posthumous investigations into his and his young son’s vast fortunes revealed the extent of his nepotism and embezzlement of public funds. Little came of the irrigation schemes that he promised to farmers, while his son’s expansive range of businesses flourished. Many villagers, nevertheless, did not seem to mind: YSR could take whatever he wished for his family, as long as he was generous to ‘his people’ too. And that he was. Much of what he had done while in office may look like corruption, but he also did a great deal for his electors, who still remember him as a ‘good man’ (manchi manishi), swearing that they would have voted for him again, had he lived. Villagers want politicians to provide (reliable supplies of water and electricity, sugar and kerosene at low prices, high pensions and interest-free loans) and they vote for those who they believe will ‘do their work’. Electors’ insistence that politicians ‘get things done’— what Ward Berenschot (Chapter 8) calls the ‘new pragmatism’ of South Asia’s post-independence politics—may look like the devolution of political life into the pursuit of purely instrumental gain. But, as Price shows, villagers in Andhra value effective politicians both as essentially useful persons and as good ones. That is, they assess them

Introduction 17

in moral rather than practical terms. The terms politicians appeal to and the terms in which they are judged by electorates are equally moral. This fact alone makes voting in participants’ eyes a choice, not a reflex of poverty or a function of greed. Cross the subcontinent and join Lisa Björkman (Chapter 7) in the slums of Mumbai, where notoriously ‘corrupt’ politicians are said to ‘bank’ on the votes of the destitute. Here Mr Kamble, an effective municipal councillor gets water pipes laid and water pressure raised in the taps; Kamble solves family disputes and helps with college applications, he saves people’s sons from police and their homes from demolition. For more than two decades he has been what they call the area’s ‘social worker’, first as an informal ‘fixer’ and later as elected councillor. He earns his keep, but after 20 years in politics he is no magnate. Marching about the slum in plain trousers and shirt (see image on p. 181, Chapter 7), he needs no flashy jewellery (or flashy guns and body guards), like the gangster-politicians described by Lucia Michelutti in UP (Chapter 12), to show that he can provide. He gets things done and he gets people’s votes. Leave the metropolis and travel north to rural Rajasthan during elections (Piliavsky, Chapter 6), where gifts and feasts remain the essence of electors’ choice. In electoral law such offerings constitute bribery, but villagers hold a much more nuanced view, distinguishing carefully between ‘gifts’ and ‘bribes’. In practice, gifts and bribes may look identical, but the contrast between them is pivotal to how electors assess candidates and in the end cast their votes. Bribes are things politicians give when they attempt shamefaced, oneoff transactions. Voters may accept ‘bribes’, but they maintain that these have no effect on their electoral choices. ‘Gifts’, conversely, are no cause for shame; in fact, voters expect politicians to give and complain when they fail to do so, calling this failure ‘corruption’. The line between gifts and bribes is not economic, but normative. Whereas ‘bribes’ are self-serving transactions meant solely for profit, ‘gifts’ create (or ought to create) mutually beholden relations which can become the basis for political ties. Despite a great deal of legal intimidation, politicians continue to throw lavish feasts, which often require increasingly ingenuous disguises like fake wedding parties or funerary feasts. Food remains the ultimate moral gift, which is as pivotal to elections in rural Rajasthan today as it was in former kingly courts and jajmani exchanges.

18  Anastasia Piliavsky

Leave South Asia’s shores and travel together with labour migrants from Kerala to the Gulf, where you will also find the familiar moral economy based on patron-client exchange (Osella, Chapter 16). Despite a great deal of effort by governmental and non-governmental organisations to liberate migrants from the yoke of ‘informal economy’, labourers seeking opportunities across the Arabian Sea prefer to sidestep bureaucracy and follow chains of relations, men whom they know from home, to the Gulf. The promise of patronage is perilous for new arrivals, as many old hands make it a lucrative, and often deeply dishonest, business. As many Malayalis on both sides of the Gulf see it, brokers ‘operate in the amoral, or altogether immoral, realm of calculated self-interest’ (Chapter 16, 366). Yet while promises of visas or good employment often come to nothing, migrants continue to prefer networks of kith and kin to NGOs. As Osella shows, despite widespread disappointment in their trade, ‘Gulf-based Malayali entrepreneurs mobilise their business skills to sustain community “upliftment” back in Kerala and many help by privileging the recruitment of migrant labour from among “their own people”’ (Chapter 16, 370; also Osella and Osella 2009). Partha Chatterjee (2004) has similarly shown that for many Indian slum dwellers it is patronage, and not the institutions of civil society and state law, that helps when the state fails to protect and provide (see also de Wit and Burner 2009; Witsoe 2011). Here, as elsewhere, the practical goods needed and the services which provide them are sought and offered through a common normative frame in a continuous process of choice by everyone concerned, however irrational it may seem to those who cannot recognise, or accept, this frame. A number of moral conventions persist across space and time in South Asian politics. The donor-patron ideal has endured for centuries (perhaps millennia) in a vastly diverse linguistic and cultural terrain, through maelstroms of political change, and across the many different, and constantly changing, styles of leadership.29 Whether in medieval Tibet (Ruegg and Diemberger, Chapters 2 and 15), 18th-century Maratha polities (Guha, Chapter 4) or recent elections in Andhra Pradesh and Mumbai (Price and Björkman, Chapters 9 and  7), munificence is the mainstay of political rhetoric, action, judgement and choice. Political giving is never only a matter of redistributing resources, it is also necessarily a rhetorical act that 29

On the variety of leadership styles in South Asia see studies collected in Price and Ruud (2010).

Introduction 19

conveys largesse as a politician’s virtue. As Price (1989) argued more than two decades ago, this idea is heir to the persistent, historically embedded kingship model. To be a raja, one had to display one’s capacity to provide in grand spectacles of magnificence (see also M. Mines, Chapter 1, in this volume). Electoral feasts in Rajasthani villages, pilgrimages undertaken by south Indian politicians and the extravagant charities of Tamil big-men all dramatise selfdenial and giving away. These dramas have discerning audiences, whose standards for judgement are often very strict. The range of presentational styles available to politicians is broad (e.g., Price and Ruud 2010), yet crucially, giving must appear selfless: boundless, from the heart, done for the people one ‘loves’. South Asian history is a parade of many different patrons, some of whom find a place on the pages of this book: Tibetan sovereigns, Maratha warlords, Tamil merchant-princes, Chief Ministers, tribal chiefs, gangsters and mafia bosses, MLAs and policemen, white-clad Gandhians and royalty, British colonial administrators, ‘every man’s’ politicians and modern India’s grandest patron of all—Jawaharlal Nehru. What makes all these characters into ‘patrons’ is the imputed obligation of selfless munificence and the duty to provide for those whose loyalties they hope to claim. What patrons give away varies, depending on their circumstances and their followers’ demands. It can be land grants made by kings to vassals, printing presses constructed by Tibetan monarchs for their lama-clients, temples built by Tamil merchants for communities of devotees, welfare and subsidy schemes initiated by Chief Ministers, money, jobs, ‘development’, protection (as in Michelutti’s and Ruud’s accounts in Chapters 12 and 13) or the ubiquitous administrative and legal favours arranged by political fixers, policemen and MLAs. A patron’s generosity, crucially, must not be generalised; it must come from one person to another. When aimed at ‘one’s own’ people, generosity forms attachments that are much more substantial than financial debt. It generates loyalties. If in the past kings gave land grants and temples, South Asia’s archetypal political ‘gifts’, the economics of patronage has grown more precarious now that most patrons are no longer sovereigns with royal coffers, but elected politicians and government employees, for whom the state itself has become the chief resource. In her study of negotiations surrounding the lodging of First Information Reports (FIRs) with the police, Beatrice Jauregui (Chapter 10) shows how policemen use

20  Anastasia Piliavsky

legal and administrative restraints as ‘gifts’ or ‘favours’, which they can choose to grant (or withhold), thereby generating petitioners’ loyalties and styling themselves beneficent patrons with the power to give and protect. The legally pivotal and notoriously ‘corrupt’ process of lodging an FIR reveals that what is at stake in South Asia’s policing practice is not just bribery or extortion (however important bribes certainly are), but always also relations with the police who ‘serve the public not as “impersonal” bureaucrats in Weber’s mould (Weber 1958 [1918]), but as patrons or providers of access to the state’s legal powers’ (Chapter 10, 257). The same logic structures the ways in which people relate to the state—via its intermediaries—in the slums of Ahmedabad (Berenschot, Chapter 8). Here, as elsewhere on the subcontinent, the state controls a great deal of what is crucial for daily survival—water, subsidies, jobs— creating legions of elected and informally acting politicians, whose work consists primarily of helping citizens access the state. In observable practice the primary job of these politicians is to help their constituents negotiate bureaucracy’s labyrinths and facilitate access to public services and goods. But in both popular imagination and politicians’ rhetoric these efforts appear as personal favours and gifts. Advocates of ‘good governance’ see this as the end of the state’s impartiality, as corruption’s reign. Yet in the slums it is precisely partiality—the exchanges, obligations and bonds among real people by no means indifferent to one another— which forces the torpid bureaucracy to do its citizens’ work. New circumstances and exigencies require new types of provision and displays of largesse. In parts of the subcontinent there has been an increasing need for protection from both criminals and the state. Lucia Michelutti and Arild Engelsen Ruud (Chapters 12 and 13) tell us that in the increasingly ‘criminalised’ political ecologies of UP and Bangladesh voters increasingly look to politicians to provide them with security instead of cash, infrastructure and bureaucratic help. Growing concerns with security in these parts of the region are part of vicious cycles of political violence among increasingly ‘muscular’ politicians who promise to defend, by whatever means and at any cost, from the threat which they themselves often pose. If the economics of politics has changed, the normative structure of relations has remained in many ways remarkably stable. New leaders who hail from disadvantaged classes do not have the means of elite politicians who dominated politics in the decades following 1947. New leaders parade munificence in other ways: by showing off muscle and courage, ‘gangster-politicians’ can appear capable of providing

Introduction 21

almost anything at will. Or, as Björkman shows, even in more dire circumstances resourcefulness is never only an instrument of security or a provision operating in an ethical vacuum. Resourcefulness remains distinctly a virtue. The show of selfless munificence is as much a part of gangster politician’s work as it is of less violent policy makers. While current talk of ‘development’ and ‘programmes’ may seem to mark a shift away from paternalism to programmatic, policydriven governance (Manor 2010a; 2013), observations show that this appearance is largely misleading: today most ‘development’ does not target generalised bureaucratic reforms, but offers subsidies and poverty-relief programmes, jobs, electricity, schools, hand pumps, paved roads, and latrines (see M. Mines and Price, Chapters 1 and 9 in this volume).30 In March 2013, for instance, UP’s young Chief Minister, Akhilesh Yadav, launched 71 development projects, which included 27 roads, a college, a hospital and seven marketing hubs. Not a single proposal was for programmatic or administrative change (Times of India, Kanpur edition, 4 March 2013, ‘Chief Minister Akhilesh Yadav Inaugurates 71 Projects’).

The moral logic and the conflict within Social scientists have long imagined patronage as a domain of transactions.31 When they use the word ‘patronage’, what they usually mean is something politicians dispense (favours, benefits, jobs) to their followers. A model of patronage as a set of transactions may show what people exchange. It may even show how resources end up being distributed. What it does not explain is why exchange takes the forms that it does, why people care about how well or badly it is enacted 30

The best known and most successful big policy scheme, Sonia Gandhi’s ‘feed the poor’ NREGA programme, guarantees a set number of days of paid work every year to every Indian citizen. When welfare schemes are not eponymous, the names and faces of their political sponsors are always prominently displayed on announcements of the giveaway sprees, which leave no doubt about their identities. Popularly, these programmes are known as ‘Sonia’s [Gandhi’s] programme’, ‘Akhilesh’s [Yadav’s] scheme’ and so on. 31 Eisenstadt and Roniger’s (1980) formidable effort to move patronage to the centre of social studies, at a time when it had already slipped to the sidelines, and to pull together the disparate case studies of the 1960s and 1970s into a coherent model is representative. Their exhaustive list of transactional characteristics of patronage runs to 19 points, but it fails to provide any more serviceable framework for understanding what makes patronage such a distinctly important socio-political form.

22  Anastasia Piliavsky

and why they are so concerned with morally appraising those who engage in it. None of this can be understood without a sense of the motivations and values behind the transactions. Because ‘patronage’ describes a particular moral idiom through which transactions are organised, we cannot understand it as a transactional network. If patronage is a moral form, the question is: whose morals are these? The model of individual profit-seeking, which shapes the idea of ‘clientelism’ in political science, precludes this question, leaving no room for understanding how South Asian politicians lead, how their followers follow, how they all decide what to do, what not to do and how to go about it in the right way. To make sense of all this we need to grasp what they value in politicians, what they think political leaders should and should not do and what relations with them should be like: what we may call a moral logic of relatedness. The calculus of rational choice offers one very distinctive set of values, which pivot around the calculating, profit-seeking individual. The actors themselves, however, insist on the opposite. They say that instrumental self-advancement is bad and that uncalculating altruism is good. As they swap gifts for services, they are not just trading, but also displaying, and hoping to be shown, the good. You can shove some cash into a voter’s pocket (as I show politicians in Rajasthan doing in Chapter 6), but this does not itself constitute an act of ‘patronage’ or guarantee electoral loyalty. Such ‘donations’ can only be thought of as patronal when the giving is perceived as generosity, and performed to this effect. The local logic of patronage is not about profiteering. On the contrary, it is a radical denial of individual self-advancement. Patronal munificence and the client’s service (seva)—both of which ought to be markedly selfless acts— affirm social relations, connectedness and society as ultimate values (see Osella’s Chapter 16 for more on that). Giving and serving as such are necessarily exertions for another’s sake. Shows of patronal munificence display the will to give away, something the picture of self-serving appropriation turns upside down, making it impossible to grasp the inner rationale of patron-client proceedings. The idea of selflessness is captured by the language of kin, which is often used in patron-client relations. The idea of patron as parent is very old, and traditionally royalty and patron-deities have been referred to as ‘parents’, or maa-i-baap (literally, mother and father). Parental honorifics have persisted in popular use

Introduction 23

and in villages across northern India people still call policemen, bureaucrats and parliamentarians just that—maa-i-baap (Roy 1965, 560; Saxena 1998, 494; Chapter 6 in this volume). Gandhi is still remembered as ‘Bapuji’, or honourable father. In some places the language of parenthood has given way to talk of dadas (‘elder brothers’ or ‘grandfathers’) (Hansen 2001, 72), but the moral significance of the new terms has remained—dadas are elder kinfolk and they ought to act as such: with selfless devotion and boundless love. Whether or not they in fact do this is a wholly different matter. The language of kinship as such, insofar as it expresses what Marshall Sahlins called the ‘mutuality of being’ (2011, 10), rejects the individual self in favour of relations in which individuals give themselves away, often quite literally via selfconstituting ‘gifts’. As in Jean-François Bayart’s Africa (2009), on the subcontinent ‘feeding-and-eating’ is the most pervasive cultural idiom (and act) of self-diffusion. At times of election, victuals are politically vital and their distribution forms the lifeline of voters’ identification with politians (Piliavsky, Chapter 6). ‘Feeding and eating’ is the central idiom in popular assessments of politicians: ‘feeding’ refers to the correct performance of patronly duty and ‘eating’ describes its betrayal, or ‘corruption’ (ibid.; also Chapter 9). The moral logic of selflessness marks the line between two classes of South Asian political personages: political patrons proper and the vast class of political ‘fixers’. In practice, patrons and fixers ply the same trade: both extract money and political support in exchange for access to state resources (Berenschot and Osella Chapters 8 and 16). Yet in popular discourse patrons and fixers stand on opposite ends of the moral spectrum: patrons are held in esteem and fixers are scorned (often with derogatory terms like dalaal, ‘fixer’, or chamcha, ‘sycophant’). The contrast has nothing to do with differences in what both do in practice, but in the moral assessment of motives that people ascribe to their work. A patron does what he does—ought to do what he does—out of selfless devotion and a fixer does it for gain. As Berenschot’s account (Chapter 8) makes plain, ‘patrons’ and ‘fixers’ are not different persons, but different evaluative terms that can be used to describe the same person under different circumstances or be put to instrumental use. An MP who should protect and provide, but who instead extorts bribes, may be dismissed as a ‘fixer’. A petitioner may plead with a professional fixer with honorifics used for patrons

24  Anastasia Piliavsky

in hope of receiving help at no cost; upwardly mobile fixers may offer services to clients at no cost to gain patronal status, and perhaps even get elected.32 The patron–client relation is not a stable arrangement or a freestanding phenomenon, but a normative formula. While the roles remain constant, the practical content of such relations alters ceaselessly and the actors are ever changing, often switching back and forth between the two roles, as suits their purposes. A head of police station in UP may play the patron with petitioners, but he is also a client of his rank superiors and the local MLA. As Jauregui suggests (Chapter 10), when police officers move in and out of roles of patron and client, the moral basis for their actions also shifts. Patrons and clients may swap parts. In Seyfort Ruegg’s medieval Tibet (Chapter 2) princes and preceptor-monks were ‘patrons’ and ‘clients’ in turn; asymmetry of status seesawed as each gave to and accepted from the other the material means for life and the religious validation for rule. During electoral campaigns candidates play the patron. They host feasts and give away hundreds of thousands of rupees. If they hope for a re-election, they will carry on in the same mode: dole out cash, build roads, ‘rescue from the police’ (thana bacchana) and arrange jobs for their constituents. But when they appeal to constituents, they turn the donor–donee relationship upside down, promising ‘selfless service’ (seva) in return for the ‘gift of a vote’ (mat daan). Or, when pressed to deliver what they had promised, politicians hide behind the rhetoric of seva, styling themselves ‘servants of state’—not patrons but themselves clients with no duty or power to provide (Chapter 6). Yet however much patrons and clients may improvise, they return to the keynotes of service and gift. This composition is not in symphonic harmony, as its lofty morals suggest, but in conflict that lies, as Gilmartin shows, at the very heart of patronage: ‘An effective patron is enmeshed in a world of perpetual calculation of costs and benefits, even as he projects an image of moral transcendence over the strategic connections and calculations on which his position relies’ (Chapter  5, 126). Generosity requires wherewithal, which patrons must accumulate, by whatever means, and so risk a clash of high morals and 32

For more on the moral complexities of political fixers’ careers, see chapters by Osella, Björkman and Berenschot in this volume.

Introduction 25

instrumentalism, which comprises its material base. To satisfy the patronal ideal, patrons must betray it.33 The rhetoric of selfless love often hides brutal distortion and cynical profiteering: and the more brutal the politics, the more politicians speak of ‘love’ (see Ruud, Chapter 13). As Diane Mines (Chapter 3) shows, the threat that lurks behind the promise of patronage and the fear of losing patronal bonds often suffuses low-caste, or tribal, accounts of their past (see also Price 2006). Patrons can never deliver as much as their clients expect them to. This is particularly so if they are not royal sovereigns, but politicians or government officers with limited legitimate means. They may use ostentation to compensate for their inability to provide, as Mattison Mines (Chapter 1) reports from Tamil Nadu. But such performances convince few, making electoral politics a desperate affair, full of hope and disenchantment, in which politicians are perennially suspected of hypocrisy, cynicism and sinister intent (Piliavsky, Chapter 6). In cities and villages across the subcontinent people say that politics is in essence degenerate—‘dirty’ (Ruud 2001; also Price, Chapter 9). This inner conflict is reflected voluminously in reports on electoral corruption and excessive campaign spending, which regularly agitate South Asia’s middle-class imagination and its projections in the international press. In patron–client relations the balance of forces is always liable to tilt, till the scales collapse, and relations which should be mutual become coercive. Then patronage turns into slavery, indenture or other forms of inequality with no legitimate moral base. When that happens, it ceases to be ‘patronage’. As John Powell noted some time ago, ‘patron-client ties clearly are different from other ties which might bind parties unequal in status … such as relationships based on coercion, authority, manipulation and so forth. Such elements may be present in the patron-client pattern, but if they come to be dominant, the tie is no longer a patron-client relationship’ (1970, 412). When the moral form falls to pieces, what it leaves behind is a skeleton of transactions—the bare payments (or  appropriations) 33

This tension is echoed in earlier writings on patronage, which remarked on the built-in contradiction between the exercise of power on the one hand and the assertion of solidarity on the other (e.g., Roniger 1994, 4). It also tracks Bayly’s (1973; 1983) distinction between dharmic and vakil patronage, which I note above.

26  Anastasia Piliavsky

and services (at times forced)—which analysts who view patronage as a set of transactions readily mistake for the real thing. The ugly practical consequences of patronage, when bonds of hierarchy become a struggle of unequals, are all too evident. As Nicolas Martin’s stark account of Pakistan illustrates (Chapter 14), a patronage which has become indenture can bring social catastrophe (see also Breman 1974). In the Pakistani Punjab patronal masters trap client slaves in cycles of chronic debted and bondage, where clients languish for generations without the protection of morals or a state law (itself a plaything of their masters). These grim pictures have long led their interpreters to see patronage as a mere gloss for abuse. But the client slaves no longer perceive their masters as ‘patrons’ who provide and protect. Fear that the normative frame is just a mask for the politics of greed and that patronage can at any moment disintegrate into coercion was there long before electoral politics reached the subcontinent. It has always been felt most acutely by the most marginal and vulnerable service-groups (often ‘tribal’). In her account of one such community, the Valaiyars of South India, Diane Mines (Chapter 3) relates how hopes for good patrons and fears of bad ones structure Valaiyars’ historical narratives and define their sense of self, present and past. Good patrons make life good and bad ones make it intolerable. Which particular circumstances precipitate momentary or longer-term collapses in patronage is a question for historians. But history shows that patronage, in its intrinsic moral ambivalence, is a highly unstable practical arrangement with material consequences which shade from the magnificent to the appalling, as people across South Asia readily recognise. Mattison Mines (Chapter 1) offers colourful snapshots of patronage’s brighter side: wealthy Tamil big-men using their fortunes to develop travellers’ lodgings, educational institutions, temples and medical services for the poor. No one demanded these from them. They were what a progressive patron did spontaneously for the common good. Another vivid example of ‘good patronage’ comes from medieval Tibet. There, Hildegard Diemberger (Chapter 15) shows, the great 15th-century efflorescence of religious and literary production emerged from relations between princely patrons who sponsored the ‘printing revolution’ and Buddhist monks who printed the books. The intense, if always unstable, patronage bonds she describes drove technological and aesthetic innovation, along with much cultural and

Introduction 27

economic advancement—as they did in Italy during the Renaissance. This account evokes the one context in Euro-American discourse in which patronage remains exempt from derision: artistic production, which still relies substantially, and without stigma, on the good will of patrons.

Patronage as politics In mid-19th-century Britain political debates focused intensely on elections and the legal and procedural standards, which were still to be clearly established. By the 1870s, the government had adopted the secret ballot and a body of electoral law, both of which travelled to India in the early 20th century, as elections were introduced. The debate hinged on a crucial distinction—between good and bad ways for politicians to influence voters. This was no easy task, and the elusive difference between ‘due’ and ‘undue’ influence was disputed for four decades. At stake was the delicate balance between voters’ freedom to choose and politicians’ freedom to sway their decisions, on which democracy rests. The voters’ freedom to choose was required by the democratic mandate of ‘people’s sovereignty’ and the politicians’ freedom to sway voters’ choice was necessary for electoral politics to work at all. As Gilmartin explains (in Chapter 5), political thinkers of the time espoused the ideal of free choice as the basis for democratic governance, but also recognised that electoral politics could not take place without voters’ attachments to politicians, most of whom were no mere equals of their constituents, but ‘men of influence’. They recognised that patronage was unavoidable—and not necessarily undesirable—in democratic practice. It was not a question of whether politicians could patronise voters, but of how they should do so. Electoral law was not intended to eradicate patronage, but to separate good patronage from the bad. George Grote, historian, radical politician and advocate of the secret ballot, argued that while ‘pressure’ or the ‘purchase’ of votes threatened the people’s sovereignty, there was nothing wrong with the ‘influence of wealth and station’, which naturally—and legitimately—drew voters to follow it. ‘“Good” patronage thus provided a model for how democratic politics ought to operate’ (Chapter 5, 130). The moral vision of liberal democrats today retains no trace of this picture. Since the 19th century, the freely choosing individual

28  Anastasia Piliavsky

has been sheltered from ‘external influences’ in the isoloir of the voting booth, has triumphed over patrons and clients. We may no longer deny the practical co-presence of patronage and democracy, old and new, but we still find it difficult to believe that the two can be morally commensurate.34 To grasp how peculiar this resistance really is, try running a simple thought experiment. Recall the liberal master principles of good democratic practice: equality and freedom of choice. Imagine a voter who decides to exchange her vote for a bicycle she wants or needs. Imagine both she and the candidate with whom she made a deal are both perfectly honest and each gives their due. Does this subvert the principles of choice by equal and free individuals? It does not. Yet does it strike us as morally acceptable? Probably not. Yet a purchase is as pure an exercise of individual free choice as there can be. What is upsetting us in this picture? Why should we think so poorly of swapping votes for bicycles? We do so because we believe both voter and politician are merely advancing their personal interests. Sound political choices should emerge from concern for the greater social good, and be driven by policies and ideologies whose benefits stretch beyond any individual’s interests or lifespan. Or so we think. Paradoxically, it is the very rational-choice theorists, and other hardened neoliberal champions of reasoned advancement of individual selves who, when it comes to democracy, believe ‘good’ political judgement must be completely selfless. What is it about democracy that makes egocentric and wholly ‘free’ choice appear so reprehensible, even to avowed egoists? Whatever institutional arrangements people live under, or ideologies they espouse, from feudal to fiercely individualist, their politics is necessarily shaped by a relational morality: a set of ideas about how those who govern and those they govern should relate to each other, and conceptions of political community which issue from these ideas. When democracy meets individualism, it generates a contradiction: an explicit denial of sociality alongside a demand for the highest levels of selfless dedication from citizens. ‘Corruption’, as defined by Transparency International or decried by Anna Hazare, encapsulates this denial. In its official definition corruption is the ‘misuse of public office for private gain’, the 34

Gilmartin makes the same point (Chapter 5).

Introduction 29

capture of government by the ‘interests’ of family, friends, tribe or any other community, and indeed by relations between one particular person and any other. Corruption is the eruption of sociality into governance. But how can representative democracy work without particular relations—relations of real human beings to other human beings? What are elections if not acts of preference in which politicians bid for voters’ exclusive loyalties and voters select the candidate of their choice? More broadly, what is democratic representation if not a social relation? Articulations of this relation vary from culture to culture (Spencer 1997), but however hard cosmopolitan political theory tries to purge politics of sociality—its own grounds for representative relations necessarily invoke society as a whole. The morals of relatedness carry special weight in representative democracies, which do not only require intense and regular involvement of the governors with the governed, but authorise such relations under the banner of ‘representation’. It is precisely because neoliberal theorists assign to political representation such great moral significance that they find patronage-based democracy so deeply offensive. There is no ready single equivalent to the English word ‘patron’ in the Indian languages. The many appellations (raja, jajman, periyar) and common terms of address (bhaiya, anndata, maa-ibaap, dada) South Asians use for ‘patrons’ cannot be captured by a single term as in the languages of Europe. The very fact that the semantic field is too broad and important to essentialise in this way suggests that the term’s socio-political implications are much broader than words like ‘sponsor’, ‘master’ or ‘boss’, the meanings routinely attached to the Latinate ‘patron’ in the Romance and Germanic languages.35 The studies collected here show that ‘patronage’ does not apply to a narrowly defined set of political relations. It encompasses the fundamental principles of social life far beyond ‘the political’, in printing in medieval Tibet (Diemberger, Chapter 15), relations between gods and humans (Michelutti, Chapter 12) or in the lives of Keralan migrants in the Gulf (Osella, Chapter 16). Political patronage is an expression of the broad moral sense that shapes the ways in which people relate across social levels 35

For more on the sense of ‘patron’ in French, see Ruegg (Chapter 2) in this volume.

30  Anastasia Piliavsky

and contexts. The essence of this moral formulation is the idea that in South Asia differences of rank do not prevent relations, but promote intimacy between parties in distinct and complementary roles. The roles themselves and the people who play them are defined vis-à-vis others, and the values their occupants espouse are oriented towards relations and sociality. Louis Dumont called this social apperception ‘holism’, contrasting it with the anti-societal orientation of the modern West (1980; 1986), whose inhabitants imagine that human dignity and respectful relations can arise only from a basic condition of sameness: the state of being individual, or identically unique. From this angle, differences of rank and power look morally distorting in themselves. Many have argued that the post-colonial democratisation of the subcontinent levelled social hierarchies in the region. South Asia’s masses, the argument goes, no longer heed traditional elites, but seek instead to topple them. Political mobilisation is no longer typically ‘vertical’, but increasingly ‘horizontal’ (intracaste) (Breman 1996, 262). The logic of caste has lost its legitimacy and whilst ‘castes’ have not vanished, the order of ranked and mutually dependent communities has given way to a collection of independent ‘societies’ (samaajs) or ‘ethnic groups’ (e.g., Chandra 2004; Michelutti 2008).36 This account captures much of what has happened in South Asia’s changing recent past. It frames the story of many new social formations, political players and parties. What it fails to reveal, and indeed obscures, is a central structuring principle at work behind the plotline. If the imagined totality of caste is no longer what it seemed to earlier lyricists of timeless ‘village India’, the relational principles of mutual dependence and intimacy across differences of rank have not lost their force. The persistence of patronage in shaping the politics of the region is the strongest possible testimony to this. Analysts of South Asian politics often say that local politicians are less representatives and more rulers who covert citizens into subjects and take away their electors’ prerogative to authorise governance. The royal stage set of popular politics—the crowns, 36

Literature on the decline of hierarchy is vast. Representative accounts include Béteille (1965), Kothari (2010 [1970]), Barnett (1975), Harriss (1982), Robinson (1988), Frankel and Rao (1989–1990), Breman (1996), Fuller (1996), Panini (1996), Seth (1999) and Michelutti (2008). For a helpful overview of this literature, see Manor (2010b).

Introduction 31

garlands and genuflections—appears to endorse this view. But are the citizens of South Asia utterly stripped of their sovereignty and the leaders they choose just sovereigns in new guise? Democratic representation is both a mode of political authorisation and a social relation through which governance is organised. In Europe and the United States that relation has been conceived, at least since the 17th century, as a contract, or explicit agreement between electors and those they elect to govern on their behalf. Representation is always an imperfect substitute for the citizens’ direct involvement in governance, a way to render most citizens who are in fact absent from government vicariously present with in it. Political representatives both stand in for and stand for those whom they represent. The substitution of one person for another is never a straightforward cognitive process, and does not explain how exactly any person can ‘stand for’—or represent—any other.37 The answer lies in the ways that people conceptualise relations on which they can rely, and more specifically, in how they imagine identifying with one another.38 Cosmopolitan political theory has privileged identification through contract, placing it above all others, which are now illegible, like the rival political systems which they uphold. In South Asia, no less than elsewhere, identification—and hence ‘representation’—forms the centre of politics, audible perhaps most loudly in the ubiquitous promise of politicians to work for their people and the voters’ stated desire to elect their men. South Asian patronage-politics is no more grotesque than politics anywhere else. Its high normative code, however often and brutally betrayed in practice, prescribes distinctive ways of identifying with politicians and distinctive ways of asserting demands on them. Local forms of political identification, which we gloss as ‘patron–clientage’, are 37

Some years ago Jonathan Spencer (1997) bade anthropologists investigate the content of democratic representation. But aside from Michelutti’s (2004; 2008; Chapter 12 in this volume) accounts of substantive (and divinely inspired) identification of citizens with their political leaders, this call remains unanswered. In the only available collection of anthropological studies of democracy (Paley 2008) the term ‘representation’ appears only a dozen times, and in none of them queryingly. 38 Some anthropologists have referred to indigenous forms of political representation and other democratic forms as ‘vernacular’ and the process of creating them as ‘vernacularisation’ (e.g., Michelutti 2004; 2008; Tanabe 2007; Neyazi et al. 2014). This leaves one wondering what kind of democracy is not vernacular or where we may be able to find it.

32  Anastasia Piliavsky

often bonds of intense intimacy: relations of common substance, shared divine origins and blood (e.g., Price 1989; Spencer 1997; Michelutti 2004; 2008; Chapter 12 in this volume). They are as substantial as contractual links (and arguably often more so) and they generate equally powerful political loyalties. It is not only that people may appeal to patronage to resist political pressures or demand resources from governments (à la Chatterjee 2004).39 Patronage itself involves entitlements and obligations, which are politically constitutive in their own right, and which oblige politicians to understand, convey and respond to their constituents’ needs. Ideally, patrons are a perfect vessel for the people’s rightful capacity to govern themselves—for their sovereignty. (If rarely or never so in fact.) In South Asia, as in Europe or the United States, the gulf between ideal and actual politicians is vast and expressed most starkly and ubiquitously in rumours of ‘corruption’—the global gloss for all that is wrong with the politics of our dreams. These dreams differ from place to place like the narratives of their betrayal. As several authors in this volume show, the meanings of ‘corruption’ on the subcontinent carry differing messages. If cosmopolitan followers of Anna Hazare in Bombay (Björkman, Chapter 7) equate patronage with ‘corruption’, for many others it is the distortion of good, proper patronage which is ‘corrupt’. Across the subcontinent, people hold few illusions about the goodness of politicians, but when leaders show loyalty to ‘their men’—by whatever means—they attract intense admiration. In Kerala, townsmen accuse politicians of corruption not when they ‘buy’ electoral loyalties, but when they fail to deliver on their promises (Osella, Chapter 16; also Osella and Osella 2000a). For villagers in Andhra Pradesh ‘corruption’ is not the embezzlement of public funds, but the politicians’ failure to share its spoils properly (Price, Chapter 9). Or, in the words of an Oriya man cited by Wilkinson (Chapter 11), ‘all politicians eat, but the BJD [regional Party] eats less’. In all of these formulations ‘corruption’ is not the misuse of public office for private gain, but the collapse of ‘good patronage’—when patron-politicians prove instrumental, selfish and tight. 39

Appadurai (1990), Osella and Osella (1996), de Neve (2000), Staples (2003), Berenschot and Osella (Chapters 8 and 16) in this volume, show that in South Asia political patronage involves greater reciprocity than earlier studies of clientelism have assumed. In a recent ethnography, Ajantha Subrahmanian (2009) showed how Keralan fishermen deploy the language of patronage—not the language of rights—to claim benefits from the state.

Introduction 33

The idea that hierarchy may form the basis for democratic governance may look less incongruous if we turn back to the Euro-American origins of modern democratic rule. When representative governments were first established in Europe and the United States, few thought of democracy as an essentially egalitarian or individualist political form. Political thinkers of the time, either side of the Atlantic, understood that the relationship between governed and governors—whether elected, divinely ordained or otherwise—is necessarily ranked. Rousseau, for instance, (1984  [1754]) described representative democracy as ‘elective aristocracy’, a system that embraced social differences and put them to political ends. The Founding Fathers who drafted the American Constitution were not egalitarians, and neither were the citizens of Ancient Athens, the only state of any scale where democracy ever existed in its ideal, direct form (e.g., Dunn 2005, 24–44). Both were slave-owning societies and both comfortably excluded women from franchise. Democracy departed from other governmental forms not by levelling differences between electors and politicians, or between people more generally, but by allowing the ordinary citizen to place the extraordinary politician above himself (Manin 1997, especially Chapters 3 and 4). The grounds for choice never included only merit (talent, charisma, experience), but also inherited status and wealth, both of which had great importance for electors in 18th-century Europe and the United States.

Idols in Westminster In August 2013 the Daily Telegraph reported an embarrassment: British Members of Parliament in Westminster were found ‘worshipping at the feet of idols like Thatcher and Churchill’ (2 August 2013). Apparently, it is the parliamentarians’ custom to rub the feet of statues of their parties’ tutelary patrons—Thatcher for Tories, Clement Attlee for Labour and David George Lloyd for Lib Dems. But in 2013 the practice was formally banned, not just because it threatened the condition of the four statues outside the Commons chamber, but because it caused a scandal like the incident in Jodhpur. Politics was marred by ‘worship’. The ‘idols’ in Westminster were only statues, not living politicians, and one could see the MPs’ practice as a charming ‘superstition’, but it pressed an awkward question: if our own parties and politicians have patrons, may our own politics too be subject to the unreason of personal bonds? In the 1830s George Grote

34  Anastasia Piliavsky

would not have opposed such a claim nor found it upsetting. But this was a long time ago. By now the idea of abstracted, impersonal, asocial governance has prevailed as the sole possible model for good governance. Bizarrely enough, it is an idea which deprives democracy of its demos and politics of its content: the ‘interests’ and ‘influences’, which we call ‘corruption’, yet without which electoral politics would simply cease. In the Summer of 2013 I met the priest who had designed Vasundhara Raje’s poster. He turned out no lunatic, but astute, witty and conspicuously sane. He cast the Chief Minister in the image of the bread-giving goddess because, he explained, she developed the midday meal scheme in Rajasthan’s primary schools. ‘In India we respect seniors and people who have the power to bring good to their people. This is an old Indian tradition. Devotion is a way to show our respect’. A year earlier, one of Miss Raje’s cabinet ministers caused similar commotion by claiming he worshiped her icon every day. He shrugged off suggestions that there might be anything perverse about this by insisting the BJP was a family and that he deferred to Raje as he would to his own mother. Indeed, what the Anglophone press may deride as ‘idolatry’ is a widespread South Asian form of deference to authority, whether democratic or divine (e.g., Bate 2009). The people cast their chiefs, patron-deities, ancestors, elders and MPs in the role of ‘patrons’, not only entitling them to special honours and privileges but making them responsible—obliging them to provide, protect and stand accountable for their actions. In practice, this logic may sometimes fare much better than other logics (see chapters by Mattison Mines, Osella and Diemberger), and sometimes much worse (see chapters by Martin and Michelutti). But however it manifests itself, and however we judge its many manifestations, we can ignore it only at the expense of comprehension. To see a similar set of principles at work elsewhere, we need only look back a century in the history of Britain. 19th-century commentators on British politics recognised and accepted that persons who were superior (whether by achievement or birth) were obliged by their status to bear a greater load of responsibility, making them the electors’ natural, and indeed most appropriate, choice (Gilmartin, Chapter 5). As Tocqueville remarked in his own memoirs, the early days of the French Republic—before the French countryside was infested with talk of ‘degrading equality’ (already rampant in

Introduction 35

the United States)—peasants attached themselves to politicians-elect (including Tocqueville himself) with exceptional fervour (1896, 127). For French peasants in the mid-19th century, inequality was not democracy’s foe. On the contrary, it provided political representation with a solid structure of responsibility: politicians were grandees who had to act in ways which were commensurate with their standing. Liberal theorists and champions of democracy may choose to tell themselves that democracy proper raises politics above inequality and particular social bonds. But if to them the politics of South Asia— and much of the rest of the world—still appears unprincipled, it is because among them its principles remain so poorly understood.

The Idea of Patronage in South Asia

The political economy of patronage, preeminence and the state in Chennai


Mattison Mines

Clients and patrons: a vignette about social mechanics


n 1967, when I began my first fieldwork in Madras, I rented a floor in T. Balasaraswati’s house. ‘Bala’ was arguably the leading Bharatanatyam dancer in India at the time and a descendant of a long line of courtesans once attached to the Thanjavur court. By the time I met her, she was a wealthy head of an extended matrifocal household composed of herself, her daughter, two of her three brothers, one recently married, and her dance master. Early on, one of the things that impressed me was that she was not a patron of others, she was a client—if of the rich and famous—and even I, a lowly graduate student, was cast in the role of a patron, a giver of gifts. Here I was a new arrival in need of services. Bala offered few answers, but down the street lived a paediatrician, Dr K, and I had an 11-month-old daughter. Dr K became my patron. In the 1960s milk was scarce and usually adulterated—groundnut paste, water and milk flavouring was the formula, which produced a telltale sediment in coffee. Dr K’s milk was pure and rich and I asked him if his milkman could supply us. Later, Dr K told me to expect milk every morning. After that, each morning the milkman arrived with a she-buffalo in tow and milked her at our gate. Guests noticed the quality of the milk in their coffee, and a fellow graduate student, recently arrived, asked me whether my milkman could also serve him. I said I would ask. But, when I did, the milkman explained that he had no milk to spare…even for me! He said that Dr K had asked him to serve me, and it was for him—Dr K—that he provided me. Dr K was the milkman’s patron and was helping the milkman’s sons gain admission to college. By serving me, the milkman reciprocated Dr K’s help.

40  Mattison Mines

Figure 1.1 The author and his assistant V. Gopalakrishnan interviewing Kaikkoolar naad.u headmen, Tirumalasai (23 November 1985). Tirumalasai was the head village (talaiuur) of the Talai naad.u territory, which included several villages now incorporated into Chennai. I got other services through Dr K—always topnotch: carpentry, an electrician, introductions to wholesalers of raisins and cashews, even weekly tune-ups for my 1949 Morris Minor. I was never taken advantage of. From these early, and for me formative, experiences I learned several things about patronage, which are familiar to all who have lived in India: 1. Wealth alone does not make a patron. Artists who perform for others are the clients of those for whom they perform. I was to learn that Bala’s patrons had numbered among the wealthiest residents of Chennai. 2. In South India, where many services and commodities are scarce, patrons have ‘good connections’ and command a clientele who supply these things. The relationship is a hierarchical one, but also a reciprocal one that is enduring: clients receive benefits from their patron, which is why they serve their patron. Selfless beneficence for the good of others is the patron’s style and is central to his

The political economy of patronage 41

or her reputation and preeminence. In 1995, while conducting research among the Komati Chettiar or Arya Vyshya merchants of Chennai, I was invited to a private concert in Bala’s home. Bala had died in early 1984, and her daughter Lakshmi Knight, her brother T. Viswanathan, and Lakshmi’s husband performed the concert. The small, select audience was composed of elderly patrons, descendants of the wealthy Telugu families who were among Bala’s patrons. They continued their patronage of these artists, who were descendants in this famous musical line, one of the most renowned lineages of Carnatic musical tradition in South India. 3. Those who provide services or commodities will not cheat persons associated with their patrons: the exchange is based on trust between patron and client. Clients dare not cheat, lest they damage their reputation with their patron. Trust is founded on reputation and on knowing who a particular person is.

Research among castes of the left-hand and right-hand moieties The purpose of this chapter is to explore the evolution of my thinking about patronage and the patron–client relationship, as it developed during the course of my several research projects in Tamil Nadu. After my initial research among Tamil Muslims, I turned my attention to Hindu castes that had belonged to the ‘left-hand’ (id . angkai) moiety of castes in the Tamil region. The first of these projects focused on the Kaikkoolars, or Sengunthar Mudaliars, a caste of traditional handloom weavers. The locus of research was the Salem–Erode–Trichengode area and the statewide cooperative system with offices in Chennai (Mines 1984). I then conducted research among the Beeri Chettiars in the petta (town-centre) of Chennai, known formally as George Town. The Beeri Chettiars are one of the great merchant castes of the Tamil South. That research culminated with the publication of Public Faces, Private Voices: Community and Individuality in South India (1994). A central focus of that book is a type of patron I call the ‘institutional big-man’ (periyar) and the role that such men play in urban sociality. Concurrently, I also studied the Achari caste, who lived among the Beeri Chettis in George Town. The Acharis, also called Kammalars and sometimes Panchalars, are the caste of goldsmiths, blacksmiths, carpenters and stone carvers. All three of these castes belonged to the left-hand section of castes.

42  Mattison Mines

Finally, in 1994 I focused on the Komati Chettiar merchants or the Arya Vysya, as they now prefer, also one of the great merchant castes in the history of Chennai. I studied the Arya Vysya in part because they are historically a paramount merchant caste of the city and in part because they belonged to the balancing side of the lefthand/right-hand moiety divide. The Arya Vysya gave me my first research experience with a caste that had belonged to the ‘right-hand’ (valangkai) moiety. My grand scheme was to study communities that were associated with artisan production and trade to develop an idea of how these groups expanded our sense of Indian society broadly. When I began my first research, most anthropology focused on village India and the dominant role of agriculturists, which certainly made sense, since India was seen as a country of villages and agriculture was the largest sector of the economy.1 But I was interested in trade and the organisation of merchants. After all, India had welldeveloped inland trade networks and had been a major participant in interregional trade for millennia. At the time I conducted research among the Kaikkoolar, handloom weaving and the textiles produced comprised the second largest sector of the state economy, and in the Salem–Erode–Trichengode region, weavers still preserved some of their regional organisation based on interlinked territories, or naad . u. When I then studied the Beeri Chettiar, I discovered that even today George Town’s layout preserves the separation of the right-hand and left-hand castes. The Beeri Chettiars and Acharis, again both left-hand castes, live to the east of Popham’s Broadway (Prakasham Road) that runs north–south through the petta. The Muthialpet and Monadi areas are the main subdivisions. Beeri Chettiars also live in the Park Town area around the Kandasami temple. The right-hand section is west of Broadway, especially the locality known as Pedda Naickenpet. To me, perhaps just as important as the selection of caste communities were the field methods I developed during my research among the Kaikkoolars. I decided to interview informants in a new way that involved minimal direction on my part, using a type of active listening developed as a psychological counselling method. After initiating an interview, subsequent questions were phrased using the informant’s own words, letting them control the conceptualisations 1 There were some urban studies, of course, including Pocock (1960), Hazelhurst (1966), Fox (1967; 1969) and Singer (1968).

The political economy of patronage 43

and the direction of their narrative. This method led me to pay closer attention to what my informants were telling me about what it was that they said they, as individuals, were up to, and what they felt they had achieved. Since people told me their plans and what they did to achieve their goals, I began to think more about individual agency and achievement and the roles that these play in social history. My informants explained the history of their community in Chennai in terms of what preeminent men and women in the past had done (Mines 1994, 53–56). This approach to listening also led me to a realisation that when people present themselves publicly in all the different guises of presentation, they are in fact making arguments designed to persuade. Public presentations themselves are arguments. What makes such presentations more or less believable to the listener/observer depends in part on how well the presentation fits accepted standards of estimation, the standards that form the background of any statement one references to express a point of view, to argue a point and to judge what we are being told (Taylor 1991). In my ethnography of big-man patronage later in this chapter, I describe the public presentations of these apical men: how they present themselves in ways designed to persuade observers as to who they are and how central they are to social life and the public good. I also realised that the outcomes of arguments were always uncertain, open to contrasting interpretations and constantly evolving. Yet almost anything presented to another person or group preserves within the presentation a form of argument. Argument for instance, frames written records such as a person’s will (Mines 2002). It is expressed in social institutions like charities, local organisations such as caste areas and denominational temples, public presentations such as temple processions or the way in which preeminent persons dramatise their public movements (Mines 2006; D. Mines 2008). As an example of argument conveyed in public presentation, think of how the Chief Minister of Tamil Nadu travels daily from his home to the Secretariat, red lights flashing, streets blocked, accompanied by an entourage. The roadway is not lined with onlookers wearing their top-cloths tied around their waists, as it would have been in the 19th century. Instead, supporters and supplicants line the way with posters depicting the Chief Minister, each with the sponsor’s name boldly printed at the bottom in the colours of her political party. In 1995, on the route from Jayalalitha’s home to the Secretariat, posters depicted

44  Mattison Mines

her as Shakti, a fierce goddess, and briefly even as the Virgin Mary. The argument in these presentations is the Chief Minister’s embodiment of power, a notion central to Tamil ideas about the cosmos and of the feminine. It is easy to interpret what people say, but interpretation injects the listener’s perspective. It is actually much harder to hear what people say and see the arguments in their descriptions of what they are doing. This combination of actions and argument is essential to dayto-day sociality and a significant feature of how informants describe themselves. When I studied the Kaikkoolar, I found that men who had achieved a certain amount of success in life and were recognised as leaders had a formulaic way of introducing themselves to me. The pattern of presentation for a successful gentleman would start with a description of his family: for example, that they were the head of their extended household and head of the family business group held jointly with their sons. They would then list their positions in institutions and what they did, political connections if they had any, memberships in clubs, trusteeships and the charitable activities they were involved in, ending with the awards and titles they had received in recognition of their service and munificence. In Chennai I found that the Beeri Chettiar and the Arya Vysya men of prominence followed the same formulaic pattern when they introduced themselves. One Arya Vysya man once gave me his calling card, which unfolded to list his offices and memberships. Another man of great accomplishment and public prominence in Chennai, and indeed in South India, gave me a three-page list of his institutional affiliations, honorary positions and public offices. Informants described him to me as a man who had particularly good ‘connections’. I would like to say that I immediately realised the significance of what these men, and sometimes women, were telling me. In fact, it took me some time to realise that each was explaining not just that they were publicly preeminent, something in any case I knew, but was mapping their place in society within a delineated set of institutional domains: family, businesses, public service and institutional management arenas, sponsorships and charity, clubs and associated public recognition. These men exemplify what subsequently in articles and my book, Public Faces, Private Voices (1994), I termed ‘institutional big-men’, a valued form of altruistic patron.

The political economy of patronage 45

A leader and his institutions are closely identified with one another, and the institution’s popularity waxes and wanes with the reputation of its head, his ability and his charisma. As Sudhir Kakar (1981, 40–41) observes, Institutions in India are…personalized to an extent inconceivable in the West; individuals who head them are believed to be the sole repository of the virtues and vices of the institutions; as human beings, such individuals in authority are thought to be accessible to appeal, open to the impulse of mercy and capable of actions unconstrained by the rule of the ‘system’.

Most big-man institutional domains are local. A few are regional. The religious leader, Kanchee Sankaraacharya Jayendra Saraswati, heads a region-wide network of institutions (Mines and Gowrishankar 1990). Finally, as others have observed, the institutional big-man is a widespread type in South Asia (e.g., Bayly 1996, 187; Banerjee 2010, xv–xvii; Haynes 2012, 89). Some institutional big-men are notorious figures, who nonetheless provide benefits for their followers who esteem them. Consider Bal Thackeray, the founding father of the Shiv Sena Party, as an example. By contrast, others are renowned for their beneficence and efforts to improve lives. Ella Bhatt, the founder of SEWA, is such a leader. Patrons come in all sizes and only a few achieve great renown. There are many would-be patrons who list their association with institutions, but who have few ‘constituents’ in their institutional domain. The Arya Vysya refer to such a man as a ‘letterhead big-man’.

The public presentation of big-men: Traveling the street An important aspect of presenting oneself as a preeminent individual is the public display of prominence, the enactment or portrayal that expresses the stature and role of an institutional big-man’s civic individuality and charisma within his domain. Except for politicians and high-level public servants, public displays of this sort today tend to be confined to temple processions and to public contexts where big-men are seen. At a public gathering the big-man may eulogise lieutenants and important figures, presenting them with garlands and

46  Mattison Mines

shawls. In this manner, he is seen as the apical person who bestows honours on important members of his domain. Travel and arrival are also used to display preeminence, as the example of the Chief Minister suggests. At music concerts, a big-man and his entourage may enter after the main singer or instrumentalist has begun to perform. He and his group will walk casually to seats reserved near the stage, so that all in the audience see him enter while the musicians pause, acknowledging his arrival. Consider the following account, given to me by an informant, of a prominent Chennai City Arya Vysya in the first part of the 20th century. The quote is from handwritten extracts from the autobiography (Chinnanati Mutchatlu) of Dr K. N. Kesari (1875–1953), himself a prominent Chennai City benefactor, in which he describes his patron (cf. Sriram V. 2012). He [Rao Bahadur Collah Venkata Cunniah Chetty Garu, d. 1912] was leading Chetty among the Komatis [or Arya Vysyas of Madras City]. All Vaisyas used to accord their first respects to him before starting any auspicious functions in their houses. He was rich, philanthropic and has a majestic personality. He was trustee of Shri Kanyaka Parameswar Temple. Now and then he used to visit the temple. During his visits he used to sport his forehead smeared with holy ash and a big and prominent red saffron mark [indicating his devotion to God]. He used to come in a Featon [Phaeton] Cart drawn by two horses with two people on either side waving fans. During such visits to temple the types of respects accorded to him are beyond those even to the Viceroy. To inform the people of the town [George Town] of Shri Chetty’s visit, the big temple bell was being rung. Hearing this all people on the way used to stand outside their houses tying their upper cloth around their waists waiting to pay their respects to Shri Chetty. To tell about the extent of his property it is sufficient to say that the entire area of the present Madras High Court and all buildings around were his.

In the above account, Dr Kesari describes ‘Shri Chetty’s’ dramatic public presentation of himself as an institutional big-man. Kesari lists

The political economy of patronage 47

markers of eminence: he is rich and his appearance is ‘majestic’, a feature of his charisma. He looks the part of a regal person: he stands out. Public displays of his wealth are the Phaeton cart accompanied by four fan-waving attendants. We are also told that he owns a large amount of some of the most valuable property in the business centre of George Town, which in itself would have made him a very wealthy man. Of more relevance to our topic of patronage, Shri Chetty is described as a philanthropist, an indication of his generosity and the ‘first respects’ or honours he receives, a sign that he is preeminent among his caste. Respects include the presenting him with a new veeshti or dhoti and a taambaalam, a salver of items symbolic of hospitality given to honour a guest whenever households celebrate an auspicious occasion. One also sees the close connection between the Sri Kanyaka Parameswari Devasthanam (temple), or SKPD, as it is known, and the man. He is the temple’s head trustee or dharmakartha, and the great bell rings at the temple when he is in route to it. He performed his passage for a constituency, for the men who, upon hearing the bell, lined up on either side of the road with their top-cloth wrapped around their waists in a traditional gesture for subordinates to show respect, and, as noted, for the households who gave him first respects or honours. When Shri Chetty travelled to the temple, his passage staged a dramatic public presentation or argument for preeminence and of his singularity as head trustee of the temple and head of his caste. Dr Kesari, who recalls Shri Chetty in his autobiography, was himself not an Arya Vysya but a client of Shri Chetty. He was a Telugu Brahmin who walked from the Ongole region of presentday Andhra Pradesh to Madras at the age of 13. He spent his early life in poverty. Fed and clothed by Arya Vysya charity, he became a famous Ayurvedic doctor and manufacturer of indigenous medicines. He ‘made a fortune, donated very liberally for all worthy causes and established educational institutions in Madras’. In short, he himself became a patron of others. Dr Kesari explains in his autobiography, During my acquaintance with Vaisyas my immediate basic needs for food and clothing got satisfied. Due to this later after I learnt medicine I could get more opportunities to treat them. They became my princely patrons. Later due to their hold, I became an employee in the Shri Kanyaka Parameswari Temple free Ayurvedic Dispensary because this was founded by the trustees of the temple. In this

48  Mattison Mines

temple poor Brahmin students were fed everyday. In those days Vaisyas had more respect towards Brahmins than now and food is given very freely. Everyday during Navarathri festival for all the 9  days Brahmins are fed. During those days everyday about 2000 Brahmins were taking food. The food prepared was very tasty.

Notice that Dr Kesari describes his benefactors as having a hold on him, one that he cannot deny, and so reciprocating for the benefits he received, he became an employee of the temple. Notice also the emphasis on giving food to Brahmins as a special form of benefaction. Feeding Brahmins is a valued form of generosity that publicly expresses the devotion of Arya Vysya and the religious style that the caste projects publicly. During my fieldwork among the Arya Vysya, I was told several times, including by a Brahmin woman whose mother was raised—‘informally adopted’, she said—by a wealthy, but childless Arya Vysya merchant, that giving food was the best charity because recipients ate as much as their hunger allowed. No one was left wanting more, a condition that had the potential of being the patron’s sin. Arya Vysya as a caste depict themselves as religious and describe patronage of Brahmins as an expression of their religious virtue. It is what might be called a targeted form of generosity or patronage. In his account Dr Kesari may be seen as a supporting lieutenant and client of Collah Venkata Cunniah Chetty (‘Shri Chetty’) who, as the dharmakartha of the SKPD, is the preeminent patron and a ‘merchant prince’. The two men publicly present their relationship and relative status during temple processions. Dr Kesari describes how when the Vijaya Dashami day (commemorating goddess Vijaya’s victory over demons) was celebrated at the temple, the deity was taken through the Varada Muthiappan Street where the wealthiest Arya Vysya merchants lived. Shri Chetty and the other trustees walked at the front of the procession, the deity ‘adorned with diamond studded jewels’ and accompanied by ‘police officials with horse-riding soldiers’. Dr Kesari and other temple employees followed them. We see, then, Dr Kesari, sustained by the charity of his Arya Vysya patron, which enabled him by dint of his own hard work to become a famous doctor, a manufacturer of Ayurvedic medicines and a founder of schools. He is a very appreciative beneficiary and loyal constituent of Shri Chetty, one who owes his own successes to the charity given him. In return, he serves the institution.

The political economy of patronage 49

Patronage and Tamil generosity Dr K, the patron who provided me the daily delivery of pure milk in 1967, and Shri Chetty express two styles of a generous man, a man who earns a reputation for being a good person. Both single themselves out from others through their actions and are respected. Dr K’s altruistic giving is a small-scale, yet important, form of patronage that is not attached to institutions and that has its roots in Tamil notions of generosity (e.g., Mines 1994, 11–12). Generosity itself is governed by a karmic logic that holds that acts of generosity counter sin and contribute to things going well for the generous person. At the level of individual relationships, one should avoid overt rejection of another, and one should acknowledge the needs of the other. Denial of another is the antithesis of generosity. And one must be careful to avoid the dangers that causing the desires of another to be unfulfilled generates.

Locating the roles of the institutional big-man To understand the significance of the institutional big-man and why patronage takes this valued form in South India, one needs to recognise several things about the organisation of local society and how things are done. To begin with, Tamil society is what I call a trust-based society. What I mean by this is that to be successful in daily endeavours, say in running a small business or in seeking employment, a person must be trusted. This means either being well known to others, so that they know personally what to expect of the person, or being known to the leading men and women of the community, persons whose reputations are known, so that they can vouch for the individual to those who seek to estimate him or her. For example, lenders will give credit only to those they know repay their loans, and employers hire only those for whom they or others they trust can personally attest (Mines 1994, 31–32). It follows that in a trust-based society people need to know and keep track of large numbers of people: who they are and what their reputations are. And in Chennai residents do keep track of a wide range of people. Indeed, in this kind of society, there is necessarily little or no anonymity. Either a person is known or others will avoid dealing with him. In 1994, when I began studying the Arya Vysya of Chennai, I rented part of a house that my landlord was remodelling. To protect the building materials stored in the compound, the landlord hired a watchman who lived and slept on the house grounds. The landlord told me not

50  Mattison Mines

to worry about the watchman, a poor man and a stranger to me. I could trust him because he was a known person. He could not steal and run away because the landlord knew his home village and so could always find him. He explained that there would be no purpose in the watchman running off to a place where he was unknown, because, if he did, then, being unknown, no one would be willing to engage him. To be hired, he would need someone trusted, a local big-man, to vouch for him. Similarly, but at a very different level of society, a prominent Arya Vysya businessman once told me that wealthy Arya Vysya businessmen used to lend large sums of money on trust, with a deed as surety or without: ‘A big-man on one side would say: “Yes, you can give the loan and he [the borrower] will return it according to the [agreed] time”. When a big-man gives his recommendation, then he is responsible. But there is pressure to repay because if [you] don’t repay, then the man’s reputation is finished’. This gentleman went on to explain that in George Town this kind of lending is much less common today because it ‘involves not accountable money’, or black money, which prevents lenders from taking legal action to recover a debt. This kind of trust-based lending slowly began to disappear after World War I and was gradually replaced by bank lending, which offered better rates. He explained that today Aggarwal, Sindhi, Marwari, Gujarati, Punjabi, and some Arya Vysya merchants still lend in this fashion. Trust went with a particular business: among rice merchants, among those in the jewellery business… [A  merchant would promise]: ‘tomorrow I will give 50,000  Rupees’. So they would borrow with interest depending on the time of the loan’. If a merchant took a commodity and then paid the loan in the evening, there would be no interest. ‘If someone cheats [fails to keep his word] cash, commodity, whatever, everyone will know by evening and they won’t do business again.

With bank lending, being known and having good connections with big-men can also be crucial. I describe in Public Faces, Private Voices (1994) how a man I call Tambi and his business partners used connections with Tambi’s elder brother, Annan, to secure bank loans for their start-up industry. Annan had established connections with the bank manager, a member of his caste, who knew Annan’s reputation as a trusted businessman. I write: In a trust-based society, when persons who control desired  resources, commodities, and services need  to

The political economy of patronage 51

choose who is to get what, then the identity of the individual seeking a benefit and the influence of the people  he or she knows can be the critical factors determining how choices are made. For example, a seller often must choose to whom among several potential buyers he will sell, and a banker has to decide to whom to lend money. Under such conditions, knowing eminent persons, people south Indians often refer to as periyar, big-men, can obviously be very beneficial…. Big-men can wield great influence among members of their communities, because people perceive them as being able to act as brokers on their behalf. The civic attention of the community focuses on the individuality of big-men when community members accord prestige to these community leaders, a public recognition that distinguishes their singularity and agency. These are the individuals who make things “happen” for their followers. By contrast, bureaucracy is designed precisely to circumscribe the influence and agency of individuals by  depersonalizing the operation of institutions (ibid., 35).

It is not that bureaucracy plays no role in South India. In fact, its role is crucial. Law and the courts, voter registration and elections, driver’s licenses, car registration, taxes and fines, examinations and university degrees, bank accounts, research visas and foreigner registration with the police fill people’s lives. Bureaucracy depersonalises relationships. But those of us who have lived in India can attest many of these forms of bureaucracy listed here, if not all of them, can be facilitated by connections to influential women and men. In 2006, for example, I  purchased a car but could not register it personally (however I tried) without using a go-between to pay the requisite bribe, which itself is intrinsic to contemporary bureaucracy and also relies on a code of trust.2

Arenas of preeminence in the 19th and 20th centuries: Temples and government honours How, then, does one become a big-man, a trusted person of influence and good connections, a person who can vouch for others and make things happen? In a trust-based society like Tamil Nadu and in the castes among which I have conducted research, a big-man makes 2 I thank Anastasia Piliavsky for making this point.

52  Mattison Mines

his own reputation and preeminence, which distinguishes him and draws attention to the presentation of his civic individuality. Holding a hereditary position of leadership, such as caste headmanship, does not in itself suffice. The big-man is a charismatic leader, a kind of virtuoso of sociality. In the early East India Company records, Arya Vysya headmen, titled pedda chetty (Big Chetty), were drawn from the Suncuvar family, and until the end of the 18th century these hereditary headmen were able leaders and wealthy merchants who commanded great respect from all, including, if sometimes grudgingly, the British who also depended on them as brokers and trading partners. In the early 19th century the Suncuvar family found its primacy challenged by wealthy merchants belonging to the Collah (also Kolla or Kollah) family. Collah Moothoorama Chetty and his younger brother Collah Ravanappa Chetty, ancestors of Dr Kesari’s ‘Shri Chetty’, built the SKPD (c. 1803) on the edge of the Arya Vysya Charity Garden. In 1804, after his brother’s death, Collah Ravanappa became dharmakartha or head trustee of the temple. In 1816 Collah Ravanappa negotiated special honours to be given him by the East India Company in exchange for a new style of benefaction. That year in Kanjeepuram he built a choultry, a large rambling rest house for Arya Vysyas and Brahmins on pilgrimage, and gave in perpetuity an endowment for choultry maintenance to the Madras Company Raj. In return for this expression of trust in the Government and the charitable nature of the public benefaction, he negotiated with the British for the Madras Government to give him public honours: a palanquin embossed with the Company’s crest, a heavy gold chain and medallion with the crest of the East India Company and silver-plated staffs to be carried before him, to ‘mark the high sense entertained of his liberality and public spirit conveyed in a Cowle granted to him by Government under date 13th September 1816’ (Collah Singanna Chetty printed genealogy). Wherever he went in public, these honours would be unabashedly on display in a show of his eminence and good relations with Government. Government negotiators also seemed to feel that being given control over the endowment embellished the Government’s reputation. After all, this wealthy man trusted the Government to manage his endowment for all times. I believe that the Government’s (Hindu Religious and Charitable Endowment) HRCE Department still does. The argument conveyed in this public display was, therefore, double-headed: first, Collah Ravanappa is of such importance that his bequest honours

The political economy of patronage 53

the Government and, second, in return the Government honours him as a public benefactor. The patron and the Government of British India scratch each other’s back, as it were. Recall how in a parallel manner at the beginning of the 20th century Collah Venkata Cunniah Chetty (Shri Chetty) also used the drama of his passage in the street to present himself as preeminent and to command the respect of observers lining the street. The British awarded Shri Chetty the title of Rao Bahadur (Hero-King) in recognition of his public charities, once again flaunting his excellent connections with Government. Notice that the preeminence of these two men is achieved in two distinct social arenas: one is located in bhakti (devotion) and temple management; the other in public display of colonial Government’s recognition. With the Company’s recognition, Collah Ravanappa Chetty presented himself as the premier leader of the Arya Vysya caste in George Town, challenging the Suncavar headman. Both men had their clientele and supporters, resulting in the formation of two factions or parties (katchis), one supporting the Suncuvar family and the other the Collahvars. Reflecting this rivalry, next to the Komati Charitable Garden (komatla tottam, later transformed into the kothawal chavadi produce market), is the oldest charitable institution founded by the Arya Vysya in Chennai, the TKKNN Vysya Charities located in Triplicane established in 1826. In the Charity’s history, one reads that in 1823 a group of the Arya Vysya residents of Triplicane belonging to Kollavari Katchi (Collah Family Faction) ‘conceived the idea of establishing an institution for the welfare of the people and for inculcating Bhakti and Godliness through Religious activities’ (‘Brief History of our Charities’, 25). Collah Ravanappa Chetty and his Arya Vysya supporters, the Kollahvari Katchi, overshadowed the Suncuvar headman with British help. In time, however, the great wealth of the Collahvar family declined and finally came to an end in 1940 when the temple’s hereditary dharmakartha was bankrupted and debarred from acting as temple trustee. The trusteeship was restored to a Collahvar descendant in 1952. The present dharmakartha has modest means, and while he is the dharmakartha and President of the temple management committee, he no longer ranks among the big-men of the caste. Similarly, although Arya Vysya still recognise the Suncuvar family as the line of their caste headmen, and so deserving of first honours at auspicious ceremonies, no one is left to carry on that role today. Among Chennai’s Arya Vysya headmen are figures of the past.

54  Mattison Mines

That preeminent Arya Vysya big-men are no longer members of these two families is not to say that the status of institutional bigman is in decline. As I have indicated above, preeminent men and women, in the guise of institutional big-men, continue to play a major role in the Arya Vysya community and city life. Indeed, members of the temple management committee are drawn from among the preeminent and are selected by the committee for their professional skills in administering the finances and charities managed under the temple’s umbrella.

The institutional big-man at the end of the 20th century For purposes of comparison and to illustrate the institutional identity of a big-man, I turn again to the Kaikkoolar and the story of Kasiviswanathan, or ‘Kasi’, whom I described in Public Faces, Private Voices (Mines 1994, 44–46; also Mines 1984). His story summarises the manner in which he introduced himself to me and explained who he was. Kasi was a prominent member of the Kaikkoolar weaver community in Trichengode, a town about 200 miles south and west of Chennai and not far from Erode, an important handloom wholesale market town and distribution centre. Trichengode is the location of a famous Shaivite temple dedicated to Ardhanariiswarar—Siva represented in a half male and half female form—which is perched dramatically on top of a rock outcropping at the edge of the town. The Kaikkoolars sponsor a temple festival and maintain a meeting hall adjacent to the temple where they hold naad.u, or territorial caste council meetings. Born in 1912, Kasi was 67 when I interviewed him in his home in 1979 (Mines 1994, 43–47): Kasi had a precise sense of what it was he needed to tell me so that I would know who he was and what he did and had done. His approach was identical to the way I had heard others describe themselves when they, like Kasi, had achieved some degree of eminence. He related to me first his status as the head of his family and family businesses. Next he listed his current civic offices, and then he described the institutional offices that he had held earlier in his lifetime. ‘He said that he was the proprietor of all his family’s businesses, meaning by this that he was still the head

The political economy of patronage 55

of his own joint family. He and his wife lived with their second son, while three other sons lived separately. Nonetheless, all five men together managed the several enterprises that they jointly owned under Kasi’s proprietorship. Kasi also ran a textile sizing business, which he owned in partnership with one of his brothers.

Kasi next described to me the offices he held in civic institutions: At the time of the interview, he was headman (naattaanmaikaarar) of the Kaikkoolar caste council in Trichengode and the joint headman of the caste’s territorial (naad . u) council for a region incorporating seven cities and their surrounding village hinterlands, known as the Seven City Territory (Eeruurunaad . u). Trichengode was the territory’s headquarters. In other words, Kasi was telling me that he was one of two preeminent caste leaders for the seven-city area. Next, Kasi said that he was president of seven religious institutions and the president of a committee that collected construction funds for the Trichengode Government High School. He told me that the committee had already built eleven rooms for the school, three in his family’s name. When Kasi spoke of his family’s name, he meant his name as family head. He had also built under his leadership and in his family’s name a marriage hall for his caste, which was located in front of the rock temple, an auspicious location. He said ten to thirty marriages were conducted there by his caste fellows every year. Finally, he said, he was also a member of the Trichengode Rotary Club, an association that kept him in touch with all the associations in town. He kept his finger on the pulse of local events, he said, and remarked that his second son had been the joint secretary of the Junior Chamber of Commerce the previous year. Next Kasi described civic offices that he had held in the past. Until 1977, he had been the vice president of the statewide handloom weaver’s cooperative society, Cooptex, an important post. A friend, caste mate, and close political ally was the society’s president. Kasi had also been a member of the state’s Handloom Board for ten years and director of three cooperative spinning mills located in Salem, Tirunelveli, and Srivilliputtur. And he had been the director of the Cooperative Union in Tamil Nadu, which is an advisory board to the government on

56  Mattison Mines

production cooperatives. Finally…when he was younger and had political ambitions, he had been the president of the Congress party Taluk Committee. At the time, I remember realizing that Kasi’s purpose in describing his institutional offices was not simply to list his accomplishments for me, but primarily to describe his role as an institutional leader with multiple constituencies, some small, some large. His description reveals the contextual nature of civic identity and its spatial dimension, which expands and contracts in relationship to the size of a leader’s constituencies…. Once he had been a state-level leader. Until his retirement from Cooptex in 1977, Kasi had served a statewide constituency that he had helped build. The handloom textile cooperative movement took its inspiration from Mahatma Gandhi [and C. Rajagopalachariya, ‘Rajaji’] and was until 1977 controlled by members of the Congress party, which used it for the distribution of largesse and the building of grassroots support. In those days, the handloom weaving industry was the largest category of employment after agriculture in the state, and so Cooptex appealed to what was theoretically a significant pool of potential political supporters. Throughout the state under Cooptex auspices, local leaders founded or managed production cooperatives, which tapped into the statewide association, enabling them to enact roles as patrons within their local weaving communities. Kasi told me that when a weavers’ housing colony was built in Trichengode with Cooptex support, it was named after him, although he himself had favoured naming it after Rajaji. As president of the Congress Party Taluk Committee, therefore, and vice president of Cooptex, Kasi was for a time a patron of statewide influence and eminence. In 1979, when I interviewed him, Kasi was a leading townsman and regional caste leader serving a much smaller constituency. As co-headman of the Seven City Territory, he shared his preeminence as a caste leader with only one other man. That night, following the interview (see Mines 1984, 124–129), I had the opportunity to observe Kasi in his role as headman during a council meeting that the leaders and representatives of the subcouncils of the territory attended. Kasi sat on a

The political economy of patronage 57

raised dais surrounded by the subordinate headmen and representatives of the Seven City Territory, and when I entered the meeting room, I had an immediate sense of his power. He was a chief among his lieutenants, a man with the power to outcaste or impose fines on any local council that failed to adhere to Kaikkoolar caste rules regarding the governing of councils…. Yet the caste council was only one of the many institutions within which Kasi held a preeminent role at the time of the interview. The other associations he described himself as heading were what Tamils term charitable institutions. The constituencies of these institutions—the seven religious institutions, the school building fund committee, and the caste marriage hall— were much smaller than that of the caste council. Each of these institutions, however, enabled Kasi to play the role of patron. As patron, Kasi cast himself in the role of an altruistic leader dedicated to the service of his community, his clients. He told me that he had donated Rs. 30,000 to the school building fund. Through these many institutions Kasi built for himself a grassroots political constituency, based on his personalised relationship with the circles of clients associated with each of the many institutions he managed.

The story illustrates how an institutional big-man presents himself as a powerful but altruistic leader. A respected institutional big-man presents himself as generous and motivated by the desire to serve others. He cannot be regarded as motivated by self-interest, which would make his word and actions suspect, uncertain—untrustworthy. He publicly presents himself as concerned for the welfare of others, and he gives generously. Kasi’s story is also instructive about the organisation of a caste and the relationship of big-men to their communities. His leadership is located in multiple, distinct forms of community, each with its own constituency: (1) religious devotional societies, (2)  charitable organisations designed to serve the common good of the Trichengode locality, including social service clubs like the Rotary Club, (3)  the school building society, (4) the Kaikkoolar naad.u system of territorial organisation, each segment of which is anchored by denominational Shaivite temples, and (5) the cooperative movement, including production and marketing cooperatives, spinning mills, the statewide

58  Mattison Mines

system allied with party politics and the distribution of political largesse which includes, importantly, housing projects. With the exception of the  naad.u system, these associations are what might locally be considered modern, progressive institutions designed to promote social upliftment through education and improved standards of living. The cooperative system is the largest of these. The cooperatives are modern, secular, bureaucratic institutions that were created in response to Rajaji’s political commands to improve the lives of the handloom weavers and for the distribution of political largesse. Kasi had met Rajaji and, as a political lieutenant, used the governmental resources channelled through the cooperative system to help provide benefits that improved weavers’ lives. The naad.u system is the other large regional structure, which connected over a large area small-scale weavers who produced handlooms for the all-India market, and on a more limited scale for the international market as well. The principal institutions associated with the naad.u system are the Shaivite temples managed by Kaikkoolars or in which Kaikkoolars have established rights of sponsorship. In the naad.u system, Kasi acts as a caste chief who has the authority to impose fines on member localities for their leaders’ transgressions. He can also collect taxes from his caste membership, which are used to fund religious festivals at the Ardhanariiswarar temple. His role as a naad.u chief demarcates him as an honoured leader within the caste in its traditional organisation. But note the modern domains that constituted in 1979 significant organising features of the caste in its modern form. The big-man of the caste is an agent who creates through his leadership the caste composed of multiple constituencies (cf. Subrahmanyam 2001).

Temples and preeminence among left-hand castes When conducting research among the Kaikkoolar, I found that temples, caste organisation, party politics and institutional big-men were of a piece. The institutions of big-men were not limited to temples, but temples were the basic major institution around which a local community and its leadership was organised. In Salem the Kaikkoolars pointed out to me that, if they did not control a Shaivite temple in their neighbourhood, they would build one of their own which they could control. This centrality of temples made sense in light of research concerning the role of temples in the kingly models described by Arjun Appadurai and Carol Breckenridge (1976), David

The political economy of patronage 59

Schulman (1985), Nick Dirks (1987), Gloria Raheja (1988) and Pamela Price (1989), among others. But I was finding that temples were not just a source of symbolic honours and public presentations of eminence. In cities denominational temples—temples controlled by a caste—were repositories of the main assets held jointly by the caste, and temple funds were used to distribute charity. Having a managing role in a temple, or other religious institution with rights in a major temple, was requisite to an institutional big-man’s identity and to his public presentation of self. Later, when conducting research in Chennai among the Beeri Chettiar, the connection between temples and institutional big‑men was once again reconfirmed. I found that the Beeri Chettiars too had once controlled a network of several denominational Shaivite temples and a caste monastery in George Town, and that the head trustee of the Kandasami temple, P. Balasubramaniyam, was the community’s preeminent big-man during the 1980s and 1990s. Nevertheless, it was also evident that the majority of temples that the Beeri Chettiars had once controlled had since been taken over by the HRCE Department and no longer played a role in community affairs. These HRCEcontrolled temples, lacking big-man control, stirred little of  the excitement and passion that Bala’s management of the Kandasami temple did. This decline in temple vitality paralleled the demise of caste headmen who once controlled these temples, and the decline in the caste’s local population, as Beeri Chettiars moved from business to government employment and other non-traditional professions. During Balasubramaniyam’s management the Kandasami temple became a place where things happened in the public eye. In addition to spectacular temple processions in which Bala walked at the front, he also used the temple to stage a number of highly popular events. In 1969 he founded the Kandan Arts Academy to promote music, drama and literature. He told me that he created the association in part to bring the Beeri Chettiar caste together again, uniting them through patronage. In the mid-1990s the Academy had 400 members drawn mostly from the Park Town Beeri Chettiar business community, in the neighbourhood of the Kandasami temple. Under Bala’s leadership, Kandan Arts annually sponsored two major public concerts, held each year on a temporary platform set up on the street in front of the Kandasami temple under a huge thatch roof ( pandal ). The concerts were grand events that attracted well known music stars, and they were immensely popular. Spectators packed the street. During the

60  Mattison Mines

concerts Bala played a prominent public role and was seen as its organising patron. Leaders like to sponsor dramatic events to attract public interest. In 1974 the Arts Academy gave two cash prizes of 51,000 rupees each to artists it judged up-and-coming. Artists like the two awardees, Bala said, are low-income people, so the prizes funded by money raised by the Academy’s performances gave them a boost and attracted a great deal of public attention as gestures of grand patronage. Since 1980, while he acted as dharmakartha, Bala also invested over 2 crore (20 million) rupees of Kandasami temple funds in projects designed to enhance the prestige of the temple and to benefit the renown of the Beeri Chettiars in Chennai. One of his projects was the construction of a gold-plated temple car (teer) covered with seven kilos of gold, which worshippers could hire as part of their private worship (arccanai) of Kandasami. The temple car, a towering chariot, carries the deity’s processional image and is pulled through the streets near the temple as an act of devotion. The golden car, one of only two of the kind in south India, broadcasts the temple’s wealth and grandeur and adds to Bala’s reputation as temple trustee. Then in 1989, Bala organised and managed the temple’s kumbaabisheekam ceremony, the elaborate and expensive re-consecration and purification ritual following temple renovation, the performance of which most temple trustees only dream of. Bala had also renewed the practice of singling out prominent members of the community by couriering to them personal invitations to temple festivals. Using temple funds, he also established a number of satellite big-man institutions designed to benefit the caste and extend its collective reputation. With the donations collected at the temple he founded and built two schools and was planning a polytechnical college, which has been founded since. One of the schools is a high school open to all communities, which enrols 1000 students and which is built on land donated by a deceased affine of Bala’s. The other is the Hindu High School, located in Park Town. Bala also founded a free library, featuring magazines, dailies and more than 4000 books concerned with literature, religion and advice on starting small industries. He founded two medical dispensaries. And in the 1980s the temple gave 500 rupees a month to poor Beeri Chettiars to assist them in starting a business, finding work or meeting marriage expenses. In addition to these temple-sponsored benefactions, as a businessman, Bala also extended his reputation as businessman by

The political economy of patronage 61

creating a number of private companies known collectively as the P.B. Group. Bala, then, is a new kind of big-man quite different from the Beeri Chettiar headmen, the caste’s ejamaanar, who had been preeminent leaders up to the late 19th century and in some cases until the early part of the 20th century. This is not to say that the old leading merchants were simply forgotten. When I had begun my research among the Beeri Chettiar, I met with a group of community leaders, including Bala at the Kandasami temple, and they excitedly recounted the names and stories of wealthy, leading Beeri Chettiars who were still remembered for the major and sometimes dramatic contributions they made to temples in their localities. Listening to them, it was instantly clear that the relationship between big-men and temples was central to the Beeri Chettiar’s sense of their caste’s history and to the role temples played in local organisation and public displays of eminence, especially during festivals when the temple gods were taken in processions through the streets (Mines 1994; 2006). I also found that the Achari caste in George Town had its own denominational temple, dedicated to Kaligambal, a pacific form of Parvathi, the Shaivite goddess. In short, in cities—Chennai, Salem, Trichengode—all the castes which historically belonged to the left-hand moiety I had studied controlled their own denominational Shaivite temples and used them as institutions of devotion and public display of leadership, and as stores of resources belonging to the community and used for charity. Given this, my hypothesis was that at least in George Town the local organisation of temples expressed the old left-hand/ right-hand moiety division, each section controlling its own temples and processional streets, and that, if I were to look, I would find that the local organisation of the right-hand section of George Town, Pedda Naickenpet, would be the mirror image of the left-hand division. Because I knew that the Arya Vysya were the competing right-hand merchants, I merely had to ask to learn about their denominational temples. When told that the Arya Vysya’s temple was the SKPD, my hypothesis seemed confirmed. True, it was only one temple, but by the 1980s the Beeri Chettiar controlled only one temple in what had once been their ‘galactic polity’ of caste temples (Tambiah 1976; Mines 1994, 68ff). This galactic polity consisted of several temples and institutions controlled by the heads of the caste endogamous subgroups known as gumbu. Headmen and subgroup endogamy no

62  Mattison Mines

longer exist. I supposed therefore that among Arya Vysyas the number of temples which they controlled had also declined. I was, however, wrong. When I first visited the SKPD, I was surprised to see that it was smaller and less imposing than the Beeri Chettiar’s Kandasami temple and some other temples in Muthialpet. This was striking as in the 19th and early-20th centuries the Arya Vysya included among them some of the wealthiest merchant-princes in Chennai. So where were their temples? It was not until I conducted research among Arya Vysya that I realised that the SKPD was a relatively recent, 19th-century innovation built adjacently to the Komati (Arya Vysya) Charity Garden only in 1803. Ravanappa and his descendants—of whom Sri Chetty (d. 1912) was one—used the temple much as the Beeri Chettiar big-men did, as an institution for the public presentation of their preeminence and for charitable undertakings. But many influential Arya Vysya explained that the temple was an innovation of the Collahvar family, who were Shaivites. 70% of the Arya Vysya are Vaishnavites and for sectarian reasons avoided SKPD temple events until recently. The temple, then, has had limited devotional appeal for the majority of the Arya Vysya, and some of my informants indeed explained that the Arya Vysya customarily shunned the management of temples. It was not their way. They held rights to sponsor festivals in the temples and provided funds for temple operations and renovations, but, with the exception of the SKPD, they altogether avoided managing temples. They built rest houses for Arya Vysya and Brahmin pilgrims at religious sites, but they did not build temples. As one informant explained, the danger of being a temple trustee is the inability of a trustee to avoid committing sins. My informant quietly explained that the Arya Vysya believed that childlessness indicated past sins committed at some stage in the life cycle. All Arya Vysya know that in the past there have been a number of wealthy Arya Vysya merchants who had no children or perhaps only a daughter. Without children, they passed on their legacy through charities they founded to memorialise themselves. Before building the SKPD, the caste’s principal charity held in common was a charity garden, the komatla tottam or Komati’s Garden, which the Suncuvar headman managed on behalf of the caste (Mines 2006). Aside from the SKPD, in the 20th and 21st centuries the Arya Vysya asserted their preeminence through feats of magnificent generosity and charity of individuals—‘the merchant-princes’—who with the fortunes they amassed established charities defined in law as supports of religious and public good. This ‘private’ charity given

The political economy of patronage 63

in the name of a merchant or his wife, nonetheless, took the form of public presentations that focused civic attention on the nature of the big-man’s preeminence. One such feature of eminence is the government’s public recognition of the great man. Recall how in exchange for granting in perpetuity to the East India Company the gift of a substantial charitable endowment, Collah Ravanappa negotiated to receive Company honours—the palanquin, gold medallion and silver-plated staffs. The golden age of individual charity among Chennai’s Arya Vysya lasted between 1915 and 1950 and was a response to British policy. In 1915 the British no longer gave palanquins but instead awarded a graded series of titles in recognition of great public benefactions—Rao Sahib, Rao Bahadur and Diwan Bahadur— which they bestowed on several of the most munificent merchantprinces. A few also had the distinction of being appointed for a year to the honorary position of Sheriff of Madras. These great men continued the tradition of using their commutes in the public streets to dramatise their preeminence and princely wealth. When automobiles were introduced, they were among the first to own them. Diwan Bahadoor Thathikonda Namberumal Chetty, the famous wealthy constructor of government buildings (including the High Court) owned one of the first cars in Madras, a Dideon, which bears the license plate ‘MS 3’, marking it as the third car registered in the city (and the first to an Indian). The Vupputur family, one of the wealthiest Arya Vysya families, owned a Rolls Royce. (I am told a Daimler was considered almost as good.) Yet constituents always describe the demeanour of wealthy Arya Vysysa big-men as unassuming, even as they earned great acclaim in life through munificence and structured their giving to secure their remembrance after death. Consider this school souvenir account of the founder of what remains the wealthiest Arya Vysya charity in the city (Centenary Celebration Souvenir): Rao Bahadur Calavala Cunnan Chetty and his brother Rao Bahadur Calavala Ramanujam Chetty earned their wealth as partners of King and Co., an enterprise dealing in high class timber from Burma and costly spirits like Scotch Whiskey from Great Britain. There was no black market then and the two brothers were content with small profit earned by honest endeavour. Their wealth increased steadily and surely and brought a wider business. Their policy of honest truthful and contented salesmanship increased their wealth largely and they came to be called ‘Timber Kings’ and Merchant princes. They were

64  Mattison Mines

held in high esteem and great respect. Both the brothers were unassuming in manner, despite their huge wealth.

The Souvenir continues: Ordinarily, a rich man can help the poor by providing educational and medical facilities. Mr. Chetty Garu [Calavala Cunnan Chetty] lavished his wealth magnanimously on these two items of charity, considering it [his wealth] as merely ephemeral, that which will not accompany us after death and which is no more than a trifle. Before his death, he was spending more than Rs. 50,000 annually on the charitable endowments instituted by him in the city of Madras and other places…. It is remarkable that he has been endeavouring for the betterment of Public life by his active participation in a big way. He is a member of the South Indian Chamber of Commerce. He is a Director of the Indian Bank. He is President of the Pappu Chetty Garu Charities [an Arya Vysya family charity]. He is a trustee of the Kanyaka Parameswari Devastanam….

Calavala Cunna Chetty’s public charity was extensive and the list of his institutions long and diverse: schools, his membership in the Society for the Prevention of Cruelty to Animals and his presidency of the Arya Vysya Sangam numbered among them. He made a large monetary gift to the Madras Social Service League and served as its Vice President; he started and supported an Ayurvedic hospital in Triplicane and made arrangements for the free distribution of milk and for the feeding of a large number of poor children. Every Sunday in Krishnappa Naicken Street, in front of his father’s house, he fed ‘thousands of poor people’, and built a rest house at Egmore near the Rail Station ‘at a cost of Rs. 72,000/- for the benefit of the passengers alighting at Egmore and struggling for want of accommodation’. He was also a member of the City Corporation Council. Childless, he wrote in his will: ‘I make this Last Will and Testament because I am anxious that the bulk of my properties should be utilised for charities and that my name should be perpetuated not by descendants but by schemes of public benefaction which I am anxious to organise and provide for’. Perhaps in a will such as his one cannot find a clearer description of the big-man as a paternalistic figure, who cares for his society as he might do for his own children.3 3 Anastasia Piliavsky suggested this to me.

The political economy of patronage 65

Conclusion The pace of change in Chennai since the 1990s has been dizzying, so rapid that even a few years make a big difference. In conclusion, I will limit myself to a few observations about what is happening to the role of trust as a force in social relations and in the way that big‑men operate. The need to be known and the absence of anonymity are requisite to trust and to how the big-men described in this chapter establish their reputations as patrons of their constituents. Even in the 1990s I could see that the need to be known and seen—the opposite of anonymity— was declining in some contexts, while still thriving in others. The move from private houses in established neighbourhoods to the new suburbs and to high-rise apartments has put many residents ‘out of sight’ and has increased anonymity. Yet in the central parts of Chennai there is still little anonymity. In mid-1990s I hired for the first time a research assistant who was native to the city, a young Telugu woman I call Priyanka, whose family had long resided in Chennai. While conducting research among Arya Vysyas, I discovered that the Telugu community kept track of one another. Whenever Priyanka and I interviewed a member of the Arya Vysya community, my informants, who were much older than she, asked her about herself. She was not an Arya Vysya, but she explained who her family was and questions often took wing, locating her parents, aunts and uncles. Some went to school with her mother or grandmother, and they knew her uncles, father or his family. There was also what Priyanka called her parents’ social circle. Everyone in this middle-class, university educated, Teluguspeaking set knew what was going on with the others. And within 5 minutes Priyanka was conceptually located and identified. In other contexts, older men known to her family, but not to me, would see us in George Town and come up to ask what she was doing in the petta. This is what people do in a trust-based society when they observe someone out of context: they try to establish the social context that occasions the chance encounter. And many eyes make for strong social control. Through all this I was thinking ‘Here I am in a city of millions, teaming with people, and yet she is a “known person”, identifiable, recognised, located among her kin and social group’. While anonymity is on the increase in the city, Telugu speakers, and not the Arya Vysya alone, continue to keep track of one another. This is true also among the Tamil-speaking Beeri Chettiars and among the Smartha and Aiyangar Brahmins. People know what is going on in their neighbourhoods and among their set. They keep tabs.

66  Mattison Mines

What, then, has happened to patronage by institutional big-men and the public presentation of preeminence? Today it is more muted than it once was. Wealthy merchants are more cautious about drawing attention to their wealth. Gangster, or goonda, politics are a factor. But so are the income tax authorities. As one informant told me, ‘If you want a visit from the Income Tax people, buy a Mercedes. They will be at your door in three days!’ These days, ostentatious public display when traveling in the streets is more the politician’s style than the wealthy merchant’s. In Chennai, nevertheless, there are many contexts in which the patronage of big-men carries on. Temples are one of them. The support of the arts is another. Medical dispensaries, hospitals, libraries, hostels, scholarship funds, marriage halls, the funding and building of educational institutions and increasingly the development of universities are all arenas in which institutional big-men play a role. Among the Arya Vysya, wealthy merchants are still the managing trustees of the big charities which the merchant-princes established in 1915–1950. These trusts fund all of the above endeavours, and they are also behind much of the commercial building in the city. Reputation is important to a businessman and underwrites his ability to obtain loans and finance. Businessmen explain that reputations are hard to acquire and easy to lose. There are some who are well known and trusted. On the other hand, as one merchant said, ‘trust is important, but in the end, whom can you trust?’ Nonetheless, patronage remains the centre of social relations in Chennai and among each of the communities described in this chapter. It is the manner in which communities like the Arya Vysya explain their history. Patrons have good connections and draw their constituents around them with the benefits provided by the institutions they manage. Among their institutions are the charities created by the now deceased merchantprinces. And these patrons are remembered. Their names and deeds appear in newspaper articles. You can also look them up on the Web.

The temporal and the spiritual, and the so-called patron–client relation in the governance of Inner Asia and Tibet


D. Seyfort Ruegg


ibet and the adjacent Inner Asian and Himalayan areas present remarkable and very telling examples of what is commonly referred to as ‘patronage’. The same areas also offer instances of a relationship between the spiritual and the temporal, or the religious and the lay, where the spiritual order represented by a religious preceptor–officiant is paramount and cannot be adequately described as the lay ruler’s ‘client’. In Tibet, the spiritual and the temporal orders have been represented by a donee (Tibetan: mchod gnas/yon gnas), often loosely designated as ‘priest’, and a donor (Tibetan: sbyin bdag/yon bdag), often referred to as ‘patron’.1 This diarchy (lugs gñis, tshul gñis, khrims gñis), pivotal to the structure of Tibetan society, is expressed in Tibetan with the compounds mchod yon and yon mchod. In a frequently occurring metaphor, the two—the (lay) donor/ruler and the (religious) donee/ preceptor–officiant—are said to be twinned, like the sun and the moon which illuminate the two halves of the nychthemeron. This metaphor refers to the fundamental Tibetan Buddhist concept of the conjoining, or synergy, of dharma and rule (chos srid zun˙ ’brel). The two concepts are attested from an early period in Tibet and Inner Asia, and they can be partly traced back to India, the source of so much in Tibetan civilisation. The Tibetan terms mchod gnas and yon gnas translate and correspond to the Sanskrit¯ıya or (dakkhin.eyya in Pali), or ‘worthy of the ritual gift/fee’. The Tibetan yon bdag is an honorific form of sbyin bdag (d¯anapati in Sanskrit), 1 For a fuller discussion of the preceptor–donee and ruler–donor relationship between the spiritual and the temporal orders, see Ruegg (1995). The meanings of the principal terms used in this chapter can also be found in Ruegg (1997).

68  D. Seyfort Ruegg

the ‘donor’ in the traditional structure of South Asian Buddhist society, where the mendicant monk, the bhiks.u, and the community of such monks, the sam . gha, receive donations or alms from the householders (gr.hapati, gr.hin; khyim bdag, khyim pa in Tibetan) which act as donors (da¯napati, or sbyin bdag in Tibetan). The relationship between the spiritual and lay orders did not assign superiority over religious preceptor–donees to lay donors, as the word ‘patron’ implies. Sometimes the two were held to be equal but different, occupying distinct domains. At times, according to the particular prevailing circumstances, one or the other was regarded as preeminent. Sometimes the two orders were concentrated in a single person and sometimes they were divided between more than one. Yet, the two orders were invariably regarded as complementary, with the spiritual normally treated as paramount. In Inner Asia one version of this religiopolitical relationship was formalised (if not yet fully theorised) in the 12th and 13th centuries, when the great abbot-hierarch of the Tibetan monastic state of Sa-skya, Sa-skya pan.d. i-ta (Sa-pan.) Kun-dga’ rgyal-mtshan (1182–1251), and his nephew, ’Phags-pa Blo-gros rgyal-mtshan dpal-bzan.-po (1235–1280), established an understanding—a kind of concordat— first with the Mongol prince Göden (Go-dan) and then with the Sino-Mongol emperor Qubilai Qan (r. 1260–1294). These princes were Chinggisids, or descendants of the great Mongol emperor Chinggis Qan, who at the time were the paramount rulers of Inner Asia. The Sa-skya hierarchs—who in the persons of S­a-skya pan.d.i-ta and his nephew ’Phags-pa entered into the mchod yon relationship with Göden and Qubilai—were themselves also rulers of a large and important polity in Tibet, suggesting that the spiritual order was not always fully dissociated from the temporal and the political.2 Examples of the mchod yon relation can be traced further back in Inner Asia, notably to the relationship between a hierarch and ruler of the Tangut state. Through time and space, variations in the interrelationship between the two orders have arisen in their practical implementation. Problems emerged from the fact that in Tibet this relationship was never clearly defined constitutionally, as it were. Some thinking on the subject nevertheless took place in Mongolia, the central military and 2 The prevailing situation has generated this and other sociopolitical relations in Tibet, which have been described as ‘feudal’, a term that does not well fit the historical and cultural context.

The so-called patron-client relation in Central Asia and Tibet 69

political power of the time and the main protagonist in the shaping of the roles of the preceptor–officiant monk as donee and the ruler as donor. For, as I noted, the mchod gnas/yon bdag relation appears to have been substantially developed under the Sa-skya hierarchs and the Mongol rulers during the Mongol paramountcy in Inner Asia and the Sa-skya hegemony in Tibet (although it had significant precedents and precursors). The historical relation between the Sino-Mongolian emperor Qubilai and the hierarch (or prince abbot) Sa-skya rulers became the model for the regional relationship between the temporal and the spiritual orders. Notwithstanding above mentioned variations, we can speak of the spiritual order as generally paramount in the governance of Tibet, however often things did not work out this way in practice. An obstacle in understanding this arrangement arises from attempts to conceptualise and translate it into Western terms. For a time, the relationship between the spiritual and the temporal orders was described as the ‘patron–priest relation’. This inadequate, and potentially misleading, characterisation generated much confusion. The mchod gnas—a religious person, often an ordained monk—is scarcely a ‘priest’. And in Buddhist polity and society the yon bdag/sbyin bdag is not a ‘patron’ in the sense of being superordinate to the mchod gnas, or the monk. In relation to the spiritual preceptor–officiant monk, the bdag/sbyin bdag is both a benefactor and a beneficiary. That is, in the ideal, the relation between donor and donee in Buddhist polity and society is shaped by one of Buddhism’s (and indeed Indian civilisation’s) fundamental concepts: the d¯ana or ‘giving/ gift, liberality, alms’. In Buddhist practice and thought the donee is the ‘merit field’ (pun.yaks. etra) who makes it possible for the donor to gain ‘merit’ (bsod nams =  pun.ya) or ‘good’ (dge ba = kus´ala); thus, in effect, the donor or d¯anapati depends on the donee, who teaches him or her dharma (or ‘the gift of dharma’, dharmad¯ana) in return for food and other requisites of material and religious life (¯¯ana). The question is: who in this binary relation benefits and who ‘patronises’ whom? And is the relationship symmetrical or asymmetrical? To ignore or from the outset set aside fundamental concepts which underpin Tibetan society in favour of Western terms and concepts is to fail to grasp the meaning and shape of the mchod yon relationship. To answer this question, we need to distinguish carefully between indigenous (‘emic’) categories and the comparative (‘etic’) categories of the modern West.

70  D. Seyfort Ruegg

The Tibetan idea of governance divides polity into two parallel branches: the clerical (ser po ‘yellow [clad]’), known as the rtse skor, and the lay (skya bo, ‘white [clad]’), known as the šod skor. In theory, this particular form of governmental diarchy—where spiritual and temporal powers are distributed between two persons or groups— informed the Tibetan polity. This articulation of diarchy was not, however, the only kind known in Tibet. In Tibetan, the phrase bla dpon has sometimes been employed to designate a lama who was also the ruler of a territory.3 The form of governance represented over the recent centuries by the institution of the Dalai Lama (t¯a la’i bla ma in Tibetan) is perhaps the best-known articulation of this general formula. As a manifestation of the bodhisattva of compassion, Avalokites´vara—regarded as both the spiritual ‘lord of the world (of living beings)’, Lokes´vara, and the ‘ethnarch’—the station of the Dalai Lama combines in a single person the spiritual and temporal orders. This arrangement was institutionalised in the 17th century, during the reign of the Qošot Mongol ruler Gušri Qan, who conquered Tibet but then passed on its rulership to the 5th Dalai Lama, the abbot of the dGa’-ldan pho-bran˙ in the ’Bras-spun˙ s monastic seminary, the head of what is known as the dGa’-ldan pho-bran˙ government in Tibet. A further complication was introduced by the fact that the 5th Dalai Lama appointed a minister (sde pa, sde srid) to perform the temporal function, even though, as a manifestation of Avalokites´vara–Lokes´vara, his person already combined the spiritual and temporal powers. The arrangement formalised in the institution of the Dalai Lama was not a ‘theocracy’, but a ‘hierocracy’, or more precisely a ‘bodhisattvacracy’, or government by a benevolent and compassionate bodhisattva. Beneath the Dalai Lama, the bodhisattva-ruler who combined in his person both orders and functions, the government of Tibet was diarchic: it comprised the clerical branch, the rtse skor, and the šod  skor branch of lay officials. Other Tibetan hierarchs have also been religious figures who ruled over territories and estates, large and small. Pan.-chen Rin-po-che was the ruler of a large part of the gTsan˙ province in southern Tibet, and the Karma-pa Lama ruled over an extensive territory. A  comparable system of governance was found in Bhutan under the Žabs-drun˙ Rin-po-ches. In Mongolia, the Tibetan-born Jebtsundampa Qutuytu VIII (Tibetan: rJe-btsun 3 Because this term has also been used in other contexts, it is polyvalent and hence ambiguous.

The so-called patron-client relation in Central Asia and Tibet 71

d­am-pa, known as Boghda Gegen), the spiritual leader of the Khalkha Mongols based in Urga, was in 1911 named the head of the newly founded independent Mongolia. Thus, through the course of history Tibetan ideas of governance have been fluid and highly dynamic—so to speak, kaleidoscopic—and the actual implementation of these ideas in practice has ranged. A more or less close relationship between the religious/spiritual and the lay/temporal orders has not been an exclusive feature of South and Inner Asia, whatever characteristic forms it might have taken there. Until modern times, Europe also had princely bishops and princely abbots, men often born into noble and powerful families. They were regarded as vassals of temporal power, as in the case of the Germanic Holy Roman Empire, in which the Pope’s supremacy over the empire was a burning issue. In 1077 the submission of German emperor Henry IV to Pope Gregory VII at Canossa had great symbolic significance. But it did not permanently settle the ‘quarrel over investiture’, the highly contentious question of whether a lay emperor had the authority to invest bishops with their crosier and ring of office, and thus to transform a bishop into a kind of feudal lord in vassalage to the emperor and symbolically suborn religious authority to temporal power. Attempts to resolve the quarrel over investiture were made at the Diet of Worms (1124) and the Lateran Council (1125), at which lay investiture was abolished. But a compromise between the two orders was reached by regaling the investiture of a bishop with a sceptre as a temporal symbol and prerogative. In mediaeval Europe, the opposition between the papal party, the Guelphs (the Germanic princely family of the Welfen), and the imperial party, the Ghibellines (the party of the Hohenstaufens of Waiblingen), was an important factor in the relation between religion and politics.4 The Bishop of Rome—the Pope— was an independent and supreme religious figure, the Holy Father; but he was also the ruler of erstwhile Papal States, and still today he is the ruler of Vatican City and thus a head of state.5

4 This opposition is reflected in Dante’s writings, including La divina commedia. 5 Perhaps unexpectedly, the modern British monarch—descended from the elector- kings of Hanover who were historically linked to the papal party of Welfs—is not only the Defender of the Faith but also the supreme governor of the (Anglican) church; thus a curious reversal of roles from the ones played by the Welf/Guelph ancestors, whose party upheld papal supremacy, has taken place.

72  D. Seyfort Ruegg

As late as the early 19th century, the French social philosopher Hugues-Félicité Robert de Lamennais (1782–1854) began by defending ultramontanism—the paramountcy of Roman papacy over French Gallicanism—only to reverse his position later, advocating a ‘liberal’ form of Christianity in which the state was to be entirely separate from the church. The relation between emperors and patriarchs—the issue of so-called caesaropapism—was also significant in Eastern or Orthodox Christendom (Byzantium, Russia and so forth). The relation here between the spiritual and the temporal was very delicate, as the Orthodox churches of Eastern Christendom were autocephalous (even if the Patriarch of Constantinople was regarded as the first among the Bishops). In the Shi’i Islam of contemporary Iran we also find power divided between the Supreme Guide (a jurist–cleric) and a President, who may be a layman and who is constitutionally subordinate to the Supreme Guide. In Ruhollah’s Khomeyni’s terms, this idea is embodied in the concept of velayat-e-faqih, or government of the learned, which gives primacy to the clerical branch. Prima facie, this concept appears akin to the Tibetan donor–preceptor relation; but the resemblance is formal rather than real, and in practice there is little in common between the Supreme Guide as a jurist–cleric and the Dalai Lama as a manifestation of the bodhisattva of compassion. In the Sunni world, the khalif—a temporal emperor governing a more or less ideal caliphate—also had a religious function as ruler over the umma, or the universal community of Muslims.6 Just as in Tibet, the religious and temporal functions have sometimes been distributed between different persons and at other times they have been in one or another way vested in a single person, when the question of patronage may indeed arise. Indian antecedents to the diarchic mchod yon/yon mchod relation present us with considerable complexity. In the Indian tradition, princes, kings and Righteous Rulers (cakravartins) have frequently been associated with a spiritual adviser, the kalya ¯ n.amitra or (r¯aja- )guru. The great Mahayana Buddhist philosopher Nagarjuna, to whom hortatory texts and letters of advice to the ruler are ascribed, is but one renowned forerunner from the early Common Era. In legends, 6 The Islamic caliphate was seen as possessing a universal religious dimension; hence, for example, the caliphate movement in India after the dissolution of the Ottoman empire/caliphate following World War I. In some quarters, the idea of an Islamic caliphate lives until this day.

The so-called patron-client relation in Central Asia and Tibet 73

the bodhisattva Maitreya is also associated with the cakravartin S´an.kha. In ancient India, in the context of Vedic sacrifice, a distinction was made between the sacrificer (yajam¯ana) (the organiser, sponsor and beneficiary of sacrifice) and the ritual priest (rtvij) whom the sacrificer employed to officiate at the sacrifice and to whom he offered a ritual fee known as ¯ . Since the yajam¯ana was the patron of the sacrifice that he offered, and also the patron of the priests who officiated at it, the word ‘patron’ here is polyvalent. The Indian jajm¯anı¯ system (see Introduction and Chapter 6) of so-called ‘feudal’ village organisation has been regarded as a variant of this ancient form. As I already noted, the Tibetan term for the spiritual preceptor– officiant, mchod gnas/yon gnas, corresponds to the Sanskrit daks.inı ‘worthy of¯ ’ (the Tibetan word ‘yon’ is . ¯ya/daks.ineya, . used to translate the Sanskrit¯ ). In the Indian and Buddhist worlds, the oppositions between the spiritual and the temporal, the sacred and the profane, or the lay and the religious sometimes break down, a fact made plain by the virtually univocal untranslatability of the Sanskrit word dharma, which may be glossed as ‘religious or worldly duty’, ‘law’, ‘morality’, and so on. The third goal of human action (purus. a ¯ rtha)— or dharma—may be either worldly or supramundane. Only in the fourth of these goals (moks. a or ‘liberation’) do we find a specifically transmundane, and explicitly religious (or spiritualphilosophical), aim. Moreover, in the Hindu staging of human life (¯as´rama) one and the same (male) individual theoretically passes through the life stages of a (celibate) religious student (brahmac¯arin), a (married) lay householder (gr.hapati) and a forest recluse (v¯anaprastha, still accompanied by his wife if still living), only sometimes to become, finally, a (celibate) religious renunciant (sam . ny¯asin). In the case of the Indian var na . (‘class, station, rank, category’) taxonomy, there is no exclusively ‘spiritual’ class, even if the first (the Br¯ahmana, who is a priest representing the god Brahman) may include this, in contrast to the second (the Ks.atriya or the warrior/ ruler class representing the ks.atra), which does not do so, at least explicitly, even though the warrior and the king are expected to conduct themselves in accordance with (their) dharma. Here again, then, the familiar opposition between sacred and profane falls apart.7 7 The question of divine kingship in India is too vast to be considered here. But see Chapter 12 in this volume on the implications of this idea for electoral politics today.

74  D. Seyfort Ruegg

In India, a partial parallel to the Tibetan preceptor–officiant monk is provided by the S´am . kar¯ac¯aryas, the heads of the four great ´ Saiva monasteries of India in the four cardinal directions, beginning with the mat.ha at S´ringeri. The S´am . kar¯ac¯arya of K¯añc¯ı is another such figure. The mediaeval Indian kingdom of Vijayanagara had similar arrangements, as did the Hindu kingdom of Nepal and the Hindu mah¯ar¯ajas of India with their purohitas and spiritual preceptors. Buddhism in India and Tibet may sometimes contrast the spiritual and the temporal in terms of the categories of the worldly/mundane (laukika = ’jig rten pa) and the transmundane (lokottara = ’jig rten las ’das pa); the current opposition between sacred and profane holds only a limited relevance to understanding these two levels. But the pairings of laukika and lokottara and of donor and donee are not truly isomorphic. Notwithstanding points of contact, the two sets of pairs do not mirror each other and they relate to distinct concerns. Thus, the level of laukika, which embraces godlings and celestials, does not completely exclude the category of the religious, and the spiritual/religious order may be concerned not only with the transmundane level but also occasionally with the mundane. This is, for instance, the case with the preceptor–officiant as a manifestation of the all-compassionate bodhisattva Avalokites´vara–Lokes´vara. The year 2011 marked a new chapter in the relation between spiritual and temporal orders for Tibetans and Tibetan polity. The distribution and exercise of the spiritual and the temporal functions once again became a central issue. As I already noted, in Tibet at least since the 17th century the Dalai Lamas combined in a single person the spiritual and the temporal powers. At its summit, Tibetan  polity was ‘bodhisattva-cratic’, headed by the Dalai Lama as a bodhisattva who ruled. Below the Dalai Lama, the government of Tibet was bifurcated into the religious–clerical branch and the temporal–lay branch. But in the spring of 2011, the 14th Dalai Lama, Blo-bzan˙ bsTan-’dzin rgya-mtsho, publicly renounced his temporal function, to be vested henceforth in an elected Prime Minister (bka’ blon khri pa, currently Blo-bzan˙ San˙s-rgyas). By stating his intent to continue to act as the spiritual guide and preceptor of the Tibetan people, the present Dalai Lama reverted to a version of diarchy (lugs gñis) in which spiritual authority and temporal power were vested in two separate persons. In other words, he returned, at least in part, to the old form of the bicephalic mchod  yon model of governance

The so-called patron-client relation in Central Asia and Tibet 75

in which the spiritual power remains paramount. Given the fact that the spiritual reigns supreme, the diarchy thus envisaged is not entirely symmetrical. In the absence of an independent (or even truly autonomous) Tibetan nation, the temporal power now vested in the Prime Minister concerns primarily Tibetans in the diaspora. The mchod yon relationship, which formerly operated between a religious preceptor–officiant and a lay donor, thus takes on a new form.8 The Tibetans have thus had to give up monocephalic governance in which the spiritual and the temporal functions are vested in the person of the Dalai Lama and to turn to the disaggregated preceptor–ruler paradigm—an updated mchod yon model where the two powers are divided between (at least) two distinct persons and institutions. There are significant differences between the European prince– bishop or prince–abbot and the Tibetan mchod yon models. In the Germanic Holy Roman Empire the ecclesiast was an imperial elector (a Kurfürst), just like his lay counterparts. A prince bishop’s function as an imperial elector depended less on his spiritual function as bishop than on his temporal role as the ruler of a state within the empire.9 By contrast, where it came to the Pope—both the Holy Father and the Bishop of Rome (and the temporal ruler of the former Papal States and, in recent times, of the Vatican City)—his temporal role depended on his religious function, which is paramount.10 In the Tibetan Buddhist model, the Dalai Lama was also authorised to rule by virtue of his being the emanation of Avalokites´vara– Lokes´vara. This seems to have already been the case when the young descendant of Altan Qan, the conquering ruler of the Tümed Mongols, was chosen to be the 4th Dalai Lama, Yon-tan rgya-mtsho (1589–1617). This choice combined Tibetan spiritual patronage with Mongolian politics, and it was Altan Qan who had earlier conferred the Mongolian title of Dalai Lama (Tibetan: t¯a la’i bla ma, meaning ‘ocean lama’) on bSod- nams rgya-mtsho (1543–1588), Yon-tan rgya 8 By a curious coincidence, Dalai Lama XIV announced his decision to separate the spiritual and the temporal orders exactly one century after Jebtsundampa Qutuyru VIII of Urga, the religious head of Outer Mongolia’s Khalkha Mongols, was declared head of the newly independent Mongolian state. 9 Were he a cardinal, however, his function as papal elector would be derived from this ecclesiastical rank. 10 In 2012, in a move that oddly parallels the Dalai Lama’s, Pope Benedict XVI announced his renunciation of worldly duties.

76  D. Seyfort Ruegg

mtsho’s immediate predecessor. As a result, bSod-nams rgya-mtsho’s two immediate predecessors, dGe-’dun grub (1391–1475) and dGe-’dun rgya-mtsho (1475–1542), came to be retrospectively regarded as the 1st and 2nd Dalai Lamas, although they never actually bore this title during their lifetimes. The preceptor–ruler relationship was subsequently established between the 5th Dalai Lama, N˙ ag-dban˙ blo- bzan˙ rgya-mtsho (1617–1682), and Gu šri Qan, the conquering ruler of the Qošot Mongols who transferred to him the rulership of Tibet. The relationship between the two orders took on greater complexity when the 5th Dalai Lama appointed a minister or a ‘regent’ (sde pa) to supervise Tibet’s temporal affairs, thereby transforming a binary relationship into a ternary one. Later, at the time of the Manchu (Qing) dynasty, another variant of the relationship was established between the emperor (identified as bodhisattva Mañjus´ri¯ı) and the Dalai Lama (a manifestation of Avalokites´vara). On one level in Eastern and Inner Asia a relation between a religious preceptor–officiant and an imperial donor (in what was theoretically a reciprocal and symmetrical relationship as the emperor was also regarded as a bodhisattva) was established. But on another level, in Tibet the two functions were combined in the Dalai Lama’s person. In other words, within the greater East and Inner Asian domain the Dalai Lama functioned as the emperor’s spiritual preceptor, but within the Tibetan cultural world the Dalai Lama combined both the  spiritual and the temporal functions. The practical relations, and the political and military situations which it engendered were complex and many-faceted. But the relationship was not unworkable, and the arrangement lasted until much later changes transformed the empire. Over the course of time, this paradigmatic relationship was the outcome of the conquest and military domination of Tibet by the Mongols, Han Chinese and the Manchus. But to be sustainable, this order had to be expressed in culturally acceptable terms. The arrangement might be regarded as mere window dressing, as putting a good face in a politically uncomfortable situation, perhaps even as ‘saving face’. The outcome might be described as a form of ‘patronage’ of a ‘client’ (and a client-nation) by a patron (and suzerain power). Yet, it was evidently conceived, perceived and received as more than this alone. If the arrangement between the religious and the temporal was to be satisfactory in the eyes of all concerned, it needed to be conceptualised and expressed in

The so-called patron-client relation in Central Asia and Tibet 77

sociopolitical and religiopolitical terms which were meaningful and acceptable to both sides. The mchod yon/yon mchod structure was flexible enough to serve the purpose. Not only was it rooted in a set of traditional concepts, which were familiar and assimilable, it could also help to root conceptually a religiopolitical accommodation and a modus vivendi in the region. This frame could then be filled in with more or less varying content, according to the particular historical circumstances prevailing at a given time. As already noted, the governance and polity of Tibet, particularly under the Dalai Lamas in more recent centuries, can best be described as ‘hierocratic’ or bodhisattva-cratic, rather than theocratic. The nature of bodhisattva-cracy evolved over time, but the basic shape of a binary pairing of spiritual and temporal orders persevered. Much earlier, in the ancient royal or ‘imperial’ period in Tibetan history at the end of the first millennium of the Common Era (when Tibet was a major independent actor in the region), rulership came to be regarded as bodhisattva-cratic, with the three great Tibetan rulers of the time being identified as bodhisattvas. The bodhisattva-ocracy in Tibet had a temporal and a spiritual dimension. In future, however, and following the recent decision of the present Dalai Lama, this bodhisattva-ocracy will be regarded as quintessentially spiritual with no defined temporal function. Thus unpartitioned, it is to be neither counterbalanced by temporal power nor indeed ‘patronised’ by anyone. Yet while remaining uncompromised, this spiritual order might still enter into an association with (perhaps even into a pragmatic accommodation of) the temporal order. This may be possible in the frame of an Inner and East Asian commonwealth of culturally autonomous peoples or nations (if not nation-states). Yet, this outcome will most certainly require fresh and inspired thinking. Such clear and forward thinking, which is not precluded by ‘traditional thinking’, is not promoted by current constitutional and international law, which has long been congealed in nationalistic legalism. Let us now return to patronage and to patrons. In Tibet and adjacent cultural areas, these have of course existed. But ‘patronage’ does not pertinently and adequately describe, much less define, the relation between the preceptor–donee and the ruler–donor. Western conceptions of patronage and of patrons do not always and neatly fit the religious and cultural concepts that have prevailed over the centuries in Buddhism and Tibet. There, a major part of what in Western parlance is known as ‘patronage’ has been understood in terms of gifting,

78  D. Seyfort Ruegg

generosity and liberality, that is, in terms of the fundamental Buddhist ideals of d¯ana (sbyin pa) and d¯anapati (sbyin bdag, yon bdag). ‘Patron’ also often refers to a nobleman or an ecclesiastic who has the right to bestow benefice or office onto a person of subordinate rank. Sometimes this became entwined with simony and the purchase of offices. But there are other uses of the term as well. A patron of the arts can be referred to as a Maecenas, who is sometimes thought of as being financially, and perhaps also socially, above the protégé, who is not only supported materially but also subordinate and perhaps even inferior to the patron. Like the French word patron, the English ‘patron’ sometimes means ‘boss’. But the French mécénat does not carry this connotation, and when the Maecenas is the state or a cultural organisation this is also not so. The terms ‘patron’ and ‘patronage’ (along with current Western conceptualisations of the spiritual and the temporal as a contrast between the sacred and the profane) do not fit without residue the Tibetan, Inner Asian, Himalayan and South Asian concepts we seek to understand. Occasionally, our terms do not at all correspond to the conceptual grids we try to translate and the societies we hope to understand. Instead of clarifying and explaining, they risk obfuscating and confusing the issues. We have to abandon the procrustean truncation and stretching of Inner and South Asian concepts and expressions to fit current Western usages and preconceptions. Such parochialism—be it of missionary, journalistic or academic origin— can only distort our subject. Provided these caveats, we can nevertheless say that examples of what we now call ‘patrons’ and ‘patronage’ could be found in Tibet and India, in cases of religious endowments, the foundings of temples, st¯upas and monasteries and in the commissioning of images, manuscripts and printed books, which in Tibet have been more often religious than not (see Diemberger, Chapter 15). The Tibetan expression employed is usually to ‘establish’, ‘found’ or ‘offer’ and, more fully, to ‘act as donor’. What needs to be explored through further research is the distinction between ideal forms of governance and society and their application in actual practices of government and society in the region. The same is also true for the study of the practical implementation of the ideal relation between the spiritual and the temporal powers, which I have discussed. There is a need, for instance, for further study of the Tibetan mi ser, the tenants or villeins of monastic and lay estates,

The so-called patron-client relation in Central Asia and Tibet 79

described as ‘serfs’ in the Marxist literature on Tibetan ‘feudalism’. The lexical definition of mi ser is ‘subject’ (’bans), or mi dman˙s ‘common person, (common) people’, and it is opposed to the terms ‘patron’ or ‘lord’ (dpon po’i ldog phyogs). The term ‘mi ser ’ came to be used as a partial synonym of mi dman˙s, a word that seems to alternate lexically with mi man, ‘the people, or masses’. Mi dman˙s dban gtso then becomes an expression used both for ‘democracy’ and for the Marxist proletariat (i.e., the ’byor med or ‘the dispossessed or opposed’ versus the ’byor ldan or ‘bourgeoisie’). The Tibetan conceptualisation of the relationship between spiritual and temporal orders in its various guises is complex and sometimes ambivalent or polyvalent. The materials assembled in this chapter point to some important Tibetan and Buddhist ideas surrounding patronage, gifting and the situation of the donee–preceptor/officiant relative the donor–ruler, whose practice needs to be examined in much greater detail, case by case.

Remnants of patronage and the making of Tamil Valaiyar pasts


Diane Mines1


t is well known that in South India patronage is a way for persons and communities (patriarchs, politicians, political parties, village headmen, moneylenders, caste associations) to generate power and influence, in part through generous grants and displays of largesse, as Pamela Price and others have shown (1979; 1996; Chapter 9 in this volume).2 Such largesse is modelled on the ruling strategies of patronkings, who gave ‘honours’ out and down from centres of power to subjects who in turn offered devotion and labour (‘service’) both in and up to the king.3 Honours took the form of land grants, temples, rights to serve the king, privileges and titles. By bestowing honours, patrons gave something to somebody in some place, thereby marking persons, places and time in any number of ways.4 Rights to perform service were granted and passed down to descendants, and such rights 1 Data for this chapter were collected in 2009 and funded by a Senior Fellowship from the American Institute of Indian Studies, with funds from the National Endowment for the Humanities. I have changed names to ensure anonymity. 2 See also Breckenridge (1977), Appadurai (1981), Ludden (1985), Dirks (1987), Mattison Mines and Gourishankar (1990) and Diane Mines (2005). 3 The spatial metaphors I use here (out, down, in and up) express a well-known dynamic of redistributive exchange in which patron-kings, headmen and other powerful people reproduce their social position both as central actors (who receive goods, devotion, taxes, etc. going in from many people and redistribute goods back out again) and as superior givers (whose gifts establish a hierarchy where they are ranked higher [up] than recipients [down]). Such exchanges produce and reproduce their centrality of power and their top positions in social hierarchies. 4 I use the term ‘marking’ the way it has been used by others analysing South Asian exchange systems to indicate how actors move a substance (such as a gift) from themselves to someone else or someplace else in a one-way, asymmetrical transaction that leaves a physical mark on the person or place in question, much as animals ‘mark’ their territory (Marriott 1990, 20). Marriott remarks that in social contexts, marking provides ‘continuity and relative stability to social relations’, presumably as these social relations tie givers and receivers (markers and markees) in exchange (ibid., 19). In this way patrons may be said to ‘mark’ their clients.

Remnants of patronage and the making of Tamil Valaiyar pasts 81

often literally marked things as they were inscribed on stone walls and etched into copper plates. Titles awarded for heroic and unheroic service marked the recipients with names, which sometimes became personal names of grantees and their descendants, and eventually even of entire sub-castes. Patrons marked the landscape by endowing the construction of temples, aqueducts, roads, mosques, irrigation tanks and so on. They endowed poets whose literary productions live to this day and continue to enhance the reputation of long-dead patrons. In each case, a physical trace of patronage was first produced and then it endured. Such a trace or remnant is, in semeiotic terms, an ‘index’. Indices are signs that are contiguous with events that produced them and which may serve as dynamic portals to past acts of social production. These traces do not exist only for historians. They also mark places and times for ordinary people. The ordinary people I describe in this chapter live in scattered villages north of Madurai, in the south Indian state of Tamil Nadu. They call themselves Valaiyars, or more often these days (especially the young people) Muttaraiyars. According to their own histories, Valaiyars received and were marked by acts of royal patronage. The honours they received—titles, land and temples—outlived kingship, but they continue nonetheless to exist now as telling remnants of the community’s past. In this chapter I take up the remnants of patronage as they persist in the titles, stories and connections to places that Valaiyars, and the young Muttaraiyars, use to narrate their pasts and to define themselves as persons and as a community. Past acts of patronage claimed by the community remain present and signify for Valaiyars not only their pasts but also their sense of place now. Patronage is, in this manner, embedded in the Valaiyar understanding of self. I argue here that past acts of patronage generated durable signs (markings, indices) that outlived the contexts of their production to persist into the present and point back, focusing attention on previous events and on one’s relationship to them. The signs are carried forward through time in narrative, titles, names, memory and places, and they are served up again and again as historical facts. In this way, patronage is significant not only as a contemporaneous element of South Asian social structure, but also as a series of markings that accumulate, endure and point back to previous actions, persons and places (or possible actions, even imagined ones; although there is no reason to think that these kinds of actions did not in fact take place)

82  Diane Mines

to generate a community’s sense of itself. It is the actual contiguity of time mediated by elements of place and story that makes it possible for such material fragments to carry forward with them the particularity of the past. These signs and their reiteration in narrative tie contemporary Valaiyars to places and events that continue to shape their understanding of both present and past.

Ambivalent clients In the Cilapatikaram, a popular Tamil epic dated between the 3rd and 8th centuries CE, the author Ilanko describes a choice of three roads that the hero and heroine, Kovalan and Kannaki, can take through the forests and hills on their journey south to Madurai. All three paths are rough and difficult. In the heat of the summer, the land is a desert wasteland (palai) ‘causing great hardship…. It is like a great kingdom whose ruler has lost his sovereignty (Parthasarathy 1993, 111). Kovalan and Kannaki take the middle, most direct, path through this parched and unruly forest landscape. They leave the settled agricultural (and ethical) countryside and walk through ‘many forests linked by villages’ of hunting peoples. In Tamil literature, the dry forest and desert wasteland landscapes (the palai tinai, from the Sangam literature) has over centuries served as a metonym for emotions and experiences related to death, suffering, separation and loss. This landscape is also a place devoid of ethical content and of patron-kings who could offer protection. It is the landscape where the Valaiyars who appear in this chapter live and into which, as one of their elders called Vanni described, they were pushed after losing their land, time and again. The dry hilly forest between Nattam and Madurai is where they have become, in Vanni’s bitter words, ‘devotees of the mountain’. Today, if you ride a moped north from Madurai towards Nattam, you will see the mountain on the horizon. To get there, you will need to make your way through an outburst of middle-class suburbs which have sprouted around former villages, competing with the dust-stirring jumble of cars, motorcycles, lorries, minitrucks, bicycles, animals, scooters, college students hanging off tilted buses and a multitude of school buses crisscrossing the suburbs to pick up and deposit children at various English-medium schools, until you break out into the cool rice fields and the smell

Remnants of patronage and the making of Tamil Valaiyar pasts 83

of green.5 In the near distance you will see a long range of hills. The largest of these is called Alagar Hill, on whose southeastern edge resides Alagar, a form of the god Vishnu. But our route steers us towards the hill’s western side. Keep going and the rice fields become more numerous until you cross the broad Periyar irrigation canal. As soon as you cross it, the green is gone, the rice is gone and the land becomes browner, drier and full of thorns. Heat rises from the blacktop. The road empties but for a few vehicles: small trucks laden with mangoes, workers or building materials, buses running between the two towns, romantic young couples riding double on Honda motorcycles to a nearby ‘go‑cart resort’ and occasional vans carrying foreigners on their way to the yoga ashram or one of the many NGOs operating out this way. Palmyra palms dot the dry earth as you approach the hills. As you keep moving north, you pass occasional villages, mango orchards, roadside encroachments and teashops, dryland gardens, goats, goatherds, small cattle and women carrying bundles of wood. At the foot of the mountain, the dry landscape grows thick with low forest. About 20 km north of Madurai, the road cuts through a narrow pass flanked on both sides by forested hills. By the late 19th century, these hills had fallen under the control of the Indian Forest Department and were designated as ‘reserved forests’. The villages strung along the road, as it cuts through these hills, are inhabited mostly by Valaiyars who associate their lives, past and present, with the forest and who often recall their former hunting, gardening and gathering practices, many of which are no longer fully practicable or even legal. This area, these villages, is where I have been doing research since 2009. Today, people here do all sorts of work. Some gather and sell forest products: they deliver medicinal herbs to middlemen in market towns and sell firewood as cooking fuel in larger villages up and down the road. Most families herd goats and a few cattle; some still try their hand at dryland farming of millet and other grains which depend only on scant rainfall, and many keep ‘kitchen gardens’ 5 Out here, amidst cultivated and fallow fields, real estate speculators have marked out with granite stakes numerous plots for future developments; even an anthropologist, who may stop on the roadside to look at a tree, will be approached by villagers who double as real estate agents.

84  Diane Mines

around their houses. But the enforcement of conservation laws has hindered many of these activities (even dryland farming is hardly sustainable any longer because of night raids by the protected forest bison, or wild gaur) and in recent years villagers have been relying increasingly on wage labour and cash crops. Many now commute to work in the nearby fields and towns and send their children to high schools and colleges when they can. Most fields have been dedicated to a single cash crop, mangos, which need little water and which are relatively safe from bison encroachments. A regional NGO has built several mango pulp canning factories in the area that provide jobs in the summer months when the mangoes get ripe. Some families tend small stands of coconut trees or teak and a few have purchased small rice fields south of the irrigation canal. Several old women and younger men run roadside tea stalls. Often, when I ask people about their livelihoods, they turn the conversation into a familiar lament about the loss of health and vitality because of their store-bought rice-based diet, which is so much less filling than the nutritious grains they used to grow. ‘Country medicine’, which they practiced in villages, has been replaced by hospitals which sap their incomes. In the past, they farmed small crops of millet and vegetables, hunted small animals in the forest, gathered tubers in the cool season, and raised cattle. They show me their tools, some no longer used or employed only rarely: a brass flute which once called cattle back from the forest, boomerangs for hunting rabbits, rabbit snares, slingshots and long iron digging sticks. This exhibition of artefacts from former forest livelihoods led to further discussion of the past, for tools were not the only mementos of earlier days. Valaiyars also keep and bring forth deity shrines, titles, and tales for display and for storytelling. These artefacts and stories about the past were not about hardships they face now but about hardships they endured before. The trope of hardship suffused their tales and conveyed a generalised sense of loss. As we shall see, the young people tell a much more optimistic story, in which their past is as glorious as their future hopes. But let us begin with the elders and their history. Valaiyar elders’ historical tales described migration and the challenges their people faced as they searched for a safe place to live. Most of these stories turned on interactions with patrons, so that even as Valaiyars told stories about themselves, they also related the dynamics of patronage from the viewpoint of low-status

Remnants of patronage and the making of Tamil Valaiyar pasts 85

forest  dwellers. On  the one hand, the stories portray independent people who fend for themselves. On the other, they narrate a search for ties to patron-kings. In this way, Valaiyar stories express ambivalence towards sovereignty—the domain of patronage bonds— as Valaiyars are both pushed away from and pulled into relations with patrons. Valaiyars themselves are the self-appointed tropes for this ambivalence, at once in pursuit and in terror of bonds of patronage. Ambivalence towards both forests and kings in the Valaiyar stories is consistent with portrayals of Valaiyars in existing ethnohistorical sources. Colonial scholar Edgar Thurston defined Valaiyars both as ‘forest dwellers’ and as ‘torch bearers’ for a local king (1909, VII, 275); historian of Hinduism, Joanne Waghorne, described them as ‘forest people’ who also took part in the royal hunts, beating the jungle for game and supplying meat, honey and medicinal herbs to the court of Pudukkottai (1994, 166–170). Nicholas Dirks likewise remarked that in Pudukkottai while Valaiyars were ‘associated with the forest’, where they carried out their ‘hunting and gathering activities’, they also performed a few ritual services for the king (1987, 127). Dirks further argued that because they were never fully incorporated into dominant political structures, and because their control over power and territory was limited to the forest, Valaiyars were only ‘passive recipients of incorporation, unable to initiate or alter these processes’ (ibid., 273). This image of passive reception of political incorporation is countered by Valaiyars’ own narratives of decisive action, whether their rapid flight from abusive patrons or re-assertion of rights under new ones.

Fleeing and settling The first of the stories is one I heard many times. This version was told by an 80-year-old man named Alagar,6 and it relays the migration of Valaiyars from north of the banks of the Kaveri river south to Ponnamaravati, a large village on the edge of today’s Pudukkottai District, near the border of Madurai District and about 60 km from the village where Alagar lives. The Kallars (who figure prominently at the start of the story) were local chieftains and the Chettiyars (who appear at the end) were wealthy merchants and landowners. 6 I inserted a few lines in brackets to add common features of other people’s narrations, clarity and common detail.

86  Diane Mines

‘We won’t give’, said our elders, our grandfather’s greatgrandfathers. They would not give the girl to the Kallar chief. Do you know what they did then? The elders tied and locked a red dog and a black dog [in the house]. [They buried the girl in a granary pit, and left her for dead.] After that, they took the women and they took the children and they all fled. The Kallars followed, terrifyingly chasing after them. Our ancestors had to get away. They came to a river, the Kollitankaveri, and when they came, a Vanni tree bent down [to create a bridge to the other side]. They crossed the river using the tree [and as soon as they had crossed it, the tree sprang back again, leaving the Kallars on the other bank]. As they were crossing the river, a huge wind and rain squall came down on them and the jewel case they were carrying fell into the river. They had to leave it behind and continue across. It was the jewel case of the goddess, Alakanacciyammal. Later, when they went back for it, they couldn’t lift it. So, they had to stop there. After a while, they decided to go into the river and try to lift it out again. It was then that the goddesses Alagarnacciyammal and Nallammal appeared in the form of two children. Near the place where they stopped, a Chettiyar was plowing his fields. There were seven plows and seven plowmen. The children were so hungry; they picked and ate some millet from the field. They scooped the grain and devoured it, and the Chettiyar grabbed them and beat them. The plowmen beat them. They beat the children and then what happened is this: those seven plowmen who beat the children, the hungry children who had come and scooped and devoured some millet, they all went blind. They could do nothing but just sit down. The lady [goddess] had put a curse on them. Our ancestors, realising these children were goddesses, put their hands together [in worship]. And the goddess restored eyesight to the Chettiyars. After this, the Chettiyars welcomed our elders with kindness! The Chettiyars wrote it out: ‘you can have the space you need, however much you ask’. And that is where the Alakarnacciyamman temple now is, in Ponnamaravati [on the lands granted by Chettiyars.] The Chettiyars wrote it out and gave us that land.

Remnants of patronage and the making of Tamil Valaiyar pasts 87

The story follows a pattern common to Valaiyar narratives: the ancestors flee because a patron in some way fails. In this case, the patron demanded too much and became what David Gilmartin (Chapter 5) refers to as a ‘bad patron’ who abuses his power, coerces his clients and takes more than he gives. Because Valaiyars refused to give their daughter to the Kallar, they faced ruin as the patron turned into a tyrant. The girl is left for dead, the patron insulted and his people give chase to the fearful Valaiyars who are now homeless until they stumble upon the landowning Chettiyar. At first, he refuses food to the would-be clients but is forced by the Valaiyar goddesses to accept them in the end.7 In this, and many other tales of flight and resettlement, patronage is a central force behind the narrative, and the historical shifts it relays, as the Valaiyars abandon corrupt patrons in search of moral ties in a new place. This is also a morality tale about good patrons and bad patrons and what those mean for Valaiyar lives. Much of what is both good and bad about their past, comes from the goodness or badness of patrons. This tension explains the Valaiyars’ Janus-faced life: one face turned to the dry mountain forest, where relations with patrons are attenuated, and the other towards a settled life that depends on patrons. This tension defines the Valaiyars’ push-and-pull relations with patrons. The tidal movement of flight and incorporation describes an ambiguous life between movement and settlement, outside and inside, forest and kingdom. This narrative and life-trope of flight and reincorporation expresses the central ‘paradox of patronage’—the tension between moral and equitable relations between patrons and clients, and the potential of inequity and power abuse—as described by Gilmartin (Chapter 5). It further indicates a deeper moral tension between belonging and not belonging, between existential connection to natu, the settled agricultural lands under royal control, indicative 7 Thanks to Ravichandran of Madurai for pointing out the pattern of demand for patronage as a prelude to settling, present also in migration stories in his own Maravar community. He also notes that the image of the Vanni tree bending across the river is not limited to the narratives of Valaiyars, suggesting that the trope of migration may be much more widespread in South Asian narratives, however little attention it may have received from scholars. Historically, lower status communities would have migrated more frequently, something noted by David Ludden (1985, 81–82) and Saurabh Dube (1998, 30–31), both of whom discuss how low caste people, when mistreated by patrons, would migrate to unsettled territories or frontiers to start up again.

88  Diane Mines

of a stable ethical order of patronage, versus the palai or katu (dried forest) with its tangled paths of moral uncertainty in the absence of sovereignty.8

Indexical density The next story takes us further south. It repeats the movement away from bad and towards better patrons, and also reveals another narrative trope: the recurrent indexing of the story’s facticity though reference to physical remnants. As patrons come and go, the grants and honours, the titles, temples and rights they give remain as marks of past relations. Remnants of patronage have a cumulative property. They add up as they recur in different times and places, and the people articulate them again and again as they tell their stories. The following story was told by T. Alagar (Alagar is a very common name) and supplemented occasionally by his wife and nephew. I asked him to tell me how his lineage arrived in the village near Alagar Hill. He demurred, saying he ‘didn’t know’, but soon decided to tell me what he ‘had been told’. In Ponnamaravati we were the sons of seven mothers [or, cousins]. Only after leaving Ponnamaravati did we split up. Each of the seven groups took one of the seven deities for their own. They split up the goddesses [and so split up into sections, each section worshipping a different goddess, so they could tell whom they should not marry]. The people all came together to Melur. They came there to live. In those days, Melur was famous for its oil presses. When our ancestors arrived in Melur, seven of the oil presses stopped working. The village people of Melur thought that our elders’ arrival had something to do with the presses not working. They complained to the king and urged him to do something about the situation, to do something about us. In the meantime, we asked our deities: ‘did you stop the oil presses?’ Well, yes, 8 I allude here to Anand Pandian’s (2009) book, Crooked Stalks, which demonstrates how the ethical sovereignty of the rich agricultural zones (nadu) in the Tamil countryside was used as the model for South Indian social civility; the dry regions (katu), meanwhile, are still construed as uncivil zones of ‘crooked’ paths and unethical activity.

Remnants of patronage and the making of Tamil Valaiyar pasts 89

they had. When the king came, the gods started them working again! That’s how we got our Ambalam title.9 In Melur, our people were Ambalakkarars (dispute arbiters) because our seven gods made the presses run again after stopping them! Even today, in front of the temple there is an ‘Ambala Stone’ with an inscription that says that our title is ‘Ambalakkarar’.

Nephew: Our elders have a copper plate there, in Melur. Alagar: So, bravely they came like that. Wife, Chinnakka: Wherever we come, we come and sit in the shade. Alagar: Whatever tree may fall, we will go find and sit under another shade. This is how it goes, changing place again and again. Alagar and his family reiterate the pattern of abandoning failed patronage relations and seeking better ones. When one shade tree falls, they find another, ‘changing place again and again’. Hunger drove them from Ponnamaravati. Another patron failed to feed, another shade tree fell. Valaiyars arrive in Melur, but to settle there, they need a patron. The deities stop the oil presses to get the king’s attention and start them up again to force the king’s hand. The king incorporates them into the kingdom by giving them the honourable title of Ambalakkarar and sealing the deal with inscriptions on copper and stone. Every migration narrative I have collected ends with the generation of some object, fragment or artefact—a copper plate, an inscription, a land grant, a title, a temple—that marks Valaiyars’ settlement and the acquisition of proper patronage ties. In the narratives, these remnants accumulate and form something like a series of stepping stones which lead back through time. In this way, patronage generates not only a social relation but also over time a set of markers for retracing and passing on knowledge about their past. Patronage becomes historical knowledge, a way of articulating belonging, territory and self. T. Alagar and his nephew insisted that a copper plate (cempu tattaku) and the Ambala stone still existed in Melur, and they encouraged me to find them. I went to Melur but found nothing. 9 Ambalam is a term that means both ‘open space for the use of the public’ and ‘headman of a village’ (Tamil Lexicon, 95). It can also be used, as in this case, to designate an honoured position of the arbiter of disputes in a king’s court.

90  Diane Mines

I looked in all the wrong places, they said. Such copper plates do exist. Samuel Sudanandha (1996, 174–176) reports finding some which granted certain rights to Valaiyars around Melur. Objects like these copper plates are not the only durable, tangible signs of former patronage bonds. T. Alagar himself, sitting right before me and holding his grandchild gently in his lap, is such a sign, heir to the title of Ambalakkarar. He and his lineage members use this title, especially in formal contexts, when they sign official documents. His caste section is also called the Meluran, the ‘Melur section’ or the ‘people of Melur’. Like stone inscriptions and copper plates, names and titles pass down and accumulate, further marking the Valaiyars’ journey from patron to patron and from place to place. T. Alagar had not one but two titles. The other, Tirumutikantapiticca Alagar, was acquired by his family when they moved from Melur and settled in the village called Valaiyappatti (Valaiyar village) near the temple of Alagar. There, according to T. Alagar, they found employment in the fort which surrounds the temple. He explained that his people received other titles corresponding to special feats they performed in the course of their service. His own title, Tirumutikantapiticca Alagar, which means ‘Alagar who discovered the sacred hair’, was bestowed on his ancestors for the feat of unearthing the statue of the god Alagar (with his sacred head and hair), which was buried deep beneath the vines of a forest tuber (vallikilunku). Every day the vine tripped up women who carried milk down the forest path. When they tripped, the milk spilled and vanished instantly into the thirsty soil. Because of this daily nuisance, T. Alagar’s ancestor, who hunted in the area, was called over to dig in that spot. He dug deep and finally he struck the statue with his digging stick. Another title, Koli Mekki (Chicken Grazer), belongs to the descendants of those who once tended chickens in the fort. And the title Kottaippuli (Fort Tiger) was given to an ancestor who killed a tiger who had been devouring married women (they found seven gold marriage pendants in its guts). ‘Like that’, concluded Alagar, ‘every title they received from there, they brought with them here. They lived there a long time before coming here’. In my previous work (D. Mines 2005) I have used the term ‘density’ to describe how in Tamil Nadu value accrues through the visually busy displays of objects, gifts, ornaments and even people, when they gather in great numbers. Density describes the celebrated crowding of people and reproduced materials. In the

Remnants of patronage and the making of Tamil Valaiyar pasts 91

ritual contexts which I have studied, this may include the crowding of gold armbands and silver bangles on the bodies of possessed ‘god-dancers’, whose frenzied, colourful decorations reflect the power of the dancers’ connection to deities tightly packed cooking pots which women tend when they make festive boiled rice (pongal) for the goddess; the number of animals tied up and awaiting sacrifice; and the crowd of festival goers all assembled in tight quarters at a particular place and time. This density conveys the greatness (perumai) of persons and communities who are putting on the display. Bernard Bate has identified the same phenomenon in political contexts: on the walls and newspapers of Madurai city, which are plastered with a proliferation of often identical images of politicians in a manner that contributes to a production of greatness for politicians, their clients and the political workers who sponsor the posters (Bate 2000, 268–274).10 In villages strung out along the road to Nattam, indexical marks of long-gone patronage accumulate over time and sediment in stories, knowledge and on the ground as they mark places with shrines and other physical artefacts. Elders and others retell stories, rename descendants, remember and revisit the places marked by patrons’ actions and grants (which include the temple to the goddess Alakinattciyammal in Ponnamaravathi). Place names accumulate too. Remnants anchor the actions and deeds of ancestors in Kolitankaveri, Ponnamaravati, Melur, Valaiyapatti, Alagar Fort, mapping the territory across which they fled in search of new patrons. Titles, deeds and places that have accumulated over time crowd the present-day narratives, packing in with the indexical densities long stretches of experience, space and time, which the Valaiyars call their community’s history (varalaaru). Acts of patronage, in other words, structure the way Valaiyars recount and continue to know their history. They mark the people and places they have passed on the way, and they are integrally related to the Valaiyar’s sense of themselves as a people who have lived on and traversed the landscape, fled and wandered, settled, suffered and belonged in turn. Remnants of patronage continue to mark the places and contexts of their production and to assert contiguity between then and now. As such, they are not part 10 Also see the documentary, ‘Writings on the Wall’, where a character G. Mani explains the relation of these posters to his reputation and political aspirations.

92  Diane Mines

of history as a sequence of past events, but of history as one’s sentient present and one’s understanding of self. Patronage gained, lost and  regained leaves residues in names, stories, temples and people’s knowledge of the terrain. They carry these remnants with them,  even as the time of their production recedes into the distant past. Remnants are neither dead nor inert. If we treat them as indices, we can explain how they exert a lively and highly dynamic force. ‘Indexical signs’, writes Charles Peirce, ‘act as a force carrying the attention…; an index ‘refers to its object…because it is in dynamical (including spatial) connection both with the individual object, on the one hand, and with the senses or memory of the person for whom it serves as a sign, on the other hand’ (1955, 109). The person whom an index serves as a sign is ‘forced by a law of mind to think [about the connection]’ (ibid., my emphasis). The intrinsic dynamism of indices invites us to think about the relation between signs (or ‘remnants’) and their semeiotic objects, in this case past places, structures, persons and the circumstances in which the signs were produced. In this way, indexical remnants of patronage focus Valaiyars’ attention on their past. Titles, land grants, temples and other remnants dynamically index past contexts of social production and beg questions like: ‘where did this come from?’ or ‘what does this mean?’ Yet ‘indices [on their own] assert nothing’ (ibid., 111). They do not make arguments or general statements about how the world works or worked in the past. They merely point. Like index fingers, they direct people to turn their heads and notice something. No remnant draws the Valaiyars’ attention as pointedly as temple deities.

What happened here? In each place I have mentioned—Ponnamaravathi, Melur, Valaiyapatti and of course Alagar Hill—temples mark events involving patrons. The next story, just like the others, involves a search for a good patron, which culminates with the production of a shrine to a deity, whose presence on the roadside continues to index things past. This story was told, once again, by C. Alagar, who said that it happened in the time of his grandfather’s great-grandfather (pattan puttan). The characters include four Valaiyars/Muttaraiyars, a tiger, an ox, two famous deities and a Nayakkar ruler and patron, Lingamanayakkar,

Remnants of patronage and the making of Tamil Valaiyar pasts 93

whose family, it is said, once lived in the area now owned by a wealthy doctor from Madurai.11 It happened a long time ago, when Lingamanayakkar was living here. In those days, our people were just wandering around the countryside. We came here and asked: ‘can we live here alongside you?’ They said: ‘fine, stay’. And so we did. We lived together, but separately. Our people separate; their people separate. Their place was where the Doctor’s house is now. We were their servants. One day, two of our men went to graze cattle…their cattle and ours, altogether 500 head. They drove the cattle onto the mountain to graze. When they were on the mountain, a tiger attacked and killed an ox. They only noticed it missing after they returned and the Nayakkar counted the cattle. One was gone. It was one of the Nayakkar’s oxen. It was wearing a bell. So, the two men, one who was our kinsman and one our affine, went in search of it. They went onto the mountain and there they saw that a tiger had killed the ox. ‘Okay’, they thought, ‘some tiger killed it, then ran off’. They reasoned that at night the tiger would return to drink again from the watering pond at the foot of the mountain near the settlement of Nayakkars. The two men then cut some vines and set a trap to catch it. When the tiger came to that place for water, it got trapped, but it pulled and pulled until it got loose and then it lay there, waiting. Early in the morning, let’s say it was 2 am, one of the men came to check. The tiger was lying in wait. It followed the man, staring intensely in to the dark. It stared and it pounced. It pounced, caught the man and killed him. When he didn’t return, the second man went looking for him. He saw what had happened. His in-law was dead. And now, he too was caught and killed by the tiger. Right there at the watering hole, the two men had lost their lives. And the tiger escaped. 11 There were several historical Lingamanayakkar chiefs in this area. The Nayakkars, who ruled the area around Madurai for about 300 years, were deposed by the British, who nonetheless kept some Nayakkar chiefs as landlords (zamindars) to administer their fiscal affairs in the countryside.

94  Diane Mines

In the morning, one of our kinswomen and one of theirs—well, they were married to the men [who had died]; you see, the two men had traded sisters in marriage… The women went to look for them, wondering where they could have gone. And then they found them. Dead. Yes. The tiger has snapped the trap and bitten up the two men, and escaped. Both were dead. They were lying dead on the raised ground near the pond. After the women found them, they started to pray: ‘let no flies or ants bother the bodies until we can return with fire to burn them’. They took off quickly for the Alagar Temple where they could find some fire for the funeral pyre. They went into the temple and Alagar said: ‘I cannot give you fire. I only give it to men. I can’t give to women. Go and ask [the goddess] Meenaksi. She will give it to you’. And so, they left Alagar and went to Madurai to see Meenaksi and they asked her for fire. They told Meenaksi the whole story and said: ‘we need to set the fire. We’ve already stacked the wood. We need to set the fire, so please give us fire’. Meenaksi asked: ‘but how will you carry it?’ (for they had nothing with them). ‘We are innocent, chaste women. Scoop it into our lapfold. We’ll carry it there’, and so saying they spread out the ends of their saris. And it is said that after they spread their saris, Meenaksi scooped and gave them a handful of embers each. They bundled them up and tied them into their waist-fold, and returned to the bodies. They carried the embers just like that, in the folds of their saris. As soon as they returned here, they lit the pyre to burn their husbands. The two women jumped onto it, too. They jumped on, and all four burned up together. All four people burned on the pyre. After this, the ruling chief Lingamanayakkar worked long with his own two hands to make a statue, which he put into place and established as a deity. ‘You people worship this’, he said to us, and then he and his people left. You see, for them [the Nayakkars], that was the end of an era. This was the end of their rule. As they left, they

Remnants of patronage and the making of Tamil Valaiyar pasts 95

said:  ‘So many years…we can’t even count them. All these years were ours. We ruled this Madurai District. Now we’ll step aside. The whole place that belonged to our people, we place with you. You keep it for your children, whose time will come’. They said that, and then they left for good.12

Malaiyammal is a dynamic, indexical force. The story explains that her presence at the roadside points to a particular set of events  involving protection, patronage, servitude, loss and land. Nowadays, if a passer-by notices, he may turn his head and ask: who is that, what happened here? If he stops at the nearest roadside tea stall to ask, he may also be told this story. He may learn that Malaiyammal, gazing out over the road, still asserts herself actively by causing accidents on the road. The wizened tea shop proprietress explained: ‘you don’t just walk past her temple swinging your arms [casually like you have no care in the world]. You have to take notice’. Those who do not attend to her presence, may find their attention drawn to it in a brutal fashion. The brutality of Malaiyammal’s intrusions into the roadways carries into the present the brutality of her own origins. One set of losses (servitude, death and suicide; the end of the Nayakkar rule and everything changing) lends force to another (car wrecks, trucks turning over, economic losses on the road), bringing the past to bear on the present as a living force. The goddess makes sure that the Event of her origin continues to assert itself and to command our attention.13 She comes alive and seizes the day and the lives (and vehicles) of those who dare not to notice.

12 It seems that they did not go too far. A local Nayakkar zamindar and his religious specialist still trace their history to this place and they continue to be involved in village affairs. Specifically, they may be called upon to ratify marriages and they attend Malaiyammal’s temple festival where, at the deity’s behest, they parade trick-performing oxen. 13 The goddess’s historical force bears on the discussion started by Dipesh Chakrabarty, who noted that in Indian peasant histories deities are historical actors, challenging professional historians’ relegation of divine agency to the anthropological spheres of ‘folklore’ or ‘belief’ (2000, 97–115; also McLean 2004). Chakrabarty concludes that the best academics can do is to regard agentive deities as reminders of their own inability to grasp the range of realities of human life. In a recent paper, I have argued that another way to grasp the reality of deities’ agency in South Asian histories is to treat them as dynamic indices (D. Mines 2012).

96  Diane Mines

As Valaiyars moved from place to place, and from patron to patron, they left traces behind and carried others with them: the Ambala Stone in Melur, a goddess temple in Ponnamaravati, Malaiyammal on the roadside, the Alagar statue, a copper plate, as well as titles, place names and stories. Tim Ingold writes that ‘in storytelling as in wayfaring it is in the movement from place to place—or from topic to topic—that knowledge is integrated’ (2007, 91). It is in this way that patronage converts into a mode of knowing, remembering and narrating the past, a way of keeping track of the places where the people settled and from which they fled. The stories also keep track of the Valaiyars’ social location as servants of patrons good and bad. The lives which Valaiyars narrate are dignified but also vulnerable to patronage ever liable to turn into abuse. The mountain forest served as a last-resort patron, but never a reliable one. Although it offered refuge from abusive patrons, it was full of dangers as well. Nowadays it is deforested and over-hunted, an even drier landscape teeming with night-raiding bison, and haunted by laws and forest rangers who chase the Valaiyars away.

Your time will come Alagar ends the story of Malaiyammal with the Nayakkar’s final words: ‘You keep [this place] for your children, whose time will come’. The children now see their time as having arrived, but not as Valaiyars. ‘The Valaiyar people have discovered that they are Muttaraiyars’, part of a ‘super-caste’ that claims membership throughout and even beyond South India (Sudanandha 1996, 132). As the youth embrace this new identity, they also learn to tell a different story about their past. Casting aside the image of vulnerable forest dwellers, ever on the move, they upend their elders’ accounts and present themselves not as clients to various patrons but as illustrious patrons themselves— heirs to chiefly ancestors who founded cities and bestowed honours on those who stood below. Yet despite this drastic reversal of caste history, patronage remains the key historical and moral trope. The politics of the new youth is rooted firmly in democratic movements and the electoral process, but the language of their political struggles relies on the idiom of patronage through which they assert dignity and establish a voice. Nowadays young Muttaraiyars actively write their caste history on the Internet, in hagiographies of the founding ancestor (the great

Remnants of patronage and the making of Tamil Valaiyar pasts 97

Chief Muttaraiyar), in historical novels and archaeological studies, monthly newsletters, speeches and demonstrations at the foot of new statues of Muttaraiyar, and in popular magazines like the widely read Tamil weekly Kumutam. For some time, Kumutam has been running a monthly column devoted to the culture and history of the Tamil people. Each month, the column features a different Tamil ‘community’ (camutayam), which it celebrates by listing its past and current leaders and heroes, and the distinctive rituals it performs. These communities, which anthropologists usually refer to as ‘castes’, are described in the column as historically deep, bounded entities with unique ethnicised identities. The Kumutam feature article on the Muttaraiyar community (Figure 3.1) opens like this: War! A Great War! A fierce war between the Pallava king Nandivarman II and the Pandya king. The situation appeared dire for the Pallava king. The cities, villages and tanks in the Pallava Country were all falling to the enemy. Unable to remedy the situation, the Pallava king asked the Chief Perumpitiku Muttaraiyar for help. Trustworthy Perumpitiku assured his allies. What happened after that is the truth of history. [These] historical events have given the Muttaraiyar community (camutayam) a place in history through the ages (23 September 2009, 132–133).

The article goes on to list Perumpitiku Muttaraiyar’s many successes in war, including the 16 titles he received for his victories and his successful takeover of the city of Tanjavur, which, the article claims, he had founded. Citing inscriptions, the article lists several poets, religious sects and communities which the line of Muttaraiyar kings patronised. In fact, the Muttaraiyar kings supported ‘the Tamil people’ at large by building temples, protecting Jains and constructing irrigation tanks that saved the Tamil country from perishing of thirst. The article lists more recent Muttaraiyar heroes who have served not just Tamil Nadu but also the Indian nation: freedom fighters, state legislators and so on. Samuel Sudanandha traces the identification of Perumpitaku Muttaraiyar as the apical ancestor of the community to a confluence of sources, ranging from an historical novel by Kalki to work by historians and archaeologists. In illuminating the names and places of

98  Diane Mines

Figure 3.1 Kumutam image of the Muttaraiyar statue in Trichy. 7th and 8th-century Muttaraiyar chiefs, they retroject contemporary identities onto the past and provide material that can be used to stake a claim to caste identity today. He quotes archaeologist Kasinathan, who wrote: ‘When we look at the contemporary condition of the life of Muttaraiyars, we begin to sympathise with them. Those who once reigned with great fame…who gave away their wealth to others abundantly, are now reduced to such a pitiful condition’ (Kasinathan 1976, 105; in Sudanandha 1996, 139). I went to interview A. V. Ayya, head of a Muttaraiyar caste association, in his headquarters on a busy market street in the town of Tiruchchirappalli. Stone lions and political posters welcome visitors

Remnants of patronage and the making of Tamil Valaiyar pasts 99

to his office (also the headquarters for his real estate business). Ayya had political aspirations. He was no historian and he did not know very much about the historical figure of Muttaraiyar. He dreamt of on obtaining for the Muttaraiyar super-caste (which includes approximately 28 different castes, including the Valaiyar) the status of a ‘Most Backward Class’, and concomitantly launched a new political party to represent the interests of Muttaraiyars. As part of his plan, he developed business and government connections, and even made forays into the film industry, in the hope of capitalising on the star potential of his handsome son.14 Back in the villages, at all entry points one now finds signboards announcing the presence of a ‘Muttaraiyar Youth Association’. These often list the names of its officers along with depictions of the herochief Perumpitaku Muttaraiyar, appearing on his royal lion with his sword drawn. Although many villages in the region have signboards like this, the associations are not always active, and their members do not always know much about the political movement of which they are part. One village I worked in did have an active Muttaraiyar Youth Association, whose members styled themselves after their chiefly ancestor and actively enacted their royal history. Like Perumpituku Muttaraiyar, they built and patronised temples and worked for their community’s economic development; like him, they embraced and protected their village from others’ contempt. The slogan on the signboard (in Figure 3.2 below) reads: ‘If you sense love, embrace it; If you sense trouble, Muttaraiyar, swing a sword at it!’15 Some of these changes in the Valaiyar sense of self and their history emerged in an interview with one youth association’s active leader, Mutturajan (3 December 2009). At the time, Mutturajan was studying history in college. He was also a founding member of the ‘VMK Friends’—Vinayagar, Muttalamman, Karuppasami Friends—a youth association (sangam) named after the three deities popular in his village. Mutturajan said that the Muttaraiyar Sangam, which was formed in 2006, quickly ran into inter-caste trouble and was expanded to include youths from other castes. He explained: 14 In the politics of Tamil Nadu a film star’s fan base converts readily into political clout. 15 Thanks to Ms Rohini, Mr Rajasekaran and Mrs Vidya from the CM Centre, Madurai, for their help with translating this awkward couplet, whose form and tone, as they point out, appear to be influenced by the writer of popular cinema lyrics, Vairamuthu.

100  Diane Mines

Figure 3.2 Signboard announcing a village branch of the Perumpituku Muttaraiyar Youth Association. Well, there were some problems with the Muttaraiyar Sangam. It didn’t really work. Maybe they’d work for a couple of days and then stop. It was supposed to organise and do small jobs in the village. So, since it wasn’t working, all the youths joined together and thought we could do it differently. That’s how VMK Friends came about. V for Vinayagar, M for Muttalamman, K for Karuppasami…. There were also Chettiyars and Nadars in the village, and so they came and said: ‘hey, we’re here too. We’re here in this village; why are you just putting ‘Muttaraiyar’ [on your board]. So, this was the kind of problem that arose. VMK is for everyone. All lineages and all jatis worship god. That’s VMK. That’s why we’re named after Vinayagar, Muttalamman, Karuppacami. We do good works. We do a lot of jobs. We are 60 people.

VMK Friends did a lot of work in the village, mostly ‘village improvement’ work that their elders suggested. They helped, for example, construct an impressive new goddess temple. Mutturajan said that VMK Friends did about half the work, including lifting and

Remnants of patronage and the making of Tamil Valaiyar pasts 101

carrying the large granite stones and pillars used in construction, saving the villagers a great deal in labour costs. They dug a drain to channel waste-water away from houses, so that it would not run down and foul the streets near the temple. They also took care of any other ‘trouble’, such as a clash over a cricket match with a neighbouring village, making good on the slogan I cite above (p. 99). Mutturajan: During a cricket match, playing against our boys, the players from village P. grabbed a bat and hit a boy round the head with it. Rajasekaran: Aren’t they also your people (Muttaraiyars)? Mutturajan: They are. They didn’t know who they were messing with. Right away, we got four vans and about 300 of us went there and beat them up. That’s maximum (a word he used in English), the VMK Friends. Mutturajan said that Perumbituku Muttaraiyar ruled Tiruchchirappalli as a great king, having first been a chief or a little king. He said that the name Muttaraiyar derives from the words mu (three) and tarai (territory) and refers to the fact that he united the three kingdoms of South India and ruled them as one. He said that Muttaraiyars ruled Tamil Nadu for 200 years, from about 400 CE. While relating this history, Mutturajan frequently referenced academic authorities like the State Archaeology Department or a particular history book he had seen. ‘The elders’, he said, ‘really don’t know much about all of this. It’s only just recent knowledge [using the English word]. If it was known before, the knowledge would have passed down, but because this is recently known, only educated [using the English word] people know about it’.

Conclusion Despite their cardinally opposed visions of life—one retrospective and one prospective—both the old Valaiyars and the young Muttaraiyars ground their historical knowledge in the language of patronage. Whether losing and winning over patrons, or becoming patrons themselves, patronage is the language and indeed form of knowledge which conveys history and marks one’s ongoing sense of

102  Diane Mines

place and belonging to it.16 In the stories of elders, relations and acts of patronage created remnants that serve as dynamic indices tying them to the region and its past. Valaiyar stories show that relations with good patrons are keystones of belonging and of maintaining a dignified existence. Carrying these remnants with them, Valaiyar elders mark the landscape they lived in, and travelled across, to produce a dignified image of their past. But patronage ties are as vulnerable as they are important. The moral order of patronage can easily give way to the tyranny of greed, the perennial threat inherent to all patronage bonds. There is no more vivid an image of extreme moral strife at the heart of patronage (see Introduction and Gilmartin in this volume) than the Valaiyar flight from the patrons who failed them into the forest with all the dangers it holds. For the Muttaraiyar youths, their ancestral patron-chief is an icon of and a model for their political aspirations. Unlike their fathers, they present themselves not as servants of kings, but as once and future patrons themselves. The stories youths tell are about the future: about hopes for a political constituency of their own, an audible voice in politics, a dignified reputation and the possibility of better employment. But to imagine this future, they look back to the patronancestor Chief Muttaraiyar, who personifies their aspirations. These aspirations can no longer find fulfilment at the foot of the mountain, a very poor patron who can no longer save its people from starvation. At the end of my interview with Mutturajan, I asked him what he thinks the future will be like for him and his children. Surprisingly, he returned to the migration theme: hope for this uncertain future lies in migration (for more on international migration, to which Mutturajan refers, see Osella’s Chapter 16 in this volume). People will continue to gather knowledge and seek out new places to live well. There is no hunting left here, no person who hunts, no livelihood, no staying put. Even now about 60 people are working in foreign [English]: Dubai, Malaysia, Bahrain, 16 The conceptual relation between Tamils and the places they occupy has been well documented. These relations are based on the qualitative entanglement of locations and persons (e.g., Daniel 1984; Trawick 1991), which grounds their place in the world. Daniel (2010) gives an account of what happens when places are no longer there.

Remnants of patronage and the making of Tamil Valaiyar pasts 103

Singapore, all those places. To study, to get government jobs, to become educated. That’s why VMK Friends was started, to help people study, to advocate for one another… People will study and gain knowledge of the world…. Of my 18 classmates, 15 are in college.

Patronage and state-making in early modern empires in India and Britain


Sumit Guha

Introduction Social science has also misled us on formal organization, having been taken in, perhaps, by what bureaucracy says about itself (and by Weber’s romance with the idea of bureaucracy) (Schneider, Schneider and Hansen 1977, 467).


he time has come to look more critically at the view still prevalent in the social sciences that patronage is ‘a perverse and backward political practice, which prevails only where modern states fail’ (Introduction, 3). This limited view presumes a bureaucratic model of a ‘legal rational’ ideal-type, as prescribed by Max Weber. Interestingly, Weber himself saw the bureaucracies which emerged before World War I, and the business corporations which sprung up alongside, as the purest examples of this type (Constas 1958). Sociological research since Weber has shown the vast distance between formal and actual functioning of these institutions. After the economic crises of the 1970s, governments handed over large areas of economic life to the corporate sector. Other types of social provision, too, have been increasingly modified on what are seen as ‘market principles’, these mimetic efforts resulting in strange, parastate institutional hybrids like quasi-autonomous nongovernmental organisations (QUANGOs). But if we look at governance more broadly, patron–client relations may well frame the political order, alongside kinship and cash payments for service. Of course our modern fondness for machine models, which goes back at least to Jeremy Bentham, makes us hostile to such formations, and we often ground our objections in comparisons of efficiency. During the 1998 Southeast Asian economic crisis, there

Patronage and state-making in India and Britain 105

Figure 4.1 ‘Dun-Shaw’, a cartoon by Gillivray showing a kilted but crowned Dundas as a ‘Bashaw’, or oriental despot, captioned as follows: ‘One foot in Leadenhall Street and the other in the Province of Bengal’. Like Atlas, the ‘Dun-Shaw’ bestrides the two continents (© National Portrait Gallery).

was much discussion of ‘cronyism’ as its source. The 2008 bank collapses and bailouts in the crony-free heartland of the University of Chicago and the Blair-liberated (London) City have discredited this picture. Furthermore, economic or political successes—like the feats of crony capitalism in China, Korea or Japan—do not require any measurable level of personal or corporate purity. Political scientists

106  Sumit Guha

note that in Italy at the time of miraculous post–World War II growth, local communities, youth organisations and professional bodies voted for parties and individual politicians in blocs in return for rewards in local projects or favourable legislation. This is what Carl Landé described when he coined the term ‘corporate clientelism’ (1977, xxx–xxxi). Yet no matter how often we see clientelism thriving, ‘not despite democratic statehood, but alongside and indeed often through it’, the dominant view that it is a ‘feudal residue’ remains (Introduction, 3). A critical review must start from first principles and the institutions of ‘formal governance’ themselves, as they are, after all, the benchmark against which we measure clientelism. If we inspect the design of the bench and do not focus only on where the mark is placed, we will need to consider the entire range of governmental practices in which patronage is embedded, especially as so much of today’s world is run not by state agencies, but by corporations, QUANGOs and other complex organisations. All forms of governance are complex hierarchies. And complex organisation both requires and generates unequal power relations. Patron–client structures offer one way of organising these relations, as are armies, bureaucracies and corporations. Some work predominantly through systems of command, others mainly by incentive or market mediation. Today hybrid forms burgeon, and states are increasingly moving back to 18th-century habits and ‘farming out’ many of their formal roles, including their right to use violence at home and abroad (Scahill 2007). The choice of ‘suppliers’ or ‘partners’ has opened a wide avenue for patronage, reminiscent of Georgian England (Namier 1957; Private Eye No. 1324 2012, 5). Comparisons of alternative governmental systems usually remain speculative and all too often end up comparing ‘model with muddle’, as the late Alec Nove called comparisons between capitalism and communism of his day. In this chapter we examine two rival political formations which contended for power in India during the late 18th century. The first part will review the logic of the two systems, which deployed different but comparable strategies of state-building in 18th-century India—the Maratha confederation and the emerging British regime. In the second part, we locate these institutions within the context of patron–client relations.

Patronage and state-making in India and Britain 107

Patron–client relations in South Asian social analysis Fredrik Barth’s (1959) classic study of politics in the Swat Valley of northwestern Pakistan showed how a state system emerged out of acephalous politics, through structures of rivalry and patronage. Swat was an upland valley where a favourable climate allowed intensive farming and supported a complex social hierarchy. Internecine conflict and a series of conveniently contrived deaths allowed one chief to dominate the others and then be legitimised by the British colonial government as the Wali (lord) of Swat. He was not the first among equals, and competition for land and power among chiefs carried on. Ultimately, each chief’s ability to retain and acquire land—and the income and patronage that it brought—depended on latent or overt military action, which required constant entertainment of a body of dependents in the ‘men’s house’. Such entertainment was costly, and Barth records several instances of chiefs who had to sell land to sustain the hospitality needed to sustain their influence. One chief explained this rationale: he knew he was constantly reducing his long-term income, but, as he said, ‘if people stop sitting in my men’s house, I shall lose the land even faster; only this constant show of force keeps the vultures at bay’ (Barth 1965, 81). The control of wealth-producing land, which had originated in the valley’s 16th-century tribal conquest, depended on the open or veiled use of power, sustained by judicious disbursement of wealth. Wealth and power flowed into each other, something that Indian political theorists have understood since ancient times, when artha referred to both wealth and power, and the science (s´a¯stra) of politics was the study of both of these (Trautmann 1971). In Swat, a stable and complex social and political structure was made of dyadic ties, most of which were forms of patron–client relations. ‘The relations which give a position of dominance and authority to one partner are occupational contracts, house tenancy contracts, membership of men’s houses and religious tutelage’ (Barth 1965 [1959], 3–4). Apart from the last, each of these relations assumed that the patron possessed superior resources—usually derived from the effective control of land and its product. But excessive liberality could erode that base and expose a leader to his enemies. Swati politics operated within an exceptionally fertile and productive agricultural ecology where almost all of the land was held by the

108  Sumit Guha

Pakhtun caste, whose membership was restricted to landowners. Land was highly concentrated and much of it was held by a few dominant landowners, who kept tenants and dependents for its cultivation. A range of specialists formed a part of this system. They provided essential services at different times in the agricultural cycle, at the end of which they were paid. The landowner was pivotal to the process, and the system was identical to the familiar ‘Hindu’ jajma¯nı¯ system (see pages 11–13 and 156–61 in this book). At the completion of harvest, the carpenters, rope makers, blacksmiths, muleteers and so on came to the threshing floor where they received shares or fixed amounts in kind or other payment. The allotment was overseen by the landowner’s special agent, who received the lion’s share of the crop. The prayer leader (ima¯m) and the barber were normally obliged to perform certain services. The former could have a land allotment or an annual grain allowance; the latter could have a grain allowance, either per person served or from each plough unit (ibid., 44–48). The landowner kept the bulk of the harvest, which (among other things) supported his men’s house and the retainers who sustained his political power. This power, as Barth’s informants observed, was what prevented his land from being seized or encroached onto by his rivals. But sacrificing land to sustain patronage meant that power was bound to shrink in future. The trade-off was stark and potentially lethal. In 1987 Nicholas Dirks published an important book which sought to reassert the importance of power in the structure of Indian society. Society depended, he claimed to show, on the centrality of the king and not the sacredness of the Brahmin. Ultimately, he hoped to replace the brahmanical myth of sacred aura with the myth of kingly cornucopia. Dirks imagined that the chief means for the formation and articulation of a ‘political community’ was the kings’ gifts of rights to land and of various honors, emblems, titles and privileges which symbolically and morally linked individuals with the sovereignty of the king. The political economy was thereby predicated on a set of moral principles and understandings (1993, 130).

An essay of my own (1994) took issue with Dirks’s gift model of 18th-century statecraft. I argued that his description was not true even of the south Indian kingdom of Pudukottai (which he studied), where the protection of the king’s person and his palace was entrusted to

Patronage and state-making in India and Britain 109

north Indian mercenaries, not grateful recipients of gifts. Dirks’s model begged the question of how the king secured a seemingly unlimited supply of land and wealth in the first place. To see the roots of such wealth and power at the time, we must turn to the most successful indigenous political  formation of the 18th century: the Maratha confederation in western India. Marathas were well acquainted with the over-generous roi fainéant: in the peninsula, this was the Raja of Anegondi, the descendant of the once-mighty Vijaynagara kings, whose domains had long slipped into the hands of insubordinate na¯yakas and invading sultans, not to speak of the Marathas from Shahaji Bhosle onwards. In Maratha legend, the once-paramount lord of the South is a comical figure who calls himself sarvabhaum, lord of the whole earth. He enters the revenue of the world on the credit side of his ledger and then expends it on the debit side. He is a fraudulent bookkeeper, not surrogate divinity. I doubt if such a production would have a good run even in the theatre states of Bali (Geertz 1980). For the Marathas, it was more a farce than a tragedy, and an early 19th-century dictionary includes the following associations with Anegondi: Anegondi ka¯rkha¯na¯—disorderly business Anegondi ka¯rbha¯r—vast and foolish proceedings Anegondi kharc—vast and foolish expenditure Anegondi boln.e—wild, extravagant speech So unaware were the dictionary’s compilers of the supposed centrality of gift-giving to Indic kingship, that they called the Raja of Anegondi ‘a soft fellow ready to give whatever is asked of him’ (Molesworth 1857 s.v.).

Maratha power: limiting patrimony through patronage Apart from such indirect linguistic evidence, we find explicit recognition of the need to husband resources and limit the growth of entitlements in the A¯jna¯patra, a celebrated text on politics written by Ramchandra Pant Amatya in 1717. Ramchandra Pant was one of the organisers of Maratha resistance during the Mughal emperor Aurangzeb’s invasion of the northern Deccan Plateau (1682–1707), and he composed this text when the Maratha successor-regime was taking shape. It was an influential work, and copies continued to be made at least as late as

110  Sumit Guha

1843. Discussing the king’s responsibilities, the A¯jna¯patra emphasises the importance of having a full treasury: Examine the income and expenditure and act so that the State treasury is enriched day by day. The treasury is the life of the State. A crisis that arises may be overcome with the help of a full treasury. Therefore understand this, and always keep the accumulation of treasure in the forefront of your mind.

The experienced minister did not need to insult his readers’ intelligence by pointing out that there was no contradiction between control over land and control over men; the latter could always be lured by cash which the former yielded. Logically, he went on to warn against making gifts from the state revenue, especially if they were hereditary. He wrote: If such gifts are made out of the revenue of the State, that revenue is reduced for generations to come. If the revenue falls off, the State decays and the royal fortune (ra¯jyalaks.mı¯ ) is lost. Therefore, if you wish to found a royal dynasty, do not succumb to anybody’s flattery or allure, and do not allow the tax collection to decrease.

A grantee’s heirs, he added, might lack his virtues; if they are inclined to betray (hara¯mkhorı¯ ), they will be all the harder to control on account of their stable incomes (vritti). For this reason, new grants should not be made. The author goes on to discourage land granting even more emphatically: ‘The king who is generous with land is the enemy of his kingdom. The king is called the lord of the land (bhu¯sva¯mı¯ ) because of his land; when the land is gone, what is he to rule over? What is he the lord of?’ Fiscality (not morality) and taxes (not gifts) were clearly at the centre of a viable political community (Guha 1994a). Ramchandra Pant’s ideal was a stable patron–client system based on a balance of performance and reward. Yet his centralising drive had to reckon with the impossibility of superseding hereditary claims. On this account, many disgruntled landholders deserted Marathas for the Mughals during the war. While he warned against fresh alienation of land, Ramchandra also deprecated efforts to erase extant claims. But such restraint was difficult to enforce. Holders of office and land alike had many would-be rulers competing for their allegiance. There were two rival Maratha dynasties, the rump Mughal regime and innumerable entrepreneurs in political power

Patronage and state-making in India and Britain 111

active throughout the 18th century. The threat of switching sides was real. In practice, offices under Marathas easily turned hereditary patrimonies, or watans. This  produced a second-best strategy for exacting fees and fines from their holders, and manipulating their internal conflict for political and fiscal advantage. Yet, as long as the institution of patrimony (watan) was visibly present, clients would aspire to own one. And once they succeeded in becoming watandars, they ceased to be clients (Wink 1985; Guha 2004b). Patronage turned entitlement ceased to be an effective form of subordination, and there were continuous efforts to counter this fact. This strategy exacerbated the chronically fissiparous character of the regime. Various potentates within it were technically subordinate clients of the supreme monarch (chatrapati). Yet the latter was controlled by the Peshwa, theoretically his Chief Minister. After 1772 this ministership became hereditary and power became vested in the hands of a coalition of ministers headed by a hereditary bureaucrat (Nana Phadnis d. 1800). The Scottish soldier, historian and statesman John Malcolm was a close observer of all things Indian during the decades that saw the decisive establishment of British dominion in India. As a military officer, he was also an interloper into what became the preserve of the Covenanted Service after 1793. He noted that Maratha appointments soon became hereditary, stressing the power of hereditary claims and noting that ‘a most prejudiced attention to which [claims] pervades the whole system of Mahratta government’ (Malcolm 1832, I, 535). As a man of his times, he thought that such limiting of patronage weakened the state by generating factions and insubordination. When a favour became an entitlement, the patron lost his power to terminate benefits and fresh incentives and disincentives were required to make the system work. In Malcolm’s view, then, patronage was a viable, and indeed preferred, way to manage a bureaucracy; he distinguished it firmly from patrimonialism, which he saw as a source of factionalism and frailty. He saw the most extreme patrimonialism in the lands of the Rajput kings, where the claims of kinsmen to genealogical parity with the king, coupled with their ubiquitous fortresses, ‘kept them in a state of constant warfare with the prince to whom they profess allegiance’ (ibid., 547). The Maratha governments did not succumb entirely to this, in part because of the arbitrary interference of favourites and because of rivalries among different classes of officials. ‘The talents of individuals and the favour of princes often exalt a person at the head of the lowest of these offices to the highest consideration’ (ibid., 534 footnote). Governor General Richard Wellesley would have loved to do that!

112  Sumit Guha

The sporadic acts of patrons attracted clients from the lower echelons of the power structure and prevented central authority from being insulated from the lower levels of hierarchy, as the classic nested ‘feudal’ model might have done. It kept intermediate holders off balance, so to speak, and made them mindful of those above and below. As Malcolm put it, the quest for patrons in Central India was largely a struggle to minimise oppression: The villages in Central India, like individuals seek a patron or protector; and a link once established with any person who possesses power is a great shield against the oppression of the manager or renter [taxfarmer]. The Mahratta princes and their principal ministers derive great profit from the fines that they impose for the errors and crimes, real or pretended, of the officers employed in the provinces. This renders the latter very apprehensive of the inhabitants of a village that enjoys the protection of any person at court. This is established in various ways: sometimes from the village having a place of worship frequented by the minister or any high officer of state, or such persons having free land or gardens within its limits; at other times from one of the inhabitants being in their service or of their chief domestics having intermarried into one of the families of the village. The slightest tie is improved into a strong bond, in a government where interference in the concerns of others is, to men in office a source of increase of income and strength (ibid., II, 52–53).

Here, the nested jurisdictions which structurally mark both patrimonial demarcation and bureaucracy are bypassed so as to prevent them from completely excluding superior authority. The resulting tension holds them together, but the balance is delicate and the whole structure is like a taut bicycle wheel, liable to fall apart in the hands of an amateur. Of the major alternative models, patronage–bureaucracy was more torpid but also more stable, whereas clan patrimony was more factious and insubordinate. The Maratha strategy thus led to the creation of parallel chains of patronage which tied the top to the bottom, if only precariously, and prevented the formation of ‘nested’ structures like the ones that damaged the Rajput states. The Rajput focus on the entitlements

Patronage and state-making in India and Britain 113

of the lineage made each king and chief chronically vulnerable to subversion. Consider the well-documented case of the Central Indian kingdom of Chanda, whose reigning family dispute led to the establishment of Maratha control. Nilakanthshah prevailed, but only with the support of the adventurer Raghuji Bhosle. The two agreed that from every 110 rupees in tax revenue, Raghuji would receive 10 as official commission (sardeshmukhi), 25 as the inferior landlord’s quarter share and then half the remainder, or 37.5 rupees, as his share as the ‘brother’ of the Gond monarch. The state seal remained in the possession of Burhanshah’s lineage, and as late as 1808 a fee of 11 rupees had to be paid when it was used. This sum was shared as follows: 5 rupees to Burhanshah’s heir, Behramshah; 2 rupees to Bhuresaheb; 2 rupees to ‘Madlesaheb or Sikandarshah’ and 2 rupees to Ranisaheb, or Burhanshah’s widow. Thank offerings from newly appointed functionaries were also shared in a similar way. In this way, the entry of Maratha power into the political structure of Central India was predicated on the successful exploitation of family broils in the erstwhile ruling houses. Such strategies were also critical to the establishment of Maratha paramountcy in Rajasthan and Malwa, where the skillful exploitation of succession disputes among Rajput ruling families greatly helped Maratha expansion.1 Initially, Raghuji Bhosle was the client or ally with an uncertain footing in the region, but he gradually extended his dominance until effective power was vested in his lineage. Technically, he held the subordinate proprietary right (zamı¯nda¯rı¯ ), which made him subordinate to a suzerain. As suggested earlier, the Achilles’ heel of the Rajput states was their inability to run their kin groups through an efficient system of patronage. In the Maratha struggles with the East India Company, internal feuding also proved the Marathas’ Achilles’ heel. This emerged in 1799–1806, at the key conjuncture when Governor General Wellesley (in office 1798–1805) had to raise every soldier and borrow every rupee he could to defeat the Marathas. While many senior Maratha commanders were suborned by the Company, the Peshwa broke away from S´inde, and Holkar remained aloof (Cooper 2003). Clientship and bureaucratic appointment quickly became patrimonial, and the Maratha polity collapsed in the face of one with a larger element of bureaucratic regulation in both military and civil life. 1 For the sources and references used in this discussion, see Guha (2004b).

114  Sumit Guha

The East India Company state For much of its career, the English East India Company (EIC) was both a political and an economic organisation, and we must recognise its considerable political and infrastructural achievements. It created an apparatus that maintained communication within and control over vast spaces, which news often took three or four months to traverse. Its expansion relied in part on its ability to use the legal system of the emerging English state to police its personnel, but it also grew from the use of sureties and recommendations, or fiscal and patronage ties. This maintained a level of compliance that kept the EIC on an even keel for decades and made its stock a most desirable investment. It had to bestow considerable rewards in India and in England to sustain its powers. But the fact that courtiers and gentry gradually bought its stock doubtless deepened its embedding in the fiscal–military structure of the British Empire. The EIC was certainly not immune to the growth of entitlements, as the iron grip of the ‘shipping interest’ throughout the 18th century testified. Until 1796, all those who benefitted from inflated shipping charges collectively mustered enough votes to block efforts at reform (Parkinson 1937, 175–190). Still, as Holden Furber wrote, though ‘the company was cheated consistently by many of its servants … its administration either at home or overseas can hardly be described as inefficient’ (1976, 201). The system worked well as long as the balance of reward and effort brought needy and pliable young men of humble origin, like Robert Clive and Warren Hastings, to India. All of this changed in the era of conquest following 1756. Suddenly, even minor posts were in eager demand, and the British political establishment pursued them keenly, disturbing the balance of power between client-servants and patron-corporation. Young sprigs of the Whig establishment made bad merchants and recalcitrant servants (Sutherland 1953, 47–50 and 80–88), and the important Scottish Whig politician Henry Dundas became a key fixer and parliamentary manager. As one of his near contemporaries observed, ‘it was not unnatural that his official favours should be confined to his innumerable and insatiable partisans’ (Fry 1992, 130–134 and passim). Wellesley’s great predecessor, Lord Cornwallis, had encountered greater problems. In 1785, he was sent out to reform the system in the crisis that followed the American War of Independence, and he complained repeatedly of the number of well-connected young men who arrived from England in the hope that he would ‘provide’ for

Patronage and state-making in India and Britain 115

them. The reforming Governor General consolidated office tenure— indeed, he created what the Marathas would have recognised as a service jagir (prebend)—as a device to protect the system. Earmarking the  most valuable of offices for ‘covenanted’ servants relieved him of the obligation to ‘provide’ personally, although it also limited his own patronal powers. He could rebuff fortune seekers, even when candidates came with the sponsorship of the celebrated jobber Dundas at the Board of Control.2 He wrote the following about a young man who was sent with Dundas’ recommendation: I do not suppose that he is of a turn or qualified to make his own way in any line of private business; and I must, my good friend, recall to your recollection that no Governor in India can confer an office or employment worth holding, or indeed any substantial favour, on a person who is not a covenanted servant of the Company without essentially injuring the public interests, and committing an act for which he deserves to be impeached.

John Shore summarised one consequence of the change to Warren Hastings in 1787: ‘The members of Government, relieved from the torture of private solicitations, have more time to attend to their public duties’ (both quotes from Aspinall 1931, 31, emphasis added). This policy was confirmed by an act of Parliament in 1793, when all posts paying above £500 per annum were reserved for members of the Covenanted Service (ibid., 169). Nevertheless, within a few decades, there developed a corporate sense of entitlement. When the South Kanara district was transferred to the Madras Presidency, the Bombay Civil Servants worried about access to well-paid appointments and wrote a collective protest to Wellesley who in turn harrumphed that this was ‘a most dangerous and irregular proceeding’ (Ingram 1970, 200). Patronage had fallen into metropolitan hands, and the sons of powerful ministers began to crop up in the Civil Service. Even the powerful Governor General Richard Wellesley complained bitterly to Dundas of the insubordinate conduct of Honourable Herbert Stuart, son of George III’s famous minister Lord Bute. Wellesley was moved to the radical—indeed Jacobinian—declaration that he did not hold himself answerable to any young man’s father, ‘however exalted by rank and character…’, but that he still asked Dundas to communicate his own 2 On the role of Dundas, see Fry (1992).

116  Sumit Guha

self-exculpatory letter to Lord Bute (ibid., 300–302). While expecting subordination from other Company servants, Wellesley saw himself as born to the purple: he complained that members of the Council affect to consider themselves in too elevated a station to condescend to pay the marks of respect due to the person invested with the supreme power, such as attending his levees, being present at all ceremonies and so on. This tends to distract the obedience and respect of the people for the government and gives it more the character of an aristocratic republic than a monarchy (ibid., 216).

He was also offended by the degree to which his immediate predecessor, John Shore, had allowed a sense of equality to grow in the Civil Service. He inveighed against Shore’s ‘low birth, Eastern habits, as well as his education in the Company’s service …’ (ibid., 68 and note). As a result, two councilors, Speke and Cowper, had established a patronage network such that scarcely a servant of the company applies to me for promotion without boasting of the support of these gentlemen, or lamenting the want of it, or complaining of its exclusive and overbearing influence in all promotions under the late government. … That system has collected a sort of party about Messrs Speke and Cowper, consisting as well of expectants, as of the obliged and their friends (ibid., 69).

Wellesley’s quasi-royalist enterprise finally foundered when the Company was nearly bankrupted by his Maratha wars. He was outraged at being raised only to the Irish peerage. To add insult to injury, he was then recalled, as Company debts in India and England ballooned to unprecedented sizes. Following his repatriation in 1805, Cornwallis’s 1786–1793 reforms gradually hardened a strong, entrenched bureaucratic elite connected to patrons in England. The sense of the Civil Service as a privileged brotherhood grew. In 1820, Bombay Civil Servants were once again sending collective protest petitions, this time against the employment of Army officers on deputation in lucrative civil posts in the territories newly conquered from the Peshwa in 1818 (Memorial 1820). Because they were nominees of influential members of the Board of Control and Court of Directors, this had an effect. The special Deccan Commission was abolished in 1825 and the area subjected

Patronage and state-making in India and Britain 117

to ordinary Civil Service government. A few years later, their privileges had solidified far enough for Governor General Bentinck to bemoan the need to draw all his key personnel from a body of less than 400 civil servants, originally selected with no regard for qualification, ‘possessing appointments, like an estate, as a matter of right; secure in their promotion, because there are no competitors, and secured in their possession by the powerful patrocinium [of sponsors in England], with which they are supported’ (Bentinck 1977, II, 977; from the letter of 21 December 1832). If the Maratha governments were crippled by the growth of lifetime prebends turned hereditary patrimonies, the East India Company too faced an entrenched corporate bureaucracy that perpetuated itself through the generations. It was nevertheless effective enough (for all of Wellesley’s fulminations) to best the Maratha imperial formation and ultimately to end it.

Markets and patronage As Barth saw long ago, patronage requires economic resources. In preindustrial systems, they mainly came from agriculture. The management of necessary, specialised services has created a division of labour in every complex society. In South Asia, the standard anthropological model for this was the so-called jajma¯nı¯ system, seen as a set of dyadic ties centred on individual landowning households. Such relationships are, as Landé observed, are essentially dyadic relations that are not unique but socially standardised. Their sustenance depends on stable inequality, with allegiance and service given in return for protection and money. If this relation remains entirely transactional, it comes close to a series of market transactions and renders them highly unstable (1977, 84–85). The resulting instability may then slip into a full-fledged market economy of venal office. A very similar logic operated around Leadenhall Street within a decade of Cornwallis’ closure of individualised patronage appointments (1787–1793). During this period the sale of military and other positions was common in Britain and elsewhere, and point-blank demands for quid pro quo exchanges of payments for services were common. In 1787 Lord Breadalbane, for instance, wrote to Dundas that unless the ‘subjoined list of friends are immediately provided for’, his attachment to the present administration would ‘be considerably diminished’ (Fry 1992, 130). Coincidentally, around that time Cornwallis began to raise defenses against such pressure. Appointments had been sold

118  Sumit Guha

as early as the 1770s, and it was now clear that the first obstacle of entering the Covenanted Service had to be crossed before reaching for the branches of the pagoda tree. Various potentates in the Company’s establishment acquired quotas for nominations of officers and some began to sell these, as the dearth of supply had significantly raised prices. By 1809, the matter and certain agents known to be in the business became so notorious, that a parliamentary committee reported on it. One Mr Stoughton testified before the committee: he heard that Mr Shee was a broker in appointments and applied to him. In November 1808, Shee replied that ‘he could get me a Bengal Cavalry Cadetship or one in the Artillery or Engineers in any Presidency; also a Bengal Writership price 3,500 guineas, the former appointments 500 guineas each …’ (Report from the Committee 1809, 76). Patronage was accepted, even desirable, but its full commerciali­ sation was seen as destructive. In large part, it seems to have arisen from a highly competitive political milieu, something we can easily see in the case of William Collett. Collett was appointed a Cadet for the Bombay Infantry by the India Board on 3 July 1805. This appointment was in the nomination of Lord Castlereagh, who, at the recommendation of the Right Honourable John Sulivan, gave it to Richard Cadman Etches, for a relation of his, on account of services performed by Mr. Etches for the Government. Mr. Etches sold it to Mr. Chaplin, an attorney, for the sum of £250. Mr. J.A. Shee was agent for Mr. Passmore, an attorney who received the money of Mr. Etches (ibid. 1809, 6).

It is evident that the appointment was a direct swap for political service, analogous to what Swati retainers studied by Barth offered to their chiefs. Even racial barriers could be circumvented in this way: Mr. Samuel Lewis was appointed a Cadet in 1800, by Sweeney Toone, Esq. at the recommendation of Mr. Evans. This appointment was passed by Mr. Evans to Mr. Sanderson: Annesley McKercher Shee seems to have procured it of Mr. Wright, and received 300 guineas from the Cadet’s father. Mr. Samuel Lewis being a Mulatto and thereby disqualified, procured a young man of the name of Phillips to personate himself, and pass the previous examinations, for which he paid 20 guineas (ibid.).

Patronage and state-making in India and Britain 119

Had the practice carried on unchecked, it might well have ended up as a version of the watan system. But unlike the Maratha government, the East India Company was able to limit these processes, which seem to have disappeared by the mid-19th century. By 1854, the institution of competitive examinations had definitively blocked such markets, though it created the new profession of ‘crammers’ who prepared candidates for the Civil Service exams—a business that still flourishes in India today. Turning back to humbler forms of patronage, Barth found in Swat that a patronage system that organised social reproduction emerged from, and was regulated by, the dominant landowners. It in turn fed the political power that ultimately enabled them to be landowners. This classic patron–client system of power inequality and personalised relations replicated in entirely Muslim Swat what William H. Wiser called the ‘Hindu jajma¯nı¯ system’. In Swat, as in the Hindu heartlands, this was how the division of labour was organised. It had nothing to do with Hinduism or the ritual order of ‘castes’. In some areas, specialists, whether providers of craft labour or personal services, were members of the village community and were entitled to fixed dues. The common Indic term for this village system—known widely as baluta in Western India—was never adopted into English and remains relatively unknown, even among social scientists studying India today. Baluta described a system in which a village had hereditary functionaries with fixed entitlements which could be levied on all of its households in addition to equally specific payments in kind (supplemented sometimes with cash) at harvest and on festive occasions. The functionaries ranged from watchmen and leather workers to powerful officers like headmen and accountants. Some were dominant, others subordinate, but all received dues in kind and other perquisites. Clientage was an entitlement which could be turned into capital and sold, under certain circumstances, for cash. Early modern records from western India are full of references to the sale of clientship, referred to as ‘baluta watan’. The same also happened in jajma¯nı¯ relations. Across 19th-century India, lawsuits for the right to perform certain functions were widely entertained in British colonial courts of law. The term entered sociological literature to describe the relationship between a patron-household and individual servant-household(s) that supplied it (and other similar patrons) with goods or services in return for payments at harvest and

120  Sumit Guha

on other occasions. In Swat, however, the dominant landlord enforced the system, preventing the rights of service from ever congealing into entitlements. The service system varied over time across the Indian subcontinent; in some places it was altogether lacking and there was nothing particularly Hindu about it (Marriott 1960; Barth 1965; Guha 2004a). I have shown elsewhere that this formal and regular creation of hereditary claims originated in the opportunity to secure rents (as fees, gifts or bribes) and the consequent preparedness of entrepreneurial individuals to pay the holders of political authority for creating and protecting such rights (Guha 2013, Chapter 3). It is significant that these village offices were called watans (sometimes vrutti, both best translated as ‘patrimony’) and were thus seen as analogous to other property rights in venal office. We can watch this system emerge by looking at sources from central India dating to around 1820. While the detailed descriptions of village servants with fixed dues in kind were being compiled in Western Maharashtra, little trace of any such system could be found in thinly populated areas where cash economy was undeveloped and the need for fixed services should have been all the greater. In the early 19th century, Chhattisgarh (formerly part of Madhya Pradesh), for example, was a landlocked region with limited trade. In 1827, Richard Jenkins, the English administrator of the kingdom in the 1820s, wrote a report about it, remarking on the absence of headmen or peasants with hereditary rights in this area, which stood in contrast to the Maratha territories further west. The establishment of village servants was equally undeveloped, and those who existed were mainly employed by the gaontia headmen (effectively tax farmers) of each village. Craftsmen and specialists were seen as dependents of village lords, exactly like tenant farmers, and only the smith and the washerman were employed for the benefit of the farming community as a whole ( Jenkins 1923, cited by Guha 2004a). In other places, it is evident that such perquisites were beginning to appear as a consequence of the creation of various offices under Bhosle’s rule, accompanied by population growth and commercialisation. As these became offices of profit, individuals began to acquire and dispute them, providing a political and fiscal resource for the regime. This stands in marked contrast to the more densely settled and commercialised Western Maharashtra, where by the 16th century, if not earlier, saleable hereditary office was

Patronage and state-making in India and Britain 121

already institutionalised. Here, the cash economy and competition made hereditary office a desirable acquisition and its creation and adjudication a source of profit to the state and its ever-hungry local functionaries. A few extracts from the surviving ledgers of several tax farmers from western Maharashtra (spanning the period between about 1750 and 1825) illustrate how the regulation of the baluta system could be used to fiscal advantage: Half a rupee [fine realised] village servant [Mahar] did not serve the village properly One rupee Leather-worker left the village of Pisarve and went to Amboli 22 rupees Saheb Mitekhan beat a Mahar One rupee … Leather worker left the village work undone, so he was made to complete it and fined One and a quarter rupees Santu Potter failed to supply the mask for Mahalakshmi, so fined (Guha 2004a)

I would suggest that the opportunities of such fees and fines, as well as the ruling classes’ demand for the reliable delivery of specialised village-based services were instrumental in transforming village specialists into hereditary office-holders. When the British legal system appeared on the scene, a considerable body of case law grew up around enforceable claims to jajma¯nı¯. Family priests often successfully sued their patrons for payment or to exclude rivals: such suits, which we commonly find in pre-colonial times, clearly indicate that the relationship was neither dyadic nor voluntary, as a patron– client system would be. It was only after the mid-19th century that the Crown administration gradually chipped away the legal basis for such suits, as it concentrated all land rights exclusively in the hands of the landowners (Kikani 1912, iii–viii and 76–80). But socioeconomic organisation is dynamic, so let me end this chapter with a contemporary example. The Conservancy (Sanitation) Department of the Kolkata Municipal Corporation has long offered relatively secure employment with minimal benefits and a pension. Such positions were desirable enough to be passed on by employees to heirs or relatives. In the early 1970s, Subrata Mukherjee, a

122  Sumit Guha

Congress Party minister in Bengal, sought to check the rise of the Communist Party (Marxist) by creating a system whereby employees could nominate four of their five closest relatives to fill the post when they took a long leave or retired. The idea worked well, and the department became a non-Communist party stronghold. A market system gradually emerged: some posts were sold, and others were rented. The nominee could be induced to pay the regular employee large sums for the job; alternately, the employee went on leave and nominated a substitute who paid half the wage to the regular holder (‘Corporation Fails to Break Job Panel Vise’, Kolkata Telegraph 17 May 1999). Patron–client systems have had a long history in India, from pre-colonial times to the present. But the competitive acquisition of power and wealth is the thread that runs through every setting wherever it may be found.

Democracy as Patronage

The paradox of patronage and the people’s sovereignty


David Gilmartin


atronage is an old and deep-seated phenomenon in Indian society.1 It has long been one of the fundamental bonds of political order, shaping relations between leaders and followers and structuring hierarchies of political power. Yet the practice of patronage has also been a focus of moral tension, for patrons embody some of the deepest contradictions that define political order. At the most basic level, relations of patronage are characterised by what may be called ‘unequal reciprocal exchange’, a phrase that captures its internal paradox. On one level, patrons’ relations with their followers are legitimised by reciprocity, and the equality which it implies, suggesting that patrons and clients occupy a shared moral world. On another level, inequality lies at patronage’s very heart: the patron’s authority depends on access to status and power which transcends that of his followers. Patronage is thus defined by a never-ending negotiation, not only between patrons and clients, but also between the paradoxical principles on which it is based. The good patron is continuously steering between the dangers of two opposing poles of action, one of which threatens his efficacy and the other his legitimacy. On the one hand, the success of a patron depends on his connections to networks of resources and power. To deliver practically for his clients, he must command the flows of resources connected to such networks. On the other hand, if a patron is to act in the name of his clients, he must maintain a claim to moral independence from these networks, a claim central to the establishment of a reputation as a protector of his 1

I thank all those who commented on earlier drafts of this chapter at a workshop at the University of Pennsylvania and at the Triangle Legal History Seminar in North Carolina. I also thank the Stanford Humanities Center for support while writing this chapter. Thanks also to Anastasia Piliavsky for her helpful suggestions.

126  David Gilmartin

client-community. His position thus involves continuous negotiations on multiple levels. An effective patron is enmeshed in a world of perpetual calculation of costs and benefits, even as he projects an image of moral transcendence over the strategic connections and calculations on which his position relies. Forms of patronage have varied vastly over space and time. This chapter focuses on the nature of patronage as it operated in mid20th-century elections in India, when elections first became the foundation of India’s political order. Patronage remained central to India’s politics and to the flow of resources in its society, but the new rules of electoral democracy and the idea of ‘the people’s sovereignty’ crystallised the inner tensions of patronage in new ways. This chapter looks to electoral law to examine the relation between patronage and Indian democracy and the paradoxes internal to both.

Patrons and politics The immediate background to this story is the earlier history of patronage in British colonial rule in India. Something of the ongoing tension in the operation of patronage in British India, particularly in relation to politics, is captured by C. A. Bayly’s (1983) comparative account of north Indian patrons in the late 19th century. Bayly underscores the different types of strategies pursued by different kinds of patrons, contrasting the Muslim patrons of the qasbas of Rohilkhand and Awadh with the Hindu bankers of the middle Gangetic basin. Yet Bayly also emphasises the common tensions these patrons faced, as both groups balanced (though in somewhat different ways) two contrasting forms of patronly connections, which were both central to patronly efficacy and reputation. These were forms shaped by what Bayly characterises as ‘vakil’ and ‘dharmik’ relations. Vakil relations structured networks through which patrons managed resource flows. Vakils were agents through whom patrons negotiated their interests in a range of arenas extending upward towards the state and the courts; downward towards networks of clients, whose support was critical to their positions; and outwards towards the market and newly emerging spheres of public debate. The effectiveness of these networks of agents and connections determined the patrons’ ability to situate themselves at the centre of the resource flows that determined their patronly standing and

The paradox of patronage and the people’s sovereignty  127

efficacy. Yet Bayly juxtaposes these relationships with the equally important ‘dharmik’ relations maintained by patrons with cultural and religious institutions.2 In the case of Hindu bankers, for example, this involved the benevolent support not only of Brahmins, temples and other religious institutions but also of public, community activities and performances like Ram Lila (the annual re-enactment of the Ramayana story), new cultural and literary organisations (such as the Hindi Sahitya Sammelan) and new religio-political movements (like Cow-protection) (Bayly 1973). Such support underscored the image of patrons as selfless protectors of their client-communities. In certain ways, these two forms of patronage were mutually reinforcing, as both were critical to the deeply intertwined material and cultural acquisition of status and wealth that was central to the workings of patronage. In that sense, both can be analysed in terms of networks, connections and strategic calculations. But, as Bayly’s analysis suggests, there is another crucial sense in which these forms of patronage operated in sharply opposing and even contradictory idioms. However much dharmik patronage may have been related to the instrumental acquisition of status, it was not usually justified in terms of self-interested calculation and the protection of material interest. Quite to the contrary, it was almost invariably cast in terms of selflessness and disinterested benevolence. The effectiveness of patrons thus depended on their ability to place themselves at the centre of resource flows, even as they presented themselves publicly as selfless supporters of ‘their people’. The development of such paradoxical framings of patronage over time provides the backdrop to how patronage functioned in Indian  democracy in the 20th century. The importance of patronage to Indian democracy has been underscored by a number of political scientists. Some, like Kanchan Chandra (2004), have indeed characterised India’s political system in the second half of the 20th century as a ‘patronage democracy’. Yet the recognition of the importance of patronage to democracy has not removed political commentators’ deep unease about the relation between the values of patronage and democracy as distinct ideal-type systems. Despite the demonstrable intersection of patronage and democracy, many have continued to see the values that they entail as embodying a fundamental moral opposition. In this frame, patronage is often viewed 2

I use the term ‘dharmik relation’ to include all forms of religious and community patronage, whether Hindu or otherwise.

128  David Gilmartin

as a holdover from the past, a remnant of a persistently ‘traditional’ worldview rooted in hierarchy, now transposed onto a modernising democracy, increasingly defined by individual choice and popular consent. In the work of Louis Dumont (1980), for example, patronage was seen as embedded in India’s deeply ingrained hierarchical traditions, with which ‘modern’ democratic values (and even market values) were deeply at odds. Others have seen the story of patronage and democracy as one of adaptation, and stressed the ways in which patronage, a ‘traditional’ and hierarchal structure of authority, nevertheless played a critical role in India’s gradual assimilation of ‘modern’ democratic forms and values.3 Yet even this work has reflected the persistent assumption that the linking of patronage and democracy joined elements that were at root conceptual opposites, a paradox echoed in the title of Lloyd and Susanne Rudolph’s pathbreaking work, The Modernity of Tradition (1967). Much recent work in political science has continued on this path, but has sought to bridge the gap between the ‘modern’ and the ‘traditional’ with ‘rational choice’, or the instrumental calculation of benefit and cost as a culturally universalisable phenomenon. In such analyses patronage often appears as an adaptive framework that has facilitated the flows of resources within electoral politics.4 Studies of Europe and the United States have thus demonstrated the persistence of patronage not only as a ‘traditional’ survival but also as a rationally adaptive system for dealing with the complex new pressures of modern economic and social change (e.g., Piattoni 2001). The particular prominence of patronage in India’s democracy has been viewed as an adaptation to India’s own distinctive political history, especially to the history of colonial bureaucracy, to India’s ethnic and religious heterogeneity, and to the distinctive evolution of its political parties over time (Kitschelt and Wilkinson 2007). This has not, however, erased the presumed conflict between the conflicting values of patronage and modern democracy. The supposition of a tension between patronage and democratic values persists in writings on ‘civil society’ in India, which often embody a normative emphasis on the tension between patronage (arising from ‘tradition’) and the 3

To some degree, one can see this in many of the works of the ‘Cambridge School’, particularly those collected in Gallagher, Johnson and Seal (1973). It was an approach also evident in many works on caste, for example Conlon (1977). 4 For a good overview of the evolution of political science scholarship on patronage, see Roniger (2004).

The paradox of patronage and the people’s sovereignty  129

more abstract, individual-oriented values, including openness, equity and efficiency, commonly associated with democratic (and capitalist) forms of rule.5 Given that patronage relies heavily on negotiations and reciprocal exchange, it is hardly surprising that rational choice may appear as a useful frame for analysing some key aspects of its operation, as South Asia’s historians of the ‘Cambridge School’ understood long ago. But a focus on rational choice alone can hardly make sense of the contradiction between calculated reciprocity and the image of patronage as a realm of moral benevolence opposed to instrumental calculation. As I show in this chapter, this tension, heightened by electoral politics, has profoundly shaped modern democracy in India—and elsewhere as well. Indeed, the paradoxical reliance of patrons on two seemingly contradictory idioms of authority suggests a parallel contradiction at the heart of the conceptualisation of ‘the people’s sovereignty’. The fate of patronage in Indian democracy opens a window onto the paradoxical foundations of democracy itself.

Patronage in electoral law A key perspective on the paradoxical role of patronage in the definition of democracy is provided by electoral law. In India the idea of the people’s sovereignty did not find overt legal expression until the promulgation of the constitution in 1950. But the roots of this concept were embedded in the evolution of electoral law under the British regime in the decades after 1920. The development of India’s 20th-century electoral law cannot, in fact, be understood without a brief detour to 19th-century Britain, from where many of its defining features and contradictions emerged. In Britain itself, conceptual distinctions between ‘good’ and ‘bad’ patrons played a critical role in debates about the electoral meaning of the people’s sovereignty in the wake of the passage of the Great Reform Bill of 1832. Arguments in Britain on the expansion of the franchise, though complicated, were widely predicated on the assumption that voting put a critical check on power by making those in power beholden to the free voice of ‘the people’. But in a 5

For a good overview of these contradictions, see Harriss (2005). For a recent argument that basically accepts this distinction, even as it seems to critique it by relabelling the distinction as one between ‘political’ and ‘civil’ society, see Chatterjee (2004).

130  David Gilmartin

society in which hierarchy and the influence of patrons had central importance, the nature of authority and reciprocity embodied in relations of patronage emerged as a central element in debates about the meaning of democracy. Few imagined that voting could (or indeed should) undercut the influence of patrons. But if the concept of ‘the people’s sovereignty’ rested on each individual’s free choice, how could voting in a hierarchical society, with its myriad forms of influence, be reconciled with the ideal of the people’s voice? At what point did patronly ‘influence’ cross the line from reciprocal exchange to corrupt coercion or become, in the legal terminology of the day, ‘undue’? This concern with the nature of ‘influence’ in a democratic order was a key preoccupation for George Grote, the philosophical radical who became in the 1830s Britain’s most forceful advocate for the introduction of the secret ballot. For Grote, ‘free agency’ was ‘the very soul of voting’. The influence of patrons therefore presented a potential problem for democracy. But, as Grote took pains to point out, what was crucial to a meaningful democracy was not whether patrons wielded influence but the way in which they did so. On the one hand, coercive operations of powerful patrons in society could undercut the very foundations of free voting. This was evident in the threat to choice posed by a landlord who put ‘the word into the mouth of his tenants’ or by urban notables who sent round ‘circulars to their tradesmen’ letting them know how they were expected to vote (Grote 1833, 10–11). These were models, as Grote saw it, of illegitimate patronage, for they were incompatible with the exercise of free choice. On the other hand, honourable forms of patronage were, as Grote noted, another matter. ‘Gentlemen do, indeed, draw a distinction between legitimate and illegitimate influence’, he declared, ‘a distinction which I heartily admit and approve of, though I think the real meaning of these important words ought to be very carefully explained’. Legitimate influence did not arise from coercion or ‘purchase’, but rather came ‘unbidden and unbespoken’. It was the result not simply of pressure or instrumental calculation but from the ‘legitimate influence of wealth and station’, which served ‘as the passport, the ally, and the handmaid, of superior worth and talent’ (ibid., 28–31). It arose from skill and virtue, which wealth and station allowed one to cultivate. ‘Good’ patronage thus provided a model for how democratic politics ought to operate—when even amidst unequal reciprocity and political deal-making the ‘free agency’ of voters was

The paradox of patronage and the people’s sovereignty  131

maintained as the free choice of the individual was magnetically drawn to efficacious and community-minded patrons.6 This vision came to be embodied in electoral law on corruption in the second half of the 19th century in Britain, and it profoundly shaped electoral law when elections were first introduced in India in the early 20th.7 In becoming a model of corruption, the epitome of the bad patron provided a critical foil for the delineation of legitimate influence as it operated in the context of voting. This was reflected in the statutory introduction of electoral rules for the 1920 provincial Council elections that moved ‘undue influence’, or a concern for voters’ free agency, to the centre of electoral law (however limited the property-based franchise). ‘Whoever voluntarily interferes or attempts to interfere with the free exercise of any electoral right, commits the offence of undue influence at an election’, the Indian Elections Offences and Inquiries Act of 1920 declared (Hammond 1936, 746). Electoral rights in this context referred not just to the individual voter’s right to exercise choice in casting a ballot, but also to the rights of candidates themselves to freely stand, not stand or withdraw from elections without excessive pressure. ‘Undue influence’ could be of various kinds, including not just the pressures exerted by superiors on their dependents, but also community and religious pressures (such as threats of social ostracism) or inducing voters to believe that they would become objects of ‘divine displeasure, or, spiritual censure’. Undue pressures also included the offering of monetary inducements, which could compromise voters’ choice by turning it into nothing more than a personal bargain, in which the free consideration of the merits of a candidate as a representative of the community was foreclosed. The offering of any ‘gratification’ as a bribe, with the aim of ‘inducing’ or ‘rewarding’ an individual for exercising an ‘electoral right’ was thus legally forbidden (ibid., 746). At the heart of electoral law on corruption lay the protection of free voter choice from what Grote had called ‘illegitimate’ influence,


For an overview of mid-century debates about the question of ‘legitimate’ influence in the United Kingdom, see Heesom (1988). 7 Piecemeal legislation on corrupt electoral practices during the 19th century was consolidated in Britain in the Corrupt and Illegal Practices Prevention Act of 1883. Central elements of this law were given statutory form in 1920. Local elections had been held in different parts of India from the 19th century, but elections emerged as a central feature of the colonial political structure only with the reforms of 1919.

132  David Gilmartin

for free choice was the key image that lay also at the heart of the people’s sovereignty. Electoral law built on, and simultaneously reframed, the paradox of patronage in a new direction. On the one hand, patronage was based on unequal access to resources and status, which played an unavoidable, and indeed a necessary, role in elections. ‘Influence’, as a form of pressure, remained ubiquitous and critical to the most basic operations of political life. Yet if free and honest elections ‘amongst unconstrained freemen’ were to occur, elections required a patron’s influence to be conceptualised as an exchange of efficacy and virtue for the ‘heartfelt esteem of a willing public’, as Grote had put it. Only then could unequal reciprocity be maintained amidst assertions of free and equal choice. Electoral law thus emerged as pivotal to imagining a ‘sovereign people’ and to providing a legal framework in which ‘undue influence’ could be adjudicated, and the contradictions of patronage negotiated in relationship to an emerging vision of sovereign choice.

The state and illegitimate influence In India the paradox of patronage was played out in a range of electoral litigation cases, both before and after the independence in 1947. As in Britain, losing candidates could file petitions in the wake of elections, charging opponents with violations of electoral law. Although these cases ran the gamut of offences, a significant number of them focused on issues that related in one way or another to patronage and the nature of a patron’s influence. No cases were more important in the early years of elections than those dealing with relations between patrons and the state. During the colonial period, the British state was itself in many ways the most powerful patron of all, collecting and dispersing resources on a substantial scale. How patrons’ ‘influence’ operated in relation to the state thus became a central issue in negotiations of electoral law. In India (as in Britain) direct state intervention in electoral politics evoked a strong sense of corruption. Direct interference in elections by state officials was in fact legally prohibited, both by election law and by ‘Government Servants’ Conduct Rules’.8 Full-time state officials 8

As the rules stated: ‘A whole-time Government servant shall not canvass or otherwise interfere or use his influence in connection with or take part in, any election to a legislative body…’ See Sen and Poddar (1951, 805n).

The paradox of patronage and the people’s sovereignty  133

could not run for office. And yet connections to the state remained a key source of resources and status in India’s patronage-politics. Two cases illustrate the ambiguities patrons had to negotiate in their relations with the state as elections were introduced. The first case, dating to 1923, indicates the critical role that state patronage continued to play in the success of patrons who competed for elective office. This case came from the United Provinces Legislative Council elections for the Muttra [Mathura] (Non-Muhammadan Rural) constituency, where an Indian National Congress candidate contested against a prominent pro-government local notable, R. B. Babu Ram Nath Bhargava. The Congress had boycotted the first provincial elections in 1920, arguing that the new electoral order, with its limited franchise, was potentially open to considerable government manipulation (and was in some ways simply a mechanism to shore up existing government influence). But when the Congress decided to contest in 1923, they used the law to attempt to limit the scope of government patronage and to redefine its cultural image. When the Congress lost the Mathura election to Bhargava in 1923, the Party thus filed a petition labelling the district Collector’s support of Bhargava in the election as ‘undue influence’. The petition charged that the Collector had written to Bhargava well before the election, urging him to run for the Chairmanship of the District Board so as to put himself in the best possible position to run for a seat in the provincial Council. He had then helped Bhargava to gain the District Board Chairmanship by privately securing the withdrawal of another candidate. Beyond this, he had written to a pro-government organisation, the local ‘Anti-Revolutionary League’ (aman sabha), urging it to scrutinise the electoral rolls and to expunge the names of Congress supporters who did not meet the required (property-based) voting qualifications.9 Here a British official had himself acted within the new electoral system as a powerful patron, using his position in the government to shepherd a prominent pro-government notable to success against the Congress. By filing a petition, the Congress appealed to the law to label such patronage illegitimate. The law now provided, at least in theory, a standard of ‘undue influence’ against which to judge such actions, that is, a framework for delineating the boundaries that state influence 9

The ‘Anti-Revolutionary League’ was a society of notables and officials organised by the government in the early 1920s to oppose Congress non-cooperation.

134  David Gilmartin

could not cross without threatening the people’s free, sovereign choice during an election. Yet, as the Congress quickly discovered, the reverse was also true. By defining the parameters of legitimate patronage, electoral law also provided a framework for delineating what forms of government patronage lay within the newly defined parameters of acceptable electoral influence. This became manifest when an election tribunal (headed by a British ICS officer) rendered its decision on the case. The tribunal’s take on Congress’s charges was clear: ‘Mr. Fremantle’s [the Collector] action did not amount to more than helping to put Babu Ram Nath in a position of influence which might be useful to him’. Such influence could hardly be termed ‘undue’, they argued, for it was deeply ingrained in the longstanding relations between local notables and the government in Indian society. ‘It might with equal logic be urged that the [government’s] recommendation for or bestowal of a title on a prospective candidate would be an exercise of undue influence’. But this was hardly tenable, they asserted, for it would strike at the normal, everyday operation of influence within India’s political and social order, in which the role of the state was a fact of life. ‘A candidate’, they said, ‘is entitled to the legitimate influence of his position and status’: The fact, therefore, that Babu Ram Nath, possibly with Mr. Fremantle’s help, attained to the influential position of Chairmanship of the District Board cannot in our opinion be regarded as an exercise of undue influence unless Babu Ram Nath is shown to have abused his position as Chairman in that election.

In other words, government intervention in the creation and recognition of influence did not, in and of itself, challenge legal assumptions about voters’ free choice. In this case, the tribunal added, the Congress had made ‘a grievance out of the most ordinary occurrences’, as it was ‘prone to look upon even the most harmless acts with the eyes of suspicion’ (Hammond 1920–1927, I, 191–197; also Narain 1930, 118–120). A second case, arising from the 1937 elections in Bengal, suggests how complex, and often explosively contentious, such issues eventually became. In this case, at stake was not simply the behaviour of British officials but the role of Indian patrons themselves, as they were integrated into the state apparatus as elected ministers, ostensibly representing the voice of ‘the people’, as they now came

The paradox of patronage and the people’s sovereignty  135

to operate levers of state power. The question erupted dramatically in 1938 with the filing of a petition against the election of Nawab Sir K. G. M. Faroqui of Ratanpur, a government minister and Muslim League candidate from the Tipperah North Muhammadan Rural constituency (now in Bangladesh) to the Bengal Legislative Assembly. The petition charged Faroqui, who was a large landowner in the Tipperah district and Minister for Agriculture and Industries in the government, with having ‘abused his position’ during the 1937 election campaign to win votes. The case turned not only on the Nawab’s direct, allegedly corrupt use of his government office to canvass support but also on the image of illegitimate influence that arose from his alleged efforts to trade on his wealth and ministerial prestige for instrumental electoral gain (Calcutta Gazette 22 January 1938, 1–71).10 Charges in the petition included the allegation that the Nawab had toured his constituency, providing ‘feasts’ to the people as a direct ‘gratification’ in return for votes and that he had promised to patronise the repairs of local mosques with similarly calculated aims. One witness quoted him as saying: ‘It is a pious work. I must do it. Please all of you vote for me’ (ibid., 6). Even more serious in the eyes of the tribunal were charges that he mobilised employees of the Cooperative Department (which was under his ministerial control) to campaign for him directly, in violation of the Government Servants’ Conduct Rules. Although the tribunal found that the petition provided inadequate proof of individual votes having been thus coerced, they nevertheless cited overwhelming evidence that the Nawab had in fact deployed Cooperative Society inspectors for electoral canvassing and condemned his ‘highly improper’ behaviour in the strongest possible terms (ibid., 11). The most sensational charges related to Faroqui’s deployment of a high-ranking civil servant in his department, the acting Director of Industries, Satish Mitter, to negotiate with other leaders as a go-between on the Nawab’s behalf, another seemingly blatant breach of electoral law. Mitter was a well-connected Calcutta figure in his own right and in 1934 had published an important book on Bengal’s depression-era economic problems, A Recovery Plan for Bengal. But Mitter’s relationship with Faroqui during the election became fraught, as the Nawab had defined for him a role that the 10

For a more abbreviated description of the case, see Sen and Poddar (1951, 802–821).

136  David Gilmartin

tribunal characterised as nothing short of an ‘unofficial election agent’ (ibid.,  10). Evidence in the case detailed how the Nawab had seemingly translated his patronal relationship with Mitter into a purely instrumental one, directed entirely toward the Nawab’s electoral ends.11 An example was the Nawab’s alleged use of Mitter to pressure his relative, the Congress leader Sarat Bose, to intervene in securing the withdrawal of one of the Nawab’s opponents. Most sensational was the evidence that the Nawab had used Mitter as a go-between to get A. K. Fazlul Haq, the leader of the Krishak Proja Party (KPP) and Premier of Bengal at the time the case was being tried, to sign a statement supporting Faroqui against a candidate of Fazlul Haq’s own party in Tipperah. Behind this stood complex personal animosities, for Faroqui’s opponent was a factional rival of Fazlul Haq within the KPP. On one level, such exposure was a public embarrassment for the Premier, who in his evidence before the tribunal could not quite remember whether he had actually signed the document or not, even as handwriting specialists testified that he had.12 But the evidence proved even more damning for Nawab Faroqui as the details emerged in the press. The aggregate import of the testimony was to portray the Nawab as a man who in pursuit of electoral victory had transformed his patronal connections into nothing but structures of official pressure, with Mitter helping him win in return for a promise of a permanent appointment as Director of Industries as a quid pro quo. The case in fact transcended questions of improper state influence to bring the nature of electoral patronage as such under the microscope in legal argument, in the court of public opinion and in the press. Competing images of the patron, as both a manipulator of resources and a servant of the community, were brought into sharpest contrast in evidence before the tribunal relating to what turned out to be the most serious charge in the case (and which led to the Nawab’s unseating): that he had deployed religious ‘undue influence’ to win 11

The most telling evidence included some of the Nawab’s correspondence with Mitter, which was sometimes couched in the language of patronly loyalty and affection. In one letter, for instance, the Nawab wrote: ‘Don’t I know you and your attachment towards me’. Yet this was contrasted with the instrumentality of the relationship that the electoral context had brought to the fore. 12 Fazlul Haq first claimed not to have signed but then stated (in the paraphrase of the tribunal) ‘that one cannot be too dogmatic about an isolated signature, and it is not impossible for busy people like himself to sign documents by mistake’ (Calcutta Gazette 22 January 1938, 32).

The paradox of patronage and the people’s sovereignty  137

the election. The legal allegation was that the Nawab had circulated election pamphlets that threatened his constituents with ‘divine displeasure’ and ‘spiritual censure’, if they dared to vote against him. The texts of these pamphlets, which contained dialogues between wise men (including the ulama, Muslim religious scholars) and the people, were most noteworthy for their attempt to align the Nawab’s image with patronly selflessness and the interests of the community as a whole, a virtue depicted here as his most powerful claim for votes. Voters should select a ‘community worker’, one pamphlet declared, and of all the candidates contesting, ‘Nawab Faroqui, alias Suba Mia Saheb, is the fittest of them all. … He is one of the real Muslims of the real Muslim party. He is a real benefactor of the tenants and for that reason he accepted the Ministry of Agriculture’. The counter-image of selfish manipulation was here reserved only for Faroqui’s opponent, a lawyer ‘unable to defray his family expenses by his profession’, who was running as a ‘benamdar’ (a shil) for the Congress (even if technically with a KPP ticket). A religious leader in the pamphlet declared: ‘Alas! These wretches are the enemies of Muslims and faithful Muslims should shun them…. Those who harm Muslims or Islam do not remain Muslims’. The peasant interlocutor in the pamphlet responded with unalloyed gratitude: ‘Maulana Saheb, after listening to what you have said I have become giddy … God has kept you alive, otherwise we would have gone to our graves as kafirs’, misled by traitors of the community (ibid., 14).13 Here the Nawab’s candidacy seemingly offered patronly protection not only for the people’s interests but also for their souls. But the irony of the legal case was that, while defending himself against accusations of religious ‘undue influence’, Nawab Faroqui was also forced to repudiate the very image of patronly protector presented in these pamphlets. To protect the Nawab from charges of ‘undue spiritual influence’, his lawyer first tried to distance his client from the pamphlets, claiming that Faroqui had not been involved in their circulation. When this failed, he attempted to dissociate the Nawab from the sentiments expressed in the pamphlets by questioning how many voters would actually take such sentiments seriously. The president of the tribunal, Justice Edgley, retorted: ‘If they were not to


The text of the whole pamphlet, in Bengali, is in Calcutta Gazette (22 January 1938, 51–54).

138  David Gilmartin

produce any effect, why were they distributed at all?’ (Amrita Bazar Patrika 8 December 1937). In fact, in setting out a framework for evaluating the charges of spiritual undue influence, the tribunal presented a model for ‘good’ patronly influence in India that offered an implicit foil for the conduct of the Nawab. This model drew on a 19th-century Irish election case (in the UK) relating to the electoral role of a Catholic priest: The Catholic priest has, and he ought to have, great influence. His position, his sacred character, his superior education, and the identity of his interests with his flock ensure it to him; and that influence receives tenfold force from the conviction of his people that it is generally exercised for their benefit. In the proper exercise of that influence on electors, the priest may counsel, advise, recommend, entreat, and point out the true line of moral duty, and explain why one candidate should be preferred to another, and may, if he think fit, throw the whole weight of his moral character into the scale; but he may not appeal to the fears, or terrors, or superstitions of those he addresses.14

Even if based on unequal exchange, the relationship between a good patron and his followers was, in other words, deeply reciprocal, with power and benevolence on one side balanced against rational, reciprocal gratitude (and votes) on the other. Behind this statement stood a vision of moral community. And with respect to Faroqui’s case, the tribunal made clear that this extended beyond the confines of the narrowly defined religious priesthood, adding that most Muslims in India had ‘no regular priests’, but looked for spiritual influence to ‘men who pursue the usual avocations of life but who have some reputation for learning or piety or have some standing in society’ (Calcutta Gazette 22 January 1938, 17). This description certainly might have fit the Nawab. At stake, in effect, was patronly reputation. But, ironically, the very process of the tribunal’s evidence-taking and deliberation, which had found the Nawab guilty and voided his election, was structured precisely to strip away the language of disinterested patronage to reveal the politics of self-interest, dealmaking and coercion that lay beneath. The law seemingly required 14

The language of this case was recommended by Hammond (1920) as a general template for these types of cases in India.

The paradox of patronage and the people’s sovereignty  139

this. Notably, the most powerful defence the Nawab’s lawyers mounted was based on their claim that the filing of the petition itself had been motivated by nothing but political calculation, and thus was little more than an attempt to extend the electoral contest into the courts. They argued that the petition was itself part of the post-polling political machinations of Fazlul Haq and the Nawab of Dacca, who in the wake of the 1937 elections had tried to bring the KPP and the Muslim League into alliance, a deal that required Faroqui to be, in the words of his lawyer, ‘left behind’ (Amrita Bazar Patrika 2 December 1937). In this reading, the petitioners were nothing but paid pawns in a larger political game, in which the processes of election petitioning, and of the law, were themselves fully implicated. But this was a reading the tribunal sharply rejected. Although court proceedings put heavy weight on the moral image of patrons within the new arenas that democratic elections (and press reporting) had introduced, they also provided an arena in which an image of legitimate patronage, even in the midst of political deal-making, was given distinctive legal form. The tribunal’s deliberations thus underscored the importance of law in giving meaning to the image of the patron as a figure defined not just by the calculated manipulation of state power, but as one bound into reciprocal communal structures (and freedom of choice) as well. The law, meant to protect the principles of the voters’ free choice, here stood conceptually apart from everyday electoral politics, for only the law could adjudicate the contradictions between influence and free choice which patronage embodied. Yet, as cases like this illustrated, in the end the law did less to reconcile the paradoxical principles of patronage than to highlight and sustain them as the foundation for democracy. Indeed, the paradoxical relationship of politics and law (which were separate but never wholly separate) mirrored the paradoxes of patronage, which can be understood as underlying the conundrums of democratic practice.

Negotiation and the morality of exchange At the core of such cases lay questions about the morality of patronly deal-making and exchange more broadly. As the case of Nawab Faroqui illustrated, one of the main dangers that elections presented to patronage lay in the apparent power of electoral competition to reduce the reciprocities of patronly deal-making (whether involving the state or not) to nothing but amoral payoffs. Unequal exchange

140  David Gilmartin

was, after all, a ubiquitous feature of the electoral landscape. Yet such dealing remained, as the principles of the law underscored, in pervasive tension with the equally critical principles of reciprocity— and community—that defined the worlds in which patronage operated. The legal interpretation of exchange was thus central to the meaning of elections as an expression of sovereign authority.15 The ultimate legal model for amoral exchange at times of election was the direct monetary purchase of votes. How common the direct purchase of votes was at this time is impossible to say, but there is little doubt that it was widespread. Strikingly, the purchase of votes was on one level completely consistent with the ideals of free agency, for it involved free negotiation. In one case from the Punjab, a witness described graphically how a vote’s price was negotiated: ‘Lala Mukand Lal was offering them [a group of Bawarias and Chuhras in a village in Hissar district] Rs. 4 each, but they stuck out for Rs.  5 each. Finally Lala Mukand Lal agreed to give them Rs. 5…’.16 Here the agency of even those at the bottom of an unequal relationship is clear; if, as Grote had put it, ‘free agency’ was the soul of voting, this was agency indeed. Yet such bargaining for votes was also strictly illegal, for it violated another essential principle of democracy and moral patronage. The purchase of votes short-circuited the freedom of moral choice that lay at the heart of voting, for it was a form of ‘undue influence’ that led voters to substitute immediate gratification for fully rational choice. It thus dramatised vividly the moral problem of a system of representation based on nothing but amoral exchange.17 Much more complicated (and legally critical) than vote-buying, though raising many of the same issues, were the various incentives and inducements involved in pre-polling negotiations among patrons themselves in the run-up to elections, for here we can see the complexity of the relationship between self-interested dealing and 15

This is not to suggest that the tension between morality and exchange was intrinsic to Indian culture (or any culture), but rather that this tension was an old one with multiple manifestations, which were now played out in new ways in electoral law. For a larger discussion of the relationship between gift-giving, commercial exchange and morality in India, see Parry (1989, 63–94) 16 Hissar North (General) Punjab Legislative Assembly case (Punjab Gazette 7 January 1938, Part I, 14). 17 The problem of vote-buying in electoral law has been interpreted in different ways. Although the outright buying of votes is almost universally illegal around the world, the problem as to exactly why it is illegal has been subject to some controversy. For a discussion of the issue in the United States, see Hasen (2000).

The paradox of patronage and the people’s sovereignty  141

the structuring of community. Negotiations among patrons were critical to the shape of electoral competition, for they determined which candidates remained standing when the votes were actually cast. At times, these negotiations involved candidates jockeying for the tickets of political parties; at times they involved balancing the interests of various castes and communities in the constituencies. It was common practice in the late colonial period for many men to file nomination papers (or to put up ‘agents’, vakils, to file nomination papers) as a way to announce themselves as ‘players’ in electoral negotiations. Withdrawals then took place as deals were struck and alliances were formed. The deals and ‘gratifications’ that this involved severely tested the limits of electoral law. The effect of these negotiations was often precisely to limit electoral choices or even to short-circuit popular choice entirely when they resulted in one candidate standing unopposed. And yet such negotiations were also critical to controlling the potentially explosive, divisive conflict that elections had the potential to unleash. Indeed, patronly negotiations before elections reflected the contradictory interplay between calculated political self-interest and a concern for community order that defined the paradox of patronly authority more generally, and which shaped the emergence through elections of a ‘popular voice’. The difficulties such negotiations presented for electoral law were displayed vividly in two connected cases from the Punjab’s Dera Ghazi Khan North (Muhammadan Rural) constituency in 1937 and 1938. The first case involved the 1937 election to the provincial Assembly of Khwaja Ghulam Murtaza, a relative of the hereditary custodian (sajjada nishin) of the most important Sufi shrine in the north of the district (located now in Pakistan), a district long dominated by powerful and rivalrous Baloch chiefs. The shrine at Taunsa had its origins in the north Indian Chishti revival of the late 18th and early 19th centuries, and had been founded by Khwaja Suleman Taunsvi (d. 1850), a Sufi whose regional reputation hinged on piety and patronage alike. In the early 20th century, the shrine had fallen prey to a bitter internal succession dispute, which, after many appeals and iterations, was ultimately decided by the Privy Council in London in 1921.18 Subsequently, factional networks radiated outwards from this dispute, 18

Khwaja Muhammad Hamid vs. Mian Mahmud and others (Privy Council Appeal no.  118 of 1921) Lincoln’s Inn Library, London. The petition contesting Khwaja Ghulam Murtaza’s election in 1937 was filed by Khwaja Sadid-ud-Din, the sajjada nishin belonging to the rival branch of the family.

142  David Gilmartin

providing a critical backdrop to the 1937 elections, with prominent local landowners and tribal chiefs across the area aligning themselves with different sides. The dynamics of the election thus disclosed the close interconnections between patronal deal-making and peacemaking among rivals in a highly factional environment. These interconnections were dramatised in the election petition filed to challenge Khwaja Ghulam Murtaza’s victory in 1937. The petition centred on two remarkable, interlinked documents negotiated against the backdrop of these factional rivalries, documents that concluded with the statement that any breaker of the reciprocal commitments contained in them would ‘be regarded as the breaker of an oath’. These were, in effect, contracts negotiated between erstwhile factional rivals, Khwaja Ghulam Murtaza on one side, and the Khetran Sardars (the dominant tribal chiefs in the north of Dera Ghazi Khan) on the other. To secure the withdrawal of a candidate that the Khetrans had initially put forward to oppose him in the election, Ghulam Murtaza promised that if this candidate withdrew, he would give full support in the next election to whatever candidate the Khetrans then decided to support. The Khetrans agreed to this and promised in a second agreement to ‘render all possible help ourselves and through our friends’ for the Khwaja’s 1937 election. Both documents were underwritten with material considerations. Ghulam Murtaza promised not only to reimburse the withdrawing candidate’s security deposit (250 rupees), but also, and most importantly, to provide a new house at Taunsa for the Khetrans. This house, which would be ‘safe from erosion’ and ‘close to a motor road’, was to replace one that the Khetrans had given to the old sajjada nishin in the late 19th century, but whose site had since been taken over by Ghulam Murtaza’s uncle and had become the focus of litigation. Material exchanges thus underwrote a larger political bargain. Given the structure of the law, however, these material exchanges alone became the target of electoral litigation. The charge was that under the law these were corrupt payoffs and illegal gratifications, which vitiated the election. The evidence presented before the tribunal did not disguise the fact that these ‘gratifications’ were embedded in larger structures of patronly conflict and obligation. In their agreement to support Ghulam Murtaza, the Khetrans promised to ‘make peace’ with other zamindars (landholders) in the area in a way that was ‘consistent with our dignity’, and to ‘consider the

The paradox of patronage and the people’s sovereignty  143

loss or gain of Khwaja Ghulam Murtaza as our loss or gain’. For his part, Ghulam Murtaza agreed to mediate a reconciliation between the Khetran Sardars and both his uncle (leader of one of the main Taunsa factions) and Nawab Jamal Khan Leghari, the district’s most powerful Baloch patron. He also agreed to recognise ‘the loss or gain of the Sardars to be my loss or gain’. But while the tribunal took note of this seemingly formulaic language of social bonding, it did so only as it was relevant to analysing the documents’ provenance and authenticity. None of this language prevented them from ruling that the charges of illegal payoffs and gratifications were genuine and that this violated the letter of election law. The result was that the election was voided, whatever the evidence as to how material exchanges were intertwined with community peace-making (Sen and Poddar 1951, 271–274).19 The case showed the deep-seated problem that exchange presented for the logic of electoral law. While the case revealed a world of patronly deal-making embedded in a larger world of community-based social relations, it also showed the critical place of monetary and material payoffs as the touchstone for corruption. Yet, as the follow-up to this case suggested, a more complex vision of patronly exchange, with all its underlying paradoxes, also found a place in electoral adjudication. This became clear in the litigation over the same Dera Ghazi Khan North Assembly seat in the wake of a by-election held the following year to fill the now empty seat. Again, patronal deal-making loomed large, prompting another petition, which focused on the material transactions leading to the withdrawal of candidates in the run-up to the election. This time the negotiations were of wider scope, reaching all the way to the leadership of the elected provincial government in Lahore. The principal player in these negotiations was Nawab Jamal Khan Leghari himself, a close ally of Sir Sikander Hyat Khan, the new Premier and provincial leader of the Punjab Unionist Party (which had won the elections in 1937 and formed the new government under provincial autonomy). The charge in this case was that in early 1938 the Nawab had organised a meeting in Dera Ghazi Khan city (in the home of the MLA representing the Dera Ghazi Khan Central constituency), at which he had convinced the assembled men of influence to appoint him the sole arbiter for selecting a candidate for the now open Dera 19

For a slightly fuller account of this decision, see also the Punjab Gazette (17 December 1937, Part I, 1724–1728).

144  David Gilmartin

Ghazi Khan North seat, whom they all promised to support. As the tribunal acknowledged, this agreement was part of a deal intended to allay the ‘animosities and dissensions’ that the new election was bound to add to an already faction-ridden environment. According to the petition, the agreement was that if the Nawab’s nominee ran unopposed, the candidate would pay the Nawab 1000 rupees toward the support of a political party, which the Nawab was establishing with himself as head. When, despite this agreement, another candidate decided to stand for election anyway, he was induced to withdraw in favour of the Nawab’s nominee with a payment of 800 rupees and a promise of nomination to the Special Jirga (tribal court) and the District Board. The arrangements suggested the Nawab’s high standing as the district’s most prominent patron, whose influence had now been bolstered by his close ties to the provincial government. Again, the broader contours of the social order in which prominent patrons were embedded framed the monetary exchanges that were central to electoral bargaining. Once again, legal scrutiny focused on the montary payments that lay at the heart of the deal. It would no doubt have been easy for the new tribunal to find, once again, that these transactions amounted to bribery, for they looked virtually identical to the ‘gratifications’ in the earlier case. In this case, however, the election tribunal ruled differently. Although it had clear evidence of the payment of money, the tribunal accepted the Nawab’s contention that this money was not for himself but for the Unionist Party’s coffers.20 Even more importantly, in assessing the case, the tribunal made it clear that the social standing and reputation of the parties involved weighed heavily in their considerations. It clearly mattered that Leghari was the district’s most respected patron. ‘The Nawab holds a high position in the district’, they declared, ‘and we are not prepared to disbelieve him without very clear and cogent reasons, particularly as most of the petitioner’s witnesses are disgruntled candidates’ (ibid., 2752–83).21 However much the letter of the law might have dictated that these be 20

This claim was supported by the testimony of Sir Sikander, the provincial Premier. Packaging payoffs as contributions to political parties in fact helped to translate these transactions from the realm of ‘individual gratifications’ to that of ‘community interests’, thus serving to legitimise them. 21 The unwillingness of recipients of bribes to testify against those buying votes was a common difficulty, noted by many tribunals, in adjudicating bribery or vote-buying charges.

The paradox of patronage and the people’s sovereignty  145

viewed as amoral exchanges, the idea of patronly reputation was now mobilised to give a different meaning to these transactions. Whatever their dealings, the roles of patrons as lynchpins of the socio-political order could not be ignored if elections were to have political meaning. This petition was therefore rebuffed and the outcome of the election was allowed to stand.22

Patronage, elections and reputation Though it was thus in the very nature of politics that negotiations among patrons were easy to view as self-interested exchanges, this case made clear that the intangible power of reputation also influenced the law. In some cases, a concern for reputation entered directly into legal proceedings as tribunals sought to assess a candidate’s (potentially corrupt) intent, in which past actions and reputation naturally mattered (Hammond 1923, 132). But more commonly, reputation entered less openly into legal reasoning, as tribunals sought to negotiate the paradoxes that patronage embodied in its relation to both the concrete realities of power and the principles defining the people’s sovereign choice. However explicit the concern with the corrupting effects of payoffs and amoral exchange, a concern for reputation also remained vital to assessing the meaning of authority within a framework of voters’ free agency. We can get a clearer sense of the importance—and ambiguity— of patronly reputation within the law if we consider how it took formal cognizance of the contrasting realms of vakil and dharmik relations, as defined by Bayly. Given the intense political deal-making that accompanied elections, support for religious and community institutions was often a particularly potent instrument for political candidates to use in countering the images of self-serving deal-making that arose from election campaigns. Yet, if dharmik patronage had the capacity to cloak the self-interested pursuit of power in the mantle of 22

Interestingly enough, the Unionists tried to broker a compromise to end the case before the tribunal reached its decision. Part of the proposed deal was, apparently, that the elected candidate would resign if the petition was withdrawn and the inquiry halted. We can speculate that Sir Sikander, Nawab Muzaffar Khan (Sikander’s cousin and political adviser) and Jamal Khan Leghari did not wish to risk their reputations by opening their deals to public scrutiny by the tribunal (and the press). The petitioners, on the contrary, tried to use evidence of this attempted deal as an effective admission of guilt by the parties involved. But this implication was also rejected by the tribunal, which ruled in the Nawab’s favour (Sen and Poddar 1951, 281–282).

146  David Gilmartin

communal devotion and service, its impact was often ambiguous, for the electoral process also had the power to expose such dharmik acts to legal and public scrutiny as themselves products of calculated, or even hypocritical, politics. How election law treated this relationship gives a window onto how the larger relationship between patronage and democratic practice evolved. The problem posed by dharmik patronage arose in a number of election cases in the years before 1947. In a petition challenging the 1937 election to the Punjab Legislative Assembly in the Amritsar and Sialkot (General Rural) constituency, for example, the winning candidate was accused of paying 50 rupees (and of promising another 150) to repair a dharamsala (pilgrim rest house) run by the Narowal Sanatan Dharma Sabha, a religious organisation in Sialkot district. This was a dharmik offering, which would normally have been considered an act of charitable benevolence. But the case exposed just how fine the line between dharmik benevolence and electoral manipulation actually was. In response to the gift, members of the Sabha initially decided to ‘vote for the respondent and to go to the polling station in a body’. But when the candidate failed to pay the full promised amount, members of the Sabha turned against him, filed a petition and testified before a tribunal that this had been nothing but a payment for votes all along, a view ultimately endorsed by the tribunal.23 A slightly more complicated situation arose in 1926, during the UP Legislative Council election in Farrukhabad. In this case, several complaints arose over gifts made during the campaign by the winning candidate, Raja Durga Narain Singh of Tirwa. Singh, who was a prominent landlord, donated both to Ram Lila committees (which staged Ramayana performances during the Dussehra celebrations) and to a fund for Hindu prisoners, who were arrested after a communal riot in Farrukhabad. As these gifts were made in proximity to the election, the tribunal found all of them suspect. Nevertheless, the judges decided that donations to prisoners were a response to the community’s current needs and could hardly be seen as electoral calculation: ‘A calamity had fallen on some of the Hindus’, they wrote, ‘and a man of the respondent’s position must contribute’ (Hammond 1936, 358). But the Ram Lila donations were a different matter: ‘The respondent had never before given any donation to the Kanauj Ramlila committee’, 23

The tribunal overturned the election on these and other grounds (Sen and Poddar 1951, 24–25).

The paradox of patronage and the people’s sovereignty  147

they noted; ‘He gave this donation during the election campaign and this generosity cannot be the outcome of any other object but to gain voters for himself’ (ibid.). As the tribunal admitted, ‘it is sometimes difficult in connection with corrupt practices to state when charity ends and bribery begins’. But in this case an otherwise charitable, dharmik act appeared as ‘nothing more than a specious and settled form of bribery, a pretext to veil the corrupt practice of gaining or securing votes of the recipients’.24 That is, in the eyes of the law, selfinterested electoral calculation corrupted what would otherwise have been perceived as community-minded, patronly benevolence. Implicit in this view of election-time patronage was the potential multi-directionality of moral flows during the special time surrounding elections.25 If electoral calculation held the potential to drain patronly acts of their moral substance, acts of dharmik benevolence at the time of elections also had the power to legitimise self-interested electoral dealings. This understanding stood behind several election petition decisions. One example comes from the 1923 election to the Saran South (Non-Muhammadan Rural) constituency of the Bihar and Orissa Legislative Council. In that case, a losing candidate charged that the winner had offered corrupt inducements to voters by distributing food as ‘relief work’ and then claiming this as grounds for electoral support. As in the case of the Ram Lila donations in Farrukhabad, the charge here was that self-interested politics had, in the process of electioneering, turned benevolent patronage into a form of bribery. In this case, however, the tribunal ruled that the moral flow was reversed. ‘Innocent benevolence’, as the tribunal put it, was a perfectly valid basis for laying claim to the gratitude and support of the people—even in the context of an election (Hammond 1920–27, II, 254–255).26 Such benevolence in fact defined the very meaning of 24

This language was taken verbatim from an earlier British case, which the tribunal cited as precedent (Hammond 1936, 357–358). 25 For some discussion of the importance of cyclical, special election time (which gained increasing importance after 1947), see Gilmartin and Moog (2012, 136–148). 26 Also see the Amritsar and Sialkot case mentioned earlier, in which the tribunal found that the candidate’s contribution of 1000 rupees and 10 marlas of land in November 1936 for a Brahmin Bhawan at Sialkot could not be seen as corrupt simply because of its proximity to the upcoming elections: ‘It is not disputed that he is by far the most prominent and wealthy member of the Brahmin community in Sialkot. His contribution towards the construction of the proposed Brahmin Bhawan had to be commensurate with his position in the community’ (Punjab Gazette 4 March 1938, Part I, 285–286).

148  David Gilmartin

‘good’ patron–client relations, based on mutual, if unequal, reciprocity. Instead of being corrupted by the electoral process, such dharmik acts legitimised it. Indeed, such decisions acknowledged the central importance in elections of a public image of selflessness that could morally tame the patronly deal-making required by electoral success. But the importance of patronly reputation (and image) was also embedded in the law of elections in a more concrete way. Electoral law specifically forbade a candidate from knowingly making or publishing false statements about an opposing candidate’s ‘personal character or conduct’ that might be ‘reasonably calculated to prejudice the prospects of that candidate’s election’ (Hammond 1936, 747). This law was, in effect, the flip side of the law of ‘undue influence’. It reflected the law’s formal recognition of the importance of images of moral character to the very legitimacy of patronly influence in elections. To besmirch unfairly the personal character of another candidate was to threaten the very principles that gave meaning to the people’s electoral voice. The law thus sought to separate conceptually the personal character, or ‘inner man’, of the patron (‘the man underneath the politician’, in the words of a later tribunal), from the calculating ‘political man’ enmeshed in strategic dealings and exchange (Aiyar 1953, 248–280).27 This represented, in effect, a framework for giving the paradox of patronage legal form as a foundation for legitimate action in electoral politics. This does not mean that the legal line between a candidate’s private, inner character and his public, political self was easy to draw. As one tribunal explained, ‘The public character or conduct of the public man is public property’ (Sen and Poddar 1951, 332–335). ‘Criticism of a candidate in his public capacity, however ill-mannered or unhappily worded or even unfair or exaggerated’, would not, another commented, ‘render an election liable to be set aside’ (ibid., 132). But the conundrums facing election tribunals in distinguishing such public actions from those impinging on private character were considerable. In a 1920 case from the Ballia (NonMuhammadan Rural) constituency of the UP Council, the defeated candidate, who was the manager of the Dumraon estate, sought to overturn an election on the ground that his opponent had destroyed his reputation by publishing false statements accusing him of evicting 27

A similar concept was embodied in the 1960s in the legal use of the phrase ‘the man behind the politician’ (Shankardass 1968, 252–258).

The paradox of patronage and the people’s sovereignty  149

tenants in the wake of severe floods and allowing cows to be sold to butchers for slaughter at a local fair. In this case, the tribunal found that these statements were about public acts. Yet it ruled nevertheless that the false statements impinged so thoroughly on perceptions of the candidate’s character that they fell under the prohibitions of law (Hammond 1920–27, vol. 1, 27–44). Similarly, in a 1946 case from the Nimar (Muhammadan Rural) constituency in the Central Provinces, a tribunal found that false statements accusing a Muslim League candidate of secretly accepting support from the Congress, though related to politics, were actionable because they were an ‘attack on the personal character of the petitioner as a man of no honesty of purpose’ (Sen and Poddar 1951, 627–633). The blurring of boundaries between public and personal aspects of candidates, which such cases entailed, was deeply frustrating for many tribunals. As one tribunal noted, almost any public action could ‘by a more or less elaborate process of reasoning, be shown to contain an indirect reference to the personal character or conduct of that person’ (ibid., 334). But if the conceptual separation of an ‘inner’ moral man from an ‘external’ political one was not always easy to delineate, the juxtaposition of these two realms was critical to the law’s structure, framing the legitimacy of patronly influence within a system embodying the people’s voice. The tension between these realms mirrored the longstanding paradox of patronly authority. But in the context of elections, the conceptual separation of these realms in the law could also be interpreted as a legal invitation to what we might call institutionalised hypocrisy as a defining feature of electioneering. The question of reputation’s role in elections was not tangential. As one tribunal put it, ‘when a candidate is attacked in his private capacity, or his honour, integrity or veracity has been assailed’, the very foundations of the electoral process are threatened, for what is challenged is the integrity of the people’s choice (ibid., 132). But the problem of personal reputation was necessarily juxtaposed against the critical reliance of candidates on amoral dealings for electoral success. From the perspective of ‘rational choice’, hypocrisy was built into this juxtaposition, for elections seemingly required candidates to strategically (and self-consciously) create—and publicly project—an image of inner morality that was directly belied by many of their political acts. Indeed, the law’s separation of a private moral essence from overtly political acts suggested a moral chasm at the very

150  David Gilmartin

heart of the political system. Little wonder that the tension between the realities of politics and the image of ‘the man underneath the politician’ shaped a popular vision of the electoral arena as morally suspect (see Chapters 6 and 9 by Piliavsky and Price in this volume). ‘Politics’, in the eyes of many, was a ‘dirty’ business, and it still is today (Ruud 2000; Harriss 2005). Yet an emphasis on the relationship between electoral structure and ‘hypocrisy’ hardly captures the political meaning of this contradiction, for to assert this would be to assume that calculating candidates somehow stood apart from the larger contradictions of political order in which patronage (and electoral democracy) were embedded. These contradictions were in fact integral to the morality of Indian political order more generally, and they encompassed politicians and voters alike. The language of ‘selfless service’ (seva) to the community, which permeated electoral rhetoric, was as integral to the self-fashioning of most patrons as it was to their constituents, who insisted on hearing it, even as they labelled ‘politics’ morally corrupt.28 The instrumental and moral realms were—in their very contradiction—both understood to be the necessary building blocks of political order in an architecture resting on the imagined sovereignty of the people. This was reflected in, and sustained by, the law. If the paradox of patronage was an old one, it took on critical new significance in an emerging electoral order defined by the contradictory roles of politics and law alike in giving shape to the people’s sovereign voice.

Concluding case To conclude, I turn to one particularly revealing (if in the end politically  insignificant) case, suggesting the connection between patronage and democracy in the early years of post-independence 28

For a discussion of the importance of the language of seva in politics, see Mayer (1981, 153–173). This dynamic may also help to explain the prominence in Indian elections of appeals to ethnicity, caste and religion. Kanchan Chandra (2004), for instance, has argued that voting for a candidate of one’s own ethnicity is a rational way for voters to deal with limitations in their actual knowledge of candidates, for it provides them the important piece of information that there is a common bond of community, transcending the self-interest of the candidate, linking the candidate to the voter. This is probably true, but it is important to see that this very perception of community bonds is indicative of an approach to voting that involves a search for connection that goes beyond calculated interest, precisely because voters recognise that ‘politics’ is a realm saturated with candidates’ self-interested concerns.

The paradox of patronage and the people’s sovereignty  151

India, a case in which the gaze of electoral law reached inside the inner sanctums of the Teen Murti House and focused on the greatest 20thcentury Indian patron of all, India’s first Prime Minister Jawaharlal Nehru. If patrons are not-quite-kings, who share something of kingship yet remain subject to an exterior sovereign, this description fully suits Nehru, who remained deeply committed throughout his years as Prime Minister to the ideal of Indian people’s sovereignty. In the 1950s, Nehru was involved in a remarkable case in which the legitimacy of patronly influence came under legal scrutiny.29 It focused on a contest for the New Delhi parliamentary constituency. In 1952, when independent India ran its first general election, two of India’s most politically prominent women, Sucheta Kripalani and Manmohini Sahgal, had vied for this seat. Sahgal was Nehru’s niece and she ran with a Congress ticket.30 But she was defeated by Kripalani, who ran with a ticket from the Kisan Mazdoor Praja Party, a party formed in 1951 by her husband, Acharya J. B. Kripalani, who split from the Congress in protest against the election of a right-wing Hindu, Purushottam Das Tandon, as Congress president in 1950. Despite running against the Congress, Kripalani’s electoral success probably owed much to the continuing respect she commanded within the Party.31 Eventually Kripalani rejoined the Congress, and when both women applied for the New Delhi Congress ticket to stand for election in 1957, it went to Kripalani.32 Sahgal was outraged by the Party’s failure to reward her long-standing loyalty, and announced her intention to run independently. Fearing that Sahgal’s candidacy would split the Congress vote, the Delhi Congress Committee formally expelled her from the Party, no doubt hoping that this would cause her to withdraw. Yet she remained adamant about contesting, declaring that ‘there is an electorate between God and [the] Congress High Command’. As a result, a meeting between Nehru and Sahgal was arranged to take place at the 29

Amir Chand versus Smt. Sucheta Kripalani, Election Tribunal, Delhi (Aiyar 1959, 209–253). All subsequent quotations not otherwise referenced are from this report. 30 Sahgal’s father was the son of Pandit Motilal Nehru’s sister. She was technically Nehru’s first cousin once removed, but he referred to her as his niece. 31 According to Sahgal’s later memoirs, ‘Congress leaders were determined to help Mrs. Kripalani at every turn’ (Sahgal 1994, 137). For a further discussion of this contest, see Singer (2007, 72–78). 32 Sahgal had in fact filed an election petition challenging Kripalani’s 1952 election on the grounds of irregularities in her election expenses. This ultimately led to the voiding of Kripalani’s election, but not until her term was almost completed (Singer 2007, 72–78).

152  David Gilmartin

Teen Murti House on 12 February 1957. According to Sahgal, she was summoned to meet Nehru; according to Nehru, it was Sahgal who requested the meeting. Whatever the case, the meeting was followed by a sudden change of heart on the part of Sahgal, who withdrew her candidacy the following day. She said: ‘Panditji, our beloved leader, appealed to me last evening and so I decided to retire’. As a result, Kripalani won an overwhelming victory over her nearest opponent. Nevertheless, a petition was subsequently filed by one of the defeated candidates alleging that Nehru had exercised ‘undue influence’ in forcing Sahgal to withdraw, thereby materially affecting the outcome of the election. As the petition alleged: ‘Shri Jawaharlal Nehru who wields a stupendous power in India both as Prime  Minister and acknowledged Congress Leader, sent for Smt. Manmohini Sehgal … and exercised undue influence. …This undue influence took the form of coercing Mrs. Manmohini Sehgal to retire from the contest’. No one else was present at the meeting, so it is impossible to know what exactly was said. As Sahgal herself remembered it, Nehru ‘used the sentimental, family approach with me since he was head of our extended family’ (Sahgal 1994, 143). Nehru later testified that, although he could not remember his precise words, ‘I must have told her that I had not liked the idea of a contest between Congressmen’. As the tribunal made clear in its ruling, it could not find undue influence when even Sahgal herself did not allege that she had been subjected to direct threats. The case was decided in Kripalani’s favour and the election stood. But in a telling twist, the tribunal’s discussion hinged significantly on Nehru’s image as a political patron. Interestingly, the tribunal referred to the same 19th-century Irish case quoted earlier which referred to the Catholic priest’s ‘proper exercise’ of electoral influence through his patronly ability to ‘counsel, advise, recommend, entreat and point out the true line of moral duty’. This language took on particular poignancy in this case because Sahgal, as Nehru’s younger female relative, looked to him, in the words of the tribunal, ‘as an elder of the family’, suggesting a relationship of sympathy and benevolence. But the counsel for the petitioners cast this against the power of political calculation, especially in the openly political electoral context, to transform these patronly bonds into something quite corrupt. Pandit Nehru is a leader of the Congress and is in a position to exercise influence on a vast field … of patronage. That a politician is always for politics … is borne [out] from the statement of Shri Nehru. … Politicians are

The paradox of patronage and the people’s sovereignty  153

more concerned with the success of politics than other values of life and in his own words, when he says that he was interested in the success of Congress candidates, including Mrs. Sucheta Kripalani, Shri Nehru [made clear that he] would do anything subject to his being a politician and a leader for the success of Mrs. Kripalani.

However seemingly innocuous the meeting of Nehru and Sahgal may have been, the imperatives of Nehru’s political world were so overwhelming that they defined the nature of all his influence in the realm of electoral politics. The tribunal was quick to reject this line of reasoning, which made Nehru’s influence ‘undue’ by definition. Yet the tribunal was at the same time thrown off balance by the petitioner’s lawyers’ attempt to extend this logic even to Nehru’s testimony before the tribunal. However ‘truthful’ a man Nehru might be, the lawyers argued, in a case like this even he could not help but be ‘overwhelmed by the spirit of his party politics’, even in the witness box. There was, in other words, no room in the matters of party bargaining and discipline for the exercise of moral character to define the ‘man underneath the politician’, for in the realm of politics political calculation always ruled, and even Nehru’s testimony had to be seen in this light. But the tribunal ultimately waxed cosmic in rejecting this. It is ‘preposterous’, they declared, to suggest that ‘a person of the stature of Pandit Jawaharlal Nehru’ could simply succumb in a case like this to the imperatives of political calculation and ‘the stress of party politics’. Here was a man who, at a time ‘when forces of power politics threaten the very spiritual values of life and when atomic desolation is threatening mankind’, nevertheless has continued to bring ‘his private conscience’ to bear with ‘truth and fearlessness’ on the politics of the world. Even if, as the lawyers argued, it is true that ‘whoever comes in contact with Panditji gets influenced, this is a tribute to his personality and by no stretch of imagination or reasoning can it be treated as undue influence’. However much the morality of political ‘influence’ was subjected to the law’s scrutiny, Nehru remained here the ultimate patron, the image of his own internal conscience, personality and selflessness serving as the imagined antidote to the inescapable operation of unequal influence within an electoral system saturated with the instrumental calculations of politics. In an important sense, his selfless identification with the nation’s community—the commitment of the ‘man behind the politician’—ultimately stood as surety for the people’s sovereignty.

India’s demotic democracy and its ‘depravities’ in the ethnographic longue durée


Anastasia Piliavsky


n the spring of 2012 the Socialist Party in the north Indian state of Uttar Pradesh (UP) won a landslide election to the State Legislative Assembly.1 The victory became India’s political event of the year, marking the return of the notoriously corrupt and criminalised political party to the helm of India’s flagship state. A year earlier in 2011 political activist Baburao ‘Anna’ Hazare had gone on an indefinite anti-corruption hunger strike, which became the most dramatic political statement made in the same country that year. The stated ambition of Hazare and his (mostly middle-class) followers was to wipe the tarnish of corruption off India’s biggest political trophy: its status as the ‘world’s largest democracy’. Celebrations of India’s democratic boom and lamentations over its corruption crisis are equally audible in political statements made at home and abroad. Reports on the UP election were but an instance of this bipolar view.2 Here commentators noted that while electoral participation reached some of the highest national and global levels, peaking in 2012,3 corruption also rose to astonishing proportions, even by South Asian standards.4 1

This chapter benefitted from discussions with Naor Ben-Yehoyada, Mattison Mines, and especially Piers Vitebsky and John Dunn. Its earlier version was presented at the Department of South Asian Studies at Harvard, whose participants I thank for their useful comments. 2 For anthropologists’ remarks on the spread of ‘corruption’ and political ‘criminalisation’ in the press, see for instance Béteille (1994, 565), Gupta (1995; 2005) and Parry (2000). On India’s democratic boom, see Banerjee (2007) and Yadav (2009a; 2009b). 3 The first round of polling in the 2012 elections in UP saw a record voter turnout of 62–64% (Election Commission of India). Compare this, for instance, to voter turnout at the local council elections in the South East of England: 44.3% (Louise Stewart. 2012. BBC. ‘England: Local Elections, One Week to Go’. 4 Between 2002 and 2007, the last time the Socialist Party was at large in UP, 51% of the Legislative Assembly had criminal histories, giving the party the accolade of

India’s demotic democracy and its depravities  155

This paradox raises several questions. How can the two Indias— the populist and the corrupt—exist side by side, and not just coexist, but also seemingly reinforce one another? Why do so many Indian citizens continue to elect so many corrupt politicians into office, supporting what looks like rampant political chaos in their state? Why do the reportedly widespread depravities of India’s political life fail to instil endemic apathy in its electorate? And what does this tell us about the relation between ‘democracy’ and ‘corruption’—and indeed about what democracy is, what it is not, and what we hope it may be—not only in India, but also in the wider world? This chapter invites readers to reflect on these questions by taking them on a tour of rural Rajasthan at the time of the 2008 State Assembly Election. As we watch the run-up to polling date in the market town I shall call ‘Danapur’, we will learn about the sentiments behind voters’ choices, and indeed about what causes them to vote at all. We will also learn about how politicians and their electors relate to one another, what hopes people invest in their political leaders, how politicians try to win their votes and what disappointments await both on the other side of the polls. To a political scientist much of what I describe—the feasts organised by politicians, the distribution of cash and gifts, and the popular choreography of supplication—will look like ‘clientelism’, or the purchase of votes. I will argue here, however, that the participants’ own sense of engagement in electoral politics rejects this view. As they see it, the choices they make are not instrumental, but distinctly moral: grounded in what they see as the sound basis for political authority, in how they imagine good politicians and good relations with them, what obligations they think these relations entail and how they envision a worthwhile political life as a whole. Needless to say, in Danapur just as elsewhere in the world, politicians constantly betray this moral vision. But the cynicism of practical politics does not destroy its normative sense. No matter how often electors may be disappointed by politicians, and indeed grow despondent about politics as such, they continue to judge political leaders and make political choices through a widely shared value set.

goonda raj, gangster rule (Financial Times 2 May 2007, 11; ibid. 6 March 2012; also Michelutti, Chapter 12 in this volume). A recent pre-election study of candidates fielded by the major political parties in the province revealed that more than a third had criminal histories (Ians 2012).

156  Anastasia Piliavsky

In rural Rajasthan, the normative logic of electoral politics is substantially articulated in the old and widespread idiom of ‘patronage’—the relation of donors and servants—which substantially defines the roles and obligations involved in the electoral process. This relational formulation shapes politicians’ styles, voters’ preferences and popular modes of political participation. It also informs the ways in which people make political demands, insist on accountability and define ‘corruption’. The normative form I call ‘patronage’ is not confined to political and economic life but shapes relations on a much broader temporal and social scale. To give a sense of this scale, I cast my ethnographic snapshots in the ethnographic longue durée, drawing on the largest archive of cumulative anthropological wisdom about patronage in South Asia. This archive is the body of jajmani studies, which has long been confined to the dustbin of South Asianists’ research themes, but which still hold lessons of vital substance for anyone who attempts to understand subcontinental politics today. I do not call on the jajmani archive in an antiquarian spirit, to rehabilitate vintage ethnography for its own sake. Nor do I argue that the same set of practical arrangements once observed by anthropologists in the villages are still replicated throughout South Asia (or that indeed they ever were). I invoke it instead to draw attention to some durable relational principles of north India’s rural political life, whose import cannot be grasped without awareness of their historical and social reach.

Know Wiser? In 1936 William Wiser, an American missionary and Chicago-trained anthropologist, described what he called ‘the Hindu jajmani system’ in a little book entitled just that. He wrote that in a north Indian village where he conducted research, castes (jatis) related to one another through formalised, durable and often inherited exchanges of services for payments and gifts: The priest, bard, accountant, goldsmith, florist, vegetable grower and so on are served by all other castes. In turn each of these castes has a form of service to perform for the others. Each in turn is master. Each in turn is servant. Each has his own clientele comprising members of different castes which is his ‘jajmani’ (1936, 10).

India’s demotic democracy and its depravities  157

In the days of positivist, village-bound ethnography Wiser’s neat formulation had great appeal and an entire generation of anthropologists followed suit, proceeding to describe jajmani exchanges in villages across the subcontinent’s length and breadth. Over the next 5 or so decades, what Wiser originally described as an order of rules and conventions was consolidated by anthropologists into a pan-Indian ‘system’ that bound villages into closed exchange communities.5 At the height of the jajmani era (1960s–1970s) this system appeared as a rigid structure of transactions with a single ‘dominant’ patron family at its head: a ‘system corresponding to the prestations and counter-prestations by which the castes as a whole are bound together in the village, and which is more or less universal in India’ (Dumont 1980, 97). This ‘system’ took on ‘a life of its own through the various simplifications and idealizations of innumerable textbooks and lecture courses’ (Good 1982, 31). The reification of relational principles into a transactional system went hand in hand with the installation of an immutable, age-old village republic—an Indian ‘village-community’ à la Henry Maine (1861)—at the centre of South Asianist anthropology (Caldwell 1991, 3). By the early 1980s, the jajmani edifice began to crumble. Critics argued that it was at once too broad and too narrow an analytical category. Some observed that it was neither a pan-Indian nor an ageold phenomenon but an institution observed only on a very limited historical and spatial scale.6 Others pointed out that its origins, once presumed to have medieval (Beidelman 1959) or even ancient (Gough 1960, 89; Wiser 1936, xxv) provenance, could only be traced back in written record to the mid-19th century, or perhaps even to Wiser’s 1936 account.7 Yet others showed that jajmani-type relations were not restricted to villages, but extended far beyond village bounds into broader political, economic and ritual spheres.8 They were as 5

For classic examples, see Kolenda (1967), Mandelbaum (1970, 161–162) and Dumont (1980, 98ff). 6 See Commander (1983, 287), Neale and Adams (1990, 52–54) and (Fukuzawa 1972). 7 See Mayer (1993), Commander (1983), Fuller (1989) and Raychaudhuri (1984, 9). 8 Chris Fuller (1977), for instance, argued that historically people made jajmani offerings to village-based jajmans as much as to supra-local military elites and that it is only colonial meddling with the local political and economic structures that truncated jajmani exchange, leaving anthropologists with the artefact of a villagebound, ‘caste-based economic system’ (Fuller 1977, 107–109; 1989; also Wolf 1966, 47–57; Karanth 1987, 2217).

158  Anastasia Piliavsky

much a feature of exchanges within castes as between them.9 By the late 1980s, anthropologists reached a consensus: given the great variation in the contexts and ways in which jajmani relations manifested themselves,10 the ‘system’ did not correspond to any actually observed phenomenon and therefore it did not exist.11 Its demise was in keeping with the spirit of the times, which saw the last days of village ethnography (Fuller and Spencer 1990). This shift precipitated broader changes in Indian anthropology, some positive and others less so. Among the former was the ousting of the myth of a timeless village republic. Among the latter was a wholesale, and rather counterintuitive, disappearance of patronage from the radar of Indianists’ concerns. One might have expected that, once rescued from its village island, patronage would acquire a new lease of life, especially given its persistence in current politics. Instead, patronage altogether vanished from anthropologists’ writings, including the burgeoning literature on electoral politics, corruption and the state. The trouble with the jajmani critics, no less than with its advocates, was that they tended to see village relations as transactional networks, or sets of exchanges with a materially predetermined form. The more they got involved in either erecting or dismantling the transactional framework, the more they neglected the shared substance of what they saw—the ideas that shape the transactions. Whereas for Wiser the ‘system’ was a set of ‘rules and conventions’ which took on various material forms, three decades later this was a rigid transactional structure. While dismantling this structure, its critics lost 9

Anthony Good showed that prestations that have been conventionally treated as exclusive to customary exchange between service-castes and their jajmans were also part of exchange within castes at various rites of passage (Good 1982, 26; also Raheja 1988b). 10 These included different identities of participants, the types of payments and services involved, the frequencies of exchange and the tenure of such relations, among other things. 11 In Simon Commander’s summary, ‘the fragmented and rather partial sense in which the jajmani structure can be found to be functioning’ bears ‘a very dim resemblance to the pure model’ (1983, 307 and 310), that is jajmani exchanges did not add up to an isolable, village-based, intercaste exchange ‘system’. On variation in payments, see Grierson (1926 [1885], 317–322), Lerche (1993, 246) and Kothari (1994). The final blow was Chris Fuller’s seminal article in which he pronounced that ‘there is no pan-Indian jajmani system of the patron-client type’ and that the term is ‘a complete misnomer’ (1989, 37 and 41).

India’s demotic democracy and its depravities  159

sight of patronage as a widespread and durable relational mode that embodies the principles of relatedness and exchange (Karanth 1987). Yet, as the critics themselves convincingly showed, jajmani relations never added up to an isolable, self-contained, material system, but permeated an extremely wide range of settings—from the formal political to the informal and intimate, between and within castes, in households and around the marketplace. These principles still inform—if often in new material forms—modern political processes. What jajmani studies crucially showed was that in South Asia ‘patronage’ was not an economically or a politically isolable institution, but a pervasive social norm that contained the key principles of relations at large. The donor–servant relation was the basic formula through which people exchanged things, exercised power and related socially, and through which their identities effectively took shape. Classic accounts highlighted the degree to which personal and group identities were bound up in this relation, which was thought to contain in a concrete and visible form the basic principles of caste and hierarchy at large. This totalising view may jar with mainstream styles of current anthropology, but it underscored an essential point that often escapes analyses of today. Patronage was not a transactional and purely economic practice but an existentially vital form. Giving was not only an act of exchange but also an intrinsic aspect of the way donors and recipients related to, and defined, one another. As the ‘Chicago school’ analysts (led by McKim Marriott and Ronald Inden) showed, gifts quite literally carried the donors’ selves to their recipients, conveying the patrons’ ‘bio-moral substance’, as Indianists used to call the sum of corporeal essence and social standing comprising Indian personhood (Marriott 1976).12 Jajmans were thus not just important economic and political agents. They comprised a socially 12

Marcel Mauss (2002 [1925], 70–77) thought that Indian society illustrated this proposition perfectly. The notion that Indian gifts, paradigmatically food, carry the giver’s nature has been discussed in great detail by South Asianist ethnographers (recent overviews include Heim 2004 and Copeman 2011), who developed Mauss’ idea of transposition of self through gift exchange into a full-fledged theory of substantive contingency. This theory was pioneered by the Chicago ‘transactionalists’ led by McKim Marriott and Ronald Inden, who argued that in India exchange was a substantively constitutive process, in which gifts (most importantly food) carried the givers’ nature (e.g., Marriott and Inden 1973; 1977; also Parry 1986; Raheja 1988a; 1988b).

160  Anastasia Piliavsky

constitutive force.13 Food was the honoured and paradigmatic gift and ‘feeding’ and ‘eating’—both the act and the metaphor—was the crucial link between donors and servants, which offered as vivid an image as there can be of the internal entanglement between those who give and those who receive. As I show in this chapter, the acts and expressions of ‘feeding’ and ‘eating’ have not lost their moral efficacy as links between citizens and political leaders. The logic of mutual constitution remains indispensable to relations between South Asia’s politicians and their followers, to the way local political communities are formed, and indeed to some meanings of ‘political representation’ in rural India today. Jajmani studies taught us another important lesson. While South Asian patronage entailed an asymmetry of status, it did not necessarily prescribe an imbalance of power. Each party depended on the other, economically, politically and ritually. As Wiser insisted, the system ‘contained a mutuality that was lacking in the [European] feudal system’ and the Euro-American notion of ‘patronage’ as top-down bossism (1936, viii). Wiser’s view was probably far too benign and it overlooked abuses present in the system, but his insistence on the basic mutuality of the donor–servant bond distinguished jajmani studies from most other accounts of patronage, which presented patrons as largely independent wielders of power over their clients and patronage as a top-down system of domination. As later ethnographers showed, jajmans often relied on servants just as much as servants depended on jajmans. The patrons’ superior standing prevented them from performing various tasks, requiring them to commission services needed to maintain their standing. Jajmans needed servants not only to uphold their wealth and political supremacy, but crucially to maintain the ritual purity which ensured their place at the top. The servants’ prerogative to transform landlords into patrons, thus authorising the patrons’ ‘rule’, gave servants a certain degree of leverage over their overlords.14 Finally, jajmani studies suggested that giving alone was never enough and that to have moral import, it had to be put on display. This idea is iconically represented by the public distribution of grain 13

On the significance of patron-deities for contemporary political parties in India, see Michelutti (2008, esp. Chapters 3 and 6; also Chapter 12 in this volume). 14 For an historical account of protests staged by the low-caste Chamars, see Dube (1998).

India’s demotic democracy and its depravities  161

to servants on the jajman’s threshing floor. As we shall see, display remains crucial to exchanges between voters and politicians and to popular deliberations about ‘corruption’, or how Rajasthani villagers distinguish between (virtuous) ‘gifts’ and (immoral) ‘bribes’. But let us now turn to Danapur to see all this at work.

Feast in the time of elections In November 2008 the town of Danapur was ablaze with festivity. Shop fronts, temple walls, billboards, lamp-posts and boulders on roadsides were plastered with the insignia of India’s two main political parties—the Congress and the Bharatiya Janata Parishad (BJP). From the early hours of dawn loudspeakers in the bazaar blared out echoing loops of party slogans and instructions to ‘vote for the hand’ or ‘vote for the lotus’ (the respective icons of the Congress and the BJP).15 All day long brightly coloured processions and jeeps full of youths raced through the narrow streets of the town, dispensing sweets, leaflets, firecrackers and caps to children who quickly disseminated them in the town. Excitement reached fever pitch when on the day before the election a retired Bollywood star descended in a helicopter onto a local cricket field. This hullabaloo marked the run-up to the Rajasthan State Legislative Assembly Election, a much anticipated and celebrated event, a ‘great festival’, as one woman put it, ‘grander than any wedding’.16 The bazaar was abuzz with talk of politics, and there was a great deal to discuss. For almost two decades the BJP had been at large in the town, and all the important appointed and elected officials were its members. The current MLA from the locally dominant farmer caste was in his third term in office, but in recent years he was losing support. Having done close to nothing for his constituents, he stopped spending time in the town and his car was sighted less and less frequently as it raised clouds of dust on its way to and from his new home in the state capital, Jaipur. Even the BJP bosses had not seen him for months on end. This left the BJP without a reliable candidate. The Congress, meanwhile, was too deeply divided to put forward a viable contender, and several independently running candidates suddenly came to the fore. Some were deserters from the fractious 15

On Indian ballots (and since 2004, electronic voting machines) parties and candidates are designated with icons which can be readily identified by illiterate voters. 16 See Mukulika Banerjee’s descriptions of elections as festivals in rural Bengal (2007; 2008).

162  Anastasia Piliavsky

Congress and others were pawns put forward by one of the parties to detract handfuls of votes from competitors. For almost a month, not a day passed in the smallest of villages without jeep-loads of party henchmen, festooned with banners and spewing mottos through megaphones, racing through. At the centre of the festivities were electoral feasts (known usually as bandaras or savamanis),17 hosted by candidates. Until 2013, when the Election Commission cracked down on electoral feasting, at times of election in Rajasthan feasts were held everywhere, in villages, town squares and middle-class ‘farm houses’. They could be simple meals for a dozen under a tree or vast banquets at which thousands might eat. Electoral feasting is a tradition dating back to the first elections held in independent India in 1952, and not only in Rajasthan.18 Today reports on electoral feeding, some of which is very elaborately ritualised, still come from all corners of India.19 Consider, for instance, this: Tribal communities in Arunachal would pledge their support to a certain candidate over a feast of roasted mithun (domesticated gaur or a wild ox) and vote accordingly on polling day. This practice had been traditionally passed down the generations. The system … is that the most influential member of the community offers to sacrifice a mithun to exhibit his loyalty to the candidate or the party. Anyone from the community who joins the feast is expected to support the candidate throwing it. At the end of the meal the local leader makes a declaration and all those who have partaken of the meal pledge their support to the candidate concerned (Rana 2006, 158).

A more recent report on the electoral race in Himachal Pradesh tells us that ‘No election victory, and sometimes even a loss, is complete without big mutton feasts’ in the state: ‘supporters of many candidates have already purchased goats from shepherds’ (Times of India 2012). 17

Savamani is formally a feast offered in the name of the gods, especially the god Hanuman known locally as Balaji. It derives from the Hindi words sava (one and a quarter) and man (a weight unit equal to 40 kg). Thus the host must offer ‘50 kilograms’, in other words, a vast lot of food. 18 An elections manual of the time noted that ‘feeding the voters is a matter of very common experience’ (Srivastava 1957, 328). 19 For more descriptions of electoral feeding across the country, see Subha (1997, 77ff ) on Kerala, and Vij (2010) on northern Rajasthan.

India’s demotic democracy and its depravities  163

The ultimate election-time feast takes place on the eve of the polling day, during the all-night electoral vigil known as the Murderous Night—or katal ki raat—a tradition dating back at least to the early 1950s.20 During the Murderous Night (or more often during 2 or 3 nights before an election) the contestants’ henchmen dash from village to village in a last-minute bid to feed and, most crucially, provide drink.21 From the middle distance, this may look like vulgar pork barrel: the buying of poor villagers’ votes on the cheap. But this is not how the residents of Danapur, including some of corruption’s fiercest local critics, see it. For them, the ability to provide is the politician’s duty, and if you ask any child in a Rajasthani village about what politicians do, they will readily tell you: ‘they feed!’ (something they experience during elections, if hardly at any other time). Politicians certainly complain that their opponents buy votes or confuse voters’ judgement by getting them drunk the night before polling dates. But when they themselves feed and water their electors, they say that this is what people want. And indeed, people are disappointed when food and drink fail to arrive. ‘Feeding’ is not just about putting food into people’s mouths. It is also crucially about generating bonds that last, what one may call loyalty. The language of feeding is the key moral idiom in which people evaluate politicians and conceive of the ways in which politicians relate, or ought to relate, to their constituents. As one old man put it, ‘When a politician puts bread into people’s mouths, people know that he is their man, they trust him, their heart rests with him. There is love (prem) between them’. Or, in the words of a young girl from a herder (Gujar) caste, ‘political leaders feed us from the heart (man se)’. One old woman said: ‘it is the duty (dharm) of politicians to feed us. If they feed us, we give them our votes’. Loyalties and their procurement are never this clear-cut, and voters cheat no less than do politicians. They may take freebies, but they never do so in public view. Children and youths may hop from feast to feast, eating from this and that candidate, but no self-respecting adults do. 20 21

Adrian Mayer, personal communication. Katal ki raat is a term adopted from Muslim celebrations of the Muharram festival, in which the vigil commemorating the martyrdom of Hussein is held on the ninth evening called the ‘Murderous Night’. The term also refers to other eves of major transitional events, whether before a wedding night or before the announcement of the Indian Administrative Service examination results. In the electoral context, the reference alludes to this being the final battle among candidates.

164  Anastasia Piliavsky

‘You can fill your stomach today’ (by feasting with different candidates) said one farmer, ‘but then you have to live in the village for the rest of your life’. As politicians pursue electors’ loyalties, they often put on magnificent shows of largesse. At village feasts they dish out not only food but also crockery, blankets, clothing, bottles of alcohol and promises of the much coveted fruits of ‘development’ (vikas): hospitals, roads, electricity and shortcuts through the bureaucratic maze. The gifts they give are themselves promises more than anything and they express a politician’s commitment to continue to give. Every South Asianist can conjure a host of colourful images of patronage as communion, be it a spectacle of royal largesse or a grain heap divvied up on the jajman’s threshing floor. Voters’ expectations of provision express unambiguously the idea that politicians stand above ordinary men and that their primary duty is to provide. This hierarchical sense is reflected perhaps most plainly in the honorifics, like maa-i-baap (mother and father), dada (grandfather or elder brother) and anndata (bread giver), which are commonly used to address politicians in rural north India.22 Taken together, the titles of parent, grandparent and feeder express the conception of patron as genitor and provider, as a source of both sustenance and personal substance for his clientele.23 Bonds between feeders and eaters run deep, and people describe them in terms of trust and love, provision and protection or, in one formulation, as ties that transcend the give-and-take logic of reciprocity: ‘a pukka politician gives from the heart—he gives all the time, before and after we give him our vote, whether or not we give him our vote’. In the ideal, the donor–donee relation creates a structure of mutual obligation, which makes politicians accountable—in hopes if only rarely in practice— and which gives people leverage to press demands on politicians and lay claims to the goods of the state (e.g., Subrahmanian 2009). 22

Thomas Hansen has argued that in Indian cities reference to political leaders as dadas (literally ‘elder brothers’ or ‘grandfathers’) has largely displaced the older language of maa-i-baap and anndata (2001, 72). In the decade of research I have conducted, I have noted that in much of rural and provincial north India this is not so; here the terms anndata and maa-i-baap are still in wide use. Besides, the term ‘grandfather’ or ‘older brother’ (which prevail in urban ‘dadaism’  ) preserve the hierarchical logic of more traditional honorifics. 23 Historically, the titles maa-i-baap and anndata have been used across northern India in reference to all kinds of patrons: landlords, gods, kings, bureaucrats and so on.

India’s demotic democracy and its depravities  165

The cosy vision of moral bonds is a far cry from how politicians in fact perceive and pursue electoral competition—as a race to win, by whatever means. Yet to advance their campaigns they appeal, unsurprisingly, to the pervasive moral logic of donor–donee relations. As one candidate put it, ‘these people are very simple, very innocent— if you feed them, they will give you their vote’. The ‘innocence’ of the villagers is their apparent preference for commensally meaningful feasts over financial profit. ‘They would rather fill their stomachs and drink their fill’, he added, ‘than put shoes on their children’s feet; but of course if they want me to act like their bread giver, I will’, he added meaningfully. As another political boss reflected, ‘A politician must feed and care for his constituents, just like a father. We treat all these people here like our own children’, he said with a wink, ‘Who will feed them, if we do not?’ Politicians ceaselessly promise to ‘get things done’ (kaam karna) for ‘their own people’ (aapane log), to deliver exactly what they think voters want: not generalised improvements to policy, but concrete, targeted benefits such as roads, water, electricity and jobs, stopping just shy of guaranteeing better harvests and heavier monsoons. But promises are not enough. Voters demand instantiations of the politicians’ will and capacity to provide. During elections, candidates need to put on a show of superiority, but this superiority must be carefully calibrated to display a proper balance of intimacy and supremacy, to be above-standing but not too distant, an attitude one might describe as populist grandeur. In the run-up to elections, the choreography of populist grandeur is everywhere on display: the politicians’ magnanimous vague nods of the head, waves of the giving palm or the superior’s form of the namaskar greeting, and the bystanders’ genuflections, accompanied by the cries ‘anndata’ (bread giver) and ‘Jay ho!’ (May you be victorious!). Despite mounting charges of corruption from the Election Commission and litigious political rivals, feasts remain the best—both the most convincing and cost-effective—ways to promise largesse. So contestants continue to bend over backwards to host feasts whose logistics require increasing ingenuity. How do you hide a party for 1000 people? In 2012 there were reports from UP that candidates organised bogus weddings in order to host banquets. In the summer of the same year then India’s Chief Election Commissioner, Dr S. Y. Quraishi, told me that one of his officers walked into one such ‘wedding’ in UP, only to find, to his great amazement, neither a bride

166  Anastasia Piliavsky

nor a groom, but a crowd of more than 1000 people enjoying food and drink (personal communication, August 2012).24 There are other ways to hold smaller-scale feasts: birthdays, for instance, which are hardly ever celebrated in Indian villages, become strikingly fashionable during elections. So why do politicians go through such Gogolesque rigmaroles to risk criminal conviction while hosting public banquets, if all they need do is buy votes? Legal pressure has pushed public feeding increasingly out of view with the result that more and more politicians nowadays offer voters cash behind closed doors. In Danapur gangs of young men often go door-to-door in the night, offering cash on behalf of their political bosses. A close friend’s description of one such nocturnal visit in Danapur is revealing: I woke up in the middle of the night because I heard knocking at the front door. I opened it and saw two young boys. I recognized them, they were our neighbours. Nice boys, you know, one of them was that Brahmin lawyer’s son. So, I let them in and asked whether something had gone amiss. They came in and at first they didn’t say anything. Then Kalpesh [my husband] and my motherin-law woke up. So, I asked the boys to come in. They sat down right here, on this couch, and one of them put down 1000 rupees. That’s when I understood what they were up to. My husband asked what the money was for, and they said: ‘BJP’. Our family has always voted for Congress, so Kalpesh refused to take it. They didn’t say much of anything else. When I was showing them out, they threw the money into my hands and ran away. Kalpesh saw this and you will not believe the row we had that night! He told me the money was dirty and that I should throw it away. This money, he said, has the filth of politics (rajniti ki gandhagi) on it. He said this money will ruin us. But I thought: why should I throw 24

See also Alekh Angre. 2012. ‘Politicians Have Ingenuous Ways of Using Black Money for Elections’. MoneyLife. In response to such reports the Election Commission of India has been keeping a closer watch on the weddings, sometimes to comic effect. In 2013, for instance, a wedding party in Bangalore was called off because the Commission suspected a feast for 1500 people of electoral corruption. Its representative said: ‘in a large congregation, it’s very difficult to determine who is [a] genuine guest, whether one is a party agent or not, etc’. (Times of India, Bangalore edition, 19 April 2013, ‘Polls Throw Spanner in Wedding Feast’).

India’s demotic democracy and its depravities  167

it away? Politicians have lots of money and they eat people’s money, so why shouldn’t I take some too? I still voted for the Congress. We always vote for the Hand [the Congress symbol].

The domestic brawl provoked by the visit dramatised the moral tension that riddles political competition. When the cynicism of the electoral game, the hoarding of votes to win the race, is laid as bare as it was in this case, people react with moral disgust or instrumental contempt. Financial accounts (no doubt incomplete, but still revealing) of the local BJP caucus, which I was shown by one of the party’s senior members, suggest that the money allotted to ‘individual donations’ quickly disappears. Few occupy the moral high ground taken up by my friend’s husband, and most take the cash. And why should they not? The shoving of cash into voters’ pockets is blatantly not a gift, but a bribe (rishwat) that obligates no-one. Although it is difficult to distinguish ‘gifts’ from ‘bribes’ in practice, villagers draw a sharp moral line between the two. Gifts are things donors choose to give. They involve the exercise of will—good will directed at others. Like the ‘favours’ described by Caroline Humphrey in the Russian and Mongolian contexts, the giving of gifts is ‘initiatory, extra, ethical, and gratuitous’ (2012, 23). In the vernacular people say that gifts come ‘from the heart of hearts’ (man se) and that, as such, they have nothing to do with selfadvancement or the helpless necessity of privation, staked by political scientists as the drivers of ‘clientelism’. Bribes are, conversely, made ‘out of helpless compulsion’ (majburi se). You can either reject or accept the payment, but, either way, the paying candidates do not deserve votes. Interpretations of the same act vary, and what a politician may construe as a ‘gift’ may be treated as ‘bribe’ by the voters. While candidates hope that their constituents will treat all donations as gifts, the only way to ensure that donations secure voter’s loyalties is to put giving on display, something that Kitschelt and Wilkinson (2007) also note. Several senior party bosses (both from the Congress and the BJP) in Danapur confirmed this. According to them, clandestine giving generally convinces few. Indeed, bosses in both parties concluded that because nocturnal visits are often perceived as attempts to buy votes, they may do their campaigns more harm than good. Having spent nearly 150,000,000 rupees on the campaign in Danapur (almost twice what the Congress spent), two-thirds of which was distributed

168  Anastasia Piliavsky

surreptitiously in cash, the BJP lost the 2008 election. It was decided subsequently at the party caucus that funds for such ‘individual donations’ should be drastically reduced and reallocated in bulk to ‘social charities’, like ‘weddings’ and ‘birthday feasts’. Despite some dramatic changes in metropolitan centres, in rural north India elected and appointed officials continue to be popularly styled patrons and are expected, as such, to act as all-giving protectors. Few politicians can live up to such expectations, and fewer yet sincerely try. But during campaigns and, if they wish to carry on, also outside them, they appeal to the widely shared moral trope.

The law of the fishes Electoral spectacles of munificence are duplicitous, and they manage to fool very few. More often than not, once ballots are cast, the rhetoric of ‘good governance’ replaces prior promises of generosity. Ordinary people see through these charades, deriding them as ‘corruption’, a term in as wide a circulation in Indian villages as it is in the international airspace (Gupta 1995; 2005; Parry 2000). The villagers are all too well aware of the gap between politicians’ pre-polling promises and their post-polling failures to deliver on them. Most promises are empty and many are in fact undeliverable, even if protagonists were inclined to try. The Bread Giver’s role is impossible. One candidate lamented that when he was first elected MLA he kept his doors open, but was soon overwhelmed by requests for latrines, money, schools, jobs and so on. He said (in English): People here have very primitive thinking. They do not understand politics and they do not understand my position—that I am a government servant only. Instead, they think I am a king or a God who can give them anything they may want. Their thinking is from the olden raja-maharaja days. This backward thinking, madam, is the biggest problem in our India, which keeps our progress behind.

While speaking to me he was doing what linguists would call ‘code switching’, well aware that ‘feeding-and-eating’ was not my idiom for political life and not the way people like me think about what politicians should do. But with villagers he speaks very differently. Politicians’ doubletalk is not just disappointing; it causes much more

India’s demotic democracy and its depravities  169

comprehensive moral upset. Villagers express the sense of betrayal through a nostalgic contrast between today’s ‘hungry’ (bhuka) politician and the apocryphal generous leader who ‘feeds from the heart’. ‘Once upon a time’, said an elderly school teacher, fingering Gandhi’s portrait, which hung slightly crookedly on his office’s crumbling wall, our Indian politicians fed the poor and the poor belonged to them, but these days politicians just buy our votes: at elections they promise villagers all kinds of things. But when the votes are cast, they do not even open the windows of their cars when they drive through the bazaar, and nobody can even approach their office.

Most do not wax as lyrical on imaginary Gandhian leaders but denounce politicians as utterly corrupt, scoffing at the neglectful MP as a kamin, a ‘servant’ or simply ‘the low’. Indeed, as several anthropologists have pointed out, India is rife with talk of corruption (bhrashtachar).25 But in places like Danapur ‘corruption’ refers to something very different from what it means for Transparency International and in courts of law. The local sense of ‘corruption’ stands in sharp contrast to the idea of the ‘abuse of public office for private gain’ (as Transparency International defines corruption). As the residents of Danapur see it, corruption is not the assault on public office, impersonal governance or disinterested abstractions of the modern state. They bemoan instead the betrayal of particularistic, and often deeply personal, relations, lamenting that politicians fail to give consistently or enough and that they take instead of giving, as they sink into greed. While the courts of law may prosecute politicians who throw electoral feasts for ‘bribery’,26 the electorate on the contrary sees ‘corruption’ in their failure to feed. When instead of ‘feeding’ politicians ‘eat’, they do not just reverse the transaction but turn the normative order of giving into the chaos of avarice. In the words of one farmer, ‘all politicians eat from everyone else. If you need any work done and you come into his office, he will take 50 rupees, 10 rupees, a bottle, anything he can eat from you—the dog!’ As an elderly Brahmin of my acquaintance noted, ‘this is how our Degenerate Era (kaliyug) is’. 25 26

For example, Wade (1982), Gupta (1995; 2005), Parry (2000) and Das (2001). Election cases brought to the Supreme Court today are as full of accusations of ‘bribery’ with food, alcohol and cash as they were in the first decades of the Republic’s existence (Lal 1973; Dundas 1998, 8–10).

170  Anastasia Piliavsky

He described this Degenerate Era as subject to the ‘law of the fishes’ (matsya-nyaya), when the big fish consume the small or, even more unnervingly, when the small fish gobble up the big (see also Parry 1994, 112–15; Peabody 2003, Ch. 1). ‘These days’, he explained, instead of feeding the small people, the big people eat— this is today’s dirty politics, child. Oh! At elections they only buy votes, but once they get them, they start filling their stomachs. Just look at our MLA. He is my friend’s son and I saw him when he was so small I could hardly see him. But his gut is full now [after fifteen years in power] and he does not even notice me anymore.

This is not about the mismanagement of public funds for private gain  but about a different moral order which villagers see being besieged. By inverting the local order of mutual interdependence through the top-down flows of gifts, politicians’ greed creates a moral horror reflected starkly in the image of ontological chaos in the degenerate age when fish consume one another. The real ‘dirt’ of politics lies in its mockery of the dearly held ideal, which candidates invoke only to pervert. The politicians’ cunning goes further, as they shift from the language of kingly largesse to the language of seva, or selfless service. Take this campaign speech by one of the candidates in Danapur: These are not the old raja-maharaja days. Everything is different now. The common man (aam aadmi) now rules. Before, the common man bowed down (dhok diya) before politicians, but these days politicians must bow down before the common man. The ‘reign of the kings’ (raajon ka raaj) is gone; now the common man is king. He has the power of the vote-gift (mat-daan) and he has the right to demand service from the government. The politician now serves the citizen. This is our new India. Victory to the common man!

The rhetoric of seva is duplicitous and the ‘common man’ standing below the podium is not taken in by such sermons. Off stage, the same politicians who style themselves ‘people’s servants’ (lok sevaks), throw feasts and promise gifts. In reality, everyone knows that politicians neither act nor see themselves as anyone’s servants. Nor are they thought of as servants by anyone else.

India’s demotic democracy and its depravities  171

Their reputations, as we have seen, rely on the opposite image of a beneficent donor, placed conspicuously and unambiguously above the ‘servants’. Talk of service is strategic and politicians shape-shift into servants as and when it may suit. In the local idiom, ‘servant’ means something very different from the servant of state. Servants are belowstanding, and they can only receive, never give. When politicians say they are the people’s servants, they relinquish responsibility and use a language that abnegates their duty to give. When politicians shuttle between the rhetoric of giving and serving, they deceive their voters in two distinct ways, lying both about what they can do and—much worse—about what they intend. The residents of Danapur thought the idea of a state representative, who is part of their government, as a ‘servant’ preposterous. They thought it inverted the politicians’ proper role and the order of government as a whole. Many were indeed furious at the deception. ‘How can big politicians serve us, poor people (garib log)?’ said one woman; ‘they do upside-down/crooked (ulta-shulta) talk’. ‘If we accept our politicians as our servants, how can we ask them for anything?’ said another, ‘They will say: I am your servant, you are the big man, so you give us money. But what can we, poor people, give?’ Village voters do not see themselves as masters over politicians. Instead, they style themselves ‘poor men’ (garib aadmi), the title of inferiority and the term central to making demands on one’s representatives.27 Everyone knows that when politicians speak of ‘serving the poor’, they are speaking the foreign, and indeed menacingly duplicitous ‘language of officialdom’ (sarkari boli). As one young man plainly put it, ‘it is the politicians’ business to rule (raj karna), so they must rule and feed, not eat or serve’. As the schoolteacher I quoted above pointed out, despite Gandhi’s language of seva, India’s great political hero is locally seen as a donor and a great father figure, not a servant. Despite much talk of ‘service’, when politicians really try to woo voters, consistently invoke the ideal of generosity.28 27

A street-beggar, a devotee, a villager beseeching a lawyer, or a petitioner at a government office calls himself ‘poor man’ in order to get what he wants. This is also the language of the claims used by voters, and not only in rural Rajasthan, but apparently all over India (Subrahmaniam 2009). In a survey conducted by the State of Democracy in South Asia (SDSA), more than 90% of all and 80% of ‘rich’ respondents presented themselves as ‘poor people’, producing, as Yogendra Yadav pointed out, ‘a significant mismatch between people’s “objective” class measured by their income and assets and their selfperceived’ level of poverty (2009, 33). 28 For a discussion of the importance of the language of seva in politics, see Mayer (1981, 153–173).

172  Anastasia Piliavsky

Hierarchy, representation and the depravity of competition Measuring the temperature of popular sentiments of this magnitude is not a task anyone can sanely attempt. What I have described in this chapter is one powerful relational formulation and a charismatic set of orientating values that have long been central to how rural north Indian society operates. Whereas much of South Asianist anthropology today is concerned with urban life, places like Danapur and the villages that surround it still house most of the people on the subcontinent (according to the last Census of India, still more than 60%), and we can ignore their thought worlds only at our peril. Over the past three decades analysts have commented on the dramatic impact of democratic governance and its attendant ideas (civil society, public sphere, good governance and so on) on the shape of Indian society, claiming that electoral politics have prompted the old values of hierarchy, or ranked interdependence, to give way to the new values of equality and competitive independence. As a result, they contend, the old order of rank has been flattened into an assemblage of increasingly separate groups wrestling with one another over political and economic resources (for an overview of this literature, see Manor 2010b). My ethnography nevertheless suggests that, at least in rural north India, hierarchy remains central to popular politics. This ‘hierarchy’, however, is not a total edifice built out of Brahminical or any other abstracted, substantive values (as described by Dumont 1980), but rests on a normative logic that casts rank difference as a sound basis for political, and indeed any other, relations. This logic finds concrete formulation in the donor–servant relation, which, as I show, is the warp and weft of political rhetoric in Danapur. The language of giving and serving shapes the ways in which people envision political authority, what they expect from politicians, the sorts of demands they press on them and the ways politicians style themselves in order to gain citizens’ votes. It constructs authority in a way that can hold it responsible by those whom it governs and represents and by whom it is in the end authorised. Acts of ‘patronage’ which I describe are derided by international critics and prosecuted in electoral law as ‘corruption’ or the immoral purchase of electors’ votes. But, as it is understood locally, acts of patronage are highly moral, and insofar as they express the values of selflessness

India’s demotic democracy and its depravities  173

and  generosity, they are antithetical to haggling and self-interested gain. While blaming one another for selfishness and for greed, both voters and politicians do not see what they themselves do as cynical and they judge one another by the moral standards that they themselves profess. This makes the view that India’s democracy has been reduced to a purely instrumental ‘give there, take here’ affair not only analytically problematic but also empirically incorrect. The very mention of ‘hierarchy’ makes contemporary social scien­ tists, no less than lay Western observers, imagine deep conservatism and authoritarianism inimical to the dynamism we associate with democratic life. Indeed, hierarchical values, which offend the liberal master principles of equality and independence, are widely thought to be incompatible with democracy. Yet, as I suggest, in Danapur (and not only there) relations of ranked dependence are fully consonant with electoral politics. More broadly, I suggest that attachments to political leaders as donors, and the hierarchical orientation on which these attachments rest, underpin democracy’s efflorescence in rural north India. These observations offer insights of broader relevance to democracy as a prescription for political life—and its relation to what we see as ‘corruption’—far beyond rural Rajasthan. They show that by placing people in mutually dependent relationships, patronage does not necessarily imperil democracy’s success, but may on the contrary sustain it. Thus, in many places around the world where people may not share the neoliberal, individualist vision of human life, democracy and what we term ‘corruption’ often go hand in hand. Indeed the more democratically engaged a political system, the more ‘corrupt’ it appears. As David Gilmartin (Chapter 5) shows, not so long ago in Europe’s own history the fact that democracy required patronage—the politicians’ ability to ‘influence’ voters and the voters’ propensity to acquire loyalties and cast votes for one and not another politician— was well-understood. But by now the belief that equality and independence are the only bases for sound democratic rule has obscured the fact that acts of campaigning as much as voting require particular, and at times deeply personal, preferences and attachments. Democracy can thus hardly do without ‘patronage’. The liberal rhetoric makes us forget another crucial fact—that electoral politics mediates an intrinsically ranked political relationship between those who govern and those who are governed, which is inscribed nowadays in the global binary of ‘the government’ versus ‘the people’. People

174  Anastasia Piliavsky

who do not find inequality offensive can cope with this fact no worse, and perhaps substantially better, than those who do. And they see no internal conflict between paternalism, with its hierarchical thinking, and democracy’s rule. What we saw in Danapur is not the ‘politics of the belly’ driven by starvation and greed, as described by Jean-François Bayart in the African context (2009). ‘Feeding and eating’—whether literally at electoral feasts or by ‘getting things done’ for one’s people—forms tight bonds between politicians and the electors they feed. As a basis for political relations, this idiom is key to the ways that the residents of Danapur perceive political representation. The cultural content of political representation, as manifest in rural Rajasthan’s electoral practices, is based on a robust theory of ‘visceral’ social contingency, which makes ‘eaters’ not just economically, but existentially bound to their ‘feeders’. It is indeed difficult to imagine a better basis for relating than the intense mingling of personal substance through the passage of ‘gifts’. We can judge what relational ideology, whether contractual individualism or hierarchical interdependence, may be better suited to democracy’s central aim—the involvement of the largest number of citizens in the governmental process. The successes of Indian democracy, like another rise in voter turnouts in state elections held in November and December 2013, show that relational logics that differ drastically from an outsider’s are no less potent a fuel for popular democratic life. So, where may the real depravities of democracy lie? Electoral politics, as described by both voters and politicians, are cloaked in laments which shade into disgust. Voters say that politicians are stingy and selfish, that they fail on their promises, that they grab and lie and that they take away from, instead of providing for, the poor. Politicians, in turn, say that voters misunderstand modern governance, demanding from them much more than they can in practice provide; and that, having thus set up themselves for disappointment, voters bewail corruption. (Both of these views are not without truth.) If you ask anyone about how they see ‘politics’ (raajniti), they will tell you that it is decidedly immoral and ‘dirty’, befitting the degeneracy of our times (see also Ruud 2001; Harriss 2005). Perhaps, as Gilmartin suggests, the intrinsic conflict between the normative basis and the instrumental requirements of patronage politics—and I would add any other kind of politics—has always and everywhere made it suspect in the eyes of the ruled. But there is a distinctive element of

India’s demotic democracy and its depravities  175

democracy that magnifies this contradiction, making politics appear more disappointing to its participants. Democracy, whether in India, Switzerland or the United States, is a moral scheme. It is an ideal. And, as I suggest in the Introduction, this ideal is necessarily collective. In Washington DC no less than in Danapur it requires people to rise above themselves and act in the name of something beyond their individual interests, whether for a concrete ‘community’ of ‘their own people’ or a more abstract social or national good. Democratic ideology of any cultural hue is thus opposed to self-interest and instrumental gain. This is precisely where conflict lies, emerging from the clash between democracy as a normative form and democracy as a competitive practice. Representative democracy is necessarily a competitive system that sets personal victory as the politicians’ goal. Competition naturally predisposes—indeed requires— politicians to pursue selfish aims, causing them to betray democracy’s society-minded moral sense. Politicians everywhere hijack society’s moral discourse in pursuit of success. This Janus-faced politics is bound to disappoint and to give politics the stench of cynicism and moral rot. This tension is nowhere more strikingly on display than during elections. It is in this conflict that the real depravities of democracy, for Rajasthani villagers and by no means only for them alone, lie.

‘Vote banking’ as politics in Mumbai


Lisa Björkman


he 2012 Municipal Corporation elections in Mumbai witnessed a remarkable re-entry of urban elites into the rough-and-tumble of city politics. Six separate non-governmental organisations (NGOs) and self-proclaimed Citizens Platforms put forward ‘Citizens Candidates’ to contest the polls in nearly a quarter of the city’s 227 municipal electoral districts. In recent years, Mumbai’s globally empowered classes have grown increasingly animated by anger and vociferous about issues related to the material conditions of urban life: intractable traffic snarls, increasingly assertive street hawkers, overburdened infrastructures and, above all, the continuing and ever-increasing visibility of the slums. Such conditions are charged with tarnishing the world-class image of the city, embarrassing Mumbai’s boosters and discouraging global investment. The deterioration of the city’s fabric, claim civic activists, is a result of corrosion in the Municipal Corporation, which has become mired in corruption perpetuated by patronage-purchased ‘vote banks’ in the city’s swelling slums. Hoping to purge the Municipal Corporation of vote bank politics, Mumbai’s civil society and media-empowered middle classes attempted to reassert control over the governmental body which makes decisions on crucial issues like planning, housing and infrastructure. Mumbai’s 2012 Citizens Candidate movement rode in on the coattails of a heady season of Anna Hazare’s ‘India Against Corruption’ campaign, in which most of the Citizens Candidate forum leaders had, from the beginning, been involved. Taking advantage of their primed readership, the English-language media furiously narrated the Citizens Candidate movement, with top reporters and citizens group spokespersons keeping one another’s mobile phone number on speed-dial. The Hindustan Times, The Times of India and the Indian Express each ran a special series of administrative ward ‘profiles’ in the

‘Vote banking’ as politics in Mumbai 177

weeks before polling day, sending young activist journalists to meet with representatives of area NGOs and ‘citizens groups’ to discuss the problems of municipal governance. Characterisations of the existing crop of elected councillors by the citizens forums and their collaborators in the English-language press accused elected officials of failing to represent the city’s ‘real citizens’. At a public event hosted by the Daily News and Analysis newspaper, a citizens’ forum activist told the cameras: ‘We represent tax-paying, law-abiding citizens of Mumbai. Slums and hawkers are planted as a vote bank. These encroachers put up shops and houses and then they say they have rights. Where are my rights?’ ‘In Mumbai’, lamented another activist, ‘it’s largely the slum dwellers that form the political parties’ vote banks. This trend needs to be changed’. ‘Just go to the ward office’, cried a third; ‘they talk to you like you’re a beggar! We don’t have respect in our own city, and it’s time we got it back! Enough candlelight vigils, Mumbai is ours’. ‘Thus far’, concluded another, ‘candidates banked upon slum voters or communal vote banks to get elected. [We now have] a great opportunity for citizens to throw out sitting corporators and elect a good candidate for the city’s development’. Politicians, activists charged, are not elected by ‘citizens’ but by ‘banks’ of slum dwelling voters, whose illegalities are tolerated and even facilitated by political leaders in exchange for block votes. Vote bank politics has hijacked the city, which is now governed in contravention to ‘development’ or any sense of the broader public good. The way Citizens Candidate activists and spokespeople characterise ‘vote bank’ politics in contemporary Mumbai overlaps remarkably with liberal political theories of patronage, albeit with predictably less sympathy for the plight of the poor. Kitschelt and Wilkinson (2007), for example, define ‘patronage’ as ‘the direct exchange of a citizen’s vote in return for direct payments or continuing access to employment, goods and services’ (cited by Wilkinson in Chapter 11, 259). Depictions of electoral exchanges as ‘machine-like’ (Scott 1969) cast voters not as agentive collectivities possessed of real bargaining power but as manipulated and exploited subalterns, victimised by what de Wit (1996) calls the ‘politics of illusion’ wherein the ‘poor are tied to the political system by (promises of) material benefit and by almost personal, emotional ties to the highest authority’. In de Wit’s account, poor voters are not modern political subjects and independently reasoning citizens, but subjects inextricably entangled in ‘emotionally inflected’ traditional

178  Lisa Björkman

solidarities and social relations that prevent them from acting (or perhaps even perceiving) their own interests as well as the real interest of society. Liberal theorists, in other words, accuse patronage of subverting the very meaning of constitutional democracy. As David Gilmartin explains, Indian electoral law ‘celebrates the legal status of the individual not as the bearer of a particular culture, but as a universal vessel of free will and legal rights. That an official, legally recognised “voter” is a rational, autonomous actor is the conceit that justifies government by consent—and defines the people’ (2007, 56). Politicians, that is, ought to be elected on the basis of what Wilkinson (Chapter 11) calls ‘programmatic’ competition, involving ‘codified universalistic public policy applying to all members of a constituency’ (Kitschelt 1999; cited by Wilkinson, Chapter 11, 259). In contrast to this ideal, writes Kitschelt, the dynamics of patronage mean that poor and uneducated citizens discount the future, rely on short causal chains, and prize instant advantages such that the appeal of direct, clientelist exchanges always trumps that of indirect, programmatic linkages promising uncertain and distant rewards to voters. Because of poor people’s limited physical mobility and clustered patterns of residence, politicians can also monitor the adherence of the poor to clientelist deals better than that of affluent individuals (2000, 857).

This kind of theorisation presents the poor as victimised dupes who are easily appeased with short-term particularistic goods and the toleration of politicians’ illegal dealings in lieu of making broader, longer-term advances through programmatic governance. Critics of liberal formulations of patronage have suggested that in contemporary India ‘vote banking’ contains redistributive possibilities and even democratising potential.1 As regional, linguistic and caste-based identities have gained salience in Indian electoral politics in recent decades, local-level party organisation has become more significant, with state and local elections consistently attracting higher levels of voter turnout than the general. Citing increased political competition arising from the proliferation of 1

See Chatterjee (1998; 2004), Jaffrelot (2007), Guha (20 Jan 2008) and Breeding (2011).

‘Vote banking’ as politics in Mumbai 179

parties, Ramachandra Guha suggested that the notion of ‘vote banking’ has now taken on a new meaning: ‘Vote bank denotes a collective political preference exercised by a particular interest group’ (2008). He suggests that the vertical patronage ties have given way to horizontal solidarities, ultimately deepening the capacity of the Indian electorate to market themselves to various parties as ‘vote banks’.2 In a similar vein, Partha Chatterjee (1998) has suggested that the exchange of votes for particularistic material benefits is part of a broader ‘strategic politics’ through which the urban poor in modern India make claims on the state. The imperatives of democratic legitimation, Chatterjee argues, produce political obligations (1998, 281) for state actors to ‘deliver civic services and welfare benefits’ to the urban poor, who he says are excluded from formal spheres of rights and citizenship because their ‘habitation or livelihood lies on the other side of legality’ (2004, 56). Thus both liberal and postcolonial scholars agree that electoral politics in contemporary India involves the channelling of material goods and particularistic benefits towards community-based, low-income social groupings in exchange for blocks of votes. Does vote banking amount to exploitative, machine-style politics (Scott 1969; de Wit 1997) which excludes the poor, who are quelled and purchased with short-term welfarism, from true democratic participation and debate on questions of governance and social policy (e.g., de Wit 1997; Schaffer 2007; Kischelt and Wilkinson 2007)? Or does vote banking allow ‘communitiy groupings’ to use their votes ‘instrumentally’ to extract concessions from the state (Chatterjee 1998, 282), thus boosting redistributive possibility or even democratising potential (Chatterjee 1998; 2006; Guha 2008; Jaffrelot 2007)? This chapter explores these questions, drawing on ethnographic research conducted between 2008 and 2012 in the eastern suburbs of Mumbai and focusing on the relations between an elected municipal councillor and his ‘vote banks’. My account demonstrates that the vagaries and contradictions of the regulatory framework that governs everyday life in the ‘slums’ of Mumbai render the lives and livelihoods of their residents both legally and materially


For a discussion of this shift from vertical patronage ties to horizontal networks of authority in Mumbai, see Hansen (2001, 72).

180  Lisa Björkman

precarious.3 Therefore, the production and maintenance of a neighbourhood’s infrastructure and physical form, as well all manner of business activity, generally requires mediation by someone who has access to various kinds of knowledge and resources necessary (or at least believed to be necessary) to navigate the city’s material, legal and economic opacities.4 I argue that while playing a crucial role at times of elections, these networks of knowledge brokers exceed the analytical categories offered by either liberal theories of patronage or their postcolonial critics, and conclude by suggesting that instead we may consider vote banking to be the very stuff of democratic politics.

Aziz Nagar One January afternoon in 2010 I was sitting at home when my mobile phone rang, its display revealing the caller to be one Mr Khan, a scrap dealer in the popular neighbourhood of Aziz Nagar in Mumbai’s eastern suburbs. I had gotten to know Khan after my here-and-there search for someone familiar with the history of the neighbourhood had resulted in a fortuitous introduction, which had over the months grown into a stable friendship. ‘Hello Madam!’ came Khan’s voice, ‘Please come to my office’. The municipal councillor, Mr Kamble, would be inaugurating a line of new streetlights he had erected using his councillor ‘slush fund’, and he had requested Khan to summon me to the event.5 At the time I was researching water politics and was often sighted around the neighbourhood, riding around in water trucks or trailing behind groups of engineers; rumour had it that ‘the foreigner’ had some kind of a connection


For a discussion of the slippery politics of classifying slums, see Björkman (2013). Also see Berenschot (Chapter 8). 5 Each of Mumbai’s 227 elected councillors has access to a district-wise discretionary fund of 35 lakh rupees (about $70,000). The funds are allocated as part of the annual budget of the Mumbai Municipal Corporation’s Maintenance Department (‘A Budget’). The purpose of the fund is to enable elected representatives to respond quickly to maintenance-related needs unanticipated by the annual Maintenance Department budgets: pavement of smaller lanes, enclosure of open drainage ditches or construction of foot bridges over open sewers (nallahs). To access these funds, a corporator must approach the appropriate department of the municipal bureaucracy. After the proposed work is approved, the concerned department sends the request to the Maintenance Department, which dispatches contractors for the work. 4

‘Vote banking’ as politics in Mumbai 181

to the local water department, and my presence presumably added weight to Kamble’s event. —When should I come? —Now.

I had thought as much. Cancelling my afternoon meeting, I threw a shawl over my shoulders and hopped into a rickshaw, arriving in Aziz Nagar to find a nervous-looking Khan surrounded by a small crowd of residents waiting for me on the side of the highway; Kamble stood impatiently in the shade of a scrap warehouse a little way off. Our little parade, with Kamble at the helm, made its way slowly along the edge of the highway towards the newly erected streetlights

Figure 7.1 Kamble with social workers.

182  Lisa Björkman

at the far end of the adjacent neighbourhood of Phule Nagar. People poured out of the narrow lanes to see what the fuss was about. Kamble folded his hands respectfully towards onlookers, announcing that out of concern for their safety he had brought them new street lamps; women would now be safe to come and go in the dark. When we reached the transformer at the far end of the road, Khan produced a coconut which he cracked to bless the grey metal box, and an electrician connected the line of lamps to the power source. In the still-light evening the effect was somewhat anti-climactic. Nonetheless, everyone cheered, and we turned back towards Khan’s office. On the walk back, a woman called out from the edge of Phule Nagar: ‘saab, now can you give us water…?’ Two years after the event, in the run-up to the Municipal Corporation elections, I asked Khan about his support for Kamble’s re-election. I pointed out that in the Parliamentary elections the previous year Khan had campaigned for the candidate from the opposition party, but now stood firmly by Kamble in his campaign. Khan: I work for myself and for my people here; Kamble supports us. LB: But how do you convince people to vote for him? Khan: I just tell them how to vote and they do. LB: Why don’t they just nod and then vote for whomever they like? Khan: They don’t know whom they want, so they ask me. They trust me to tell them who will protect us—who will do their work. In the run-up to polling day I heard similar explanations from all corners of Kamble’s working-class constituency: community leaders, social workers, business owners and even some opposition party workers. Kamble went on to win his 2012 re-election with a comfortable margin. On what basis did Kamble’s voters ‘trust’ the advice of people like Khan? What kind of ‘support’ has Kamble offered to Khan and to the residents of that area? What kind of ‘work’ has he done for Aziz Nagar (and other neighbourhoods) to inspire such strong support for his re-election? Khan explained to me that his own confidence that Kamble would ‘do their work’ is based on Kamble’s two decades of ‘social work’ in the area, including 10 years as an elected councillor. Indeed, Kamble’s relationship to the neighbourhood is almost as old as the neighbourhood itself.

‘Vote banking’ as politics in Mumbai 183

Sitting in the office one evening after the steady stream of visitors seeking Khan’s assistance with one thing or another (fixing a misspelled name on an identity document or looking into low pressure in a water tap) had slowed to a trickle, I asked him to tell me the history of the neighbourhood, and of Kamble’s relationship to it. Khan smiled: ‘from the beginning?’ I nodded. The neighbourhood of Aziz Nagar, which is a stone’s throw away from the city’s largest dumping ground, was originally settled by Khan’s father in the 1980s. Khan himself was born in Crawford Market, a neighbourhood in South Mumbai where his father ran a tar business. His father’s family was originally from the north Indian state of Uttar Pradesh, but had moved to the working-class Crawford Market soon after independence. In the 1960s, when a warehouse fire destroyed their neighbourhood, the Mumbai Municipal Corporation resettled all the affected families into a cluster of concrete tenement homes in the city’s newly annexed eastern suburbs, not far from the dumping ground. Proximity to the dump inspired Khan’s father, whose tar business perished in the fire, to try his hand in the scrap dealing business, buying in bulk from collectors and selling the materials to traders in town with whom he had long-standing contacts. Khan’s father’s soon-flourishing scrap business got another boost when a family friend working as a low-level labourer in the Municipal Corporation told him about a parcel of land near the dumping ground which seemed to have been overlooked in the 1967 Development Plan. ‘It was shown on the plan as just an empty space’, Khan smiled, ‘so my father decided to claim it’.6 A decade had passed since the release of the Development Plan and no efforts had been made to put the marshy plot to use. Khan’s father invested his savings from scrap dealing in the construction of warehouses for storing and sorting scrap materials–spaces that he soon began to sell and rent out to dealers and brokers from all over the city, for whom a facility so near the dumping ground promised significant savings in transportation and real estate costs. ‘My father was smart’, said Khan, explaining how his father formed a society of warehouse owners, registered it with the state-level Charity Commissioner, and began paying property 6

The 1967 Development Plan sheets confirm that the land where Aziz Nagar is found was indeed unmarked by any plot-level reservation.

184  Lisa Björkman

taxes. ‘Of course they didn’t object to receiving tax money! And my father collected tax receipts. My father wasn’t educated, but he was smart about these things’. Trouble began a few years later when the society’s elected chairman stopped paying the assessment taxes. ‘We didn’t know’, Khan recalls, ‘because we were still paying taxes to the society, but he was just filling his pockets and drinking’. Suddenly, in 1984 Khan’s father received a notice from the Municipality stating that they were illegally occupying the land; the bulldozers arrived soon after. ‘My father was very angry, but he hired a lawyer, took his tax receipts and went to court’. In the interim years, while the case was tied up in court, Khan’s father rebuilt the area, this time not with warehouses but with residential structures. Over the preceding decade a number of residential developments had come up in the area, including two enormous municipal housing colonies meant to house low-level municipal workers and working class families whose neighbourhoods had been displaced by urban development projects during the Emergency. Finally, in 1990, while Khan’s father’s case was still languishing in the court, the Government of Maharashtra issued a circular declaring that unauthorised structures whose residents could prove that their homes were there before 1985 could only be demolished if the land was needed by the government for some other public purpose (and then only against proper compensation). ‘Our lawyer told us that we were safe’, Khan explained, ‘but the people who had bought or rented the new houses, most of them had come only after the demolition, around 1987’. Since only 21 families living in the 1200 homes held pre-1985 proofs of residency, in 1991 the Municipal Corporation once again razed the neighbourhood. A few years and many demolitions later, Khan—now a young man  running his family’s scrap business—was approached by a young politician named Kamble, who was standing for election to the Municipal Corporation. Khan tells me that when Kamble approached him for support in his bid for office, he said: ‘Ok, I’ll help you with the election but you have to promise that we won’t be demolished again’. Kamble agreed. However, a week after Kamble’s election, Aziz Nagar was once again served demolition notices. Livid, Khan approached Kamble: ‘I went to his home with the notice and said: ‘what’s this!?’ Kamble picked up the phone and called the Housing Minister, handing me his mobile so that I could tell him what happened’. Kahn recalls that the Minister finally phoned back with

‘Vote banking’ as politics in Mumbai 185

news that he had spoken with the Deputy Municipal Commissioner and that the demolition had been called off. ‘They consulted the new Development Plan’, Khan remembers, which as of 1991 had zoned the area for residential use; the absence of any reservation for a public facility (a school or a hospital) meant that the warrant for demolition was both questionable and unnecessary. These accounts reveal a politics of contestation, negotiation and claims-making that does not fit comfortably with the analytical categories offered by theories of ‘patronage politics’ which tell us that political claims can only be made properly either through appeals to legally enshrined rights or through ‘patronage’, by making claims to goods or services that cannot be rightfully and legally claimed by all citizens. The claims Khan’s family laid to the piece of land that would become Aziz Nagar are not clearly classifiable vis-à-vis existing regulatory and policy frameworks.7 At the time of the first demolition, the Society had an established record attesting to their successful payment of property taxes and indicating the recognition of their land claim by the state. ‘Patronage theorists’ may counter that the state’s acceptance of property taxes could simply have been a bureaucratic error committed by an inefficient (or corrupt) bureaucracy. Yet, given the multiple (and sometimes contradictory) ways in which land claims have been made in Mumbai, it is equally plausible that Khan’s father’s payment of property taxes was simply seen as legitimate.8 The 1966 Maharashtra Land Revenue Code, for instance, established a way in which parties ‘encroaching’ on public land could have their claims normalised through a relatively straightforward process of ‘regularisation’.9 While Khan’s father may have built the structures without reference to any existing policy framework, the land claim was as likely to result in legitimation as in demolition. Indeed, it was probably the existence of so many 7

For a fuller treatment of these issues see Björkman (2013). Mumbai was not unique in this respect; Benjamin (2007, 550) lists 15 separate ways by which land has been claimed in Bangalore. 9 According to Section 51 of the Maharashtra Land Revenue Code, If the person making the encroachment so desires, [the Collector may] charge the said person a sum not exceeding five times the value of the land so encroached upon and to fix an assessment not exceeding five times the ordinary annual land revenue thereon and to grant the land to the encroacher on such terms and conditions as the collector may impose subject to rules made in this behalf, and then to cause the said land to be entered in land records in the name of the said person. 8

186  Lisa Björkman

possible responses to such claims that kept Khan’s father’s case (along with many others) in court for so long. The already-mentioned government circular prohibiting the demolition of structures built on public land prior to 1985, however, did nothing to clarify the legal standing of neighbourhoods like Aziz Nagar; instead, it laid out conditions under which contested lands could legitimately be claimed by various state authorities. The case of Khan’s father was dismissed by the court not because the legal standing of his land claim was clarified by the new rule, but because the circular made it apparent that the municipal authorities had no right to destroy his father’s properties. Until the cut-off date was updated in 1995, however, residents of Aziz Nagar who did not have pre-1985 proofs of residency continued to rely on Kamble’s contacts in the Housing Ministry to support their structures’ tenuous claims to stand. Not only did Kamble make good on his promise to hold back the bulldozers, which were periodically mobilised by workers from various opposition parties seeking to assert the strength of their own networks, over the years he also chanelled significant Municipal resources into the growing neighbourhood’s infrastructure. In 2001, for instance, Kamble formally demanded that the Water Department include in their annual budget, and plan a provision for, an underground water main. Since the late 1990s, a handful of local entrepreneurs (including young Khan) capitalised on the absence of a formal water distribution network in Aziz Nagar. They set up a private extension of the municipal water grid and tapped into deep-buried mains along the highway. At first Khan and his friends supplied residents with water free-of-charge, taking money only from customers who came from other areas to buy water at a by-the-can rate on the soonbustling water market in Aziz Nagar. ‘Our best years of business were 1998–2000’, Khan recalled, ‘For eighteen months we were living large’. He laughed, a little embarrassed: ‘We were just kids! And of course we didn’t save anything. We just had fun—we bought bikes, roamed around the city, drank liquor. We didn’t get rich, but we got happy with the money. We also got greedy’. Instead of providing precious water to local residents, Khan and his friends increasingly diverted it to retail businesses and eventually started to charge residents for the use of the pipes. The residents, led by rival social workers and brokers, soon began to complain to the local office of the water department, which responded by cracking down on the Aziz Nagar water market. It was in the context of a protracted cat-and-mouse

‘Vote banking’ as politics in Mumbai 187

game between water vendors and municipal engineers that Kamble pushed through the proposal to lay the new water main. As a subengineer who remembered the planning of the main remarked, there was nothing irregular about this sort of proposal: such plans are ‘ultimately written by us’, he told me, ‘but every plan needs someone to initiate it’. The new pipe presented new possibilities, but it also posed challenges. While the main meant that it was now possible to get water, applying for metered water connections was an expensive process—an investment that low-income families could undertake only incrementally, as and when their means allowed. In 1996, just a few years after the new pipe was commissioned, the Government of Maharashtra issued a General Rule restricting provision of municipal

Figure 7.2 Water pipes in Aziz Nagar.

188  Lisa Björkman

services in the slums to residents who met the new 1995 cut-off date. Previously, a billed water connection could generally be secured without much hassle. As one senior engineer explained, since the Municipal Corporation Act of 1888 gives the Corporation the right to sell piped water as ‘moveable property’, the water department can provide water to anyone willing to pay for it. Sometime in the 1960s, the engineer recalled, the water department ‘decided’ that the BMC Act gave the water department the right to sell water even to residents of unauthorised structures. While Aziz Nagar comfortably makes the 1995 cut-off date, because of the extremely liquid nature of Mumbai’s real estate market (particularly its working-class housing stock) a large proportion of residents have purchased their homes after the date. While most of the homeowners have proof of residence (ration cards, for instance, or voter identification cards), these documents do not always predate 1995. Besides, like many popular neighbourhoods in Mumbai, Aziz Nagar houses a large number of renting tenants whose identity documents bear a different residential address. This situation has left the residents of Aziz Nagar with a few ways to obtain water. Some make monthly payments to neighbours in exchange for the daily use of their taps. The neighbourhood’s growing population, however, created the need for additional taps, procuring which relies on networks of knowledge and authority accessible through ‘social workers’ like Khan. These days, when the residents of Aziz Nagar decide to invest in a new water connection, they usually do not approach the water department directly, but go to Khan’s scrap shop, where he sits every day (between 11 am and 3 pm, and again between 6 pm and 10  pm) processing applications not only for water connections but also for school admissions, birth and death certificates, and identity documents of all kinds.10 Indeed, for residents without pre-1995 proof of residency approaching the Municipal Corporation directly is a waste of time. Even for residents who hold this proof, the politically charged nature of water supply to slums means that an application for a new connection has a better chance of being processed in a timely fashion (if at all) through a specialist: someone familiar with the procedures of the Municipal Corporation and with engineers at the ward office, and (ideally) fluent in Marathi, the language of municipal 10

See also Berenschot’s discussion of such ‘fixers’ and ‘social workers’ in this volume (Chapter 8).

‘Vote banking’ as politics in Mumbai 189

administration less well-known in Hindi-speaking neighbourhoods like Aziz Nagar. Khan maintains close relations with a handful of such water specialists in the area, men referred to as ‘plumbers’ who have specialised knowledge of the underground network of pipes, and of the ins and outs of the water department bureaucracy. Khan’s preferred plumber, a native Marathi speaker named Bansode, also maintains close relations with Kamble. In the event of any gaps in documentation, Kamble can provide a letter of support for the new metered water connection, thereby allowing the local engineering staff to extend infrastructure into the neighbourhood without exposing themselves to politically motivated corruption charges. Bansode’s reputation as someone who can make water flow out of the municipal water system gives his regular appearances in Khan’s office a particular weight. Voters therefore have very good reasons to trust Khan’s advice at times of election. The residents of Khan’s area had additional reasons to vote for Kamble in 2012. With the locality in negotiations for redevelopment under a Slum Rehabilitation Scheme, three local leaders (including Khan) divided the area into three housing societies, each of which was involved in discussions with various builders over the terms of redevelopment and about who would be included. A social worker affiliated with an opposition candidate explained: ‘they’re coming up for redevelopment, and all these builders have party connections. People will vote for [Kamble] because with all difficulties of proving residency, they’re afraid of losing their homes and not getting rooms in the new building’. In Khan’s area many residents, including some long-time residents, do not have the documents needed to prove eligibility for a Slum Rehabilitation Authority (SRA) scheme. As one resident explained, for those who bought their homes more recently, or are living as renting tenants, ‘even if [they have] been here for twenty years, what proof do [they] have?’ In the housing market of Mumbai’s popular neighbourhoods, it is common for pre1995 structures to have changed hands in the interim. There has thus come into being an unofficial practice of getting lawyers to draw up sales documents shored up with court affidavits. Whether or not any of these documents, or any combination of documentation, is accepted as adequate proof is determined largely by further, and often politically inflected, negotiations. For this reason, the worker from the rival party commented, residents ‘need [Kamble] and his social workers’.

190  Lisa Björkman

Khan himself explained to me the central role played by ‘social workers’ like him in brokering redevelopment schemes. The Slum Rehabilitation Authority requires the agreement of a minimum 70% of the area’s residents to proceed with a scheme, so securing the signatures of ‘eligible’ families is the crucial first step. In Khan’s area, Khan himself (together with Kamble) prepared the list of families, which the builder took to the SRA to ‘pass in one go’. For the residents of Aziz Nagar, a Slum Rehabilitation Scheme is a risky affair, which offers the possibility of capitalising on the city’s real estate boom, while rendering families vulnerable to the whims of notoriously slippery (and sometimes disposessionary) politics of redevelopment. The burning questions include: will the family actually get a room in the new building? Will the building be of decent quality and in the promised on-site location? Such risks are managed by social workers like Khan who have earned a measure of trust over many years. In Mumbai, the work of local-level leaders like Kamble and Khan to support claims to housing and infrastructure made by newer homebuyers has had political results that stretch beyond any particular neighbourhood. In January 2012 the Government of Maharashtra passed a resolution that made significant headway in legitimising the status of newer homeowners who claimed inclusion in the rehabilitation schemes. It emerged that more recent buyers of the pre-1995 structures may become eligible for redevelopment schemes by paying a ‘transfer fee’ and by providing a few kinds of proof.11 While the resolution is a significant political accomplishment, until (and if) the Development Control rules are amended, the residents of transferred homes will remain only ‘conditionally eligible’ for inclusion in a redevelopment project and will, thus, for the foreseeable future continue to rely on people like Kamble and Khan to mitigate the risk of being left homeless by a rehabilitation scheme. When the Congress Party swept the 2004 Maharashtra Legislative Assembly Elections in Mumbai on a promise to shift the cut-off date from 1995 to 2000, the party leadership presumably hoped to please both property developers (who would receive a windfall of potential slums 11

First, the owner must prove that the structure existed before 1 January 1995. This proof could take the form, for example, of a property sales agreement supported by a court affidavit and accompanied by the pre-1995 residency proofs of its former owners. The new owners would also need to provide proof that they have occupied the structure for at least a year.

‘Vote banking’ as politics in Mumbai 191

to redevelop and valuable development rights) and the vast numbers of voters who may now prove themselves eligibile. The promise met with predictable fury from opposition leaders who accused the Congress of pandering to  Muslim and north Indian ‘vote banks’ in slums; but a stronger and ­unexpected critique emerged from the ranks of elite civil society. In 2006, the upper-class residential welfare organisation Janhit Manch filed a Public Interest Litigation against the Government of Maharashtra, arguing that the Slum Rehabilitation Scheme was destroying their neighbourhood—a posh area in the western suburbs. An overwhelming amount of the ‘transferable development rights’ generated by Slum Rehabilitation Schemes across the city, they argued, had been utilised in their neighbourhood, where residential property values are among the highest in the city. The resulting proliferation of construction, maintained Janhit Manch, violated the height limits specified by the Development Control Rules of 1991, leading to population densities that could not be supported by available infrastructure—particularly transportation and water systems. Janhit Manch argued that the extension of the cut-off date to 2000 would exacerbate this problem and violate their rights as property owners. The Mumbai High Court was convinced by these arguments and it ruled in favour of preserving 1995 as the cut-off date. While the Congress-led Government of Maharashtra appealed to the Supreme Court, the 1995 cut-off date has come to seem increasingly arbitrary, both since the electoral mandate of the Congress government was wrapped up in the promise to update the cut-off date and because a number of slum rehabilitation projects have already been implemented using 2000 as the cut-off date.12 As a result, local political leaders increasingly railed against projects that excluded households whose proofs of residence fell between 1995 and 2000. With the cut-off date thus in limbo, attempts to adjudicate between ‘legal’ and ‘illegal’ neighbourhoods, buildings, families and water infrastructures came to appear increasingly farcical. In 2010 municipal bulldozers arrived in Ganeshwadi, a neighbourhood in Kamble’s area adjacent to Aziz Nagar, but which dates back to some time between 1995 and 2000. Following its demolition, local social workers and activists led hundreds of the displaced on 12

Rehabilitation projects carried out in conjunction with World Bank–funded initiatives to upgrade infrastructure carry their own rules for eligibility that do not map neatly onto those of the Slum Rehabilitation Authority. See Randeria and Grunner’s (2011) account.

192  Lisa Björkman

a march to the nearby luxury township of Hiranandani Gardens in Powai and demanded the demolition of the township, which—in a­ tit-for-tat move—they claimed was ‘illegal’. Brandishing information obtained through a Right to Information petition, activists informed the news media that Hiranandani Gardens had been built on 300 acres of public land which was zoned in the 1991 Development Plan for lowincome housing but which had been given to the developer at a hugely reduced rate. Instead of constructing the mandated subsidised housing for the poor, the developer built a high-end commercial and residential complex. The displaced families declared that they will rebuild their demolished homes in the verdant spaces of the Hiranandani Gardens, in accordance with the Development Plan zoning scheme for lowincome housing. While the police stopped the protesters from actually carrying out these constructions, their media spectacle worked well enough; the families were allowed to return to Ganeshwadi and to rebuild their homes in peace. By the same token, the destabilisation of the cut-off date has inspired the renewal of infrastructural activities: in February 2010 social workers from Ganeshwadi sent a letter to the Mumbai Suburban District Collector requesting that a list of neighbourhoods in the area, including Ganeshwadi and Aziz Nagar, be declared slums in accord with a largely forgotten but still existing policy framework buried in the Maharashtra Slum Area Act of 1971. ‘According to the Maharashtra Slum Area Act 1971 section 5A’, explained the letter, slums have to be provided with drinking water, sewage line, drains, toilet, bathrooms, concrete roads, gardens and parks, social welfare centre, school, hospitals etc. There is no cut-off date to provide these basic amenities as per the Slum Area Act [the letter concludes] but the government speaks of a cut-off date—which is illegal.

In the week before elections to the Municipal Corporation in 2012 I met with a Ganeshwadi social worker named Suresh in his motorbike repair shop on the side of the highway. Suresh and his colleagues were poring over a hand-written list of demands they intended to present to Kamble: 1. 2.

Another block of toilets Help with registration of a local self-help group of abandoned and widowed women for public assistance in accordance with government schemes targeting residents who live ‘Below Poverty Line’

‘Vote banking’ as politics in Mumbai 193

3. 4. 5.

Arranging an electrical substation Water issue: water is a necessity and should be available to everyone. Make effort to get rid of the 1995 cut-off rule Work towards getting Ganeshwadi declared as a slum

I asked about the last item on the list and Suresh said that 2 years had passed and the Collector’s office had not yet responded to their request; so they would enlist Kamble’s help. If Kamble agreed to work towards these goals, Suresh explained, they would support his bid for re-election.

Conclusion On the eve of the polling day in February 2012 the city’s largest Citizens Candidate movement plastered the city with fliers bearing the following message:13 There are NO LEGISLATION in BMC. NO MAJOR POLICIES FRAMED in BMC NO GOVERNMENT TO BE FORMED in BMC A ‘Nagar Sevak’ (City Servant) Corporator is an elected public representative whose main role is to: 1. Help facilitate the basic civic amenities Sadak Bijlee Paani gatar (road, electricity, water, sewage). 2. Be the guard to protect Public Properties including green space and public lands. 3. Monitor the expenditure of the BMC to make sure public monies are spent honestly and justfully for the betterment of the society as a whole. 4. To spend the budget given to him/her for the betterment of the ward. Limited to guidelines and rules. The Municipal Corporation, the flier announced, needs to be depoliticised. The existing crop of representatives must be replaced by ‘guards’ who will ‘monitor’ how the councilors spend their budgets. The Citizens Candidate movement characterised the kinds of activities 13

The fliers were printed in English, Hindi, Marathi and Gujarati.

194  Lisa Björkman

accomplished through networks of social workers and politicians like Kamble and Khan as clientelistic exchanges between politicians and particular ‘banks’ of voters. They described ‘vote bank’ support for candidates as purely pragmatic, with electoral support traded in for short-term gains rather than broader ideas, issues or lasting solutions to problems which short-term ‘work’ was said to address. In this characterisation, vote bank exchanges were counterposed to the so-called programmatic politics. Corporators, the Citizens Candidate flier suggested, should limit their activities to facilitating public investment ‘honestly and justfully for the betterment of the society as a whole’. That is, prevailing interpretations of the city’s notoriously opaque and contradictory regulatory frameworks (the ‘guidelines and rules’) were to be followed unquestioningly, in the interest of the broader public good. Yet as we have seen, few of the prevailing interpretations of the regulatory framework governing land and infrastructural access are neutral vis-à-vis ‘society as a whole’; instead, they reflect deeply unequal distribution of urban land and resources—the maldistribution that representatives like Kamble challenge and sometimes unsettle. My accounts illustrate how in Mumbai the work of politicians like Kamble involves not simple concessions but real political contests. The 2012 Citizens Candidate efforts to supplant politicians with guards was a spectacular flop. The movement did not only fail to win a single seat, but barely registered any votes at all.14 Given the convergence of the Citizens Candidate mandate with the key tenets of liberal theories of ‘patronage’, these results call for a serious reconsideration of prevailing characterisations of patronage and ‘vote banking’ as a system of instrumental transactions (for more on this see Introduction, 21ff). Ethnography presented in this essay suggests that vote bank politics helps non-elite Mumbaikars make distinctly political claims. My analysis challenges not only the liberal view on patronage but also its critics who locate any redistributive and democratising potential of such politics in the legal exceptionalism of poorer voters’ claims. By contrast, my account reveals that ‘getting work done’ is not about securing exceptions to policy or to law, but about successfully navigating the legal opacities, silences and contradictions of everyday life in the city’s popular neighbourhoods. 14

In Kamble’s district, the citizen candidate that contested the seat received less than 1% of votes.

‘Vote banking’ as politics in Mumbai 195

Voters give electoral support to candidates who demonstrate access to specialised knowledge and effective sociopolitical networks, exhibit both a willingness and a capacity to act on voters’ stated preferences and can mobilise available resources—legal, political, economic— towards these much desired ends. Particular localities tend to vote ‘in blocks’, coordinating their votes on the advice of trusted local leaders. But political theorists who interpret such relations as coercive or corrupting of voters’ freedom to articulate autonomous political preferences on programmatic matters of enduring and broad-based concern are deeply mistaken. Voters coordinate locally because political matters of shared concern are often circumscribed by locality. The advice of people like Khan is not coercive, but heeded as long as Khan’s access to networks of knowledge and authority remain demonstrable through his work.15 Electoral politics in Mumbai thus inhabits a deeply political landscape in which the most pressing issues faced by the residents of Mumbai—the regulatory and policy frameworks governing access to urban land and housing, infrastructure and health, employment and economic opportunity—are interpreted, contested and also no doubt banked upon.


This point echoes Anirudh Krishna’s (2002; 2007) discussion of ‘new leaders’ (naya netas) in rural north India, and Berenschot’s point about the ‘new pragmatism’ of Indian politics today (Chapter 8).

Political fixers in India’s patronage democracy


Ward Berenschot


olitical fixers are intermediaries who use political contacts and knowledge of official procedures to help citizens, particularly the poor, deal with state institutions.1 Variously referred to as dalaals, pyraveerkars or, more denigratingly, as chamchas or taporis, these brokers have acquired a key role in the flow of information and resources between state institutions and India’s citizens, and have accordingly been described as ‘lubricants’ or ‘enablers’ of India’s democracy (Reddy and Haragopal 1985; Manor 2000). While the often-lucrative nature of their work attracts derogatory comments about the selfishness of the ‘services’ (seva) which they provide, political fixers are generally indispensable for both citizens and power holders. They channel the demands of citizens to alien and often unresponsive state institutions and supply government officials with the information necessary to implement policies and uphold regulation. They also help politicians to trade access to  state resources for electoral support. Such brokers are central to many post-colonial democracies, but it has been argued that they play a more prominent role in India than anywhere else (Manor 2000, 817).2


I thank Ghanshyam Shah, Mario Rutten, Jan Breman, Abram de Swaan and Anastasia Piliavsky, whose insights have contributed to this chapter. This chapter is a substantially rewritten version of a piece published in the Journal of South Asian Studies (Berenschot 2011b). 2 Although beyond the scope of this chapter, quite a number of interesting parallels could be drawn between India’s political fixers and the functioning of ‘brokers’ observed in countries like Argentina, Italy, Russia or Greece. For Argentina, see the outstanding account of the functioning Peronist networks of brokers in Auyero (2001). For a comparative discussion on the functioning of brokers in Russia, India and 19th-century Greece, see Jeffrey et al. (2011). See also Boissevain (1974) and Blok (1974).

Political fixers in India’s patronage democracy 197

Given the centrality of political fixers to the everyday functioning of India’s democracy, there are surprisingly few studies devoted solely to them. They make regular appearances in writings on other subjects, like the literature on India’s political parties, which highlights how in India’s ‘patronage democracy’ brokers help politicians convert state resources into electoral support.3 They also appear in writings on governance and service delivery, which show how networks of politicians and political fixers interfere with state policies (Mooij 1999; Veron et al. 2003; Baud and de Wit 2008).4 The ‘anthropology of the state’ has highlighted the role of political fixers in shaping the way common Indians see, imagine and experience the ‘everyday state’.5 Few studies nevertheless focus on India’s political fixers themselves. After Reddy and Haragopal (1985) inaugurated their study, Manor conducted an excellent comparative study of political fixers in different Indian states (2000,  818). But his ‘invitation to more thoroughgoing research’ has not been taken up until recently (Krishna 2007; Simon 2009). This chapter builds on this work by focusing on the interaction between elected politicians and political fixers. I argue that an important reason for the pervasiveness of political fixers lies in the way the mediated character of India’s state institutions creates a need for them to facilitate clientelistic exchanges between voters and politicians. Because political actors have a great deal of control over the everyday functioning of state bureaucracies, voters need political fixers to gain access to politicians and through them to state resources. On their part, politicians depend on political fixers to ensure that their efforts to meet these demands will in fact translate into electoral support. In this way fixers operate as ‘lubricants’ in clientelistic exchanges between voters and politicians. Consequently, I will argue, the informal networks of political fixers play a crucial role in India’s electoral outcomes. Focussing on the decline of Congress and the subsequent rise of Hindu nationalism in Gujarat over the past three decades, I show that this shift can be attributed to the changing nature of patronage channels through which political fixers provide access to state resources. 3

See for instance Bailey (1963), Weiner (1967), Chandra (2004), and Wilkinson (2007). These networks of politicians and political fixers have been referred to as the ‘shadow state’ (Harriss-White 2003) or ‘political society’ (Chatterjee 2004). 5 For example, Gupta (1995), Fuller and Bénéï (2001), Hansen and Stepputat (2001) and Corbridge et al. (2005). 4

198  Ward Berenschot

I illustrate my arguments with a discussion of political fixers mainly  in Isanpur, a Dalit-dominated area in eastern Ahmedabad that was at the centre of the 2002 Hindu–Muslim violence in the city and that still suffers from widespread unemployment and acute poverty, following the closure of local textile mills in the 1980s. This  chapter is based on fieldwork conducted between January 2005 and March 2006, when I spent two months living in and around Isanpur. The names of the locality and its inhabitants have been changed.

Social workers and party workers Jagdish-bhai has a small office inside his chawl (housing block). The room is made of loosely stacked bricks and decorated with flags and posters depicting Congress politicians. Above the door, a small signboard identifies Jagdish-bhai, in English, as a ‘social worker’. Jagdish-bhai sits inside, on a large bed that is almost always covered with papers. I sit on the chair where his neighbours usually sit. Throughout the week, people drop in to present their troubles: they ask Jagdish-bhai to help unclog a gutter, arrange a hospital treatment, get someone a job, or often prevent an arrest by the police. Many come to settle a dispute; Jagdish-bhai arbitrates quarrels between neighbours, performs marriage counselling and when a boy and a girl elope, he works to bring the families to agree a marriage. Today I came to speak to Jagdish-bhai about how he is handling a conflict over a boundary wall. A relatively well-off resident had been renovating his house and in the process shifted one wall 5 inches into his poorer neighbour’s land. The latter demanded that the wall be removed. The richer neighbour declined and the dispute ended in a complaint being filed with the police. This is when Jagdish-bhai got involved. ‘The police has called me’, he said: ‘If we can reach a compromise, it is much less work for them. They regularly call me for these disputes so that they have less work. Earlier, I told the poorer man that it would be stupid to go to the police. I told him not to spoil relations’. Jagdish-bhai had gone over the case, together with a local politician and a few other ‘prominent people’ (aagal padato manaso), proposing a compromise to both sides: the wall will stay, but the richer inhabitant will provide a written promise to conduct the remaining construction work properly.

Political fixers in India’s patronage democracy 199

The poorer neighbour refused to budge. ‘He is just jealous of his neighbour, I think’, Jagdish-bhai tells me. Now the case will go to Court, under article 107 of the Indian Penal Code. He further explains: When there is a fight, the case should have been registered under article 151 [i.e as a conflict involving violence]. Now [since there was no fight] it is registered under 107, which means that the case goes directly to Court without any arrest. The police would have asked for money to decide under which article the case is registered. They [the complainants] called me because they know I would complain about the police. It is my area, so I would come to know. Without people like us, they would simply lock them up.

There is one sure way to recognise a political fixer: they all have detailed knowledge of the intricacies of the Indian Penal Code. As Beatrice Jauregui explains (in Chapter 10), police constables exercise considerable liberty when people come to report a transgression. Often they simply refuse to register complaints in formal First  Information Reports (FIRs), particularly when complainants lack political backing or when the accused have influential contacts. When police officers do register an FIR, they can complicate matters by lodging a case under a ‘heavier’ article of the Indian Penal Code, or alleviate worries by lodging it under a ‘lighter’ article or by registering an arrest—if the applicable article requires it—without actually arresting anyone. Once an FIR has been lodged, the article under which a case has been registered defines the scope that police officers and social workers have for negotiating, say, whether an arrest needs to be made or whether a dispute may be settled informally. Because police officials have considerable power over the outcomes of disputes and in the adjudication of criminal acts, inhabitants prefer to involve someone like Jagdish-bhai to ease negotiations. He can prevent excessive police harassment. The police in turn rely on him to settle local issues and occasionally to arrange the bribe for their cooperation. This is where, for Isanpur’s inhabitants, the value of political fixers like Jagdish-bhai lies: he can invoke political contacts to put pressure on public servants and strengthen their grip on the bewildering and often seemingly arbitrary operations of state institutions.

200  Ward Berenschot

The pervasiveness of political fixers in India can be attributed to a combination of the character of the Indian state and the nature of its economy. On the one hand, the prominence of political brokerage is related to the relatively large role the state plays in the local economy. Both the lack of secure alternative livelihoods (particularly after the demise of Ahmedabad’s textile industry) and the discrimination that Dalits experience when searching for jobs in the private sector, make government jobs and resources highly sought-after. In this sense, the gradual liberalisation of Gujarat’s economy since the 1980s has contributed to the proliferation of political fixers; relentless pressure to reduce labour costs has increased dependence on the state as a source of income. The limited and often precarious sources of income available to Isanpur’s residents and the relative importance of state jobs, loans, welfare schemes, contracts and so on have thus contributed to the proliferation of those who provide access to them. The prevalence of political fixers also reflects the mediated character of the Indian state. One could interpret the prominence of people like Jagdish-bhai as an indication of the weakness or ineffectiveness of the state: the ability of neighbourhood leaders to steer the implementation of state regulations may be attributed to the state’s limited capacity to implement its injunctions and to the ‘fragmentation’ of state authority (cf. Hansen 2005). But this view overlooks the fact that Jagdish-bhai also derives his authority from the state and its capacity to deliver services. Local residents need people like him to help them gain access to various resources and services which the state offers. Political fixers thrive in what Susanne and Lloyd Rudolph have called the ‘weak–strong state’ (1987): the need for political fixers is the combined product of the state’s omnipresence and ambition to regulate many spheres of life (making access to the state valuable) and its limited capacity to meet the demand it generates (making access to the state exclusive). Jagdish-bhai’s capacity to facilitate his neighbours’ interaction with state institutions relies in turn on the control exercised by politicians over this weak–strong state. The relentless efforts of political entrepreneurs to acquire some degree of control over the distribution of scarce state resources has given the Indian state a mediated character, in the sense that the capacity of political actors to manipulate the implementation of policies and laws is by now firmly enshrined in the habits and regulations of state bureaucracies. Numerous policy regulations—varying from the provision of public services to the

Political fixers in India’s patronage democracy 201

selection of welfare recipients and local dispute resolution—stipulate formally the involvement of elected politicians (Berenschot 2010). The mediated character of the Indian state is further enhanced by the rich arsenal of incentives and disincentives at the politicians’ disposal: ‘sticks’ (like control over bureaucrats’ and policemen’s postings) and ‘carrots’ (facilitating rent seeking, providing desirable postings or aiding policy implementation). As a result, bureaucrats, particularly at middling levels, find it almost impossible to avoid politicians’ interference in their work (de Zwart 1994). Protestations and a strict adherence to a universal, impersonal implementation of policies and regulations would risk them deportation to a ‘punishment posting’ or obstruct their rent-seeking activities. As a result, the demands and wishes of political actors substantially shape the everyday operations of state institutions. Political fixers thrive on this politicisation of state bureaucracies. Fixers acquire the necessary leverage to facilitate interactions between their clients and state institutions through connections with influential politicians (on which more below). Quotidian instantiation of the ‘mediated state’ generates further demand for fixers’ services. Political meddling in state bureaucracies further impedes their effectiveness, perpetuating an impression that little gets done without political support. There is a Gujarati expression that describes the experience of repeated visits to government offices before anything gets done: dhakka kavaadave chhe (‘being pushed around’ or, literally, ‘being made to eat’ your repeated attempts). More often than not people who need to arrange, say, a government loan or a widow’s pension face civil servants who need to recover the money they paid for their postings, and the seekers of services usually walk away feeling it would have been cheaper and more expedient to involve a fixer. Observers have sometimes thought of political fixers as actors who maintain only loose and temporary associations with political parties (Manor 2000, 823). But this picture obscures the fact that a great deal of ‘fixing’ is affected by political parties and is also often done by party operatives themselves (de Wit and Berner 2009, 933). Some political fixers are more closely allied to a political party than others, and the character of the political alliances that they form has a considerable impact on how they operate and present themselves. I propose a somewhat stylised distinction between two types of political fixers: ‘social workers’ and ‘party workers’, terms that people often use in Isanpur. Both are

202  Ward Berenschot

political fixers who make use of political contacts to help people interact with the state. The difference between the two lies in their self-presentation and the terms of their exchange with politicians. Party workers openly support political parties and perform various organisational tasks for politicians, thereby gaining preferential access to them. In contrast, social workers like Jagdish-bhai present themselves as neighbourhood leaders (stanik netas) and community representatives, whose support for a political party is less open and more conditional on (promises of) access to state resources.6 Party workers gain access to state resources and develop their capacity to ‘fix’ problems and deals by repeatedly performing organisational services, which put elected politicians in their debt. Whereas party workers are expected to attend various party functions and perform routine tasks, social workers are only called on to support politicians during elections. To get a sense of the contrast between social workers and party workers, consider the everyday routine of Mahesh Varma, a party worker for the Bharatiya Janata Party (BJP). Every morning Mahesh-bhai can be found at the side of one of the main roads in Maneknagar, a lower-middle-class neighbourhood in the old city of Ahmedabad. He usually sits with Pravin Dalaal, the local Municipal Councillor, and four other party workers. This roadside bureau receives a daily flow of local residents who seek Pravin’s help in dealing with government institutions: securing a loan, repairing a broken pavement, settling a police case, getting proof of residence or reducing one’s hospital fees.7 The party workers help Pravin, each in his own way: one personal assistant (PA) handles (almost) all of the incoming phone calls, another passes complaints about sanitation and basic amenities to relevant officials, a third helps to fill out official forms, and a fourth deals with requests which involve welfare schemes like widow’s pensions. Mahesh is in charge of making ‘true copies’ of examination results, death and birth certificates, or tax receipts, which he turns into ‘official’ documents by marking them with Pravin’s stamp. During elections he helps to organise rallies and tours the neighbourhood. ‘I go from home to home’, he says to me, ‘and I tell people to vote for the BJP, 6

Krishna (2002) used the term naya neta (new leader) for political fixers operating in Rajasthan and Madhya Pradesh. These new leaders are comparable to the social workers I describe here (although I have never heard this term being used in Gujarat). 7 For more on Pravin’s roadside ‘mediation of the state’ see Berenschot (2010).

Political fixers in India’s patronage democracy 203

because of the work they have done’. On polling days, he oversees voting as the BJP ‘booth representative’. Mahesh-bhai has a cheerful manner and jerky movements; he is likeable but lacks the charisma that he, at nearly 50, might have been expected to have. The stamping of papers is hardly a prestigious task and it ranks him low in the subtle hierarchy of party workers. Despite having served the BJP for 25 years, he has never held an official post, not even in the party’s local ward commission. There is a hint of bitterness in the way that he speaks about his work for the BJP: There is a lot of competition in the BJP because the party is big. It is the high command who decides [i.e., to give a ticket to someone to stand for election as the party’s candidate]; they call the workers and ask what work they have done. If I want to get a ticket, I need the support of a bigger leader, or get people together and show the support I have.

Mahesh can nevertheless get a lot of small things done, and he regularly conveys his neighbours’ and family members’ problems to Pravin and the local Member of Legislative Assembly (MLA). Mahesh says he started working for the BJP in order to help his neighbours: ‘I joined because I wanted to help people, to get people services, water, gutter and to get work for young people. When there is a vacancy [for a government job] somewhere, we give reference’. This motive may well be genuine, but it is often tangled with more selfish aims. Party workers with more charisma than Mahesh develop close bonds with politicians in the hope of starting a political career of their own. And they often exploit these intimacies to make money. Many political fixers make a living out of the fees that their neighbours pay them for removing various bureaucratic hurdles. The fixers closest to politicians can earn a lot by fixing more complex and lucrative bureaucratic problems involving real estate and business permits. These fixers mainly use the petty services (their seva) that they offer to neighbours to legitimise more profitable dealings.

Clientelistic exchanges The distinction between party workers and social workers (however schematic, as many political fixers fall somewhere between these two ideal types)—helps to understand the nature of clientelistic exchanges

204  Ward Berenschot

that take place between fixers and politicians.8 Both kinds of fixers do favours for politicians to increase their access to state resources. Yet while party workers develop their access to state resources by performing various organisational services, social workers more typically make use of political channels because of their capacity to deliver votes—a capacity an influential politician can develop by responding to his people’s requests. Jagdish-bhai can put pressure on state officials because of his closeness to the local municipal councillor, Vinod-bhai. This is how he describes his efforts to arrange a budget to improve paving and water provision in the area: Poor people do not get help from officials. The officials will say: ‘we’ll come and see’, but they don’t come. Without influence (laagvag) the work does not get done. When we go directly to the officers, we can use Vinodbhai’s letterhead. If they don’t listen, then we go to the corporator [Municipal Councillor]. The officer solves politicians’ problems quickly.

Political backing increases social workers’ capacity to deal with government officials; and social workers can use the influence of politicians over bureaucrats to further their aims. Social workers use this influence directly and indirectly: they may ask local politicians to deal with uncooperative government officials or secure the budget to improve local amenities, but they may also negotiate with bureaucrats directly by using a politician’s name or letterhead. To a civil servant, association with an influential politician conveys both a threat and a promise: if you do not cooperate, I might ‘get you transferred’ or lodge a complaint against you. If you do cooperate, I might use my political contacts to arrange a better posting, a hospital bed for your father, and so on. The combination of threat and promise enables political fixers to pressurise bureaucrats or prevent harassment by the police. This is the crux of daily exchanges between social workers and politicians: by supporting social workers, politicians boost the workers’ status and give them the authority to influence the way their neighbours vote. This mechanism of exchange is referred to by the 8

In this chapter I use the term ‘clientelistic exchange’ to refer to the exchange of political support—whether direct (votes) or indirect (organisational support or money)—for access to state resources (Kitschelt and Wilkinson 2007).

Political fixers in India’s patronage democracy 205

common expression ‘to have a hold’ (prabhav chhe): ‘You get a hold if you are forthcoming, if you organise religious events etcetera… If you can solve problems with water, or when the gutter is overflowing, then you become a leader. So people start to think that what this person says is true. So they follow the opinion of this person’. ‘To have a hold’ is to have the power and the authority to condone or sanction people’s behaviour, to have political importance and the capacity to influence a sizable number of voters. These exchanges comprise the bulk of campaigning frenzy. At times of election, all social workers and party workers need to show their support for a candidate. If a social worker has backed a winning candidate, in the future this support may translate into preferential access to state resources. In the run-up to elections, Jagdish-bhai accompanies Vinod-bhai on his rounds of the neighbourhood (lok sampark or ‘meeting the people’) and delivers speeches praising the candidate’s virtues; he attends rallies and, on election day, makes sure that his neighbours turn up to vote. After elections, both candidates and social workers study the results in great detail. If a social worker has managed to deliver a large number of votes from ‘his area’ to the winning candidate, the demands he will make on that politician will be reinforced. A proven ‘hold’ over a large group of people can, in the long run, enable a political fixer to make the much coveted career move: to acquire the party ticket to fight local elections and become a political patron in his own right. Once a political fixer acquires considerable standing and a record of having done favours for others in his area, he could cash in on these favours to gain the necessary financial, organisational and electoral support to contest elections. For many fixers, everyday deal-making is a means to a political career and more independent control over state resources. In this sense, cooperation between elected politicians and political fixers is always uneasy: while politicians need political fixers to maintain their electoral base, by responding to their requests they risk empowering their future competitors. Women fixers are at somewhat of a disadvantage in these exchanges. In Isanpur there are a few women who operate in a manner similar to Jagdish-bhai and Mahesh-bhai. But social restrictions on their movements hamper the female fixers’ capacity to develop useful contacts with politicians, bureaucrats and the police. Respectable women should not be seen sitting around, wandering outside late at night or chatting with men. Interactions with influential men, who may ‘want something

206  Ward Berenschot

from you’ are frowned upon with the result that the more effective female fixers tend to be thought of as ‘loose’. In Ahmedabad female municipal councillors rarely get a party nomination for two consecutive terms (and generally much fewer than men), which suggests that the normative restrictions imposed on women limit their capacity to widen networks and develop an independent power base. What does work in their favour is their capacity to shame uncooperative government servants: some female political fixers in Isanpur were known for their harsh scolding of bureaucrats and police officers, which can be more effective when it comes from a woman who has the moral authority of a sister or mother. Exchanges between politicians and social workers are not as unequal and exploitative as the patron–client relations described in the older literature on clientelism (e.g., Scott 1972; Boissevain 1974), where the client is much more dependent on the patron than vice versa. In Isanpur social workers need the backing of politicians to develop their capacity to solve local problems, but once they get a hold over their people, politicians become dependent on their votes. If a politician does not deliver on his or her promises, social workers can switch loyalties and support another candidate. This gives them substantial leverage over politicians. Nor can one say that Jagdish-bhai’s neighbours are hapless bystanders: most residents of Isanpur say they voted for those ‘who did the work’, their choice contingent on a political leader’s proven capacity to provide (see Björkman and Price, Chapters 7 and 9). Electoral choice is based both on the candidates’ perceived capacity to deliver services and on their perceived accessibility, often via fixers. Voters will only be inclined to heed Jagdish-bhai’s voting counsel if he has proven to be effective in persuading politicians to do their work, and if another fixer is not likely to be more successful (see de Wit 2013). In this sense, voters’ choice is to a great extent an expression of trust or distrust in the political fixer. As de Wit and Berner (2009) have argued, for poorer citizens patronage of this kind might be a more attractive avenue to state resources than ‘civic’ forms of collective action (see also Chatterjee 2004); demonstrations, rallies or letter writing are time-consuming and often less effective. A comparison with older literature on Indian politics suggests that there are two important changes in the ways that fixers facilitate political exchanges, which have taken place since the dismantlement

Political fixers in India’s patronage democracy 207

of the one-party ‘Congress system’ in the 1980s (Kothari 1964; Weiner 1967; Brass 1984). First, contemporary political fixers differ from ‘local notables’, on whom the Congress party used to rely to deliver votes. Earlier, fixing was done mainly by traditional upper-caste patrons, village strongmen and other ‘big men’. By now, however, these older political fixers have been largely displaced by people whose authority and reputation rely more on a proven capacity to get things done than on age, wealth or family standing. Second, increasingly intense political competition that resulted from the demise of the Congress system has spurred the proliferation of political fixers, enabling inhabitants to choose between workers and to employ those who actually live up to their promises (Kohli 1990; Krishna 2007; and Wilkinson, Chapter 11 in this volume). This has made the system of exchanges more pragmatic, as considerations of ideology or policies have given way to more instrumental considerations of a capacity to arrange a streetlight, a hospital bed or admission to a college. This new pragmatism constitutes an important reason why studying the changing strength of the patronage networks of political fixers can help understand electoral outcomes. As access to state resources figures prominently in voters’ considerations, a change in the capacity of political fixers and their patrons to provide it can substantially shift electoral preferences. While transformations of patronage are more opaque than, say, changes in political discourse or the rise of new political movements, they are often related to, and sometimes fuelled by, underlying shifts in patronage networks on which political fixers draw. A dramatic political shift that occurred in the 1980s in Gujarat offers an illustration.

The BJP will help you In the mid-1980s a sizable section of Dalit voters moved away from the Congress towards the BJP, enabling the latter to establish political dominance that has lasted until this day. As we turn to the career of another political fixer, the BJP supporter, Jayent Parmar, I  will show that it was because of their dependence on political fixers that Dalits could be swayed to support a movement that in terms of both its ideology (the defence of caste hierarchies) and policies (opposition to reservations for backward castes) offered little to Dalit communities.

208  Ward Berenschot

In the daytime, Jayent works as a lab assistant at the Gujarat University, but in the evenings he can often be found receiving guests or holding meetings with the local youth on a charpoy outside his home in Isanpur. Some of his visitors call him ‘professor’, a title that inflates his position at the university, but which goes well with the learning and thoughtfulness which he exudes. He is the chairman of a city-level Dalit BJP committee and one of the most prominent local BJP workers. On his charpoy Jayent-bhai busies himself with his neighbours’ problems: ‘people come to me’, he says, ‘because of my political connections. And people believe that an educated and a good person should solve these issues’. One of these issues was a form filled in by one of his neighbours in application for the 12th-grade board examination, the exam that determines whether a student can enter a good college or university. According to the boy, the clerk who had filled in the form made a mistake; as a result, his application was rejected and he was debarred from sitting the exam. His parents approached Jayent-bhai, who offered to meet the school’s upper-caste principal (‘a very influential man’) to solve the matter. Jayenthbai said: ‘I went there and talked with a lot of respect that “sir, it is your mistake, you should help make arrangements for this person to take exams”’. The conversation ended in a shouting match and the exchange of caste-related curses, and someone called the police. Jayent-bhai continued: ‘So the police came and asked me first: “what is the matter Jayent-bhai?” (because the entire police staff know me). They know that I would not do babaal (shouting) just like that’. This lowered the pitch of the confrontation, but the principal still refused to let the boy sit the exam. The next day Jayent-bhai lodged an FIR with the police. The principal was not amused and got some political leaders to ask him to withdraw the complaint. But as the principal still did not budge on the issue, Jayent-bhai refused and even organised a small demonstration, which got some news coverage. In the end they reached a compromise: the boy’s family received 25,000 rupees (about $550) as compensation, while the boy had to wait another year to take the exam. Jayent-bhai said that he did not take any of this money. Jayent could book this minor success because of his good connections, and his experience of putting pressure on unwilling civil servants; the boy’s family depended on him because they lack connections with the police and the capacity to mobilise people.

Political fixers in India’s patronage democracy 209

Jayent-bhai may not have received any financial compensation, and the incident certainly soured his relations with an influential principal (something fixers usually avoid as their livelihoods depend on such ties), but the case still offered ample rewards. The incident gained fame and could be presented as a fight of a lower-caste student against an oppressive upper-caste bureaucrat; it enhanced Jayent-bhai’s image as his community’s defender. Apart from their earning potential, problems like this one offer fixers opportunities to gain local fame and strengthen their hold over a constituency. As Jayent-bhai put it, ‘when there is a water problem, an educational problem, [you have to] take an active part. This is what makes a person popular: If I can solve a problem in 24 hours, people will think: “he is an active person, he is efficient.”’ His neighbours thought that Jayent-bhai’s day-to-day activism destined him to a prominent political career. One said: ‘at every small issue he would be in the front and he was very well known. He could have become an MLA but somehow he could not use his political power’. Jayent joined the BJP in the 1980s. He was still in college when a BJP worker approached him. He recalled that the worker said to him: ‘our party believes in nationalism. You are a Hindu, although you are an SC [Scheduled Caste]. From now on there will be no more untouchability. We will help you’. Before 1985 the BJP in Gujarat was largely an uppercaste party and it mobilised voters around the issue of reservations for members of Scheduled Castes of places in government and higher education. This made it difficult for the party to get Dalit votes: in Isanpur there are still stories about leading BJP politicians decorating a statue of Ambedkar with garlands of shoes (a derogatory gesture) during an anti-reservation rally in the early 1980s. But by the mid-1980s, the BJP shifted its strategy in an attempt to gain Dalit support, and in 1985 the Vishva Hindu Parishad (VHP) (World Hindu Council) asked its youth members to dedicate themselves to the abolition of untouchability (Nandy et al. 1995, 103). The 1985–1986 riots dramatised this shift: during the first days the fighting was between Dalits and upper castes, but in the latter phase of the riots Dalits joined the upper castes to fight the Muslims. One of the standing questions in the literature on Gujarat’s politics is what drew lower castes to the Hindu-nationalist movement that had campaigned against reservations and defended caste hierarchies (Ghassem-Fachandi 2010). During the outbursts of communal violence

210  Ward Berenschot

in Gujarat in 1985–1986, 1990, 1991 and 2002 Dalits began to side with upper-caste Hindus and turn against Muslims (who were often their neighbours), helping the BJP establish and maintain political dominance.9 Some have related this political shift to the ‘informalisation’ of the labour market after the demise of Gujarat’s textile industry in the 1980s: Jan Breman in particular has argued that the resulting scramble for jobs and resources enabled the BJP and its affiliates to tap into growing Dalit–Muslim tensions (2004, also Shah 2002). Others connect this important shift to something they call the ‘politics of recognition’ (Shah 1996; Yagnik and Sheth 2005; Shani 2007, 155): they say that Dalits turned to the BJP out of a desire to be recognised socially and to gain respect and social mobility through the Hindu nationalist movement, and the Hindu identity that it promoted. Shani (2007) describes the upper-caste politicians’ alliance with Dalit leaders as a more or less conscious strategy to suppress tensions between upper and lower castes, which by the 1980s began to threaten existing social hierarchies. She writes that upper castes, anxious over the increase in Dalits’ social mobility and their reservations in government jobs and college seats, came to support ‘ethno-Hinduism’. But her analysis is less convincing when it comes to explaining why Dalits, who did not share this anxiety, could be drawn in to support an ideology and a political party that had often openly opposed reservations for them. We cannot understand this turn of Dalit loyalties away from the Congress and towards the BJP through ideology or anxieties over identity and recognition alone. It needs to be seen also as a result of the growing dependence of Dalits on political fixers for obtaining state resources, a dependence that has fostered a highly pragmatic attitude to politics. As voters rely increasingly on politicians for the provision of state resources, promises of tangible, short-term gains (an electric connection, a repaired gutter, a hospital bed) feature much more prominently than abstract policy proposals or misgivings about ideological commitments among their considerations. Take, for example, the way one Isanpur resident reasoned about the vote: ‘I vote for the active worker, and the person who understands people 9

During the 2002 riots violence was most rampant where Dalits lived alongside Muslims, and Dalit youths seem to have been the main perpetrators, egged on and supported by upper-caste party leaders (Engineer 2003; Bunsha 2006; Berenschot 2012).

Political fixers in India’s patronage democracy 211

and somebody who solves problems for people, who can pass on a problem to top-level people and solve it for them. It does not matter which party [the candidate belongs to]’. Another said: ‘We see who does the work, we roam around a lot, so we know what work people have done. If the politicians do not do the work, we will vote for someone else’. And a third: Ideally you should vote for the person who does the work. If that person is clean, there is always the question if he will do the work or not. For example if there is some problem in the street, like if the stones have come out … then a corrupted person will send someone and that person will do the work for 50 rupees and show the work of 200 rupees. But in the end our work will be done, no?

When state resources are mediated by political go-betweens, there is little scope for considerations of ideology, policy or ‘corruption’; what matters is being able to ‘bring a problem to the top people’, to ‘do the work’. This means that when the older Congress fixers lost their capacity to contact ‘the top level people’, the BJP could start to woo Dalit voters by offering them the much-needed alternative for getting things done through new workers like Jayent-bhai.

Changing channels of patronage This is how Jayent-bhai describes the launch of his political career in the late 1980s: They [the BJP] thought that if we are to come to power, we need to have the support of these low castes. So they took more interest in this area. They started contacting people, educated persons who know the political situation. They told me: ‘we will help you if you join’ and then they did help. Even I could call [a leading Gujarati BJP politician] in the middle of the night. I remember I called him at 1:30 in the night once and he immediately helped me to get blood for a patient in the hospital. At that time the BJP took interest in educating people, and it slowly became popular in Isanpur. We told the people: ‘the BJP will help you’. That was the strategy. For five years [after that] they did work for us. Then the communal problems started and this led to more support. Now they do not respond as much to our requests.

212  Ward Berenschot

Expectation that the BJP might be more willing than the Congress to ‘do their work’ are related to gradual changes in the political economy of Gujarat. Until the late 1970s, many inhabitants of the textile mill neighbourhoods like Isanpur interacted with state institutions via the Textile Labour Association (TLA), Majoor Mahajan Sangh. At that time, the TLA formed the textile mill area fixers’ main channel for securing access to state resources. This labour union, which was set up by Gandhi in 1918 to represent textile labourers vis-à-vis owners of textile mills, played a prominent role in Gujarat’s politics because of its large membership (said to approach 100,000 in the 1960s), helped the Congress to win elections. In their turn, the TLA used elected Congress politicians to gain access to the state: while the Congress used the TLA to capture working class votes, the TLA acquired through the party useful contacts to help their members deal with expanding municipal institutions. The TLA and the Congress set up ‘worker-voter associations’ with both Hindus and Muslims as members. During elections, they actively campaigned for the Congress candidates, and after elections lobbied for the improvement of local facilities (Spodek 2011, 101; also Breman 2004). This political deal enabled the TLA to help its members by taking up various public and personal problems. They set up an ‘area department’, the latta khatu, to relay requests and complaints to relevant authorities. People could report disputes with the management of the mill, problems with basic amenities, police harassment, paperwork troubles, and so on. The TLA tried to solve these issues using its contacts with leading Congress politicians and influential officials in government departments, appointing for this purpose special representatives (pratinidhi in Gujarati) in every textile mill neighbourhood. Until the 1970s the pratinidhi were the main political fixers in places like Isanpur, and it was their duty to communicate local concerns to the main TLA office. Unlike present-day fixers, the pratinidhi were not directly dependent on politicians; their access to public services and politicians was mediated through the TLA and its ‘area department’. These patronage channels started to wither in the 1970s. The split of the Congress, engineered in 1969 by Indira Gandhi, badly damaged the party’s grassroots network. As many local Congress workers had supported the Congress (O), the faction which ultimately lost to Indira Gandhi’s Congress (R), this episode seriously undermined

Political fixers in India’s patronage democracy 213

the once-strong network of Congress fixers in Gujarat (Kohli 1990, 238–367; Yagnik and Seth 2005). Subsequently, in the 1980s, Ahmedabad’s textile industry also went into decline; many mills were downgraded or shut. The retrenchment of thousands of workers and the failure of the TLA to represent them properly weakened the organisation. As the TLA lost credibility with the textile workers, its membership dwindled and the privileged access to key politicians was also lost. The network of pratinidhi disappeared, and the latta khattu became ineffective. The decline of the TLA and the Congress grassroots network meant that the residents of Isanpur could no longer rely on the Congress and TLA patronage channels to solve their daily problems, and the textile labourers were left without dependable access to the state. This was precisely when the BJP and its affiliate militant groups—the VHP, the Rashtriya Swayamsevak Sangh (RSS) and the Bajrang Dal—started to gain Dalit favour. These organisations realised that they could not win elections and ward off the threat to upper-caste dominance without winning over the Dalits. Despite the dwindling of its ground-level networks, until then the Congress had succeeded in winning elections by strategically targeting a mixture of caste and religious groups referred to as KHAM (Kshatriyas, Harijans [Dalits], Adivisis, and Muslims). The BJP and its affiliates realised it could break this coalition by recruiting Dalit support. Not only did they proclaim they were fully committed to ending caste discrimination, they also actively recruited lower-caste leaders to the party. Hindu nationalist organisations had by then much to offer: gradually, the RSS, the VHP and its many affiliates had managed to place their members in government institutions, universities, local village and municipal councils as well as the broad range of para-statal organisations that play an important role in everyday life. They took over cooperative banks, farmer cooperatives, educational institutions, credit societies, milk cooperatives and agricultural produce market committees (Seth 1998, 28; Sud 2007). Nowadays membership in the VHP or the RSS, through which ‘influential people’ at various levels of the government can be contacted, is very useful for securing a job, arranging a licence or solving a tax dispute.10


For a further discussion of changes in the channels of patronage in Gujarat, see Berenschot (2012, 59–77).

214  Ward Berenschot

These changes in the political economy of Gujarat—the failure of the Congress network to provide access to state resources and the increased strength of the Hindu nationalist patronage channels—gave lower-caste leaders like Jayent-bhai good reasons to respond to the BJP and its affiliates. The many social and party workers in Isanpur who joined the BJP in the late 1980s and early 1990s did so because they could benefit from the growth in BJP’s patronage channels. To further illustrate this point, I will quote one of the oldest BJP workers in Isanpur, Harjivan, who spoke at some length about his reasons for joining the party: So we used to take work [to the offices of the municipality] as ordinary citizens. We had to go back and forth between the municipal officials (dhakka khaava padta hata). But the work would not get done. For weeks, months, years the work would not get done. Take, for example, our attempts to get benefits from the welfare schemes for us, the people of the backward and scheduled castes. The officers did not give the benefits. They would not even answer. But if some political party or some well-known social worker— who can put some pressure—went [to the office], then he would quickly get a form. His work will get done. As an ordinary person, you had to face this difficulty from different state departments. That is why it seemed to me that if I don’t want to suffer all these difficulties and if I want to do the work of the people properly, I have to be with a party. So in the coming days, whichever party is coming to power, why should we not do the work of that party? Okay? So that if that party [which we have supported] comes to power, our work will get done. A  friend of mine told me that in the coming days, this party [the BJP] will come into power. So you catch such a party and your work will get done.

Harjivan’s tale shows how the BJP gained popularity among Dalits. As local political fixers succeeded in boosting their status and effectiveness by associating with the BJP and other Hindu nationalist organisations, they could convince their neighbours that by supporting these groups they could improve their access to state resources. As I discussed in the first part of the chapter, because people depended on local political fixers for solving everyday problems, the fixers gained influence in their neighbourhoods. As workers like Harjivan and Jayent-bhai developed their capacity to

Political fixers in India’s patronage democracy 215

solve problems for their neighbours through contacts in the BJP, they expanded their status and authority, and their capacity to convince Dalits to vote for the BJP. At the same time, older fixers affiliated with the Congress lost some of their influence and the incentives needed to court Dalit votes.

Conclusion I have discussed in this chapter how political fixers help India’s patronage democracy work as they facilitate clientelistic exchanges between voters and politicians. They enable citizens to pool their votes and maximise their pressure on politicians, who in turn rely on fixers and their local ‘hold’ to ensure that efforts to meet the demands of constituents will translate into votes. I have argued that the pervasiveness of political fixers and their patronage networks is a crucial determinant of electoral outcomes. I have also suggested that it was because poor citizens depended on fixers that the BJP, a party with an upper-caste image and outlook, could attract lower-caste voters in Gujarat. As the patronage networks of the Congress Party collapsed after the 1970s, the expansion of Hindu nationalist clientelistic channels offered the residents of poorer neighbourhoods a much-needed alternative access to state resources. The willingness and capacity of the BJP and its affiliates to provide this meant much more for the daily life of many local inhabitants than the party’s social conservatism or its opposition to the policies of reservation. These observations also illustrate how the poorer citizens’ dependence on fixers limits their capacity to change the overall distribution of power in society. As marginalised communities struggle to benefit from India’s economic growth and face discriminatory hiring practices in the private sector, access to state resources remains vital to livelihoods in neighbourhoods like Isanpur. The resulting dependence on political mediation limits their capacity to act on policy-related considerations. It may be true that, as Partha Chatterjee argues, poorer citizens can sometimes circumvent pernicious laws and policies through the wheeling and dealing of political fixers, expanding ‘their freedoms by using means that are not available to them in civil society’ (2004, 67). But this is an overly optimistic assessment: in practice, dependence on political intermediaries means that for the sake of maintaining a capacity to

216  Ward Berenschot

mitigate the implementation of state laws and policies, marginalised communities sacrifice their capacity to shape their drafting. While economic and social changes are altering the role and character of political fixers, it is unlikely that the dependence of poorer citizens on them will diminish soon. The decline of the one-party ‘Congress system’ made political fixing a much more competitive enterprise: to attract clients today fixers have to deliver. This ‘marketisation’ of political fixing has boosted the pressure that voters can put on politicians, making clientelism less exploitative as a result. These developments, alongside the growth of India’s middle class, may slowly open up the space for more programmatic, policy-based voter considerations, which may in turn lessen the demand for fixers. As James Manor (2010) argued, the recent boost in state revenues has enabled politicians to implement various ‘post-clientelistic’ programmes, like NAREGA (National Rural Employment Guarantee Act), which aim to provide a more universal, impersonal access to state resources and to diminish reliance on fixers. As Steven Wilkinson furthermore argues in this volume, India’s economic development might reduce the demand for clientelistic services (Chapter 11, 278). I nevertheless suggest that in the near future both economic growth and the occasional adoption of more programmatic electoral strategies are unlikely to undermine the importance of political fixers, which is largely premised on the politicisation of state bureaucracies and the control political actors exercise over civil servants. This control means that poorer citizens rely on political intermediaries to access most state resources; it also continues to make apparently ‘post-clientelistic’ programmes vulnerable to political manipulation (Elliott 2012). As long as state bureaucracies remain susceptible to political interference, the need for political fixers and patronage in India’s democracy is unlikely to subside.

Patronage and autonomy in India’s deepening democracy


Pamela Price, with Dusi Srinivas



heorists of liberal democracy assume that voters have the right to exercise independent judgement, and interference with individual choice through threats or bribes is as illegal in India as it is in other democratic states.1 But voters’ choices everywhere are invariably influenced by politicians, even when direct payments or coercion is not involved. Critics condemn such ‘undue influence’ as undemocratic. They say they deprive voters of the full rights of citizenship and turn them into subjects of political lords. This chapter explores the morals of electoral choice in the Telugu-speaking region of Telangana in South India to demonstrate that while voters may be influenced by political patrons, they are not merely ‘dependent clients’, but mature political agents (e.g., Heller 2011, 157). Examining electoral reasoning among farmers in a village I call ‘Balapalle’, I show that their assessment of actual and potential politicians is guided by the value of ‘goodness’, not by helpless compulsion. I argue that in clientelistic politics electoral choice is not delimited by bribery or coercion, but remains conspicuously a choice arrived at through moral deliberation. If to a liberal theorist client-voters appear like bound subjects, the actors see themselves—and act—as citizens who make reasoned decisions. Following India’s independence in 1947, the Congress Party ruled both at the centre and in the states. Under the first-past-the-post regulation, the party ruled at large, even when it did not receive the majority of votes. By the late 1960s, however, the dominance of the Congress across the country was on the wane, and both new and old parties began to challenge the party in urban and rural areas. 1

Many thanks for comments from David Gilmartin, Anastasia Piliavsky and the Alamos (Mexico) Writers’ Group led by Diane Carpenter. I am grateful to M. Raju, Praveena Rao and Dusi Srinivas for help with translation. The authors assume full responsibility for the contents of this chapter.

218  Pamela Price, with Dusi Srinivas

Under the Congress regime, party bosses could rely on ‘vote banks’ controlled in rural areas by dominant-caste landholders who dictated to their poorer, lower-status clients how they should vote. But the late 1960s brought greater choice to the ballot. Various processes of change held the potential to turn clients into citizens. National and regional parties emerged, seeking to promote the interests of Dalits (former untouchables) and lower castes. Clientelism has nevertheless remained central to party politics and the governance of the state. The term ‘patronage democracy’ is shorthand for the broad discretion exercised by Members of Legislative Assemblies (MLAs), Members of Parliament (MPs) and petty politicians in the distribution of state resources and services to constituents (Chandra 2004). As Sanjib Baruah and others have argued, in a patronage democracy voters attach much greater significance to the person of a politician than to the formal policies of his or her party (Baruah 2010, 188; Kitschelt and Wilkinson 2007). While patronage persists in India’s electoral politics, analysts have remained largely deaf to the tone of India’s patronage democracy. We still know little about the meanings attached by ordinary people to the vote, their understandings of political processes and conceptions of leadership. Such comprehension is slow in coming partly because it requires laborious forms of investigation. While rapidly conducted surveys with standardised forms supply some of this information, uncovering issues of meaning often requires long-term observation and loosely structured conversations, which take time. This chapter draws on extended interviews carried out in 2003–2004, 2007 and 2012 together with Research Associate Dusi Srinivas.2 All interviews were taken in ‘Balapalle’, 80 km south of Hyderabad and 10 km away from the nearest market town.3 2

Between October 2003 and February 2004 every 2 weeks I conducted semistructured, open-ended interviews with approximately 35 people, focusing on informants’ general conceptions of authority, morality, honour and respect. Dusi Srinivas was one of the interpreters who assisted me. Some of the findings are presented in Price (2013b). In 2007 and 2012 Dusi Srinivas carried out interviews on his own, staying approximately 2 weeks each time. He followed my questions, both times focusing directly on notions of leadership and responses to the promises made by candidates in their 2003–2004 campaigns. We discuss results from the 2007 interviews in Price with Srinivas (2013) and Price and Srinivas (2014). 3 In 2003 Balapalle housed 800 families, including a nearby hamlet. I talked mostly with medium and large landholders. A medium landholding included between 3 and 8 acres; large holdings were over 8 acres in size. The farmers came from a list of randomly selected men, supplied by an agricultural research institute. Other people

Patronage and autonomy in India’s deepening democracy 219

Figure 9.1 Map of Andhra Pradesh showing the state’s three main regions. Map drawn by the Cartographic Unit of the Department of Geography in Cambridge. Our interviews give a glimpse into Balapalle’s experiences of clientelism in democratic process, and show voters involved in relations of patronage articulating a desire for autonomy. This desire I met casually or selected because of their social and political significance. Those who were interviewed included, among Scheduled Castes (ex-untouchables, Malas and Madigas), 4 men and 2 women; among lower castes, 15 men and 3 women (mostly from the Toddy Tapper caste); among upper castes, we interviewed 9 men and 2 women (mostly Reddys); and 2 Muslim men. The term ‘Dalit’ (referring to ex-untouchables) was not used in the village at this time. In 2007 Srinivas interviewed 26 people, including mostly those we had met in 2003–2004, with approximately the same spread of castes. In 2012 he interviewed 23 people, including several politically active men whom we had not met before. Our informants included literate and illiterate people, the politically quiescent and the politically active, the old, the middle-aged and the young.

220  Pamela Price, with Dusi Srinivas

pervaded their comments on the ethics of human transactions and their moral assessments of political leadership. These conversations also provide clues to the cultural logic of local political campaigns. Our conversations focused on the 2004 election to the State Assembly in Andhra Pradesh, and on the main contenders to the post of Chief Minister: the incumbent Chief Minister, Chandrababu Naidu of the regional Telugu Desam Party (TDP), and the Congress front-runner, Y.S. Rajasekhara Reddy, better known as YSR. Both candidates tried to meet their constituents’ wish to vote for a ‘good person’. But what was the ‘goodness’ that they appealed to? Politicians’ performances usually reach for sentiments, symbols, rhetoric and visions of community, which appeal to the electorate’s targeted segments.4 Yet, as I show, rhetoric and performances in both campaigns focused on conceptions of personal, moral constituents of ruling authority. The two main contenders for the Chief Minister’s seat were in many ways poles apart. The TDP was in its second term as the ruling party and, as the incumbent regime, was on the defensive, while the Congress attacked. Chandrababu Naidu, who attracted national and even international attention as a champion of neoliberal policies, claimed that ‘development’ in the state had moved forward considerably during his time in office. Indeed, he presented himself as the embodiment of development in Andhra  Pradesh. YSR came forward as a representative of the farming sector, at that point battered by three consecutive years of drought. YSR dressed like a farmer and talked with villagers as if he were one of them, while Naidu wore long-sleeved shirts and western-style trousers and had a more formal way of carrying himself. YSR and other Congressmen castigated him for neglecting destitute agriculturalists. Instead of expanding irrigation infrastructure in rural areas, they charged, Naidu was busy developing ‘hi-tech’ governance. Parts of the capital city were beautified and Naidu declared that the future of agriculture lay with corporate farming. He demurred that he had an ‘all-around focus’ on development, which did not exclude the rural sector. It would appear, then, that these elections were more about policies than personalities.5 A major event, which played a role in the framing of the TDP campaign, was the attempted assassination of Naidu by members of a Maoist group on 1 October 2003. Naidu narrowly escaped death 4

See the Introduction and several essays in Price and Ruud (2010). See Price (2010; 2011) for a further description and analysis of the two campaigns.


Patronage and autonomy in India’s deepening democracy 221

and suffered fractured bones. In his subsequent campaign, Naidu invoked his brush with death to tell voters that by trying to kill him, the guerrillas aimed to thwart development in the state. In October and November his survival became the central slogan of the TDP campaign: Naidu spoke about the pain he endured and said that he would sacrifice his life to serve the people. Late in November he stated: I survived only because of the blessings of [the god] Lord Venkateswara and the people of the state…. I think I survived because of my good deeds…. Since I survived the attack because of the blessings…, I have made up my mind that I should do something good for the people for the rest of my life (Deccan Chronicle 23 November 2003, ‘Life is a Matter of Moments: N. Chandra Naidu.’).

By late November, however, this focus gave way to the TDP-led campaign aimed at convincing farmers that their interests were primary for the Chief Minister. Meanwhile, Naidu continued to counter accusations of selfishness by periodically reminding his audiences that he had almost given his life in their service, and would gladly do so again. He frequently spoke about how the incident resurrected him and refocused his dedication to the people. His new persona was decisively pro-poor. YSR gave Naidu stiff moral competition with regular statements of his dedication to the poor and his will to self-sacrifice. He had strengthened his profile by going on a 1500-km trek in the hot season of 2003. He walked, dressed as a farmer, talking with rural people about the trials in their lives, especially water scarcity, made worse by the drought. The padayatra (foot march), as the trek was called, attracted a great deal of media attention. This gave YSR opportunities to reiterate the Congress Party’s criticisms of the ruling government and to state his party’s plans for an agricultural revival, including free electricity for farmers and a full-scale development of irrigation. At the same time, YSR had to counter the mounting suspicion that the main beneficiary of the padayatra was his own political career. He told villagers that he could not as yet do much to aid them, but that he wished to give them hope and alleviate their suffering. He was generous with his time and reckless with his health. When speaking about his endurance of the summer heat, he said: ‘I am not bothered about my comfort. I am ready to sacrifice my life for the welfare of the poor’ (The Hindu 1 May 2003, ‘Govt. Forcing Riots into Debt Trap, Says YSR’.).

222  Pamela Price, with Dusi Srinivas

In 2004 the Congress won the Assembly elections, capturing 185  seats, while only 47 went to the TDP. However, in terms of votes cast, their advantage was trifling: the Congress received 38.5% as opposed to the TDP’s 37.5%.6 The Congress multiparty alliance received 48.37% of total votes in the state.7 Voter choice showed no resounding preference for either party. While some observers interpreted the Congress victory as the repudiation of the TDP’s reform agenda, election analyst K. C. Suri suggested that there were many different motives among voters in casting their ballots (2004, 5496). Others argued that the TDP’s incumbency contributed to the party’s electoral failure, as was often the case across the country.

Changing relations of patronage in Balapalle Balapalle lies in one of the driest and poorest districts in the state and suffers regularly from droughts, and YSR’s promises to expand irrigation and especially to provide free electricity for extracting ground water had strong appeal. In the constituency where the village lies, the Assembly seat went to the Congress candidate and a year later, a Congress supporter was elected as Village President over the Telugu Desam incumbent. During the preceding 30 years in Balapalle, as in other parts of Andhra, traditional relations of patronage had been substantially eroded (Rao and Reddy 2008). Previously these had been based on landholdings, caste status, control of village offices and dominance over the local distribution of credit. By 2003, however, the structure of authority in Balapalle had been considerably flattened. Big Persons, pedda manushulu, were no longer primarily high-caste major landholders, many of whom, in any event, had sold their lands and moved to cities. Expanded literacy and access to media, selfassertion through party politics, increased landholding and new sources of credit (among other changes) expanded and heightened the sense of personal dignity and independence among lower castes (Price 2013b). Farmers from lower and scheduled castes reiterated 6

In the first-past-the-post system, the candidate with the highest number of votes wins the election in a particular constituency. With skilful alliances and the careful selection of candidates, when several candidates stand for the same seat, a party can secure more seats than the percentage of votes cast in its favour indicates. 7 Congress allies, including the Communist Party of India (Marxist) and the Communist Party of India, together won 15 seats and the separatist Telangana Rashtra Samiti won 26. The TDP’s ally, the BJP, won only two seats (Suri 2004, 5495).

Patronage and autonomy in India’s deepening democracy 223

pride in their freedom from the subordination they suffered under old-style patronage. They said that people were conscious of making their own decisions and of not being told what to do by somebody outside their family. A scheduled-caste farmer asked me rhetorically— ‘Who are you [to tell me what to do]?’—and many insisted that ‘there is a lot of change because now people lead their own lives because they have their own land and work’ (Price 2013b). As upper-caste families gave up farming, land came on the market and both lower and scheduled castes expanded their holdings or became landowners for the first time. Although the old regime of upper-caste landlords had drastically decayed, the resources and services provided by the state in Balapalle continued to flow along lines of patronage.8 Village factions were attached to political parties, through which local party leaders and party workers endeavoured to secure state resources and services for their supporters. In 2003–2004, when the TDP ruled the state and the Village President was a TDP party worker, Congress led the rival faction in the village. By 2007 two Congress factions had formed, one supporting the Congress Chief Minister YSR and the other, a Congress MP. In 2007 an illiterate woman from a scheduled-caste family loyal to the TDP said, when asked why she voted, that ‘if a person from our party wins the elections, then he will work for us. With this promise, we vote’. Most villagers said that they voted for the party associated with the faction that they supported, and that they supported factions mainly for the state resources and services these promised to provide. Informants’ comments suggested that politicians of all stripes distributed state benefits in the following way: first they saw to their own financial recovery from heavy campaign expenditure and secured their personal finances; then they rewarded their close followers and party workers; then they heeded the requests of those they believed to be party supporters; finally, if anything was left to spare, they could assist the rest. Politicians worked in a challenging setting. As informants from both upper and lower castes reported, villagers had developed increased expectations of what the state could and should do for them. This heightened sense of entitlement was attributed to increased ‘awareness-knowledge’ (telivi), which was supposed 8

This and the following sections build in part on Price and Srinivas (2013) and Price and Srinivas (2014).

224  Pamela Price, with Dusi Srinivas

to have been rising particularly since the establishment of the TDP and its coming to power in 1983. The founder of the party, N.T. Rama Rao (NTR), a famous film star and an upper-caste Kamma man made successful appeals to his caste-mates, who compete with Reddys for dominance in state politics. He built low-caste support bases through populist appeals, famously promising to provide rice at two rupees a kilo and to abolish traditional village offices dominated by upper-caste Big Men. NTR carried through legislation which stipulated that the position of Village President should be reserved periodically for scheduled and lower castes, as well as for women. In 2007 a lower-caste informant reported that when the first scheduled-caste man became Village President, he felt that some day it might be possible for someone like himself to achieve recognition. When informants talked about effective governance, or its lack, they spoke mainly of politicians’ personal qualities, rather than about party policies or political processes. In this way, patrimonial notions to greater or lesser degrees characterised most political visions. We can nevertheless refer to patronage democracy because it was crucial for informants that they could register their political preferences through the electoral process, with the hope of electing better, more moral leaders. In 2007 most of our informants, representing a wide range of caste, economic and political backgrounds, said that they had two motives for voting. First, if one did not vote, one risked being struck off the electoral roll as dead. Witnesses said that being on the electoral roll and casting one’s vote established one’s continuing (formal) existence and, hence, one’s rights to benefits offered by the state. This conviction comprised two related conceptions: (1) one’s vote was one’s civic identity and (2) if one did not vote, in a significant sense one ceased to exist in the village. Even villagers with strongly patrimonial views had a conception of rights to state resources, secured by one’s inclusion in the electoral roll and confirmed by the casting of the ballot. Secondly, most stated that they voted in the hope of helping to bring into office a ‘good’ person—someone who would do something for them. A lowcaste farmer explained: If I don’t vote, I don’t have any relation with the village. I  vote because I think someone who gets elected will help us, do something for us and the village. He will help us if any help comes from the government.

Patronage and autonomy in India’s deepening democracy 225

When speaking about office-holders’ performance, the villagers thought in terms of their elected representatives as ‘doing something’ for the locality or for them personally. They contemplated whether a given politician had ‘worked’ for them. A ‘good’ (manchi) officeholder did something; and an office-holder who did not was ‘bad’ (chedda). Hope for effectiveness was an ambivalent motivation. Informants did not expect elections to result in changes for the better. In 2007 one elderly Muslim man indicated the limitations of hope: ‘I vote with the hope that at least the [candidate] would do something good—only with hope’. He went on to observe that in elections farmers were like insects that get attracted to a streetlamp: ‘[they] get attracted for its redness, thinking that it’s edible, come near and die’. Several informants thought that elections brought positive changes to the village. They argued that parties promised to do better than their predecessors and that they might fulfil some of their campaign promises for fear of not being re-elected. In 2007 a prosperous young farmer from an upper caste said: ‘People are more conscious now, so the leaders have to do something for people these days. They simply can’t leave without doing anything as they were doing earlier. They can’t survive for long if they do that’ (see Berenschot, Chapter 8, on this new mood as the ‘new pragmatism’). While some said they believed the promises made by the Congress Party in its 2003–2004 campaign, most did not trust politicians. Many thought elected leaders were ‘selfish’ (svarda) and ‘corrupt’ (avineeti), two tightly connected qualities. They said that because most leaders were selfish, they ‘ate’ (tinnadu) funds instead of distributing them to the villagers. The literate wife of an uppercaste medium landholder observed that ‘only one among hundred is honest nowadays. Even if there is an honest leader, once he gets into office, he will change’. A few villagers produced more nuanced accounts of the failures of government assistance and the lack of politicians’ cooperation. Some said that Village Presidents and MLAs could be prevented from doing good work by the government’s funding limits. Some thought that Chief Ministers had to wrest funds for the state from the central government. But even those who commented on the formal limits to MLAs’ generosity said that, apart from being corrupt, MLAs in general were not very interested in exerting themselves for villagers’ sake. They simply ‘did not bother’ ( pattinchukovadam ledu). In 2007 the Village President, a prosperous, middle-aged, upper-caste landholder,

226  Pamela Price, with Dusi Srinivas

gave a strikingly full account of why an MLA may choose not to attend to a village or his constituents: There may be a funding problem. Sometimes he may [not] have funds. Some MLAs are active and can manage funds and resources and do some work. But all MLAs cannot be equally active. Some may be dull…[and] can’t get more funds. An MLA may be incompetent. Or he may think that [he] wanted to serve only for one term as MLA and think: ‘I’m not bothered about the next term, so let me make as much money as possible in these 5 years’.

The earlier TDP government and the current Congress administration had both established various programmes for village ­welfare and development, but in 2007 informants reported that their effects were uneven because government officials and state and village ­politicians took a share of the funds. One informant reckoned that 15% of the allocated money actually reached the village, while others gave more optimistic estimates of 30% to 50%. Another issue was access to these programmes, which could be confined to the office holder’s party supporters. A TDP-loyal scheduled-caste woman said that government money and programmes ‘[are] not reaching rural areas. Even if they are reaching, only 25% is reaching the villages. And even in this 25% a lot is spent on their own supporters or followers of their own party’. Villagers’ comments were often conflicted between the desire to vote for a selfless person and the compulsion to show allegiance to a party that could provide for them, should it win. This tension was sometimes expressed in contradictory statements about general preferences, in which the alleged prevalence of corruption and lack of initiative among politicians became key concerns. In 2007 Srinivas asked villagers what most affected their votes: candidate, party or the promises made during campaigns. The Village President, a prosperous upper-caste farmer, said simply that he looked first at the party and then at the candidate. He was attached to Congress and the party came first, and he surmised that most people in the village voted according to party preference. An upper-caste landholder and contractor with influence in Congress politics in the district also said that for him party attachment was foremost. However, he and several other informants’ comments revealed clearly the tension between party loyalties and high moral

Patronage and autonomy in India’s deepening democracy 227

standards in one’s choice of candidate. This tension ran through many statements: Srinivas: What kind of a person will you vote for? What qualities do you seek? Landholder: I will see whether he is corrupt or not. I will not vote for a corrupt person. Apart from that, I will see whether that person is available to people…. Srinivas: What is [more] important to you, the candidate, a party or party promises? Landholder: I vote according to party. A party will have certain policies, ideologies and so on, which are always constant. They will form the core principles of the party and they would not change. People come and go. Sometimes even in a good party, there may be a bad candidate who can come to power. Then we all go and talk to the party high command and say that he is bad, and then he may be changed.

The most prominent TDP activist in the village thought along similar, contradictory lines. He was a low-caste farmer who had lost several contests for the position of Village President, selling part of his landholdings to fund his campaigns.9 He maintained simultaneously that he always supported the TDP and that he would never vote for an MLA who had neglected to serve him and the village: Srinivas: If you elect someone as an MLA and he does not work for you, will you vote for him again? Farmer: Never. Not at all. But if the party gives the ticket to him again, then we will consider whether to support him or not, because we have to support the party. If the elected person didn’t work, then all of us would go and talk to party leaders [saying] that he did not work and he should not be given the ticket [to stand for election]. But if he is given the ticket, we have to support the candidate. It’s a difficult situation. But we will try our best not to give him a ticket. 9

His wife had been elected as Village President earlier through the state system of women’s reservations in village government; during that time he was often spoken of as the (actual) Village President.

228  Pamela Price, with Dusi Srinivas

Villagers consistently reported that they would not vote for an incumbent who had failed to help them, but struggled with the choice if that politician came from the party which they supported. In 2007 an old upper-caste farmer (and father of the Village President) expressed the tension between voting for a deserving person and adhering to a party, while at the same time claiming his autonomy in casting his vote: We will consider both the party and the candidate. But mainly whether a particular person can work for us, whether he is capable of doing something for us or not, and vote accordingly…. I would vote more for a person who can fight for the country’s progress and future. I don’t belong to any party nor do I get influenced by alcohol, money, caste or anything.

But then he added: I will vote for a good party, whoever the person is. Even if the person is good and his party does not have good policies, then there is no point in voting for him.

Moral elements and politicians-elect Villagers were quick to remark that politicians ‘ate’ state funds and that they were not concerned with their constituents because they were selfish. A Muslim smallholder voiced a common conception of a good leader: [A leader] should not be selfish. Even if he is selfish, he should use just 1% or 2% for his sake and do the rest for people. He should come forward and help people. He should be able to donate, even without taking for himself…. [He] should tell what is good and bad. If there is a crisis, he should be near [to support] you.

Notions of ‘goodness’ provide guidelines for correct behaviour in Balapalle. In rural South India generally, shows of honour (maryata) and offerings of respect ( gauravam) are essential to everyday interactions (Price 2013b; 2013d).10 Analyses of state-level politicians show that receiving respect and honour are centrally constitutive of their careers (Bate 2009; Price 2013a; 2013c). Balapalle informants’ 10

Villagers use a wide range of words, with differing nuances in meaning, to speak about respect and honour (Price 2013b).

Patronage and autonomy in India’s deepening democracy 229

moral logic of honour informed their assessment of politicians. Generosity expressed with gifts and loans, and personal exertion on behalf of others, made people deserving of honour and of being addressed with respect and treated with special shows of politeness (Price 2013b). Among politicians, ruling class honour was due to those who exhibited selflessness and largesse. In 2007 villagers expressed the desire to vote for candidates who were generous, who would exert themselves on their behalf and who were not (excessively) corrupt. During the earlier reign of village lords in Balapalle, upper-caste men and women were addressed with the honorifics Patel (if they were Reddys) and Dora (if they were Velamas). As villagers increasingly felt their independence, these titles fell out of use. People continued to strive to become Big Persons; but now this status was based more on achieved qualities than on inheritance and birth. What was striking is that villagers expressed this achievement in the old idioms of respect and honour. People showed respect and honour for a range of activities and personal qualities, including generosity, influence and wealth, education, the acquisition of political or administrative office or the protection of others. Some villagers remarked, however, that proper shows of respect were not forced, but should come ‘from within’. ‘When a person has good character’, said one Brahmin, ‘the respect that you have for him is not out of fear or compulsion, but is whole-hearted respect. But when we respect people with money or power, it is out of fear or compulsion’ (Price 2013b). An aged farmer and his college-educated son, both from a scheduled caste, agreed: There are [these] two types [of respect]. The first is much better, where you act out of free will. In the second you act out of fear of another’s position and power. More specifically, if we do not treat them with respect, they will keep this in mind and, when we go to them for getting our work done, they will not do it. Even if [the son] does not belong to or support a political party, [which a more powerful man belongs to], he will have to show respect [to the party], because he may not get the benefit or get his work done.

Displays of honour, including elaborate exhibitions of courtesy, recognised the superior moral character of those who were willing to protect or otherwise engage on the behalf of others. Several informants explained, however, that a politician (or anyone else) who

230  Pamela Price, with Dusi Srinivas

lost honour by being unhelpful, stingy or selfish could win it back by changing their ways. One could regain the status of a good person, a manchi manishi. In 2007 an elderly scheduled-caste smallholder commented sardonically on the villagers’ attitude to MLAs: After winning, a leader may do something if he is good. Suppose an MLA would have around 200 or 300 villages in his constituency. Can he give benefits to all these villages equally? Who can do that? At most, he’ll do something for 10 people in one village and 10 people in another village. So these people call him ‘good’, whereas the rest call him ‘bad’.

Ideas about ‘good’ people in Balapalle partake of the cultural logic deployed by Chandrababu Naidu and YSR in their campaigns. Both rivals claimed utter selflessness, periodically reminding audiences of their willingness to sacrifice their lives in service. The candidates’ promises and heroic postures appeared in part as their responses to voters’ criticisms of do-nothing politicians and  their desire for generous, effective and honourable leaders. In 2003–2004 it was reported that Naidu maintained an extensive intelligence network, which carefully tracked voters’ moods and desires. YSR’s attempt to reach out to his constituents also presumably gave him insight into the most persuasive ways to present himself. In this patronage democracy candidates attempted to stand forth as rulers of superior moral worth in terms familiar to the voters. I say ‘rulers’ because of the element of personal protection which the notion of leadership entailed for many people in Balapalle. Most informants personalised elected office holding in a way that emphasised the discretion officials could exert in responding to the demands of their constituents.

A hero falters Dusi Srinivas returned to Balapalle in February 2012, 2 years and 4  months  after YSR died in a helicopter crash, when he was only 4 months into his second term as Chief Minister. While Srinivas sought villagers’ views on both Naidu and YSR, YSR’s recent death made him the bigger hero. In the 2009 Assembly election the Congress had retained power, but with fewer seats and a smaller proportion of votes. In 2012 most villagers claimed that the Congress won once again because of the broad approval of YSR’s record in office. Once in power, informants

Patronage and autonomy in India’s deepening democracy 231

said, he confirmed his image of a manchi manishi by delivering on a number of promises he made during his campaign: free electricity, reimbursement of college fees for poor students, higher pensions for the poor, loan waivers for farmers and a low-cost health care programme (Aarogyasri). But, as we noted earlier, the local implementation of these policies had weaknesses, and YSR’s grand plans for expanding the irrigation infrastructure had come to nothing, despite the vast sums that his government poured into it. Villagers were still waiting for the completion of a lift irrigation project which was to provide water for their parched land. Nevertheless, most informants expressed fondness toward the Chief Minister who had died in service. The popular image of YSR as a good person was, however, muddled by a series of charges of gross corruption, which were levelled at him posthumously. Allegations which most preoccupied villagers involved YSR’s only son, Y.S. Jaganmohan Reddy, popularly known as Jagan. Shortly after YSR died, there were calls within the Congress for Jagan to become the next Chief Minister. But he was relatively new to politics, having held a political post only once, in 2009, when he became the MP representing the family bastion in the Kadapa district of Rayalaseema. The Congress leaders in Delhi were unwilling to allow someone so inexperienced and relatively unknown to lead one of the party’s most important states, and the suggestion was rejected. A YSR family loyalist reacted by starting a break-away party, the YSR Congress, which at first existed mostly on paper. Disagreements between Jagan and the Congress intensified and in November of 2010 he and his mother resigned from the party and from their elected offices. Jagan became the head of the new YSR Congress Party in March 2011 and in May of the same year he was re-elected to parliament, this time as a member of the newly formed YSR Congress. Rumours of Congressmen switching to the newly formed YSR Congress filled the state media. Jagan and his followers argued that the ruling Congress was failing to implement YSR’s programmes and that success could be achieved only if his son became the Chief Minister. Only a son could truly defend a father’s legacy. Before his father became head of the state, Jagan had a modest business career in energy. But after the Congress won in 2004, his enterprises expanded rapidly into cement, infrastructure and media, and he allegedly accumulated great wealth (NDTV 27 May 2012, ‘Who Is Jagan Mohan Reddy?’). In the course of events following his father’s death, Jagan acquired powerful political enemies who made his wealth a target for suspicion. One Congressman and one

232  Pamela Price, with Dusi Srinivas

TDP politician filed petitions against Jagan, alleging that YSR had pressured several companies to invest in his son’s businesses ‘in return for land allotments, mining leases, and other such favours from the government’ (Deccan Herald 11 August 2011, ‘CBI Told to Probe into Jagan’s Wealth’). The Central Bureau of Investigation (CBI) carried out preliminary investigations, and in August 2011 the CBI was ordered by the state High Court to make a thorough inquiry into Jagan’s business career (ibid.). In May 2012 Jagan was arrested on charges of embezzlement and in February of 2013 the Enforcement Directorate announced that they would confiscate 122 crore rupees’ worth of property from Jagan and his associates (The Financial Express 20 February 2013, ‘Jagan Mohan Reddy’s Rs 122 cr Worth Properties to be Taken Over by the ED’).11 In August of 2013, at the time of this writing, Jagan was still in jail. In February 2012 the residents of Balapalle were aware of investigations into the sources of Jagan’s wealth, to which they gave a wide range of responses. Some were troubled that YSR should be associated with corruption, while others refused to believe the charges. Several upheld the opinion expressed by a major Congress Party worker—a lower-caste medium landholder—who stated that all major leaders made fortunes in politics without being investigated by the CBI; thus, the charges against Jagan were politically motivated. It was the activist’s further conviction that villagers felt affection (abhimanam) for Jagan: ‘Even old people are cursing the government’s action. I witnessed one old woman cursing the government, saying that it is creating troubles for the son after his father’s death. These allegations are only at the political level. People do not believe them’. This view was reflected in the comments of a life-long Congress supporter, an elderly upper-caste man: I like YSR and I would defend his honour…. Nobody will believe if somebody says he is corrupt. People will not tolerate any such arguments about YSR. He did a lot for us. So how can we tolerate anybody speaking against him?

Srinivas: How did YSR and Jagan make so much money? Congress supporter: Because he treated people with love. He loved people. 11

A crore (cr) equals 10,000,000.

Patronage and autonomy in India’s deepening democracy 233

Unsurprisingly, major TDP activists in the village did not remember YSR with the same warmth. The TDP faction leader, a lower-caste farmer, thought that YSR’s death saved the Chief Minister’s reputation. He said that in 2003 the padayatra had provided him with the image of being ‘willing to work hard for the sake of the common man’, but ‘once he was in power, he committed many irregularities which are coming out now. If he was alive for another six months, people would have criticised him. But because of his death, he remains forever a manchi manishi in the hearts of people’. Another TDP party worker was also of the opinion that villagers ignored the charges levelled against YSR: All the charges of corruption cannot penetrate below the sub-district [mandal ] level. At the village level they will be irrelevant, where he will always be regarded as a good person. Whatever irregularities he may have committed, most people think that he has done them something good and that in comparison he was better than others.

These TDP activists exaggerated the extent to which YSR’s reputation survived unscathed. Several informants thought that the dead Chief Minister’s reputation had taken a dent, but were uncertain about how to interpret the media accounts. A sense of this comes from a conversation that Srinivas had with a lower-caste medium landholder, his college graduate daughter and his undergraduate son who was then studying engineering. Considering whether one could still say that YSR was a good man, the father commented: ‘Now it appears that he gave a biscuit to the dog and stole what he could in the house’. His daughter interrupted to disagree and compared YSR with Naidu, who she said had left the state with huge loans, having done nothing for the poor: ‘he only promoted IT. YSR was inclusive in helping the people in society. So it doesn’t appear a problem to me even if he is corrupt. I still feel he is a manchi manishi’. The son was less certain: We worshipped YSR as a God, no less, but now as the scams are coming out, we don’t know how to react. Now the media people are also showing what his motive was in introducing some so-called popular schemes. We believed he was God, but now I think that he made fools of us.

234  Pamela Price, with Dusi Srinivas

A new expression of autonomous thinking In 2007, while the residents of Balapalle were politicians’ clients, they showed a great deal of political autonomy in the ways that they reasoned about politicians’ morality and government ineffectiveness. Failures in the provision of welfare were widely conceived of as the result of elected politicians’ selfishness. Five years later, in 2012, the focus of blame had shifted to politicians from the coastal area of the state known as Andhra. The southern Andhra coastal districts were better off than those in Telangana in terms of agriculture, industry and education, and Telangana separatists argued that their region’ ‘development’ (abhivriddhi) had suffered because of Andhra domination. Telangana could prosper only as an independent state. Informants reported that support for separatism emerged over the previous 2 or 3 years and a new village faction, associated with the separatist party, Telangana Rashtra Samiti (TRS), had emerged.12 The other three factions were still linked to the Congress or the TDP, but their leaders said that they also supported the movement for a separate state. Some informants believed that the stature of YSR, who argued for a united Andhra Pradesh, had earlier inhibited the rise of separatism in Balapalle.13 His death and the inability of the Congress to produce a similarly popular Chief Minister created new possibilities for the leaders of the separatist movement. In November, just 2  months after YSR’s death, the founder of the TRS, K. Chandrasekhar Rao, went on a hunger strike and university students in Hyderabad demonstrated, and some even immolated themselves. The organisational capacity of the movement expanded and its speakers appeared at meetings across Telangana. The residents of Balapalle had attended such meetings and learned the arguments for a separate state. Villagers were particularly preoccupied with the lack of water for irrigation and work for their children. They believed that Andhra politicians were not concerned with water scarcity in Telangana and that better-educated Andhra candidates took jobs in the region away from Telangana youths. Some did not even spare YSR from blame. He had not come from the coast, but from Rayalaseema, another semiarid region, but several villagers argued that YSR gave higher priority to irrigation in his own constituency, leaving Telangana to suffer from 12 13

TRS was founded in 2001 to agitate for a separate Telangana. Support for a separate state existed mainly in the northern part of the region before the winter of 2009–2010.

Patronage and autonomy in India’s deepening democracy 235

increasing drought. An upper-caste large landholder in Balapalle, who previously worked for Congress, had become a supporter of the TRS because, as he explained, ‘if you visit YSR’s village, it has roads, water, all is good…. If you look here, there is no water, it’s becoming like a desert. So that’s why I shifted to the TRS’. Support for a separate Telangana took hold in Balapalle at a time when the Congress regime was experiencing increasing difficulty in implementing its welfare and development programmes. After the death of YSR, state politics had become drastically destabilised, a result of the rising regional tensions mentioned earlier and the split in the ruling party with the emergence of the YSR Congress. The administrative capacity of the state was undermined and there was widespread agreement among villagers that the Congress would not be re-elected because of decline in its performance. An important Congress activist in the village said that the government’s failures since YSR’s death were so drastic that he could not even get ration cards for the Congress party workers. A year and a half after Srinivas’s last visit in 2012 the Congress leadership in Delhi decided to create a separate state of Telangana, leaving it unclear at the time of writing whether any Rayalaseema districts would be included in the new state. Political observers concluded that electoral considerations related to the upcoming 2014 General Election played a major role in the decision. Shortly after the separation, P. Sainath, agrarian affairs reporter for The Hindu, spoke to agriculturalists, including landless labourers, in both Telangana and coastal Andhra, reporting that the rural poor in Telangana hoped that now they would get water and work, especially for their children (Sainath 13 August 2013, ‘Their Hands Will Look Different’, The Hindu). On their part, the rural poor along the coast feared that they stood to lose water and work to the western region (Sainath 14 August 2013, ‘Can We Send Them to America?’ The Hindu). Earlier, when our informants in Balapalle explained why they hoped for a separate state, they expressed the two most widespread concerns about the outcome of division, for better or for worse.

Conclusion Conversations with villagers over the course of almost 9 years illustrate people reasoning about moral behaviour as they discuss politics and leadership. Villagers valued generosity and selflessness, and most of our informants interpreted the actions of both local and

236  Pamela Price, with Dusi Srinivas

state politicians through that moral lens. It was moral failure on the part of elected officials which resulted in the lack of welfare and development experienced by informants. When ruling regimes were stable, the conditions of patronagedemocracy applied in the distribution of resources and services by the state. Villagers were clients in factions associated with political parties and they voted, hoping not only that their party would win, but also that the winning candidate would be a generous officeholder, would ‘bother’ and do good for them and the village. It was commonly said that since the 1980s telivi (awarenessknowledge) had increased in the village. In applying telivi in the evaluation of their leaders, informants avoided total dependence on party patrons. One result of growing telivi was that villagers exercised the right to vote with the conviction that casting the ballot confirmed their right to benefits from the state. In their understanding it was selfishness on the part of most politicians that denied the distribution of state benefits. Conceptions of corruption, then, did not involve distinctions between the public and the private (as in ‘public office’ versus ‘private gain’). Corruption spoke more about weaknesses in a ruler. Patrimonial thinking existed alongside the exercise of telivi. The conviction that morally good, selfless politicians met voters’ needs contributed to confusion in sorting out the meaning of accusations of corruption involving YSR and his son.14 However, such was the emphasis on elected officials ‘doing work’ to assist their constituents that some villagers could tolerate embezzlement coming from an otherwise ‘good’ politician. In the new political climate following YSR’s death, patterns of thinking about governance and welfare had not altered radically. Patronage remained the central idiom and informants spoke about future development in terms of discrete provisions, not policy trajectories. Because villagers already experienced political autonomy, the worsening capacity of the Congress to implement its programmes only resulted in greater freedom for villagers to seek new political solutions to old and unrelenting problems of their lives. As the villagers puzzled over what an autonomous Telangana could mean for them, they exerted a great deal of autonomous will.


See Piliavsky (Introduction) and Gilmartin (Chapter 5) on the tensions and contradictions inherent in a patronage democracy.

Police and legal patronage in northern India


Beatrice Jauregui



 he Police in contemporary India are notorious for being ‘politicised’ and for engaging in ‘corruption’ and ‘clientelism’ as part and parcel of their everyday practice.1 But while common academic wisdom takes these to be coercive, top–down practices, the associated transgressions and transactions are in fact often initiated by or involve the active participation of local leaders and common citizens. Interactions with the police may moreover also include forms of exchange and compromise which follow from historically rooted and culturally widespread relations of provision and protection. This chapter draws on two years of ethnographic fieldwork among the police in the north Indian state of Uttar Pradesh (UP), to theorise ‘legal patronage’ not (only) as the lawful practices of donor-servant relations, but as exchanges that provide people with access to the criminal justice system. I focus on the quotidian instances of negotiation between police and other people over the production of First Information Reports (FIRs) to understand how police officers participate in broader systems of exchange and brokerage, simultaneously occupying the roles of patrons and clients because of the particular forms of power that they are charged with deploying. Expectations and fluctuations of police officers’ patron–client practices reflect a social order in which what I call ‘provisional agency’ ( Jauregui 2014) contours concepts of moral and legal virtue. Provisional agency has a dual meaning that entails both a morally valued capacity to provide for others, and a form of temporarily effective, instrumental action that allows for progress into (and for) the future. This form of agency and its mutable moral valence


See Baxi (1982), Wade (1982), Brass (1997), Visvanathan and Sethi (1998), Singh (2000), Hansen (2001), Verma (2005), Jauregui (2010a) and Anjaria (2011).

238  Beatrice Jauregui

emerge in a context where social position is always already unstable (cf., Dumont 1980). Expressions of provisional agency may involve the transgression of moral boundaries that generate ambivalence and ambiguity; but more importantly, they may also reconfigure the moral boundaries themselves through social improvisation and recombination in everyday interactions. In India, as in most countries today, the police have two primary legal duties: (1) the maintenance of public order and (2) the investigation of crime.2 In practice, these duties translate, respectively, into the powers of (1) serving as the state’s primary means of coercion, and (2)  investigating crime, which entails the production of FIRs, the primary documents necessary to initiate and follow through any prosecution via the criminal justice system. Coercion suffuses routine police practices of arrest, interrogation, provision of security and crowd control.3 FIRs are registered when the police ‘take cognisance’ of a case, or determine that a ‘cognisable offence’ has taken place. After filing the report, the police can investigate an incident and arrest suspects without a judge’s warrant. Coercion and criminal investigation may be deployed in defensive and offensive ways—as protection for potential or actual victims and as a response to their complaints, or as a form of harassment and intimidation. I focus here on how the police decide whether and how to frame an FIR, in part because while FIRs are central to Indian policing practice—and to peoples’ conceptions of what the police are for—they remain strikingly understudied. The intimate microlevel  interactions which surround the production of FIRs have important implications for broader questions of state legitimacy and democratic politics in contemporary India. I take my cue from earlier anthropological work on everyday legal disputes and compromise in northern India (especially Cohn 1956 and 1965) and underscores the point that theorisation of patronage in South Asia should reach beyond the idiom of gifts, exchange and distribution 2

I say ‘legal duties’ because the police in India and elsewhere also necessarily engage in a multitude of extra-legal or quasi-legal practices that may be considered ‘duties’: anything from emergency medical treatment or cleaning up messes following accidents, to serving tea or providing private security to ‘VIPs’ (see Jauregui 2013a; 2014). 3 In contemporary South Asia, coercion may also take on the deadly form of encounterkillings, or alleged self-defence shootings by police that are often pre-meditated and performative state murders ( Jauregui 2010b; Belur 2010; Eckert 2005).

Police and legal patronage in northern India  239

of resources to understand the ways in which people approach the law, to what we may call the ‘morals of legality’. Patronage in India, and elsewhere, has long been discussed as something intrinsically extralegal or even criminal, as a cultural residue or a subversive informal response to formalised state procedures in contexts of legal pluralism (cf., Galanter 1972). However, relations and exchanges which embody long-standing patterns of patronage are just as frequently articulated through idioms of legality as they are through other cultural forms. Idioms of legality take precedence when various parties demand that the police arbitrate or mediate a dispute. Indeed, the murky and mutable distinctions among the legal, the illegal and the extralegal may hold the clue to what the power of the police really is. Even more broadly, manipulations of law within wider relations of patronage shed light onto how cultural forms configure governance across time and space. My ethnography of two negotiations over the production of FIRs in northern India helps to track some of the complex ways in which police officers act as ‘little kingpins’, whose capacities to negotiate and make legal decisions both reflect and shape modes of patronage that extend beyond the police institution particularly, or institutions of law and governance at large. The term ‘little kingpin’ is used by the UP police to describe themselves and fellow officers who occupy certain posts, particularly (Sub-)Inspectors, who serve as chief Station Officers (SOs). Other police chiefs in field postings—which are substantively different, if not completely divorced from, administrative ‘desk jobs’—are also conceived as kingpins. These include district (Senior) Superintendents of Police, (S)SPs, range Deputy Inspectors General (DIGs), zone Inspectors General (IGs), or the state Director General of Police (DGP). But SOs are different. They are the chiefs of local thanas (police stations), which are the nodal points of contact between the public and the criminal justice system. The vast majority of people who hope to register an FIR must meet the SOs, who are the primary sources of decisions about how the charging and the investigation of crimes proceeds. So much power vested in individuals in this position means that they receive frequent visits from elected village headmen or other political and community leaders, who themselves act as ‘patrons’ for the people under their charge. These leaders often negotiate with the local police on behalf of their followers, and are thus expected by the

240  Beatrice Jauregui

people they represent to forge and maintain good relations with local law enforcement. The fact that the police identify themselves as ‘kingpins’ is significant. The concept of kingpin is multivalent and it reflects the shifting position of police in relation to various people and channels of power and capital. Apart from its connotations of outstanding leadership or influence (to say nothing of its associations with organised crime), the idea of kingpin highlights the mechanically conductive role of the police in larger social and political systems. That said, its superlative and sovereign qualities also indicate that the roles of police are rather different from the well rehearsed notion of a bureaucrat serving as little more than a cog on a static and endlessly spinning wheel, which mindlessly intersects with other cogs to keep a system in motion. SOs and other well positioned police in India are, in fact, cognising subjects with the power to recognise and authorise action, and this is why they are regularly entreated for assistance as patrons of state law. This is the key distinction between the police officer’s role and that of a broker, or a go-between who facilitates exchanges between political patrons and clients (as described by Berenschot and Osella, Chapters 8 and 16). They play a unique role in offering people the possibility of obtaining legally recognised documentary ‘proof’ that a crime has occurred. Crucially, the kingpins themselves also beseech persons with legally and culturally mediated power over them: elected officials, bureaucrats, corporate heads, media moguls and other ‘influential’ people, including any number of local criminal bosses or religious leaders. They are just as often ‘clients’ (recipients) as they are ‘patrons’ (providers) of access to law. In the vignettes that follow—the first of which involves an officer’s refusal to register an FIR, and the second of which describes a negotiation between a police officer and a village leader over the terms of an FIR that has already been registered— I examine how everyday practices of police legal patronage reflect a social order in which morality and authority intersect via expressions and experiences of provisional agency, as both the capacity to provide and as temporary mobility in a larger system.

Patrons and clients in the police Reports by human rights and watchdog groups, news headlines and common Indian citizens who have tried to have FIRs registered by local police, all provide the same basic account: lodging an FIR is

Police and legal patronage in northern India  241

next to impossible without being socially/politically well connected, offering a hefty bribe or being able to mobilise pressure by a senior police officer or somebody else with ‘influence’ (Patil 2008; TNN 2008a).4 It is especially difficult to have an FIR registered if the accused are well connected or otherwise influential. In such cases, even if an FIR is actually lodged, it is often incomplete or diluted, and the process of filing it often entails hardship and humiliation for the complainant (TNN 2006; 2007). In UP and many other states, low numbers of registered FIRs are widely understood not as an indicator of actually low crime rates, but as an index of the amount of pressure that the ruling government is putting on the police to register or not register cases, according to its own politicised and ever-changing rules. The government may issue instructions to keep crime figures low in the run-up to an election or, conversely, demand that the police ‘do their job’ and stop ‘shying away’ from registering cases (TNN 2008b; 2008c). The police themselves generally do not deny these sobering facts ( Jauregui 2010a). But reducing the explanation for their frequent refusal to take cognisance of cases to vague charges of ‘corruption’ or the ‘criminalisation’ of governance obscures the broader cultural and historical forces that shape their specific reasons for doing so, some of which are based on professional experience and discretion, and some of which result from personal, bureaucratic or political pressures.5 The reasons given by the police themselves for refusing to lodge FIRs include insufficient evidence, disbelief in the complaint following further inquiry, insufficient staff and resources for investigation, fear of reprisal if the complaint is made against a powerful individual or a group, or the compulsion to follow orders ‘from above’ to keep the rates of reported crime to a minimum (Dhuru and Pandey 2007; Kumar et al. 2008). The last reason becomes particularly pressing immediately before municipal, state and national elections, when the incumbent party or coalition tries to demonstrate the positive effects it has had on crime rates and ‘law and order’, and puts immense pressure on the police to show 4

According to Section 154 of the Criminal Procedure Code of India, it is a statutory duty of the police to register an FIR for each and every case in which there exists prima facie evidence of the occurrence of a crime listed as a cognisable offense in the Indian Penal Code (CrPC 2005 [1973]). 5 For more on the multiplicity of motivations for police behaviour, and the implications of this multiplicity for ethical police practice (and ethical ethnography of police practice), see Jauregui (2013b).

242  Beatrice Jauregui

‘only good work’. We can observe some of the forces and relations behind police refusal to register FIRs by examining some incidents that occurred at a station I will call Chakkar Rasta Thana (CRT) in a rural area of Lucknow district, which was my main base of ethnographic operations. It is an unusually busy day at the CRT. The SO Y.K. Yadav6 has been fielding public complaints all morning and afternoon, while simultaneously overseeing a pileup of end-of-year administrative work. Around 4 pm four men arrive at the thana: an elderly father and three of his sons (all in their mid-forties or early fifties). The Shuklas, as the family introduces itself, look different from most of the other villagers who tend to come to the thana with their complaints. They drive a blue Maruti sedan and wear relatively urbane clothes compared to the usual dhotis (traditional waist wrap) and chappals (sandals) which villagers wear, if they have any footwear at all. This is only the second time in several months of my research that I witness people who appear to be ‘from town’ rather than ‘village folk’ coming to CRT with a complaint. The SO’s demeanour has been caustic and exasperated for much of the day; but when the Shuklas ask to speak to him, he comports himself differently, appearing much more ‘all business’, and addressing them with the honorific ‘ji’. Plaintiffs must offer their complaints to the police in the form of a written application, much to the chagrin of the many illiterate people who are forced to employ scribes for the purpose. First thing on arrival, the Shuklas present their written petition to the SO. Having been observing at the CRT for some time now, I can tell that their written petition is more sophisticated than most, which tend to be hand-written sheets of paper signed or marked with a fingerprint. The Shuklas’ application includes many official-looking documents and it appears that they have already consulted a lawyer. They tell the 6

Names of persons and places (with the exception of Lucknow) have been changed to protect the confidentiality of interlocutors. I usually distinguish police officers by rank; but in instances in which an individual may be identified by assigned department or some other means, I use a more general term like ‘senior officer’. When changing names, I preserve caste identification (Brahmin) and religious affiliation (Muslim), which are integral to the constitution of personal, professional and political interactions in India. Interlocutors often spoke to me in a mixture of Hindi and English. While most of the very senior officers (in the Indian Police Service, or IPS) spoke in English, those in subordinate ranks usually spoke Hindi, which I have translated into English.

Police and legal patronage in northern India  243

SO that they wish to lodge an FIR against a local man called Trivedi under Section 420 of the Indian Penal Code (IPC) for the crime of fraud, a cognisable offence. I instantly recognise the name Trivedi, a well-known ‘big-man’ in the area, a large landowner and a powerful figure in the local Hindi language press, who has made several ‘casual visits’ to CRT during my stay ( Jauregui 2013a). As I indicated earlier, casual visits are common at most thanas, especially in smaller rural stations. Pradhans (elected village headpersons), local corporate heads, civil service bureaucrats, police informants and others often ‘drop in’ to say greet, shake hands, drink chai or chew paan (beetle nut/tobacco) with the officers, and to exchange ‘gifts’. More often than not, these gifts are not material, but consist of information, generally shared informally as gupshup (gossip), which keeps police up to date about the goings on in their jurisdictions. Casual visits happen all the time, but become especially frequent when there is ‘a new sheriff in town’, after a new SO has been transferred into the post. (The same happens at the district level when there is a new Superintendant, and at the state level when there is a new Director General.) In their turn, the police spend a fair amount of time cultivating relations with local movers and shakers, making it a point to meet and periodically visit people they feel they need to know or with whom they should maintain good relations. On the very afternoon that the Shuklas come to the station, I have just returned from accompanying a group of officers from CRT (excluding the SO) to a party hosted at an old colonial bungalow by the very man in question, Mr Trivedi. We had been invited to the gathering because one of the CRT constables, Prithvi, had formed a relationship with Trivedi over the period of his posting in the thana.7 Several hundred people had attended the party, which involved a lot of networking, public presentations of accolades by and to the host, and an enormous feast consisting of fine, locally grown and prepared 7

Importantly, Constable Prithvi is also the CRT SO’s main ‘go-to-guy’ for making a variety of arrangements with local interlocutors, including ‘licit’ transactions like arranging security for public events; quasi-licit transactions like organising police rides in private taxis (when the single government-issued jeep is out-of-station) for favours to the vehicle owners (sometimes including releases from license and registration requirements) and ‘illicit’ transactions like organising meetings for exchanges of bribes ( Jauregui 2014).

244  Beatrice Jauregui

vegetables and spices, rich lassi (yoghurt drink) from the best healthy cows and sumptuous sweets, all served on traditional pattal (mahua leaf plates), dona (mahua leaf bowls) and kulhan (red clay bowls). Besides the small CRT police contingent, there were many local media personalities, police men from neighbouring stations and local dignitaries and politicians as important as the Mayor of Lucknow district, with an entourage of peons and security guards. Unfortunately for the Shuklas, the object of their charges has significant clout. They raise this point with the SO almost immediately, appealing to morals to persuade him to ‘do the right thing’ and help them because Trivedi is a ‘bad character’. They note that Trivedi has a substantial amount of influence over many bureaucrats because he is friendly with politicians in the ruling coalition, who can have officers at almost any level transferred into and out of posts at will.8 In regard to the case at hand, for which they wish to register their complaint, the father says that he has been involved in a protracted land dispute with three other men, whom I will call collectively ‘G’.  Some time ago, father Shukla and G acquired some land together, two thirds of which allegedly belonged to Shukla and one third to G. However, according to Shukla, G sold off their plot to Trivedi using forged documents, and what they sold to him consisted of more than their share, about one half of the total holdings, instead of the third which was rightfully theirs. Now Trivedi has a claim on a tract of land which Shukla argues belongs to him and should have never been sold to Trivedi in the first place. Both parties have documents to prove their ownership of the disputed territory, and have been fighting it out for the last two years in court with no resolution to date. Meanwhile, Shukla decided to cut down some trees on what he claims is his land in order to sell the wood; Trivedi was now threatening to lodge an FIR against him at a neighbouring thana for encroaching on his land. Shukla tries to persuade SO Yadav that he needs his help, because Trivedi ‘already has the police at the neighbouring thana under his thumb’. So Yadav listens to the Shuklas patiently and asks some further questions, telling them he will inquire into the matter, and that they should return tomorrow. He says he will not register an FIR just now 8

‘On the books’ police officials are supposed to be protected from dangers like arbitrary transfers and compulsion to follow illegal or unethical orders, but in fact they are subject to these and other pressures on a routine basis (Wade 1985; Verma 2005; Jauregui 2010a; 2013a).

Police and legal patronage in northern India  245

because he needs to ‘make some phone calls’ and ‘acquire further information’. The Shuklas are obviously displeased and they linger for almost an hour pleading their case and trying to convince the SO to lodge the FIR immediately. They say they fear that Trivedi will ‘use his influence and money’ to take their land and possibly other assets, and that as they do not have anything like his connections and power, they have no option but to ‘use legal channels’.9 With an air of desperation, they entreat the SO to speak with his boss, the Senior Superintendent of Police (SSP) of the Lucknow district, whom they had already visited earlier that day. This was technically ‘out of order’ and, as per standard operating procedure, the SSP told them that they ought to have visited their local thana and registered an FIR first.10 The Shuklas say to SO Yadav: ‘we will call [the SSP] now’, but the increasingly impatient SO says: ‘No! Do not do that. Just wait and let me inquire. That is the procedure. Leave these documents with me and come back tomorrow. This is all I can do right now’. Looking defeated, the Shuklas leave a stack of photocopies of land titles and such with the SO and sullenly drive away. The case of a pre-existing land dispute with questionable documentation leading to threats of FIRs and counter-FIRs is a common occurrence, and an apt illustration of a postcolonial outgrowth of what Bernard Cohn (1956) described as a clash between imported or invented British legal procedures and indigenous systems of justice, particularly the settlement of disputes by village elders or other ‘wise leaders’ who help to broker compromises in disputes. Cohn noted that from the 18th century until the early 20th, cultural forms that had been common in India’s ‘little kingdoms’—caste-based hierarchies of authority and adjudication of disputes by local panchayats 9

Quoting an interview with a member of a Muslim group organising legal advice for their community in Bombay, Julia Eckert writes that ‘while disdain for legal procedures is high among those classes that have rather easy access to the courts and can afford a lawyer, those social groups that need to defend themselves against the state, and particularly against the police, put the highest hope in legal means. “We have no alternative” they feel, since most other means (such as political patronage or money) are not at their disposal’ (2005, 191). 10 I learn later that the SO’s boss, the SSP of the Lucknow district, was invited to Trivedi’s party earlier that day. But for whatever reason, he decided not to attend because he reportedly thought it ‘might look bad’. This may have been related to the dispute with Shuklas, or to other factors, I never managed to find out; but knowing the reputation cultivated by this particular SSP for being ‘tough on crime and corruption’, it stands to reason that he would wish to appear ‘above the fray’.

246  Beatrice Jauregui

(village councils)—were combined with constantly shifting colonial means of administration derived from both British and Muslim law (see also Fisch 1983). Cohn argued convincingly that these different articulations of adjudication led people to view procedural law and the apparatuses of state order (courts and police) not as loci of ‘authority par excellence’ (cf., Dumont 1980 [1966], 167) there to protect their rights, but as new types of instruments for harnessing and expressing prestige and power in line with long-established cultural norms of reciprocity and compromise. Cohn writes that as a result, following the establishment of courts in colonial India, disputes took years to resolve as cases involved ‘too many appeals… [and] forged documents and perjury… [which] became endemic’ (1956, 90). He argues further that, [colonial] courts did not settle disputes but were used either as a form of gambling on the part of legal speculators who were landlords or merchants and who turned to the courts to wrest property from the ‘rightful’ owners, or as a threat in a dispute… [Cohn argues further that as of the mid-late 1950s, when he was conducting his field work in UP] there is no apparent abatement in this cycle of false cases and what an historian, Percival Spear, has termed the Indian peasants’ ‘slot machine’ attitude toward the courts (ibid.).

Indeed, in their multiple visits to police stations across Lucknow district, the Shuklas are frantically gambling, wagering that they may win over an officer who will register their FIR before Trivedi lodge his. But for several reasons their bets backfire. In addition to being rather unconnected and impotent, the Shuklas also appear to the SO as at best untrusting and unaware of proper procedures (which would entail visiting their local thana before appealing to the district SSP) and at worst devious. After they leave, the SO indicates that he is suspicious of their ‘out of order’ attempts to pre-empt Trivedi’s FIR with their own. This is important, because it suggests that while the Shuklas may have approached the SO as a potential legal patron—a wise man in uniform who could decide of his own good will to intervene on their behalf and bestow on them the beneficence of an FIR—Yadav may have interpreted their appeal to formal law as a an informal machination and an impropriety. Even though he too engages in informal (and sometimes illicit) activities, Yadav generally comports

Police and legal patronage in northern India  247

and speaks of himself as a virtuous officer who does his duty in line with legal procedures and not under political or economic pressures. The next day Yadav tells me that he had spoken to the SO of the neighbouring thana where Trivedi purportedly plans to lodge an FIR. He says that his fellow-SO told him that the Shuklas are ‘hiding a lot of things’, including the fact that they had approached this other SO first, even though they should have gone to Yadav (another act ‘out of order’). The fellow-SO also reported that the Shuklas had agreed they would cut X number of trees on the disputed land, but proceeded to cut X+ number of trees and then covered up the stumps with mud (in an attempt to ‘deceive’ the officers). Yadav’s suspicions about the Shuklas are growing. Later that day another man arrives in CRT to report that he was told that the Shuklas are claiming his signature as witness on the land deal with Trivedi was forged, and that he now wishes to verify its authenticity.11 He brought no legal proof of his signature or identity, but simply signs a scrap of paper, with which SO Yadav seems satisfied. This, in combination with news about the cutting of extra trees, confirms to Yadav that the Shuklas are lying, and he decides not to register the FIR. Information about the breaking of an informal pact to cut a limited number of trees, and the quasi-formal rebuttal of the accused ‘false witness’ to the land deal fraud, hardly constitute sufficient evidence for the Shuklas’ dishonesty. Yet it is enough for the SO to reject their FIR. While he frames it as an issue of professional judgement and experience, it is also clear from the way that he talks about Trivedi that he does not wish to get involved. He is not especially forthcoming when I ask why he will not inquire further. One of the SO’s more sympathetic associates speculates, out of Yadav’s earshot, that Trivedi could make things difficult and the SO ‘only wants to show good work’ (not start an investigation he cannot resolve). This is certainly likely. But having seen the lavishness of Trivedi’s party and his distinguished guests the day before, I wonder if other forces may be shaping the SO’s decision. I  ask the SO directly whether Trivedi’s status and influence may have affected his view and he answers somewhat vaguely that one must ‘be careful with powerful people’. He then repeats a ‘wise saying’ in Hindi, which I have heard 11

For further reading on the way reality is constructed through the movement of state documents and signatures see Hull (2003) and Mathur (2012).

248  Beatrice Jauregui

before: ‘paani mein reha kar magar se bair nahin karthe’ (when you live in the water, you don’t make an enemy of the crocodile). Whatever other reasons he may have had, SO Yadav’s prosaic and poetic comments suggest that he does not wish to offend a local power holder who could give him grief. Trivedi could pressure the SSP to have Yadav transferred out of his ‘kingpin’ post, which Yadav ultimately manages to enjoy for more than 12 months.12 Although the formal minimum limit for the SO’s tenure in post is two years, one year is actually quite a long time to remain in the coveted position of SO in UP. (Sub-)Inspectors are constantly vying, and often bidding, for these posts because they entail the authority to decide the fate of all cases within a jurisdiction, and offer ample opportunities to make political connections, to say nothing of the extra income (Jauregui 2014). The position of SO allows an officer to serve as a patron who offers provision and protection to his constituents, who in exchange may offer useful information, social connections, material evidence and, sometimes, payments in cash. Importantly though, as the Trivedi case also shows, at the same time as the SO provides legal ‘goods’, he is also a perennial client, either directly or indirectly. Police are not nearly as powerful as the other state agents, especially elected politicians and their associates, when it comes to just about any political-legal decision, apart from registering FIRs. In fact, the SOs do not even fully control the power they have as ‘legal gatekeepers’. As the Shuklas’ ‘out of order’ navigation of the political–legal terrain demonstrates, an SO’s decision to lodge or not lodge an FIR may be overridden by the (S)SP of his district (or by any other senior officer with the positional power to do so). Besides, people are learning that they may altogether bypass the difficult process of getting police to register FIRs by invoking Section 156 of the Criminal Procedural Code (CrPC), and making a case before a judicial magistrate, who ‘under Section 190 may order an investigation [by police into the case]’ (CrPC 2005 [1973], 101). The police in UP complain that more and more people are using this judiciary bypass to initiate criminal investigations.13 12

The district SSP eventually ousted Yadav from the post just before the state assembly elections, under pressure from the Central Election Commission, following charges by a candidate from an opposition party that he is working for the incumbent. 13 The nearest to statistical evidence of this increase in cases of people bypassing the police comes from an initial report made in 2008. This report was compiled

Police and legal patronage in northern India  249

In addition to legalistic modes that detract from the power of the police, there are significant pressures to register, or not register, FIRs by various local powerholders, especially though not exclusively Members of Legislative Assembly (MLAs) and other elected politicians. While SO Yadav does not say directly that he fears the well-connected Trivedi, he is forthcoming about police subjection to elected politicians—and to electoral politics as patronage writ large. He changes the subject from the case of Trivedi and Shuklas to relate a story meant to illustrate how he avoided the wrath of one of his ‘bosses’, the previous SSP of Lucknow notorious for extracting large sums of money from SOs in return for obtaining and maintaining their postings: I didn’t pay a bribe to [the former SSP] for this post of SO. Instead, I went to a ruling party politician, a fellow Yadav like me, and asked him to help me get the post… because for all of his power, [the former SSP] is scared of the politicians, since they could have him transferred out of his post at any time, and he wants to hold onto his money-minting position for as long as he can… SSP Lucknow is very powerful and can make a lot of money. [This former SSP] is known as a crorepati (multi-millionaire). He demands money from everyone to make things happen… After I became SO, he started to demand thana ‘monthly’ or else he would have me transferred; but I would not give it.14 He was transferred out of the SSP Lucknow posting a month and a half later, and that’s the only reason I’ve kept in response to a request made under the Right to Information Act (RTI) by a local activist named Izhar Ahmad Ansari, who runs an NGO called the Action Group for Right to Information. The report says: ‘In 2006, 258 aggrieved people knocked on the doors of judiciary to get the police to register an FIR… and in 2007, this number doubled up to 578 [sic]’ (TNN 2008c). These figures reportedly came directly from the SSP Lucknow office, which also revealed that the number of complaints received by them in which no FIR was registered was 30,356 in 2006, and 26,303 in 2007. Police were not penalised for this, except with a bit of bad press; the Principal Information Officer of the SSP office was fined 25,000 rupees for ‘not releasing the information within the stipulated time period’ (ibid.) 14 It is widely known that in order to remain in their posts, SOs often must give their bosses weekly payments called hafta (after the Hindi for ‘week’), or monthly (in English) payments ( Jauregui 2014). Officers do not always demand up-the-ladder payment. The demand is largely dependent on individual officers and those to whom they must in turn answer. SO Yadav said he has had several bosses who did not demand regular payments.

250  Beatrice Jauregui

this job. If he stayed longer, I would have been kicked out as soon as he could manage.

While Yadav prides himself on refusing to become his former police boss’s ‘client’, he makes nothing of the fact that he is now obliged to serve the politician as patron in whatever way that he may demand, by ‘fixing’ investigations, providing official or unofficial protection, using other police resources to acquire information or ‘get something done’. In return, the politician protects Yadav against unwanted transfers and other pressures, and may perhaps offer further favours, not least a transfer to an even more desirable posting later on. This jockeying for power among elected and appointed  leaders happens continually and en masse ( Jauregui 2010a). New relations are born, old relations die and change their shape and character, sometimes taking on an entirely new moral valence depending on context. The cast (and often the caste) of characters is constantly shifting as local, national and global histories advance, and the same individuals in the same posts may occupy either or both roles of patron and client; but the logic of exchanges of provision and protection remains. Out of this logic emerges a moral force which I call ‘provisional agency’ ( Jauregui 2014). The shifty moral valence of Yadav’s refusal to register the Shuklas’ FIR is at once a function of pressure from the state government to under-report crime, of intercaste hostility, and indeed of his ‘gut feeling’ that the Shuklas were lying. At the same time, his decision-making process and actions are embedded in the broader relational logic of patronage. He may eschew his oppressive boss, but to do so he must seek the protection of an MLA, a much bigger patron in the hierarchy of Indian state and politics. ‘Good patron’ trumps ‘bad patron’, but the ranked order of clientage remains. A multiplicity of motivations create enormous tension in the experiences and expressions of police powers and limits ( Jauregui 2013b). This multiplicity involves an entanglement of concepts of moral duty, professional discipline and state law, to which many officers, including SO Yadav, appeal as crucial aspects of their sense of practice and identity. The social fact of this tension is deeply integrated with whether and how the police decide to dole out FIRs, the ‘legal beneficence’ over which they have primary control. In the next case we will see how this embedding of police in

Police and legal patronage in northern India  251

overlapping systems of patronage manifests not only in decisions about registering an FIR, but also in the process of negotiating its content.

How little kingpins keep their little kingdoms in order The Shukla case demonstrates how the registration of FIRs as an ‘end’ in itself may be conceived as a form of legal beneficence, even as it is also a means to navigate various formal-legal systems, including but not limited to the criminal courts. Veena Das (2004, 228) shows how in some instances people seek to register FIRs less as a gateway to the judicial system (which is almost as notorious as the police for corruption and egregious backlogs, especially at the district level), and more ‘to obtain official proof’ that a violation had occurred, so that the makers of claims may gain recognition as victims and receive reparations from bodies partly or often fully beyond the criminal justice system. Das has observed that the police insistence on ‘dictating the framing sentences of the First Information Report… is a normal practice in police stations’ (ibid.), something I found was true over the course of my fieldwork ( Jauregui 2010a). I also found that it is equally ‘normal practice’ for officers to negotiate the terms of an FIR with a host of other authority figures and other types of actors with whom they must reckon.15 Long and overlapping chains of patronage, which complicate the production of FIRs, configure the routinisation of another practice— the ubiquitous use of representatives, or mediators, to negotiate on behalf of complainants with the police. The Shuklas did not employ a mediator or a dalaal (broker) and instead tried to speak for themselve By doing this, they probably substantially undermined 15

Das comments on the politics of FIRs in the context of mass violence against the Sikhs following the assassination of Indira Gandhi in 1984 by her Sikh body guards. She analyses FIRs as texts that have already been produced, and that reproduce ‘the state as neither a purely rational-bureaucratic organisation nor simply a fetish, but as a form of regulation that oscillates between a rational mode and a magical mode of being’, a claim about ‘the signature of the state’ that Das extends beyond the Indian context by drawing on examples from other places, like the US (2004, 225 and passim). By contrast, I am discussing police production of FIRs not as a mirror or signature of an oscillating rational-magical state, but as an active process of negotiation, which reproduces patterns of patronage.

252  Beatrice Jauregui

the chances of their success. I never saw them again, nor did I hear any more about their case. What I do know after watching many similar interactions is that if Shuklas had wanted to improve their chances, they would have enlisted the aid of someone who could ‘pull weight’, whether because he or she holds political office, has an established relationship with the police or just has more education or experience with state law. In rural India, people regularly appeal for help to the figure of the pradhan, the elected village headperson. On my very first day at the CRT, I watched a pradhan and SO Yadav negotiate a compromise over the terms of an FIR, something that, as I learned over time, was a routine occurrence. This particular case began several days before I arrived at the thana. It started when one man  attempted to irrigate his sugarcane field using a pump which extracted water from a nearby pond considered as common village property (graam samaaj sampatti). While watering his fields, he was suddenly attacked by a group of men (at least one of whom was allegedly carrying a sword) who hurt him so severely that he was hospitalised for several days. As the incident occurred in their jurisdiction, the CRT police were summoned and they registered an FIR, charging as culprits the persons named by witnesses with offences under the IPC section 324: ‘voluntarily causing hurt by dangerous weapons or means’ (CrPC 2005 [1973], 322). While investigating the motives for the attack, the police were told that the culprits were in the ‘beef business’ and thus violated cow slaughter laws; villagers claimed these men threw cow carcasses into the village pond to hide evidence of their trade, and that this was probably why they attacked the farmer. As I begin my observations at CRT on the day in question, I see a man in the station lockup who is, I learn later, an employee of one of the accused in the case. This man has been illegally (without charge) detained for questioning overnight. Later that day, another man, identified as a pradhan, comes to the station with the victim’s cousin to discuss the case with the SO. SO Yadav sets up a rickety wooden table and chairs under a large tree in front of the station, and sits behind it to begin the discussion. The pradhan begins by saying he has been petitioned by families of both the victim and the detained to change some of the content of the FIR lodged the previous day, which they believe is inaccurate and incomplete. First, he asks the SO to re-frame the charge: the

Police and legal patronage in northern India  253

crime is currently filed under IPC section 324 (assault with a deadly weapon) and the pradhan demands that it be upgraded to section 307, attempted murder. The SO silently shakes his head: ‘no’. The pradhan says that the man was beaten so badly that he was on the verge of death: ‘they tried to kill him, you see…’; but the SO counters that a harsh beating does not indicate intent to kill. Instead, he says, the assailants simply ‘went overboard… [but] there is no clear evidence of a motive to kill the victim, like a previous threat or the use of a firearm’. The pradhan retorts that some witnesses said the assault weapon was in fact a katta (country made handgun), not a sword, as described in the FIR, ‘and that must also be changed’. But the pradhan has not brought any witnesses and the SO dismisses his claims with a wave of the hand. Shifting in his chair, the pradhan then says that he is friendly with  a  prominent state cabinet minister, and that he will ‘get this minister involved’ if the SO does not ramp up the charges. The seemingly unperturbed Yadav declares rather dramatically that he is going to carry out his duty ‘according to the Law and the Constitution… [and that] even the Prime Minister of India himself could not pressure’ him into framing false charges. The pradhan tries a new approach. He says that he has conducted an inquiry, and that several witnesses claimed that the list of culprits named in the FIR was incomplete. He also tells the SO that the unnamed attackers are ‘strong men’ associated with the primary political party opposed to his own, and that they attacked their victim instead of trying to ‘reason’ with him because he was associated with the pradhan’s party. He insists their names should be added to the FIR. The pradhan adds that the accused are not only in the beef business but also in the pork business, something he presses should also be added to the report.16 SO Yadav smiles and then mirrors the pradhan’s move. He says that one of his mukhbirs (informants) mentioned one person who has not (yet) been named in the FIR, but whose name he plans to add to the  list. The  pradhan hesitates because it turns out that this other man has party ties (bandhi) with


The mention of pork business adds another nuance to the charge, making the assailants guilty of offense against Muslims who avoid pork, not just a legal offense of cow slaughter.

254  Beatrice Jauregui

the pradhan (both are members of the Samajwadi, Socialist Party).17 The pradhan says nothing definite about adding this man’s name, but as the conversation proceeds, he begins to invoke this new name along with the other four already mentioned in the FIR, essentially deferring to the SO at the risk of betraying someone with whom he has political ties. One of the four culprits named in the FIR is called Rahul. He too has ties to the Socialist Party and he has been on the ‘most wanted’ list at CRT for a long time. Sensing an opportunity to catch a ‘history sheeter’ (also ‘charge sheeter’), or a recidivist, the SO turns the conversation to Rahul to make his own appeal to the pradhan and the victim’s cousin (who has mostly sat quietly until now), both of whom agree that Rahul is one of the culprits. The SO then says: We have done raids in your village many times before, trying to find Rahul, and you have always helped him and worked against us. Now suddenly you seem to have changed your mind about him. Well then, you are going to have to cooperate with us. Without the help of you and the other villagers, we cannot catch any of these culprits. Please work with us now…

Yadav goes on to tell them that capturing Rahul is urgent and that if they provide information to help in his arrest, he will add the names provided by the pradhan and upgrade the charge to ‘attempted murder’. The pradhan and the victim’s kinsman exchange glances and hesitate; but then, following the pradhan’s lead, they admit their ‘fault’ (galti) of previous non-cooperation and promise to work with the SO from now on. From a rational legal perspective, this apparent collusion in framing, and possibly exaggerating or even falsifying, criminal charges appears as a perversion of formal procedures. But if we 17

Though beyond the scope of this paper, the politics of caste and religion colour this interaction and the context in which it occurs. The victim and the accused are all Muslims; the detained third-party informant is from a Scheduled Tribe, and the elected pradhan who formally represents them all is from one of the Other Backward Classes (OBCs), and a card carrying member of the Socialist Party, the ruling party in UP at the time. The SO also belongs to an OBC group. Though he generally denies doing his job in ways that favour the ruling party, the SO is assumed by most people to be a ‘party man’ because he is from the caste that dominates the party. (Remember that Yadav obtained his post by beseeching a ruling party leader.)

Police and legal patronage in northern India  255

view it as an interaction between a little kingpin as patron and those he is meant to provide for and protect, we may see that, as in the previous case, the officer is the grantor of legal beneficence. In this case, the SO grants the beneficence to the pradhan who, like the police officer, must balance his shifting position as patron and client. The pradhan is patron to his constituents, whom he provides with access to government development monies and protection from the police (among other malevolent forces).18 He is also a client, not only of party leaders, with whom he exchanges the votes and loyalty of his constituents in exchange for support necessary to remain in post, but also of the SO, with whom he must exchange information and assistance in return for legal beneficence, which makes him appear effective and raises his standing in his people’s eyes. The pradhan tries to shift from the petitioner-client’s role to the patronal by mentioning his ties to the minister, the SO’s big boss who could transfer him to a ‘punishment post’. But the pradhan fails. Pradhans are often treated as patrons by both police officers and constituents because they act as higher-level politicians’ ‘polling agents’ or brokerclients who garner votes, a position which gives them privileged access to local top-level patrons (for more on this see Berenschot, Chapter 8). In this instance, the legal beneficence which the pradhan and his clients seek from police is not an FIR, but a harsher sentence for the attackers: a conviction under IPC 307, which entails a sentence of ‘imprisonment for life [if such act causes hurt to any person] or imprisonment for 10 years and a fine’ (a conviction under the original IPC 324 would lead only to ‘imprisonment for three years, or fine, or both’).19 The pradhan’s patronage as ‘provisional agency’—in both helping his constituents gain access to criminal justice, and showing himself as effective in this particular moment—depends on receiving the ‘gift’ of the more grievous charge. But the SO’s ability and desire to make the change is both legally limited (he is ultimately responsible for the provision of robust ‘evidence’) and professionally motivated (he wishes to show his ‘good work’).


See Michelutti’s Chapter 12 in this volume for more on the growing significance of the protective functions of political patrons in urban UP. 19 IPC 324 may also be tried by ‘any magistrate’ while IPC 307 must be tried in sessions (district) court.

256  Beatrice Jauregui

Compromises between police and other co-beneficiaries are routine and have been so for quite some time (Cohn 1965). ‘It is not always a legal process’, the SO explains later, ‘but much of our work is done this way’. He says that both the police and the people they serve often prefer intimate negotiations and compromise to following strict rules of evidence admissible in court. His assertion is supported by the host of ‘compromises’ I watch him negotiate over the course of my fieldwork ( Jauregui 2010a). People prefer informal negotiation and compromise partly because it is usually more expedient and partly because mutual understanding has more potential to render an outcome that benefits all and endures. The same cannot be said of the judicial process, which takes a notoriously inordinate amount of time, is often prohibitively expensive and has many more opportunities for ‘corruption’ over time than a direct (and hopefully one-off) discussion with a police officer. This is not to give the impression that negotiations with the police and consequent compromises are always productive or satisfying for all. In plenty of cases I observed, Yadav and others used much more force to obtain information or they simply dictated the framing of charges in an FIR without negotiating with complainants or their representatives. I also watched Yadav and others arbitrate disputes in ways that clearly favoured one side, on the basis of caste or communal biases, exchanges of material goods, simple apathy or something as elusive as a ‘hunch’ that someone was, or was not, telling the truth. Such practices are problematic and rife with inequality, discrimination and potential injustice. But they also contain and often follow conceptions of justice, which we cannot ignore. Western-centric legal scholarship tends to idealise concepts  of fairness  and equality before the law, what is known as  ‘procedural justice’ (Sunshine and Tyler 2003). In the cases presented here, a different legal morality is expressed. The police and the people with whom they interact routinely strive to manipulate the law and demonstrate provisional agency. This is less a manifestation of a ‘culture’  of corruption or cynicism, and more a function of how continuously shifting social positions and capabilities are co-configured with moral good. Interactions that involve provisional agency disrupt conceptions of, and consensus on, a clear boundary between depraved and virtuous practice. Thus, what appears as ‘corruption’ or ‘collusion’ from one angle, from another looks like the fulfillment of one’s duty as patron or client to provide, protect or serve.

Police and legal patronage in northern India  257

Conclusion The production and negotiations of FIRs discussed here demonstrate how the police in northern India serve the public not as ideal ‘impersonal’ bureaucrats in Weber’s mould (Weber 1958 [1918]), but as patrons or providers of access to the state’s legal powers. This legal patronage reflects the historical development of law in India as an instrument for realising personal, professional, political, economic, social and cosmic goals (Cohn 1956; 1965). It works through relations among people who conceive of the state and its agents as useful tools in performances of, and contestations over, provisional agency as virtue, whether in a dispute over land, gathering support for an upcoming election, or the pursuit of any other aims. The Shuklas failed in part because they did not follow correct legal procedure, in part because the officer demurred at offending a political patron, and perhaps also in part because the SO was insulted that the Shuklas did not acknowledge his own role as the kingpin of his jurisdiction, but first went above him to the SSP. The pradhan saw it fit to come to a compromise with the SO, promising to hand over one or more of his party’s strong men in exchange for an intensified and expanded formal legal charge, which would likely improve or reinforce the pradhan’s reputation in his constituency. The refusal to register an FIR in the case of the Shuklas, and the negotiation over the terms of an FIR between a policeman and a village headman demonstrate how people simultaneously occupy the roles of patron and client, and how the vernacular ‘morality of legality’ is imbedded in the ability to protect and provide. The police officer offers certain types of legal access to those under his charge. Yet he is also a client to other patrons and a negotiator of patronage bonds outside his sphere of influence. He navigates a social order in which expressions and experiences of provisional agency are evaluated as virtuous. The concept of provisional agency is in dialogue with, though categorically distinct from, two principal paradoxes of patronage highlighted by Gilmartin (Chapter 5) in this volume: (1) ‘unequal reciprocal exchange’ and (2) an irresolvable tension between a negatively evaluated, self-serving, rational, instrumental logic and a kind of grace ordained by one’s demonstration of socially beneficent wealth and connectivity, on the other. While unequal reciprocity may be a general principle of patronage games, it is also important to underscore that

258  Beatrice Jauregui

the inequality among actors is itself an unstable condition, as many people simultaneously occupy the roles of patron and client, or may switch back and forth between the two. On the second point, the cases which I describe, especially the compromise between the SO and the pradhan, show that calculation is neither necessarily aligned with instrumentality, nor is generosity always treated as virtue. Indeed, these cases show that instrumentality and virtue are not necessarily opposed, especially if utilitarian negotiation is conceived as patronly dharma (moral duty). Shifts in social position and capability, and the contingent (though not necessarily congruent) transformation of moral valences in everyday practice, expose the inherently mutable quality of morality, legality, authority and justice. If the police are conceived, and conceive themselves, as duty-bound to deliver good to society, then instrumentality is their virtue and it may manifest itself in various ways, including patronal manipulations of law.

Patronage politics in post-independence India


Steven I. Wilkinson



olitical scientists used to view patronage politics as an unwelcome legacy of traditional societies, something that would give way to ‘programmatic politics’ as countries became more developed.1 This modernist perspective, however, has been challenged over the past 20 years, as scholars have realised that patronage politics is alive and well in many of the world’s most economically developed societies, including Italy, Belgium, Austria, Japan and the United States, and that there seem to be many good reasons to think that it will not diminish any time soon (Piattoni 2001; Kitschelt and Wilkinson 2007). In the United States a hundred years of good government movements, anti-corruption initiatives and civil service reforms have made a dent in political patronage, but have failed to put an end to it. Tellingly, in August 2003 the Chairman of the Democratic Party in Philadelphia warned thousands of city employees that if the Republican candidate was elected as mayor, their jobs would go to Republicans.2 There are many reasons for patronage to persist. Electoral systems which place a high premium on voters’ assessments of individual candidates, as opposed to proportional representation systems in which voters vote for a party list, are more likely to encourage politicians to differentiate themselves from each other by delivering patronage 1

By ‘patronage’ or ‘clientelistic politics’ I mean a transaction or series of transactions, the direct exchange of a citizen’s vote in return for direct payments or continuing access to employment, goods and services (Kitschelt and Wilkinson 2007, 2). ‘Programmatic’ party competition has been defined as a system in which politicians are elected on the basis of programmes and policies that they implement ‘as a matter of codified universalistic public policy applying to all members of a constituency, regardless of whether a particular individual supported or opposed the party that pushed for the rent-serving policy’ (Kitschelt 2000). 2 ‘Democrat says win by Katz would imperil patronage jobs’ (The Philadelphia Inquirer 27 August 2003).

260  Steven I. Wilkinson

goods to voters. Preference systems in which politicians from the same party compete for votes—as in post-war Italy’s 32 circoscrizioni electoral districts or in Japan’s multi-member districts—have a similar effect (Walston 1988, 21; Scheiner 2007, 276–297). Recent research on countries as diverse as Kenya and the United States also suggests that it is very difficult to persuade voters in ethnically diverse societies to agree on which public goods their governments ought to provide, to have voters cooperate across ethnic lines to monitor the delivery of these goods, and to persuade them that government bureaucracies will deliver these in a non-partisan way (Alesina, Baqir and Easterly 1999; Miguel and Gugerty 2005). These difficulties may lead to a preference among voters, as well as politicians, for quid pro quo exchanges in which at least some clear benefits trickle down to a community. In this chapter I ask two central questions: why is patronage so prevalent in India and what are some sources of change? I will first lay out some of the reasons put forward by economists and political scientists for the prevalence of patronage politics in India: the relative timing of mass political mobilisation, India’s great ethnic heterogeneity, and the effects of increasing intensity of political competition on raising both the demand for and the supply of patronage. I will then examine some potential causes of change. The 2011 mass protests by Swami Ramdev and Anna Hazare and his supporters (the latter leading to a government cave-in and promises of reform); India’s increasing overall levels of (unequally distributed) wealth and its growing educated and mobile middle class; the growing number of voters who tell pollsters that they are concerned about corruption and maladministration; the prospects of the Lokpal (Citizen’s Ombudsman) Bill and other anti-corruption measures; and the growing number of good government groups all seem to promise to have an effect on patronage politics. I nevertheless argue that in India, although these changes will probably, in the medium and long run, lead to substantial reduction in patronage politics, there are few immediate prospects for a quick shift to programmatic politics.

Why does India have high levels of patronage politics? Three key factors encourage patronage. First, at independence India lacked the governmental and political elements which Martin Shefter (1977) influentially identified as the requirements of patron-free

Patronage politics in post-independence India 261

politics: namely, bureaucratic autonomy and the presence of ‘external’, programmatic political parties which do not command state resources. Second, ethnic heterogeneity is an important factor that promotes patronage. In many different countries with high levels of ethnic heterogeneity programmatic politics and the neutral delivery of public goods appear more challenging. Third, the rising levels of political and electoral competition following independence have perpetuated rather than thwarted clientelism in the country. In the 1960s and 1970s, the state nationalised resources and distributed them to voters, and from the 1970s onwards both central and state governments also came up with a large number of development schemes enabling them to deliver the goods to an increasingly demanding and restless electorate. The decline of established rural and caste notables as a political force and the emergence of ‘new leaders’ (naya netas) has moreover made for a much more open and competitive ‘market’ in the delivery of goods and services in return for votes, further increasing the demand for patronage (Krishna 2002; Jeffrey, Jeffery and Jeffery 2008).

The timing of mass political mobilisation vs. the professionalisation of bureaucracy Perhaps the most influential work examining the historical antecedents of clientelism is Martin Shefter’s (1977), who tried to understand why some states in Western and Northern Europe had largely managed to avoid patronage politics while others, like Italy or the United States, had not. Shefter drew attention to the importance in the 19th and 20th centuries of the relative timing of bureaucratic reforms aimed at modernising the state and ‘reforming’ the civil service, as well as at the timing of mass party mobilisation. Where civil service reforms were carried out prior to mass party mobilisation, as they were in Germany and Scandinavia by monarchs who wished to fortify their states against internal and external threats, bureaucracy became largely resistant to patronage. In these countries, bureaucracies were able to withstand subsequent attempts to incline them to patronage because they had powerful political allies in society: the landed and professional elites who staffed their higher levels. Starved of state resources, external political parties in Germany and Scandinavia were forced instead to rely on ideological and solidary appeals. Shefter argues that where civil service reforms had not taken place before the development of mass politics, or where the franchise was extended by ‘internal’ parties

262  Steven I. Wilkinson

which controlled substantial state resources used to attract support, the result was patronage politics (Shefter 1977). Shefter’s thinking goes a long way to explain why in India patronage rather than programmatic politics dominated after independence. First, in 1947 the Indian state bureaucracy was neither highly professionalised nor independent (at least below the most elite Indian Civil Service, Indian Police and provincial civil service cadres). Second, between 1919 and 1947 the Congress Party had developed substantial experience as an ‘internal party’ which used state resources to reward its loyal supporters. Following independence, this practice of winning over voters (who were given votes by the 1950 constitution) carried on. The Indian civil and provincial services were far from being the independent and professional ‘steel frame’ of government, which their rhetoric suggests. The colonial state in fact paid the great mass of its officials (below the Indian Civil Service and Indian Police) too little to encourage honesty on the job. In the case of the police, every colonial police commission and inquiry raised the issue of insufficient wages and housing for the ordinary constable, and proposed (unsuccessfully) large-scale reforms to improve the honesty of the force and to avoid its ‘capture’ by the criminals they were meant to police. In 1902–1903 the Indian Police Commission found that the police were ‘generally corrupt and oppressive’, and that they had ‘utterly failed to secure the confidence and cooperation of the public’ (Razvi 1961, 192–193). On the eve of the independence, one senior British police officer in Uttar Pradesh (UP) told an inquiry into government pay and conditions quite frankly that only around 1% of his Station House Officers were honest (‘Report of the U.P Pay Committee’ 1947, 379). The colonial state itself was deeply clientelistic. British adminis­ tration rewarded its loyal clients with jobs, contracts, land grants and various honours, while withholding these from its opponents.3 The classic case of patronage by the colonial state was the immense irrigation project, indeed one of the largest-ever civil engineering projects, initiated in the Punjab between 1885 and 1947. Here officials gave large portions of the 11 million acres which were newly brought through these projects into production in the Punjab, to serving and 3

These patterns of favouritism were formalised in recruitment rules (e.g., the 1906 rules in UP), which specified the exact percentages or upper limits for the recruitment of different religious and caste groups.

Patronage politics in post-independence India 263

retired soldiers, key ethnic groups and rural notables in order to reward or purchase their support.4 The largest single award of nearly 8000 acres was granted to the prominent Sikh leader, Baba Sir Khem Singh Bedi, whose support was seen by the British administration as particularly valuable (Ali 1988, 73). As a result of an enormous investment of resources through patronage channels in the Punjab, this was the only province where a political party allied with the British won the first genuinely competitive elections in 1936. Bengal, a ‘disloyal’ province, was by contrast consistently denied public investment and raided for resources, and Bengalis were systematically excluded from service in the army.5 The Congress Party was also, crucially, not just an ‘external’ opposition party, but—as a result of the many elected bodies created by the British in the three decades before independence—also an ‘internal’ party which possessed substantial access to state funds. The long and hard Congress campaign for independence and the sufferings of many individual Congress workers should not obscure the significance of the way that state resources influenced Congress electoral strategies and the position of Indian bureaucracy following independence. After 1919, as party politics expanded, Congress politicians used whatever control they had over elected bodies to deliver benefits to their key constituencies. Myron Weiner reports that in 1923, when the first elected municipal corporation was established in Calcutta, ‘political patronage rather than merit became the criterion for appointments’ and ‘payrolls [were] padded with the names of nationalist political prisoners, under-qualified men appointed to many posts within the municipal administration and councillors more powerful than the municipal administration became features of municipal government’ (1967, 327–328). In 1962 one councillor with experience in the corporation during this period said to Weiner: Yes, it is true that the old administration was corrupt. The reason was that the political sufferers [nationalist workers 4

David Gilmartin has shown how the Punjab government made grants of land to district chiefs (zaildars) and other landlords effectively to ‘bind this class of rural patrons to their regime’ (1988, 26). 5 Between 1916 and 1934, 100% of the very substantial jute duty raised in Bengal was spent by the Central Government. For Bengali complaints about the misuse of Bengal-derived taxes by the centre, see the 27 February 1935 debates on the Bengal budget in the Council Proceedings of the Bengal Legislative Council (1935, 421–423).

264  Steven I. Wilkinson

who had been in jail] were appointed to the corporation. Big businessmen who gave money to support the nationalist movement got contracts from the corporation. The corporation was the only thing we controlled. The corporation maintained the families of political sufferers by giving jobs to sons or brothers. Political sufferers got full salaries while they were under detention. When the new act went into force in 1952, maybe 70 per cent of the corporation were political sufferers. Even now 40 or 50 per cent are political sufferers. But all of this corruption was for helping the nationalist movement (ibid., 328).

David Washbrook (1973) tells us that colonial district governments were ‘enormous pools of patronage’: in the South between 1909 and 1919 their budgets increased by 70% (to an annual average of 1,100,000 rupees per district), doubling again between 1919 and 1929. Provincial governments, which were established in 1919, channelled an even larger pool of patronage resources to Congress and its rivals, as they now had not only substantial tax-raising powers but also control over education and (after 1935) agriculture. In the period of Congress provincial governments in the late 1930s and mid-1940s, the ‘opposition’ Congress party mobilised voters as much through the distribution of resources which they controlled as through programmatic appeals. At that time, many joined the party not simply out of concern for freedom but also and significantly because they sought jobs, contracts and political influence. Nehru himself expressed deep reservations about this change in the character of the party in the late 1930s, when following its victories in the 1936 elections, its membership shot up to 4.5 million and a large number of fixers and favour-seekers swelled the party’s rolls. Congress politicians increasingly threatened civil servants and policemen with retaliation if those did not do their bidding and, especially as independence approached, the colonial state was increasingly likely to allow interferences with bureaucratic autonomy in the broader interests of decolonising India with minimal conflict. Late in 1946, for instance, the British accepted the forced resignation of the top police official in UP, who had tried to stop Congress politicians from contacting lowerranking police officers directly to request transfers and other favours, and thereby interfering with his chain of command. Initially, both the British Governor of the province and the Viceroy attempted to defend the officer, but after threats from Nehru, Govind Ballabh Pant

Patronage politics in post-independence India 265

and Sardar Patel, the Viceroy gave up for the sake of keeping broader constitutional negotiations on track, and the officer was sent home.6 In the system which emerged in the late 1940s and 1950s, many Congress politicians already had extensive experience of state patronage, and many had joined the party in search of it. What bureaucratic autonomy did exist could also be reversed by politicians, a process also traced by Piattoni in France (2001). But in India, as in the other post-colonial states, insistence on bureaucratic autonomy was initially even more vulnerable because here it could be easily derided as an illegitimate, foreign imposition.7 One retired police officer from Madhya Pradesh (MP) describes how in the early 1960s, after he refused to drop criminal charges against allies of Arjun Singh, later a senior Congress politician and a member of the MP Legislative Assembly at the time, the Chairman of the local Congress Committee delivered a public address against the official, in which he said: ‘It is we [the people] who elect the MLAs [Members of Legislative Assembly]. The MLAs then elect the Chief Minister. Therefore the Chief Minister and the Ministers…are our servants. And the government servants are servants of our servants whether it is the Collector or the S.P.….Hence, the servants of our servants can never be our masters!’ (Sharma 1991, 91). As many have already observed, Indian politicians, through their power to punitively transfer and otherwise discipline state officials, have many weapons which make bureaucrats and police subject to their will (Wilkinson 2004, Ch. 4; also chapters 8 and 10 by Berenschot and Jauregui in this volume).8 In some states, the power to transfer officials is explicitly recognised and rationed as a scarce political resource: from the 1990s, the leaders of the ruling parties in MP and Rajasthan have been giving their MLAs the (formally illegal) right 6

‘Sir Philip Measures, Inspector-General of Police in UP: circular instructing Superintendents of Police to correspond with Ministers through him; objections; resignation of Sir Philip’ IOR/R/3/1/191 Aug 1946-Apr 1949 and Dayal (1998, 80–81). 7 Admittedly, this was less often the case in India than in most other colonies, such as Kenya and Tanganyika or the Sudan, which had hardly any indigenous presence in the government until shortly before independence. In India, on the contrary, a decade before independence Indians constituted 51% of the Indian Civil Service and there were elected Indian governments with substantial powers. 8 Not all officials submit to subjection. In 2002, for instance, 27 were allegedly ‘punished’ by transfer by Narendra Modi’s government in Gujarat for their refusal to allow communal violence to carry on. Most officials nevertheless are naturally reluctant to resist, as that comes at a high cost to their careers, families and sometimes also personal safety.

266  Steven I. Wilkinson

to transfer up to 4 senior officials every year. A central government investigation into police effectiveness in the late 1990s also found that punitive transfers and political interference in the states had become so widespread that the average tenure of key district-level officers was about 4 months (Times of India 31 August 2000).

Ethnic heterogeneity India’s extraordinary ethnic diversity may also promote the politics of patronage. Ethnic diversity creates disagreement about the basis on which goods ought to be delivered to voters. It also promotes uncertainty over whether the government will really deliver what they promise to ethnic groups which are not represented in the governing coalition, and thus encourages clientelism.9 A number of recent studies by economists and political scientists suggest that it is more difficult to get people to agree to the provision of public goods, to support greater public spending on things like education, roads and water supply, and then to monitor and ensure the delivery of public goods, in places with higher levels of ethnic heterogeneity (Alesina, Baqir and Easterly 1999; Miguel and Gugerty 2005; Habyarimana et al. 2009). Alesina, Baqir and Easterly found that the more ethnically diverse a district is in the United States, the less government funding is spent on roads, sewers, education and welfare (1999). The authors analysed data from all American counties in the 1990s and found, consistently, that more diverse counties spent less on roads and schools, and that ‘a move from complete ethnic homogeneity (ethnic=0) to complete ethnic heterogeneity (ethnic=1) would lower the road spending share by nine per cent’.10 Studies conducted in Africa, among different groups in Uganda and Kenya, and studies examining changing preferences for the provision of public goods over time, as countries become more diverse, generate broadly the same findings.11 According to a large-scale 9

This is not to argue that ethnic diversity has no benefits, including economic benefits, as explored by Page (2007) and Rauch and Trindade (2002) in terms of macro-economic payoffs. 10 The research controlled for income per capita, city size (big cities may be more fragmented with more ghettos), age structure (which is well known to influence support for public goods like education) and income inequality (to control for the possibility that race might be just a proxy for income inequality). 11 On welfare spending, see Luttmer (2001) and on trust and participation Alesina and Ferrara (2000).

Patronage politics in post-independence India 267

survey conducted by Habyarimana et al. (2009) in Uganda’s capital Kampala, the actual public goods provision seems to be consistently lower in more diverse areas. One reason which they identify is the difficulty in more diverse areas of disciplining non-coethnics who do not ‘play by the rules’ or refuse to cooperate in efforts to provide public goods for the larger community. Gugerty and Miguel’s (2005) study of community school funding in Kenya further confirms this: both cooperation in raising money and then sanctioning those who do not contribute to the provision of public goods appear more difficult in more ethnically heterogeneous environments. The growing economic literature on the effects of ethnic heterogeneity on the ways in which resources are distributed focuses largely on macro-economic outcomes. But taken together, it also suggests three mechanisms that may help to understand why patronage politics may be more prevalent in ethnically diverse societies, especially (in Shefter’s terms) in those which have not developed a professionalised bureaucratic state prior to mass democratisation. Disagreement among different groups over just which goods and services they expect from the government, a lack of voters’ confidence that a government administration dominated by one group will deliver goods or job opportunities to the others,12 difficulties in monitoring the delivery of services and sanctioning those who violate the rules or do not deliver as they should (because of the added potential for conflict when members of one group discipline or challenge members of another) all promote clientelism.13 These mechanisms all find some support in the work of historians and anthropologists. Given the sharp differences between access which different caste and religious groups have to wealth 12

After the Congress first came to power in the 1930s, it was ineffective in its (halfhearted) efforts to root out caste bias in the administration. In 1938 the elected Pant government in UP ordered police recruitment to be henceforth made on grounds of merit rather than caste quotas established in 1906. But my interviews in the 1990s with officers in charge of recruitment in the post-partition period made it clear that caste ties continued to play an important role. 13 This mechanism has been extensively documented in a recent series of studies of Uganda by Habyarimana et al. (2009). The authors found that in experimental situations co-ethnics, in particular ‘egoists’ who might otherwise be least likely to play well with others, cooperate more with each other in discrimination games because they believe that members of their own ethnic group are more likely to punish them in some way if they don’t. They argue that fear of identifiability and punishment by co-ethnics is what drives norms of co-ethnic reciprocity (ibid., 125–130).

268  Steven I. Wilkinson

and services, and residential segregation by group in many areas, Indian groups differ in their assessment of which goods they need: what benefits Jatavs may not benefit Jats, Brahmins or Kurmis. It  also seems clear that the Indian state is not widely perceived as a socially neutral political entity, but as one that is dominated by particular groups. Consequently, most believe that one can only secure the services of the state via patrons (ideally, members of one’s own group). Moreover, attempts to remonstrate with government servants from different groups in a society with deep caste and religious cleavages are frequently deflected by those on the receiving end as ethnically motivated, making coordination more difficult and increasing the risk of conflict as well as diffusing the pressure for reform. Because of these mechanisms, arranging direct transfers between politicians and voters, patrons and clients, might be preferable to relying on state spending that may never reach one’s own group, or may be spent on things one does not want in the first place.14

The effects of political competition Competition among Indian political parties, which has been burgeoning since the 1960s, has spurred greater demand for and supply of patronage.15 In every state, the average electoral volatility—the percentage of seats that change hands at each election—has doubled since the 1950s, rising from 21% to 42%. The number of effective parties competing for office (calculated by using a fractionalisation index16) has also grown. In Bihar, for instance, it rose from 2.6 in 1966 to 4.3 in 1967 and 5.2 in 1970. Inter- and intra-party competition, which has expanded since the 1950s, increased both the supply of clientelistic programmes and the demand for them. After the death of Nehru, a major conflict 14

I do not suggest here that ethnic heterogeneity is the only cause of uncertainty about the neutrality of the state. Walston, in his study of the much more homogenous society in Calabria, notes that clientelism continues there ‘at least partially not because there is no state (patently untrue in modern societies) but because its uncertainty and inefficiency render the intervention of another figure essential or at least useful. On the local government level it is not surprising that clientelistic practices are almost unknown where there is an efficient bureaucracy’ (1988, 36). 15 We develop this point further in our introduction to Kitschelt and Wilkinson (2007). 16 This is measured by an index of party fractionalisation, 1/Σsi2, where si is the proportion of seats held by each party.

Patronage politics in post-independence India 269

erupted inside the Congress, pitting party President Indira Gandhi and her supporters against well-established patronage networks in the states led by ‘the syndicate’ of established party leaders. Social mobilisation and increasing education among middle and lowercaste voters in urban and rural areas made their community leaders increasingly unwilling to accept the very low payoffs they had received from the Congress, which in the 1950s was dominated by the upper castes. Middle-caste leaders who were not satisfied with the Congress support of caste-specific and farmer-specific subsidies, founded several new political parties around farmer (kisan) and middle-caste identities to advocate increased spending on their particular needs. In the 1967 state elections held across northern India, these parties, such as the Bahujan Kisan Dal (BKD) in UP, did very well.17 Increasing state control over the economy after land reforms and the second five year plan (1957–62) had further alienated private businesses and former large landholders, who collated their opposition in the Swatantra Party during the 1967 elections. The Party won 44 seats in parliament (enough to form the opposition) and sufficient seats in the state assemblies of Bihar and UP to form coalition governments there. Most of Swatantra’s funding came from big businesses, which could legally donate 5% of their profits to political parties without the shareholders’ approval, and unlimited amounts with it (Desai 1999). In the late 1960s and early 1970s the Congress responded to these challenges by expanding its own resources and striking at the resources of its opponents. It nationalised banks and insurance companies, creating new loans and investment funds which could be channelled to voters and party donors. The 14 major banks nationalised in 1969 were instructed to pool 40% of their funds into ‘priority areas’, usually identified by politicians, sometimes into the infamous ‘loan melas’ (loan fairs) where loans were handed out to voters as gifts. The Congress also established a whole new panoply of centrally and state-sponsored anti-poverty schemes, which proved highly inefficient in ameliorating poverty but which were very effective in channelling resources to politically connected groups of voters and politicians. In 1969 the Congress also passed


For details on the very low representation of middle and lower castes in the Congress in the 1950s and 1960s, see Jaffrelot (2002).

270  Steven I. Wilkinson

various laws to ‘clean up’ elections, thereby denying its opposition legal access to electoral funds from private business.18 Electoral competition has clearly driven the growth of Centrally Sponsored Schemes (CSSs) and state-sponsored patronage schemes since the 1960s. Many of these schemes had names, like the Jawahar Rozgar Yojana ( Jawahar Employment Programme) or the Indira Awaas Yojana (Indira Social Welfare Programme), which linked them explicitly to the Congress party or its leaders. Rival parties mimicked this practice, setting up their own schemes under their own names when they displaced the Congress in the government of a state. In Orissa the Biju Janata Dal (BJD) state power ministry launched the Biju Gram Jyoti programme (Biju Rural Light, named after the party icon, Biju Patnaik) for the ‘electrification of villages / habitation which are not scheduled to be covered under Rajiv Gandhi Grameen Vidyutikaran Yojana (RGGVY)’.19 As Anirudh Krishna (2002) and others have argued, over the past two decades voters have become less rigidly ‘fixed’, or reachable through established patron–client relations or political party links. Bonds between parties and voters have become much weaker than they were under the Congress one-party system in the 1950s. Relations between rural people and parties and politicians are now more likely to be mediated by ‘new leaders’ (naya netas) or ‘fixers’. Many of them are educated villagers, some are tied to parties and others yet operate as brokers in a market (see Berenschot’s Chapter 8 in this volume).20 Krishna found that 60% of the villagers he surveyed in 69 villages across MP and Rajasthan turned first to the naya netas for help. Yet the netas themselves often rely on ties with government and party officials to get their clients’ jobs done (Krishna 2007, 150–156; also Björkman’s Chapter 7 in this volume). Clients can abandon their fixers for others better fit to deliver goods and services, the demand for which has been augmented by rising levels of education and wealth, higher professional and economic aspirations and the emergence of a more transparent ‘market’ economy (in place of earlier hierarchies). 18

A detailed account of this is contained in my chapter in Kitschelt and Wilkinson (2007). 19 20 Often, as Craig Jeffrey’s (2008) work shows, naya netas come back to the village because their sub-standard college education, important though it has been in giving them status as well as political and administrative connections, has not made them competitive in the general labour market.

Patronage politics in post-independence India 271

On their part, politicians have been doing away with some of the ‘permit raj’ since 1991. Yet they still substantially access voters by facilitating access to the many state services, like policing, medical care and education, which remain largely discretionary. They lure both middle classes and poorer voters with the anti-poverty schemes, which every year circulate around 72,000 crores of rupees (16 billion dollars) of highly politicised funds, and the large-scale subsidies for food, fertilizer and cooking oil.21 The Targeted Public Distribution System (TPDS) worth an annual 36 crores of rupees (7.1 billion dollars) is but one major example. Several important audits performed by the Comptroller and Auditor General (CAG) of India and the Planning Commission have exposed how these programmes work in practice. The 2003–2004 audit conducted by the Planning Commission in 60 districts in 18 states found that only 42% of subsidised grain actually reaches households Below the Poverty Line (BPL); ‘political connections’ frequently helped to determine which of these households received a BPL card, which enables access to subsidised food, and which did not (only 57% of eligible households have a BPL card). More than 20% of the subsidised food is skimmed off by households Above the Poverty Line (APL) (42% in Karnataka and 50% in Tamil Nadu), with political connections again determining access to the card. Politically connected owners of Fixed Price Shops (FPSs) or people further up the distribution chain likewise embezzle a great deal (Indian Planning Commission 2005). Figure 11.1 gives some sense of how errors of inclusion and exclusion affect the total distribution of food to BPL households through the TPDS schemes. In states like Bihar, UP and Punjab more than 60% of food is diverted from its beneficiaries, and whatever does reach BPL households usually goes only to families sufficiently connected to political patrons to receive the card in the first place. District-level micro-audits have also produced ample evidence that political connections play a decisive role in accessing state funds and development projects. In 1998–1999 an audit of the Integrated Rural Development Project (IRDP) spending in the Gonda district of UP found that the largest single category of Scheduled Caste (SC) programme beneficiaries (31%) gained access 21

For a critique of these programmes, data on their extent and a plea for Direct Cash Transfers as a solution, see Kapur, Mukhopadhyay and Subramanian (2008).

272  Steven I. Wilkinson State


Tamil Nadu


West Bengal


Andhra Pradesh








Himachal Pradesh




All India










Uttar Pradesh


Madhya Pradesh






Source: Leakage and Diversion of Subsidized Food Grains in TPDS (Targeted Public Distribution System), Indian Planning Commission, Performance Evaluation of TPDS (Indian Planning Commission 2005)

Figure 11.1 Percentage of total misallocated or lost food in the 2003–2004 Targeted Public Distribution System, TPDS (as % of BPL quota). to it not through formal channels—which themselves may have been subject to political influence—but through the intervention of ‘others’, like the ‘functionaries of political parties’.22 The proportion of SC voters who got IRDP benefits varied widely depending on how electorally pivotal the SCs were in a given village: ‘There are only 22 percent of IRDP beneficiaries in villages with a low concentration of Scheduled Castes, whereas there are almost 40 percent in villages with a higher concentration of Scheduled Castes’ (ibid.). This is also how 43% of Other Backward Caste (OBC) households in Gonda gained access to the programme. 22

Planning Commission audit of IRDP, Gonda district, UP, 1998–1999, viewed in July 2013

Patronage politics in post-independence India 273

The rest were selected at the village meeting (gram sabha), an intensely political process. So, what political logic drives the distribution of state goods? Political scientists who write about patronage are divided into two camps: those who see party patronage as primarily a way to pay off core party supporters and increase voter turnout, and those who see it as a means to buy swing and opposition votes.23 But a large number of quantitative studies of clientelism have failed to prove either theory. There are 4 main reasons for this. First, accurate data on patronage are hard to obtain. Many transactions are illegal and politicians have an obvious interest in keeping the extent of their patronage secret; as a result, voters are never sure whether they are being underpaid or overpaid relative their peers. (Mayor Daley of Chicago was thought to be the only man in the city who knew exactly how many jobs his machine controlled.)24 Second, in practice, political parties use patronage both to pay off core supporters and to attract swing votes. In an earlier article (2007), I analysed how in 2000–2002 an MLA in one sub-district of MP disbursed funds from the Local Area Development Scheme. I found that the MLA’s funds first went to three of the 10 village councils (panchayats) which supported her in previous elections (amounting to 72% of her overall expenditure) and that nothing at all went to the panchayats which did not. But two years after her election she began to increase the number of grants she gave to the other panchayats, whose support was now pivotal to the elections ahead. The caste leaders in one panchayat held a meeting at which they publicly switched their allegiance to the block’s dominant faction, which supported the MLA. As a direct reward for this shift, and also presumably to encourage others to defect from rival factions, the following year the MLA gave the largest single grant of 225,000 rupees to this panchayat. Third, the effect of patronage on votes is difficult to assess empirically because patronage is rarely a single, quid pro quo transaction but often operates over time through a large number 23

Stokes (2005) looks at the use of patronage in luring opposition votes, and Cox (2006) analyses the effects of patronage on the participation of existing supporters. 24 See Royko (1971, 98) and extracts in Haeger and Weber (1979, 95–104; also 1979, 395). This does not, however, mean that politicians necessarily wish to keep all evidence of their patronage secret, because of the value of having a general reputation as a powerful donor and ‘fixer’.

274  Steven I. Wilkinson

of interactions, most of which are not easy to trace. People receive benefits through ties of friendship, family or other kinds, not just for casting votes. Even in Michael Johnston’s (1979) comprehensive study of patronage in New Haven, where he collected meticulous information on the recipients of city summer jobs in 1974, it is hard to say whether ethnic particularism is the best explanation for the distribution of jobs, not least because we do not know whether the jobs represented all or a small portion of the total number of patron-client exchanges. Fourth, even if a patron gives something to a voter who then votes for him, this does not necessarily mean that the voter’s choice was ultimately determined by this gift. Both the gift and the vote may be part of complex, much broader, non-economic webs of relation, obligations and motivations. These complex webs have been vividly described in some studies of patronage in political science, which were based on long-term fieldwork and significant immersion in local culture (e.g., Scott 1972; Auyero 2001). Scott emphasised that patron–client interactions are ‘whole-person’ relations rather than explicit and impersonal contract bonds. A landlord may, for example, have a client connected to him by tenancy, friendship, past exchanges of service, the past tie of the client’s father to his father and ritual ‘co-parenthood’ (Scott 1972, 95; in Mintz and Wolf 1950). But the distinction between what Scott terms the ‘affective’ and the ‘instrumental’ aspects of clientelism has been lost in recent political science, no doubt in part because it is not easy to quantify.

Prospects for change In what ways and to what extent may India’s incentives for political patronage change in the future? In the introduction to our book, Herbert Kitschelt and I (2007) argued that a country’s level of patronage, once conditioned by the historical legacies examined by Shefter (1977), is determined by its degrees of electoral competition, economic development and ethnic heterogeneity. The overall effects of interaction between these factors are sketched out in Figure 11.2. I predict that, as economic development increases over the decades, the overall presence of patronage in Indian politics will decrease. Though the large number of poor voters, continued corruption in India’s bureaucracy, and the country’s intense ethnic heterogeneity will ensure that patronage will not altogether disappear.

Patronage politics in post-independence India 275 12

Policy Mix



Programmatic Linkage through Policy Goods Clientelism due to Ethnocultural Social Networks


Clientelism due to Electoral Competition & Economic Development



0 1950









Figure 11.2 The policy mix in India at different levels of competition and economic development (from Wilkinson 2007, 111). Why should the demand for patronage decline as economic development advances? First, as economies become more developed, the kinds of goods that people demand from the government— whether better regulation or the provision of services and complex technologies—become too difficult to secure from individual patrons in the traditional manner. Second, economic development increases the number of middle-class voters—like the many who came out in favour of the Anna Hazare movement and voted for the Aam Aadmi Party in Delhi in 2013. They have the leisure and organisation to protest effectively and are also more likely to be less perturbed by the costs of opting out of patronage networks. As Banfield and Wilson (1963) pointed out some time ago, patronage does not die because patrons stop offering, but because increasingly prosperous clients no longer wish to receive. The Indian poor, on the other hand, though they may care about the evils of the system just as much as, or more, than the urban elites, may prefer targeted handouts to the distant benefits of policy change (Kitschelt and Wilkinson 2007, Introduction). Some have argued in fact that the Indian poor vote more than the rich precisely because they benefit more from patronage. But the levels at which the poor vote have probably been exaggerated. The oftencited surveys conducted by the Centre for the Study of Developing Societies (CSDS)/Lokniti claim that on the whole the poor vote 1%–2% more than those who are better off. One problem with these polls is that they are based on voters’ claims, rather than independent

276  Steven I. Wilkinson

verification from voter rolls. But voters are not always honest about whether they have voted or not, either out of embarrassment or to preserve the exchange value of their votes. Nirmala Ravishankar tells us that 86% of people interviewed in the 2004 CSDS survey reported having voted. Yet we know that the actual turnout rate was closer to 60% (Ravishankar 2007, 81 and 101–103).25 Ravishankar’s close examination of polling data reveals that neither the voter’s social group (except for Muslims, who voted less than the Hindus), nor their wealth seemed to have a significant effect on how likely they were to vote (ibid., 103). Because CSDS surveys do not show us which classes of voters are more likely to exaggerate their turnout, we do not know whether the actual voter turnout among any of these groups is in fact higher or lower than average.26 We can nevertheless say that the poor in India do not vote notably less than the rich. This is no doubt in part because they value democracy, but it is also likely that their vote promises more immediate, tangible and relatively valuable returns from patrons or parties. Were it not for the large amount of patronage, Indian voter statistics would probably look much more like the other countries’ where the poor vote notably less than the rich. Thirdly, economic development may reduce the demand for patronage because a growing educated middle class, with the leisure to read books and newspapers, and to watch televised investigative reports on corruption and malfeasance, is likely to be more informed about the negative aspects of patronage and possess more information on the advantages of generalised reforms. Throughout the 2000s, opinion polls have shown that increasing numbers of Indian voters have mentioned corruption as a major concern, with 43% mentioning in a 2005 survey increasing corruption as the single worst aspect of India’s democracy (State of Democracy in South Asia 2009, 245). Historical evidence shows that once a substantial portion of the electorate becomes aware of and inclined to reform, high levels of electoral competition begin to reverse the political trajectory: they cleanse the system instead of increasing the supply of patronage. In the United States the advent of good government movements in the 25

In the 1996 CSDS survey 87% of voters also claimed that they voted in that year’s election, reporting 20% more than the polling roll. 26 The best source on this is Ravishankar’s Harvard Government Ph.D. dissertation, ‘Voting for Change: Turnout and Vote Choice in Indian Elections’ (2007).

Patronage politics in post-independence India 277

1890s and 1900s, combined with high levels of political competition, forced Democrats and Republicans, in at least some of the towns and states, to support reforms to win elections. Good government groups refused to support parties unless they promised reform in the form of open primaries, civil service reforms and town charters, forcing the parties to oblige. This suggests that in a competitive party system, once a significant portion of the electorate (about 20%) chooses generalised reforms over patronage, political party dynamics will do the rest. Individual parties will calculate that they have more to gain by offering pro-reform agendas, or (more likely) a mixture of reform and patronage, than by engaging in traditional patronage politics. These parties will enact reforms, which will limit (although not end) patronage, in turn reducing the number of people who will demand access to it. This will consequently increase the political pressure on the patronage redoubts that exist for change. In 2009 I travelled to Orissa with the hope to better understand the reasons for the landslide re-election of the BJD in the state the previous summer. My visit made it clear that the party had won by combining traditional money and patronage politics with a more programmatic development platform. They had built a reputation for being more efficient and ‘clean’ by delivering electricity, roadways and water more effectively than their predecessors had managed, and voters I spoke to in Bhubaneshwar and the rural areas of Cuttack district all agreed that the BJD had done better than previous regimes in providing low-priced food and roads (generally excellent in Bhubaneshwar and the main parts of Cuttack). Bishwanath Panda, a priest at the Lingaraj temple in Bhubaneshwar spoke, unprompted, about the achievements of the Chief Minister Naveen Patnaik’s administration: smooth roads, better water supply and improved anti-corruption measures. Many other voters I met remarked on the same.27 Several also approvingly mentioned the fact that Patnaik was a bachelor, which meant that he had fewer family members to ‘feed’, and several said that he was doing a good job even though he spoke little Oriya.28 One man summed it up: ‘All politicians eat, but the BJD eats less’. 27 28

Interviews were conducted on 13 and 14 December 2009. I was told that his cabinet meetings are held in English, and that on the rare occasions that Patnaik gives speeches in Oriya, they are written out phonetically in Roman script.

278  Steven I. Wilkinson

The BJD government clearly made it their priority to improve the universal delivery of basic amenities, and to be publicly seen as improving the lives of the people. They widely publicised the rice at two rupees a kilo scheme (at a ration of 25 kilos per month for the BPL families). The Chief Minister and other state ministers monitor the scheme’s implementation at weekly meetings, and several able officials have been placed in charge of its efficient delivery and measures that minimise leakage (Aggrawal 2011, 21–23). Judging by recent independent surveys of food distribution under the Public Distribution Scheme, these efforts have been broadly successful. Shortly after taking office, the former Roads Minister A.V. Singh Deo made a show of suspending a large number of Public Works Department officials to penalise them for poor road quality and to signal the government’s low tolerance for the Department’s notoriously corrupt road-building process.29 It would be a mistake, however, to think that the BJD has magically become a ‘good governance party’ and that it now runs on its party platform without the support of patronage. While the party (and the World Bank) has promoted good governance initiatives and basic infrastructure delivery to one part of the electorate, it has also continued to raise enormous sums of money through control of the mining licenses and government land sales in order to fund patronage politics elsewhere, where the traditional way of running the state anti-poverty schemes continues as usual. While better roads and electricity help to win over voters in thriving urban areas around Cuttack and Bhubaneshwar (where the party won 71 out of 77 seats in 2009), old-style politics seems much more prevalent in the poorer hinterland. A recent CAG audit found that in the administration of the massive Backward Region Grant Fund Programme (BRGF), in 19 of the state’s most ‘backward’ districts, projects were hardly ever developed as planned or in consultation with village assemblies, but were channelled instead through party ministers and MLAs.30

Conclusion In 2011 there was an active discussion in India about whether a Lokpal/Jan Lokpal (Ombudsman) structure proposed by the United 29 30

Interview with A.P. Singh Deo, 13 December 2009. state_audit/recent_reports/Orissa/2010/Civil/Chapter_2.pdf, 17–18.

Patronage politics in post-independence India 279

Progressive Alliance might reduce patronage and corruption in India. At one level, the Lokpal/Janpal Bill looked like it made sense. The basic problem with India’s current anti-corruption machinery is how little constitutional authority and legal bite it has, compared to its counterparts, like the American Federal Bureau of Investigations or the Corrupt Practices Investigation Bureau in Singapore.31 The Indian Constitution makes law and order the subject of the state and it gives the central Controller and Auditor General (CAG) and state officers the right to investigate the misuse of state funds. The CAG has indeed identified numerous instances of embezzlement in its excellent reports. But the CAG has no legal power to do anything about the misuse it uncovers, as the discretion over investigation and prosecution of such cases rests with the very state governments and politicians who misused the funds in the first place. India’s leading anti-corruption investigative body, the Central Bureau of Investigation (CBI), similarly operates under the feeble legal authority of the 1946 Delhi Special Police Act, which limits its jurisdiction to cases which directly involve the centre, or which a state government gives the permission to investigate and prosecute.32 In practice, state administrations have granted such permission selectively, allowing the CBI to proceed against political enemies and the opposition and not against their own political allies. In the spring of 2013 resistance by the Congress coalition partners to an investigation into the 31 billion dollar ‘Coalgate’ scandal almost brought the United Progressive Alliance government down. The centre’s aversion to strong central anti-corruption organisations is perhaps understandable, given the long history of misuse of Article 356, which allows the government of a state to be dismissed and replaced by central ‘Governor’s rule’ against opposition. But it hampers any larger anti-corruption efforts. 31

Singapore has a very comprehensive Prevention of Corruption Act (1960) and a highly effective Corrupt Practices Investigation Bureau enforcement (http://app.cpib., which makes it easy to complain and which is widely regarded as impartial and efficient. All complaints are processed for further action within two days, with a final decision about whether to go forward with an investigation made within a fortnight. 37% of complaints in Singapore are now submitted by email, and 43% by email or fax. 32 Clause 6 of the Act specifies that ‘Nothing contained in section 5 shall be deemed to enable any member of the Delhi Special Police Establishment to exercise powers and jurisdiction in any area in 5 [a State not being a Union Territory or railways area, without the consent of the Government of that State]’. dspe.php.

280  Steven I. Wilkinson

Dissatisfaction with current political machinery and its corrupt outcomes was one of the major reasons behind public pressure to pass the Lokpal Bill. In the summer of 2011 Anna Hazare and his followers succeeded in pressing the bill on the centre—reflecting the tougher Jan Lokpal proposal rather than the original government bill—in that autumn’s Lok Sabha (Lower House) session. Supporters of the Bill claimed that the People’s Ombudsman (Lokpal) in the centre and Anti-Corruption Ombudsmen in the states (Lokayukts), with new statutory investigative powers over politicians and bureaucrats, would drastically deter patronage and corruption. But the proposed Bill— still languishing at the time of writing—was to staff the new bureau with personnel from the existing and already overburdened Indian Administrative Service, Central Vigilance Commission (CVC) and police cadres, which would limit its capacity to process a large number of complaints. Many constitutional aspects of the Bill remain unclear: which punishments the Lokpal could levy, who would prosecute the cases once they were investigated, and whether a central body could prosecute state officials and politicians without infringing on the states’ constitutional powers over law and order.33 The Lokpal Bill is not a panacea. As a recent article reminded us, the Indian Penal Code has already more than enough provisions under Chapter IX: ‘offences by or relating to public servants’ to take action against all offenders; what’s missing is the political will (Banerjee 2011). As I have argued here, this determination will not come from legislation alone. It will ultimately have to come from politicians’ responses to the growing number of voters who make the concern with patronage and corruption a serious determinant of their political choice.


Since the 1994 Supreme Court Bommai decision, the power relationship between the centre and the states tilted in the other direction, allowing state parties pivotal to the national United Progressive Alliance and the National Democratic Alliance coalitions to extract greater fiscal transfers and recognition of the states’ constitutional prerogatives in return.

Prospects and Disappointments

Kingship without kings in northern India


Lucia Michelutti


 his chapter examines the relation between patronage, democracy and ‘criminal politics’ in northern India by investigating the ways in which lower castes, which recently entered the political elite, had vernacularised the concepts of ‘political representation’ and ‘the people’.1 Democracy claims to deliver rule by ordinary people, but elected politicians are necessarily extraordinary (Dunn 1992, v). The question I ask, then, is: what happens to political patronage when the ‘common man’ becomes ‘king’? More specifically, I explore the implications of being ruled (and patronised) by ‘ordinary people’ from historically lower castes, and the ways in which ordinary people are transformed into the extraordinary with divine, muscular and expressly ‘democratic’ qualities. Contemporary forms of political patronage in the north Indian state of Uttar Pradesh (UP) continue to be substantially articulated in the idiom of kingship divested of its royal qualities by secular and democratic charters. I argue that when politician-‘kings’ are stripped of royal qualities, they are transformed into ‘extraordinary kin’ understood to be related to their electors by ties of ancestry, blood and shared divine origins. They gain status by appeals to stories of past glory and the ideal of a strong and fearless patron-protector who can deliver to ‘his people’, by whatever means. Divine ancestors may include kings, queens and revolutionaries, deified ‘gangsters’, Robin Hoods and other communal patrons. Today the archetype of a semi-divine, brave patron-protector is in wide circulation among politicians, their followers and electors who invoke this image to 1

I have developed my ideas about the relation between caste, Indian ‘patronagedemocracy’ and ‘divine kinship’ over the course of several projects funded by the British Economic Social Research Council (ESRC) (2004–2008, ‘The Vernacularisation of Democracy’: Comparisons Across India and Venezuela), the Nuffield Foundation (2009, ‘Elections, Society and Culture: Understanding Voting Behaviour in the 2009 Indian General Elections’) and more recently by an European Research Council—Starting Investigator Grant (2012–2016, ‘Anthropology of Muscular Politics in South Asia’).

284  Lucia Michelutti

contest elections, make political choices and indeed refashion old conceptions of justice and political community. I have proposed elsewhere that ‘divine kinship’—the relation between humans and gods—is a crucial vernacular idiom through which we can examine electoral representation and sovereignty across cultures (Michelutti 2013; Michelutti and Forbess 2013). In the anthropology of South Asia the relation between humans and deities has long been the subject of intense interest.2 Yet the place of such relations in politics—where they have been strikingly prominent— remains conspicuously under-analysed.3 In this chapter I suggest that the lens of divine kinship can help us find important clues to understanding links between politicians, parties and citizens, as well as the violent manifestations of, and fears unleashed by, the politics of patronage in north India today. Appeals to divine kinship in the constructions of charisma of politicians like Mulayam Singh Yadav or Mayawati inject politics with energy, hopes, dreams and fears whose potency cannot be appreciated without grasping the indigenous idioms on which they draw. Importantly, divine kinship also infuses ‘the people’ themselves (in this case Yadavs and Dalits) with power and charisma (cf. Michelutti 2013). The consecration of kin has also produced new folk theories of social justice based on what local communities think they are entitled to and others are not (Michelutti and Heath 2013).4 Scholars have already noted that in South Asia increasingly fierce electoral competition has expanded political clientelism, creating demand for extra-legal economic and muscle resources, and generating greater rates of criminality among politicians (e.g., Wilkinson 2007). Democratic practices—voting, electoral campaigning, the organisation of political meetings, and so on—are increasingly seen  as direct or  indirect ways to extract state resources and obtain or maximise 2

Dumont (1980), Babb (1975), Fuller (1979; 1988), Bouglé (1992) and Gellner (1992). I discussed the lack of scholarly engagement with divine kinship in a paper co-written with Edward Simpson and entitled ‘Gods and Men: A Fuller View of Society and Religion in India’ at the workshop in honour of Professor Chris Fuller, ‘Identities, Inequalities and Politics in Modern India’, London School of Economics, 22 September 2009. 4 Increasingly, castes claim a share of state resources proportionate not to their status, as they did in the traditional jajmani system, but to the size of their population. Voting is not just about getting what one can for one’s own people, but also about trying to prevent other communities from taking away what one already has (Michelutti and Heath 2013). 3

Kingship without kings in northern India 285

authority within India’s booming ‘patronage democracy’ (Chandra 2004, 133).5 Work on political patronage has nevertheless paid little attention to communities’ antagonisms and the political styles which arise from ethno-populist governance in India and the region at large. Although since the 1990s politics in UP has substantially hinged on the struggle between Yadavs and Dalits who compete for political, economic and symbolic privileges, the sources of this competition remain little understood. Drawing on ethnographic insights gained over 15 years of research in the town of Mathura in UP, this chapter focuses on the struggles and shifting strategies of two major political groupings—the Yadavs and Dalits—and the two political parties with which each is respectively associated, the Samajwadi Party (SP) and the Bahujan Samaj Party (BSP).6 Over the last three decades, both Yadavs and Dalits have benefited a great deal from affirmative action, rising up to be the key protagonists in India’s ‘second democratic upsurge’ (Yadav 1996) and in the rise of its ‘patronage democracy’ (Chandra 2004). This upsurge, which dates back to the 1990s, brought leaders from some historically lower and ‘backward’ castes to the fore of politics, and has become associated with electoral mobilisation of poorer, less educated and otherwise lower-standing communities. Nowhere has this phenomenon been more prominent than in UP, where politics is increasingly perceived as an arena of ‘plebeian’ self-assertion and a profitable ‘caste business’ of the newly risen elites (Michelutti 2008, Introduction; also Jaffrelot and Kumar 2009). In this chapter I show both how Yadav and Dalit politicians are legitimised as divine kin, and how castes are fashioned and refashioned vis-à-vis patron-deities, something that gives rise to the ‘communal sovereignty’ of castes, clans and families in place of the ‘popular sovereignty’ possessed by ‘the people’ (bahujan) at large. I argue that in this environment voters style their political leaders


Chandra defined ‘patronage democracy’ as a democracy, in which the state monopolizes access to jobs and services, and in which elected officials have discretion in the implementation of laws allocating the jobs and services at the disposal of the state. The key aspect of a patronage-democracy is not simply the size of the state but the power of elected officials to distribute the vast resources controlled by the state to voters on an individualized basis, by exercising their discretion in the implementation of state policy (2004, 6, emphasis in original). 6 Yadavs are a low to mid-ranking caste of pastoral agriculturalists and Dalits are former untouchables, in this case mainly from the Chamar leathersmith caste.

286  Lucia Michelutti

as neither providers nor representatives, but as fearless, and often necessarily violent, protectors.

Patronage democracy, representation and divine kinship In this volume, David Gilmartin shows that at the heart of relations between patrons and clients lies a puzzling contradiction: On one level, patrons’ relations with their followers are legitimised by reciprocity, and the equality which it implies, suggesting that patrons and clients occupy a shared moral world. On another level, inequality lies at patronage’s very heart: the patron’s authority depends on access to status and power that transcends that of his followers (Chapter 5, 125).

A similar contradiction also marks relations between elected politicians and the people they represent. The legitimacy of democratic governance derives from the ability of governments to act ‘in the name of the people’ (Laclau 2005). Democracy, however, comes with a doubtful promise ‘that rule ought to be by ordinary rather than extraordinary people’ (Dunn 1992, v). As Jonathan Spencer put it, democracy aspires to make ‘all men and women the subjects, rather than the object of power’ (2007, 137). At the same time, democratically elected leaders are extraordinary—literally chosen—people who nevertheless share their special standing with the ‘common people’ they represent. How this political vision and the ties which bind people and governments are formulated varies across the world. Spencer writes: We all know there is a link between representative and represented, but we cannot specify what form that link may take. It may be a link of common substance: fathers and kings may like to think of themselves as embodying those they are said to represent. Or it may be a contractual link, in which the representative is only temporarily mandated to put forward the views of those she represents, while those represented retain the right of recall at the first sign of their views being misrepresented… there is a huge scope for different ways in which to construe the idea of the ‘people’ as well as the idea of ‘representation’ which supposedly binds them to the government (1997, 12).

Kingship without kings in northern India 287

Whatever sort of relationship ‘representation’ may be, how the ordinary articulates with the extraordinary in electoral governance is never straightforward (see Introduction, 29ff). Democracy may put an end to traditional kingship, but the enchanted imaginative tropes of divine kinship and ideas about leadership which underpin the power of kings persist in the mundane political realities today. The crux of kingship is the separation of one human being, the king, from the rest. This leitmotif recurs in the rituals of kingship which extract one person from the economic, political and kin relations which bind everyone else, refashioning kings into extraordinary persons who stand apart from society (Quigley 2005, 4). In this sense kingship and kinship are fundamentally opposed: kingship separates and kinship unites (de Heusch 1997). In democracy, links of kinship are not ritually severed. On the contrary, links of common substance are often created when the contractual link which underpins our idea of electoral representation is vernacularised through local socio-cultural and religious formulations, the idea of representation generates powerful theories of popular sovereignty, leadership and charisma (Michelutti 2008). Understanding how representation is rendered into the vernacular is crucial for grasping how democracy works in South Asia (see Introduction) and why patronage and cultures of violence and criminality may be so central to it.

Elected representatives, beneficent donors or violent protectors? Indian political leaders have often been described as having the duty ‘to care for the material interests of [their] followers’ (e.g., Brass 1990, 96). Pamela Price has described this idea as heir to the persistent, historically embedded models of lordship, pointing out the longassumed duty of the king to provide (1989, 47). The raja’s ‘beauty and dazzle’, she wrote, ‘symbolised the potentialities of wealth for the community as a whole’.7 The raja was also a servant of dharma (virtuous duty) and ‘his major two obligations were to protect his realm from outside aggression and to ensure that his people lived together peacefully and in a way conducive to the rita [truth/order]/dharma complex’ (Menski 2004, 14). 7

On ‘bandit-kings’ and lordship styles of rule see also Shulman (1986), Dirks (1987) and Peabody (1991).

288  Lucia Michelutti

In contemporary north India a ‘tough reputation’ and the proven capacity to exercise violence bring together the two long-valued potencies of a politician—to protect and provide. Politicians often choose to project an image of ‘men of action’: men who can ‘get things done for their people’ (Witsoe 2011; also Berenschot, Björkman, Piliavsky and Price in this volume). The reputation of a gangster—goonda, dada or mastan—often attends the projection of a leader’s effectiveness. The qualities of effectiveness, potency and courage also signify the wealth, status and political empowerment of the caste which politicians represent (Michelutti 2010). Thomas Hansen explains: ‘It is the performance of a certain style of public authority—generous but also with a capacity for ruthless violence— that determines who can defend and represent “the community”, defend neighbourhoods, punish and discipline’ (2005 136). Politicians are known to be corrupt and to resort to violence, but if seen as ‘our men’ loyal to ‘our community’, they may often command wide support in majoritarian electoral politics. Elected politicians thus become guarantors of provision and protection who often share imagined kin relations with the people they represent. Indeed, for castes, clans and kin groups protection is an identityshaping value. People see ‘kin’ who protect them as the best guarantee in a world where authorities and the legal system are treated with growing distrust. In this I am in agreement with Hansen (2001), Hansen and Stepputat (2005;  2006), Price and Ruud (2010, Introduction) and Ruud (Chapter 13 in this volume), who show that goondaism or dadaism is widespread in South Asia. But the following question remains: is this something new and, if so, when and why did this new political style come about? Also how is this trend related to the processes of ‘ethnicisation’ of caste and the production of new conceptions of popular sovereignty? Since the 1970s, anthropologists have observed shifts in the structures and practices of political authority and indeed in the fabric of Indian society as a whole. After India’s independence, the  Congress Party appealed to ‘traditional’ political systems like jajmani patronage and caste-based village factions to mobilise its electorate and transform relations between village dominant castes and their clienteles into relations between Congress politicians and their vote banks (Brass 1965; Rudolph and Rudolph 1967, 36–64). Soon after independence, however, the cumulative effect of rising electoral competition, spreading education, affirmative action and land reforms prompted great transformations in the local structures of dominance

Kingship without kings in northern India 289

and authority, opening political spaces for formerly marginalised groups. Marguerite Robinson, who conducted field research in the villages and district centres of Andhra Pradesh for 25 years, showed that by the early 1970s people substantially stopped voting for traditional elites (1988). She concluded that this change reflected the broader delegitimation of caste hierarchies and the political rise of the landless (ibid., 248). Jan Breman similarly argued, on the basis of extensive research in Gujarat, that in post-independence India inequality lost its social legitimacy, precipitating a shift from ‘vertical’ (inter-caste) to increasingly ‘horizontal’ (intra-caste) patterns of political mobilisation (1996, 262). Policies that reserved places in institutions of higher education, civil service and politics for historically downtrodden communities (Scheduled Castes, Scheduled Tribes and Other Backward Classes) further reshaped how people related to old-time patrons and prompted the rise of ‘new’ patronly classes. Others yet argued that the levelling of hierarchy changed the very idea of justice and the way people came to perceive it in relation to communal differences. André Béteille argued that in India the claim that the state should distribute the benefits of employment and education equitably to different castes and communities, according to their particular circumstances and requirements, was part of the age-old Indian social order of distinct and interdependent status groups (1991, 541).8 In the jajmani system of village patronage it was acknowledged, for instance, that every caste had a claim to its own share of the village’s total social product, which was not equal to the share received by other castes but appropriate to its relative standing. In democratic India this has changed. Nowadays, castes claim an equal share, proportionate not to their status but to the 8

The notion of community-based entitlements was heavily reinforced in colonial times. In many ways the British Raj set up the conditions under which different groups of Indians could participate in the political process by politicising caste and community identities through enumeration practices, the institution of separate electorates for Hindus and Muslims, and reservation policies for marginalised sectors of the society (e.g., Appadurai 1993; Cohn 1996; Bayly 1999, 97–144). The attachment of benefits to these identities changed the ways in which people thought about themselves. From the colonial period many castes began to transform themselves into increasingly larger communities (samaajs) in which internal hierarchy was downplayed, endogamous marriage rules abandoned and the idea of united and numerous ‘imagined communities’ created. Post-colonial India inherited this legacy, and the Indian Constitution explicitly recognises communal entitlements through affirmative action policies.

290  Lucia Michelutti

size of their  population. As Kanchan Chandra (2004) has argued, Indian citizens make electoral choices on the basis of ‘ethnic head counts’: they vote for parties that contain most members of their own community instead of comparing their policy platforms. Because parties also count their voters’ ‘heads’, it follows that the main demand of caste-based political parties like the SP and the BSP is distributive ‘social justice’ related to the narrowly defined ‘social upliftment’ of their voter communities (jobs, access to welfare programmes and a sense of dignity and empowerment). Hansen captures the ways in which old forms of patronage— the donor-servant ideal—and the values attached to it had been adapted to and transformed by democracy, and vice versa. He argues that by the 1980s ‘India’s democratic revolution gradually undermined the idiom and practices of ma-baap’ism [paternalism/ patronage]’ and that ‘In Bombay of the 1980s, the so-called dada culture, which enjoys a long history in Bombay and other Indian cities as a popular model of authority and power, acquired new prominence as a style of public and political conduct, as dadaism’ (2001, 72). Analysts tell us that now the ‘common man’ no longer simply appeals to the idioms of traditional patron–client relations, but engages in what he thinks of as a separate domain of ‘politics’ (Khilnani 1997; Chatterjee 2004). The ‘silent’ democratic revolution ( Jaffrelot 2003) has not been very silent at all, and the noises it has emitted have not all been benign. Political democratisation has been paralleled by a ‘democratisation of violence’, which is now an ordinary feature of South Asia’s political landscape (e.g., Ruud in this volume). Politically motivated violence is no longer perpetuated on behalf of powerful local landlords. In UP, as elsewhere in northern India, upper castes have lost the monopoly on violence which they once enjoyed. Violence is now the resort of a great number of people from a wide variety of castes. Anyone with some money can nowadays buy protection from autonomous ‘violent entrepreneurs’ (Blok 1974; Gambetta 1993; Varese 2001; Volkov 2002). Since the 1990s, an increasing number of these entrepreneurs have entered politics, becoming the new ‘rajas’ and creating what locals call ‘mafia political dynasties’.9 9

The ideal of kingly protection here is combined with forms of protections that are at the heart of mafia organisations (Gambetta 1993). I am currently conducting fieldwork on the refashioning of kingly ideas of protection vis-à-vis the emergence of mafia-like protection styles in UP.

Kingship without kings in northern India 291

By 2002, almost 50% of political candidates in UP had criminal charges registered against them. ‘Criminal’ candidates won 206 out of 403 seats, an absolute majority of 51.1%, earning for the SP, which governed the state between 2002 and 2007, the title of ‘goonda raj’ (gangster rule).10 The victory of the BSP in the 2007 UP State Assembly election ended the rule of SP. But counter to what many had hoped for, this election was no less criminalised. Indeed, in 2007, the number of ‘tainted’ candidates from both parties increased, with BSP fielding the highest proportion (34%) of criminal candidates. Just as in 2002, five years later, newspapers and television channels ran reports  on candidates who conducted election campaigns from prisons and broadcast speeches through illegally acquired mobile phones connected to speakers. In 2012 the SP returned to power in UP. Despite a vocal anticriminalisation campaign launched by the Election Commission and several civil society groups (including the Anna Hazare movement), a total of 189 legislators, or 47% of the newly elected State Assembly had criminal charges pending against them (according to the affidavits they were now required to submit by law). The Chief Election Commissioner said to me that, despite anti-criminalisation campaigns, criminality among politicians is on the rise. The following sections explore political criminality and the standoff between Yadavs and Dalits, as well as the folk theories of sovereignty and social justice which emerge from it.

Divine genealogies, caste and the collective nature of patronage The shift from vertical to horizontal structures of political mobilisation and from the ideal of beneficent donor to the ideal of violent protector reflects the broader transformation of hierarchical ‘castes’ to independent and competitive ‘ethnic groups’ through a processes of caste ‘substantialisation’ (Dumont 1980, 226) or ‘ethnicisation’ (Barnett 1975, 158–159). The ‘caste system’ has been levelled into a set of ‘horizontal’, disconnected and increasingly 10

I use the term ‘goonda’ (gangster or muscleman), and the associated concepts of ‘goonda politics’ and ‘goondaism’, in a loose way describe individuals, activities and political styles that rely on the use (or threat) of force to protect personal and communities’ material and symbolic interests. The term ‘goonda politics’ is also widely used to denote corruption, bribery and, in general, all illegal activities linked to patronage.

292  Lucia Michelutti

egalitarian groups with their own distinct culture and way of life, a shift that has been marked in the vernaculars by the increasing substitution of the word jati (caste) with the word samaaj (community or society) (Mayer 1996, 59). I have shown elsewhere in some detail that, while ritual hierarchy has been substantially weakened as a principle of social stratification, other aspects of ritual life—particularly divine kinship—have facilitated a dynamic adaptation of old forms of social and political life to modernity (Michelutti 2008). I argued that to make sense of these changes we must attend to the logics of descent and kinship (cf. Kolenda 1978) and to the ways that traditional religious ideology of hierarchy (Dumont 1980) has been usurped by an ‘egalitarian’ and ‘democratic’ ideology of descent. In the construction of ‘horizontal groups’ or ‘alliances’, divine kinship plays a crucial role. Popular Hinduism makes no categorical distinction between divine and human beings (Fuller 2004, 3). Virtually all Hindu ‘gods are ancestors and ancestors can become gods’ (Michelutti 2008). The deification of human heroes relies on a rich repertoire of stories (e.g., Blackburn 1985; Coccari 1989), which connect their protagonists to lineages of various local rulers (Raheja 1988; Quigley 1993). In the early 1990s, the rise of Hindu nationalism was attended by a mass politicisation of popular religion and the nationalisation of deities, particularly Ram, the hero of the great Indian epic Ramayana. Regional parties have local gods and the Yadavs have their own. A joke that circulated in Mathura during the 1999 parliamentary elections went like this: A man trying to persuade the god Krishna to stand for elections tells him that he is the most suitable candidate. —After all, he says, you are a Yadav. —I am a god and do not belong to any particular caste, says Krishna. —Sir, even a god could not succeed in the present political arena. At least try to hold on to your caste!

For the Yadavs, the ideal ruler is not Ram but Lord Krishna, the most mischievous god of the Hindu pantheon. No more than a 150 years ago there were no Yadavs, but various herder communities, which

Kingship without kings in northern India 293

over time united into a single caste that traces descent from the eponymous Yadu dynasty and its most illustrious member, the cowherd, prince and warrior Krishna. Over the years, Yadavs adopted higher forms of Hinduism, substantially standardising their caste pantheon and shifting (over the past three generations) from the cult of deities who demand blood sacrifice to the veneration of vegetarian gods. In this process of ‘sanskritisation’ (Srinivas 1952) Krishna displaced lesser sub-caste and clan deities (kul devatas) as the main ancestral god. The Yadavs have been so forceful in appropriating Krishna into their pedigree, that Krishna is by now generally accepted in UP as the original Yadav. The Yadav Krishna possesses special martial and political qualities: he is a ‘muscular’, socialist and democratic patron-god. He is ‘the first fighter for social justice’, befitting a ‘caste of politicians’ and ‘natural leaders of the oppressed’, which the Yadavs in Sadar Bazaar (where politics is indeed many Yadavs’ main business) claim to be (Michelutti 2004). The Yadavs share their ‘political qualities’ with their patron-god: they say they have ‘politics’ and ‘social democracy’ in their blood, which entitles them naturally to govern. Many in Mathura say that the Yadav leaders, in the name of Krishna, ‘the prophet of social justice’, represent the Yadavs as the natural leaders of the OBCs. Krishna was the ruler in Dwapara Yuga, and the Yadavs, as his descendants, were supposed to lead the society in the Kali Yuga (Degenerate Age) and fight the injustice and the exploitation of the backward classes by the forward classes (Rao 1979, 214).

Divine kinship and being extraordinary, or deifying ‘the people’ The myth of Krishna is not only a narrative of Yadav origins, it also undermines hierarchy and differences within the community, offering a powerful idiom for articulating notions of protection and dharma. ‘Krishna/Yadav’ substance is not passed from the patron-deity to the devotees through gifts (as it is in Rajasthan, as described by Piliavsky in Chapter 6). At the level of ideology, there is asymmetry between ‘Krishna the ancestor-god’ and the rest of Yadavs. Marshall Sahlins described kinship as the ‘mutuality of being’, which brings ancestors, gods, spirits and kin in to relate to and become part of persons (2011a; 2011b). Sahlins shows that the

294  Lucia Michelutti

Maori chief has more fellowship, more mana (power, influence, magic) and more occasions for forming kinship bonds than others because of greater connections to ancestors (ibid., 229). In a similar way, Yadav kin are all encompassed in Krishna and their political leaders, who provide them with substance and protection. Let me explain. Contemporary Yadav politicians attribute to themselves, and are attributed with, particular qualities suited for statecraft vis-àvis links to Krishna and deified caste heroes. This phenomenon is neither new nor particular to Yadavs and it has been described by Price (1989) as a widespread aspect of the contemporary adaptation of Hindu models of divine kingship. In the case of the Yadavs, however, this phenomenon is reinforced by continuous references to the substance or character inherited from ancestors and gods as the source of political skills and knowledge. This conceptualisation of how knowledge and skills are acquired is part of the ambient ideology of caste, in which members of each community are believed to have a special aptitude for their caste occupation, a propensity transmitted ‘in the blood’ (e.g., Parry 1979, 85). In their rhetoric, Yadavs present relatedness as being rooted in descent. ‘Essences’ and ‘qualities’ are hence inherited from previous generations. And the very fact of descent from Krishna makes Yadavs the privileged vessels of ‘democratic’ knowledge. In Sadar Bazaar people generally believe that skills and aptitudes are passed down together with physical traits ‘in the blood’, and informants often described their predisposition to politics as innate (Michelutti 2004). They said ‘they learned it in the womb’ (pet se sikhte hai), having been born to be politicians. They also spoke of the ‘womb’ when speaking about apprenticeship, especially about the business of herding cows, the skills involved in ethno-veterinary practices and the special sign language used by brokers at cattle fairs. How did people learn all these skills? The answer was, almost invariably: in the womb. Conceptions of relatedness and knowledge transmission are reinforced by local festivals which reconstruct links of substance between Krishna and Yadavs (ibid.). The myth of Krishna enacted at these events says that thousands of years ago the throne of Mathura was usurped by Krishna’s tyrannical uncle, Kamsa, whom Krishna killed (once Krishna grew up) to liberate his people from oppressive and illegitimate rule. The Festival of Kamsa’s Destruction (kamsa

Kingship without kings in northern India 295

vadh ka mela) in Mathura was first organised by the local Yadav caste association 30 years ago to celebrate democracy’s—and the Yadavs’—triumph over monarchic despotism. During the festival, Yadavs dress up two young boys as Krishna and Balram (Krishna’s elder brother) and carry them in a procession through the streets of the town. The two boys are not merely ‘acting’ but in fact become the gods, and are worshipped as such during the procession. I was told that only the sons of Yadavs, as descendants of Krishna, could legitimately play the roles of Krishna and Balram. The same is true of celebrations of the Krishna Lilas (Krishna’s Plays), which accompany the Kamsa Festival, and in which the actor who plays Krishna’s part must be a Yadav. As I already noted, in Hindu cosmology any human person can be deified and turned into a hero. But in northern India members of some communities, like the Yadav, are thought to have ‘heroic substance’ (Coccari 1989, 260) and be particularly susceptible to becoming hero-gods. The Yadav hero-gods (kul devatas) are often cowherds deified for defending their community, and its herds, from higher castes or Muslims. Contemporary Yadav politicians and local leaders like to advertise this heroic heritage. One informant in Mathura said: ‘“social justice” is Krishna’s message, and it is also the message of the Samajwadi Party … Lord Krishna had always helped the poor and needy people. There is a similarity between the ideas of Mulayam Singh and the god Krishna’ (Arun, 35 years old, SP activist). Mulayam Singh Yadav and other Yadav leaders celebrate Yadav heroes who fought against the ‘royal heritage’ of the British (Michelutti 2008, Chapter 6). They draw on their ‘pastoral roots’, presenting themselves as rustic cowherd-politicians while at the same time claiming aristocratic (Kshatriya) descent from the prince Krishna. The former Chief Minister of Bihar, Laloo Prasad Yadav, often gives interviews to the national press while tending his cows and buffaloes, and describes himself as an avatar of the cowherd-prince Krishna. Mulayam Singh Yadav’s biographical anecdotes, which occasionally appear in the press and the Yadav caste literature, portray his childhood as resembling Krishna’s: he grew up amongst cowherds and was a mischievous child who loved sports, and especially wrestling. Most importantly, Yadav politicians present themselves as uniquely capable of getting things done for ‘their people’ and for all those

296  Lucia Michelutti

who had been socially side-lined and politically oppressed. The Krishna they invoke in political speeches is his epic, ‘martial’ avatar who ‘defends the poor and the marginalised’. Yadav politicians thus continue a long tradition of protector-patrons: Mulayam Singh and Laloo Prasad are Robin Hoods who steal from the rich and give to the poor. In the eyes of their supporters, they are heroes who fight for their community, defending it from its enemies. The following is an excerpt from a speech given by Mulayam Singh Yadav (current President of the SP, former Chief Minister of UP and former Union Defence Minister) to the Yadav Caste Association Meeting in Delhi in 1999: I believe that whenever we refer to Krishna we cannot avoid talking about politics. Lord Krishna fought for the underprivileged groups. Today we (descendants of Krishna) fight for social justice and for the social wellbeing of India’s backward communities.

Mulayam Singh Yadav made his career during India’s ‘second democratic upsurge’ in the 1990s (Yadav 2000). He began as a wrestler and eventually established himself as UP’s chief Yadav leader. Links of substance connecting Mulayam Singh and his caste supporters have long been consolidated. In 1993 and 2002, when he was elected as Chief Minister, the Yadavs in Sadar Bazaar shouted: ‘I am Mulayam!’ More recently, they celebrated his son’s election as UP’s new Chief Minister, shouting: ‘I am Akhilesh!’ and ‘I am Mulayam!’ Whereas for Yadavs—who see themselves as a natural community with natural leaders who naturally lead the oppressed—the past is a positive inspiration, Dalits see their past as something they need to contest, redress and overcome. Regaining pride and overcoming humiliation is one of the main themes in the political rhetoric of the Dalits. The central political claim is: because Dalits have for centuries been oppressed, it is now their turn to put their political skills and heroic, democratic ethos into practice. Nevertheless, just like the Yadavs, Dalits invent their own hero-gods, ties to whom form the basis of communal formation and give substance to the ‘socialist muscular charisma’ of their leader, Mayawati. Badri Narayan (2005) has shown, for instance, that Dalit womenheroes (viranganas) who participated in the 1857 Rebellion are

Kingship without kings in northern India 297

now invoked as symbols of Dalit assertion and used to construct Mayawati’s charisma. ‘Mayawati’, he writes, was depicted as a brave and wise leader who is also highly patriotic, having devoted her life to the service of the nation … during public meetings with bureaucrats and government officials, [Mayawati] always projects her leadership qualities by highlighting her masculine characteristics of aggression and arrogance. She tries hard to subvert the upper-caste notions of femininity and lady-like behaviour, which are associated with meekness, mildness, subservience, docility and hierarchical form of address (ibid., 158–159).

Mayawati’s self-fashioning project culminated in her presenting herself as a living Goddess (ibid., 162). She was widely criticised for spending millions on statues, parks and other displays of munificence, and her government was condemned for countering socio-economic inequalities symbolically and not through policy change.11 Mayawati and Mulayam both built their images of courage, populism, virile strength (in this case not an exclusively male virtue) and active prowess on a wide repertoire of cultural and religious ideas which create bonds between them and their followers. Descent from a shared line of divine ancestry—or divine kinship—has been the central trope. People see their charisma as a quality inherited from hero-gods, sanctified by anti-colonialist (and anti-imperialist) heromartyrs and enhanced by their feats as leaders of the underclasses’ democratic upsurge. Political cosmologies of Yadavs and Dalits accommodate divine and human beings, leaders and their supporters. The human-divine nexus forms the structure of intimacy with political authority and with the people in power. In Mathura, ‘mini-Mulayams’ make their reputations by assuming Mulayam’s muscular image and by cultivating friendship with him or his circle of acolytes, or at least by boasting of such friendships (Michelutti 2013). These infrapolitical connections, with their appeals to divinity, kin and heroic 11

However, as Nicolas Jaoul (2006, 175) has argued, the statues of Ambedkar are one of the main features of the Dalit movement and the focus of renewed democratic aspirations. Ceremonies often organised around the statues have provided deprived citizens with opportunities for building social and political support in the state.

298  Lucia Michelutti

and muscular knowledge have a ‘magicality’ (Hansen and Verakaaik 2009) on which the charisma of many local Yadav leaders rests. This charisma—spun substantially out of bonds of divine kinship—is a powerful mechanism of political mobilisation.

Fears, violence and strategic caste-based voting In UP, no less than in other parts of South Asia, voters are keenly interested in the material benefits that politicians can provide. But the polarised political climate in the state riven by rivalries between Yadavs and Dalits makes voters more interested in having ‘their man’ (hamara aadmi, referring to both women and men) as a leader who can provide in the long term than in immediately gratifying gifts like alcohol, food, money or clothes (as in Piliavsky’s Chapter 6). What they look for, most urgently, is security and protection, which is as much a symbol of empowerment as a practical defence from real and imminent threats. In 1998–2000, when I spent 22 months in the town of Mathura, antagonism between Dalits and Yadavs was already intense (Michelutti 2008, Chapter 5). When I returned in 2009 to conduct a study of the parliamentary election, the conflict between them was much fiercer than before. The first thing I was shown on arrival in Mathura, were paved roads in the neighbourhood’s Dalit section. The rest of the neighbourhood remained untouched by progress. This show of grievance was meant as an example of the new pro-Dalit BSP government’s discrimination of Yadavs, and it drew my attention to one key development. My informant said: ‘The only new construction in the neighbourhood is the thana [police station]. Do you remember? It used to be a small thana; now it is a palace!’ Not only was the building new. Even more conspicuously, so were the people working with in it. In 1999, 90% of the staff of the station were Yadavs; 10 years later, the station was dominated by Dalits (mostly Jatavs and Chamars by caste). Yadavs and others (particularly the Bania merchants) complained that the officers refused to register their reports. This state of affairs encouraged extra-legal forms of justice, making the residents of Sadar Bazaar more ready than ever to take dispute settlement into their own hands. Political interference in policing and bureaucratic administration is widespread in northern India (see Jauregui, Chapter 10 in this volume).

Kingship without kings in northern India 299

Politicians have considerable power over bureaucrats, sometimes even over the highest-ranking of civil servants. They can transfer a civil servant from one post to another, and promote or demote them, almost without challenge (Krishnan and Somanathan 2005, 292–299). This was a regular feature of Mulayam Singh’s SP government in 2002–2007 no less than it was of Mayawati’s in 2007–2012. The police are seen as henchmen of the ruling party and caste, an idea that further exacerbates tensions between rival communities. Growing antagonism between Yadavs and Dalits is borne out by the surveys I conducted in Sadar Bazaar in 1999 and again in 2009.12 In 1999 I asked a series of questions about relations between different communities in the neighbourhood. People from different castes expressed broadly similar opinions, and Dalits held a slightly more positive view of social relations in the neighbourhood. But by 2009 the mood had changed and opinions split. Upper castes were now more positively disposed towards neighbours from other religions and castes, and they consistently showed high levels of trust in communities other than their own. But the Yadavs, Dalits and Muslims were much more despondent. Both Yadavs and Dalits now regarded each other with suspicion, which both reflected and further intensified caste-based voting. Most people no longer vote for parties in order to gain selective benefits (as they did 10 years ago). They now vote against parties to avoid losing benefits. During the general election campaign in 2009, the BSP candidate was sent away from Mathura—‘he was literally kicked out’—said P.P. Yadav (a 55-year-old shop keeper), when he arrived to campaign. In Mathura, the Yadavs who supported the SP and went to campaign for the party in the nearby district voted for the Rashtriya Janata Lok Dal (RJL)13 candidate Jayant Singh, grandson of India’s former Prime Minister Charan Singh and son of Ajit Singh, the old enemy of the SP and Mulayam Singh Yadav. The Yadavs, who were deeply unnerved by the prospect of Mayawati’s ascent to power, voted strategically and en masse for Jayant Singh to prevent the BSP


The survey I conducted with Oliver Heath in 2009 was part of a project entitled ‘Elections, society and culture: understanding voting behaviour in the 2009 Indian general elections’, funded by the Nuffield Foundation. 13 The RJL was then allied with one of India’s two main political parties, the Bharatiya Janata Party (BJP).

300  Lucia Michelutti

candidate from winning.14 And they spoke openly about preventing Dalits, ‘who could not be trusted’, from running the state. They joined the Jats, Banias and other upper castes in voting for the RJL. ‘We are doing “social engineering” in reverse’, they said, mocking Mayawati’s slogan, which saw Brahmins and Dalits coming together to ensure majority support for the BSP in the 2007 state election. The popular BSP campaign slogan ‘haathi nahin, ganesh hain, brahma, vishnu mahesh hain’—‘The elephant [BSP logo] is not just an elephant; it is Lord Ganesh and the Brahma-Vishnu-Shiva Trinity all in one’—invited higher castes to ‘come ride the elephant’. Across UP more generally, the Yadavs were much less likely to support the SP if there was no strong party candidate and if the BSP threatened to win. In the politically polarised ecology of UP, where ‘the enemy of my enemy is my friend’, people increasingly vote not for their preferences, but against the worst of all evils (Michelutti and Heath 2013). Political exchanges become less encouraged by reciprocity and more by fear of losing protection and benefits.15 Strategic voting practices suggest that instrumental clientelistic exchanges between political leaders and voters (the purchase of votes with alcohol, food or saris) have a marginal impact on electoral outcomes when compared to fears of being ‘patronised’ by lower castes or losing control of the patrimonial-democratic state. If 15 years ago local communities were busily acquiring entitlements, they now increasingly focus on preventing other communities from taking these away.16


In 2009, 60% of Yadavs in Sadar Bazaar said that they voted for RJL, a figure substantially higher than in any other community. 15 Linking the political ethnography of Mathura with the 1999 and 2009 Mathura election studies, and data collected by the Centre for the Study of Developing Societies in its National Election Study in 2009, Oliver Heath and I noted that the insights from Mathura were consistent with patterns observed across the state (Michelutti and Heath 2013). 16 The growing antagonism between Yadavs and Dalits is also apparent in increased tensions between Yadavs and Muslims in Mathura (Michelutti and Heath 2014). In the past year, caste or community-based patronage politics of the SP has further reinforcing both communal identities and differences. The idea of patronage as an essentially vertical, dyadic relation obscures the fact that in South Asia political patronage often rests on ‘horizontal’ kinship ties between communities and their representatives.

Kingship without kings in northern India 301

Conclusion Democracy claims to deliver rule by ordinary people, but elected politicians are necessarily extraordinary. How exactly people reconcile this paradox of electoral politics differs across the world. Divine affinities, which voters claim across South Asia, help us follow the mental leap, which participants in democracies on the subcontinent make. In this political environment, Mulayam and Akhilesh are not just mascots or ‘symbols’ of the Yadav community. They substantially embody the people they represent. One may say that the voters are ‘internalised’, or in Dumont’s (1980) language ‘encompassed’, by politicians—they are substantively part of those they elect. The force of this idiom of political representation is easy to feel, and it is just as easy to see why this relational logic may be so compelling for those who vote. Mulayam Singh, Akhilesh, Mayawati and other leaders are not representatives elected to speak and act on behalf of their electorates (Michelutti 2004) by ‘a contract’, a much flimsier connector than the substantive bonds of divine kinship they form with their constituents. Affinities between political leaders and their voter-communities run much deeper than the very idea of elections assumes, and they reach far beyond the electoral context (see Introduction, 31–32). When I returned to Sadar Bazaar in 2012 and asked who the new leaders were, the unanimous answer was: ‘everybody thinks they are Akhilesh’ and ‘Yadavs are all chiefs’. Unlike Rajasthani villagers (Chapter  6), the Yadavs in Sadar Bazaar do not style themselves ‘ordinary people’, ‘poor people’ or ‘servants’ of overlords.17 Instead, the town is full of ‘mini-Akhileshes’, ‘mini-Mulayams’ and even ‘miniMayawatis’. The same is true of the Yadavs’ Dalit competitors. What both nowadays want is to protect their increasingly consolidated and airtight communities’ economic, political and symbolic attainments. When each party, and the caste-community it represents, find themselves out of power half of the time, both factions increasingly hope for protection that is increasingly available only when their own leaders are at the helm. In this environment, ‘democratic rajas’ must not only be generous benefactors (Björkman, Mattison Mines, Piliavsky and Price in this volume) but crucially also protectors who can look after their people, defending and punishing, as the need arises. This desire finds expression in the widespread idiom of 17

I shall emphasise that the community under study is relatively wealthy and politically well connected.

302  Lucia Michelutti

fearless virility, which increasingly surrounds political life and fills politicians’ rhetoric. Despite ubiquitous violence, voters do not simply succumb to politicians’ command. Politicians need to win their votes, using violence not only to threaten and subordinate, but also to display it as a guarantee of their capacity to rule. Increasingly, in UP voters do not only ask their representatives to help them get jobs, access health provisions, resolve disputes or arrange marriages, they wish to be assured that the leaders will put in place ‘safeguards’ which will guarantee that their promises materialise. More often than not, these ‘safeguards’ take the form of a ‘democratic’ version of danda: the threat of punishment, or the fear of being punished, which is also an old Indic term for ‘justice’. As Werner Mensky summarises, ‘Danda is about accountability at every level’ (2004, 12). Thus, during speeches he gave in Mathura in the course of the 2012 election campaign, Akhilesh Yadav routinely brandished a sword. Dark glasses and leather jackets are no longer sufficient symbols of a tough reputation, as was the case in the late 1990s. In the past 15 years, weapons have become increasingly visible and politicians have larger and larger entourages of heavily armed bodyguards and supporters, who not only display the might of their followings, but also protect from potential attacks. Perhaps most notably, violence has become part of a growing number of elected politicians’ Curriculum Vitae. The patron here morphs into the Mafioso—the bahubali, the local dabbang, or mafia don—displaying vividly the scope for violence in  India’s patronage democracy and some of the ways in which democracy has been vernacularised.

The political bully in Bangladesh


Arild Engelsen Ruud


he collusion of criminals and politicians in Bangladesh dates back at least to the years following the 1971 war of independence.1  In  the war-ravaged country armed gangs roamed amidst the political chaos and dire scarcity into which the country had sunk (Mascarenhas 1986,  22). Tens of thousands of men carrying some of the 150,000–200,000 arms, which had come from India to East Pakistan during the war, were on the loose, many forming militias which did not recognise the government’s authority. Lawlessness was rampant and party  activists were issued with arms ‘to defend themselves’ (Ali  2010, 96–97). At the time, the line between criminals and political activists was faint: street gangs were often indistinguishable from some of the private armies associated with the Awami League at the helm of the newly formed Bangladesh. By 1974 the situation deteriorated so much that the government asked the army to intervene. However, to the frustration of army officers who led the intervention, anyone affiliated with the ruling party was exempt from prosecution, leaving many brigands immune to the rule of law (Mascarenhas 1986, 43ff). Later military governments acted in similar ways. In the late 1970s the Ziaur Rahman administration established the so-called youth cooperatives throughout the country: ‘In practice the 93,990 members [of the cooperatives] spread throughout the land were essentially groups of young toughs and thugs kept on a leash for use in elections, demonstrations, referenda or simply to smash the opposition…’ (ibid., 129). Political rivalries which raged in the 1970s and 1980s encouraged what Atul Kohli (1990) called the ‘deinstitutionalisation’ of the state, expanding the arena for thugs, something that to varying degrees has been 1

I am greatly obliged to my research assistant, Md Abu Bakar Siddique, for his invaluable assistance, and to the senior party man who prefers to remain unnamed, for his assistance, hospitality and insight. I am also heavily indebted to Anastasia Piliavsky for her critical reading of drafts.

304  Arild Engelsen Ruud

true in much of South Asia.2 After democracy was restored in 1990, individual politicians belonging to the two main political parties, the Awami League and the Bangladesh Nationalist Party (BNP), have engaged increasingly, and often quite openly, in corruption, extortion and related violence, which occasionally claims lives. Today political bullies, or mastans, play a central role in the politics of Bangladesh, where they are an inescapable feature of street-level political life (Khan, Aminul and Imdadul 2008, 187).3 Almost every week, newspapers carry reports on political killings, the number of which escalates dramatically in times of political unrest. Like political fixers (see Björkman, Osella and Berenschot in this volume), mastans engage in political brokerage, although often through more ‘muscular’ means.4 Like the fixers, most mastans enjoy politicians’ protection (Banks 2006); while some have stronger political links than others, none can persist in their business without some form of political ‘cover’. All are effectively attached to one or another political party, or one of their wings. This chapter examines the role of political bullies in Bangladesh and probes the nature of their allegiances to political movements, 2

In parts of north India politicians do not only use gangsters as their henchmen, but are themselves increasingly gangsters, or at least act as gangsters, cultivating a certain ‘goonda [gangster] look’, which invokes criminality for political legitimation (Michelutti 2008, especially 181; also Chapter 12 in this volume). 3 Almost 90% of respondents to a survey conducted in Bangladesh in 2007 said that mastans played an important role in local party politics (Khan, Aminul, and Imdadul 2008, 188). 4 The literature on the South Asian state often describes the role of mastan (goonda in Hindi) as distinct from that of a broker or middleman (dalaal) who mediates between citizens and the state. In his book on Bangladesh, David Lewis refers to the kind of politics that mastans engage in as the ‘everyday state’ (2012, 104), which Barbara Harriss-White calls the ‘shadow state’ in the Indian context (2003). Whereas middlemen are said to ply their trade through connections, money and favours, mastans employ violence or its threat. ‘Dalals [brokers], middlemen’, writes Stanley Kochanek about Bangladesh, ‘get things done, manipulate, obtain… When necessary, they also provide for mastans to facilitate enforcement’ (2000, 549). But the difference between the political fixer and bully is overstated. Even if the mastan employs different means, his pursuits are not altogether dissimilar from the middleman’s. Ward Berenschot (2011) has argued that in Gujarat goondas essentially play a mediator’s role, filling the gap between slum dwellers and the Indian state (see also Berenschot in this volume). In Gujarat people depend on goondas, particularly in poorer neighbourhoods where they lack alternative means of approaching the state. The same is true in Bangladesh, according to a survey conducted in the slums of Dhaka. Nicola Banks suggests that here the gap between slum dwellers and politicians is bridged by mastans, who mobilise vote banks in exchange for improved services or other benefits (2006, 8).

The political bully in Bangladesh 305

organisations and leaders who employ and protect them. It suggests that although political bullies serve political masters and are protected by them, they are largely independent operators who start their careers in politics but in the end operate quite freely. They play their own game with its own rules, rewards and pitfalls, all the while maintaining links with party politics, which afford them protection and legitimacy. As they expand their business or seek refuge, they move from place to place; thus their theatre of operations requires patronage not by individuals or even groups, but by the umbrella organisations of political parties. And yet, despite their involvement in politics, political ideologies and political strife play a relatively minor role in their lives. The image of a political leader as a master-godfather who uses and protects gangsters is not correct. Mastans serve, and are in turn helped and protected by, a network of ‘contacts’ within, and sometimes without, political parties. Yet as they mature from lowly ‘party activists’ to full-blown mastans, they depend increasingly on their own income, motivation and entrepreneurial spirit to stay ‘in the game’. The arena of politically motivated crime has many points of contact with formal party politics, but the connections are neither mechanical nor systematic. ‘Crime’ is rather an independent ‘field’ of action, to borrow Craig Jeffrey’s reading of Pierre Bourdieu ( Jeffrey 2010, especially 19ff), to which the mastans refer as the ‘underworld’. Each field has its own rules and stakes, enabling those who are engaged in it to compete for economic, social and cultural capital. People make their careers in the criminal field and gather the tools necessary to engage in politics, and vice versa. Political connections are maintained to ensure a successful career in crime and, at least in Bangladesh, not substantially the other way around. This chapter is based on the biographies of three mastans—Ruby, Babul and Alamgir—whom I interviewed extensively in Dhaka in 2006. The accounts they give of themselves help to understand the nature of bullies’ attachments to their political patrons—the political movements, leaders, parties and organisations by which they are protected and employed.

In the beginning I first came across mastans in 2005–2006, while studying student politics in and around Dhaka University. All three of the self-avowed bullies described themselves as ‘party activists’. Politics was how I  met them in the first place: Ruby and Babul were active in the

306  Arild Engelsen Ruud

student branches of their parties (Chhatro Dol and Chhatro League, respectively), and Alamgir was a party activist in the workers’ wing of the BNP (although by the time we met he had largely withdrawn from politics).5 Ruby and Babul were in their late twenties and both sported sunglasses and fashionable shirts. Ruby was especially menacing, erratic and at times aggressive. Babul was more relaxed and he bragged about his exploits with a broad smile and conspicuous pride. Alamgir, a man in his fifties, was much more subdued and on his way out of crime and politics. In 2006, when I carried out the interviews, Khaleda Zia’s second government was still in power, or had just resigned in preparation for the national election.6 The election was derailed and a military-backed coup was followed almost immediately by a crackdown on politicians and their criminal followings. Both Ruby and Babul went ‘underground’ and Alamgir ‘retired’ to his village. Suddenly, none of them answered my calls. Meanwhile, student activists let me in on some of their activities, including ‘hall capturing’ and ‘seat politics’, which revealed the darker side of the country’s political life (Ruud 2010). ‘Hall capture’ was a routine practice at the university, where after each national election the winning party’s student wing evicted all prominent losing party activists from halls of residence. Politics in Bangladesh is effectively a two-party system and every national election until 2014 has led to a change of government, which in turn led to ‘a change of guard’ in the university student halls (Muhith 2007, 45–46). Over the years, hall capture has become increasingly extensive and violent: following the 2008 election, the student wing of the winning party swept through all student halls except ones which were occupied by women and ethnic minorities. The capture usually begins with an assembly of winning party activists who shout slogans, make noise and sometimes brandish weapons just outside the halls. Occasionally, there has been


Chhatro Dol is the student wing of the BNP and Chhatro League is a nominally independent student organisation that is effectively the student wing of the Awami League. 6 Political parties in Bangladesh have an agreement that three months before the election the incumbent government resigns and leaves the government to an interim caretaker government.

The political bully in Bangladesh 307

open violence; guns have been fired and handmade ‘bombs’ thrown, even inside the halls.7 Activists capture halls to control the ‘seats’, that is to sway the distribution of permissions to the much-coveted student residence (which is subsidised, conveniently located and exciting to live in). Those who capture a hall, cram it with students, sometimes two or more to a bed. In exchange, they demand political loyalty and as a result political rallies are often full of residents of student halls. Successful rallies add political stature to ambitious student leaders; many senior political leaders began their political careers in student halls. The halls also offer convenient lodging to students who ‘muscle’ people around instead of engaging in academic pursuits. Everyone in the halls knows who they are, which rooms they live in and when they come and go from the halls. Some are friendly and chat with other students, but a certain sinister aura overhangs their presence and their business is discussed only in whispers. Student leaders use these mastans in several ways: to keep order in their organisations, ensuring for instance that students who are reluctant to take part in the demonstrations (michils) ‘change their mind’ in time. Mastans also help to keep rivals, within and beyond their parties, at bay (Moniruzzaman 2009). The bullies in student halls do not start off as criminals. They come to the university just like all others, but once they join the rough culture of student politics, those with the right ambitions and capabilities take up the role of mastan.8 Initially, they thrive in processions and demonstrations, for which they are hired as recruiters or foot soldiers (Rashiduzzaman 1997, 257). But the pay is poor and many eventually shift to the much more lucrative work in construction business, which thrives across the country. Over the last few decades, Bangladesh has been swept up by an economic boom. The healthy growth of the Gross Domestic Product (currently at an annual 6%) has placed it among ‘the next 11’ economically upcoming states (in World Bank parlance). 7

These bombs are relatively small homemade explosive devices, which nonetheless can create much mayhem and indeed sometimes kill. According to a recent newspaper report, they are available for about 500 taka (just over 6 dollars) a piece. 8 Student politics is more prominent in Bangladesh than in most of South Asia, although across India colleges and universities also function as nurseries for politicians, especially of a rougher sort.

308  Arild Engelsen Ruud

Construction cranes pierce the skyline of Dhaka and in practically every neighbourhood new buildings are springing up and old ones are being extended. In this building boom mastans play several roles. First, they protect construction sites, workers and machinery. Security guards at these sites are usually poorly armed, and a night guard or two cannot stand up to a dozen armed thugs sent by a rival entrepreneur or a local leader dissatisfied with his cut of the profit. Businessmen need gangs of mastans like Alamgir to protect their premises. Extortion is another side of the protection business. In the 1990s local businessmen and small-scale constructors were often kidnapped for ransom, usually at a rival’s behest, a practice that somewhat abated later. You can also find mastans at ferry landings ( ghats), in residential neighbourhoods wherever there are small food shops, eateries or workshops, in the markets of rural towns, at bus depots and so on. In each of these places they run a racketeering business, ‘controlling’ and protecting an area or a business, whether a large construction site or a market stall, for a fee.

Party affinities Mastans are free to operate as they do because they enjoy political protection. Some have only an indirect link to politics, through someone else, and others are involved in political organisations more directly. Political bullies invariably belong to this or that political ‘camp’, and party affiliation looms large in their presentations of self. Many speak at length about their party’s politics, the wrongs of the rival party, the leaders they know, the demonstrations in which they have participated and the parties’ political slogans to which they appeal to justify what they do. Bangladesh’s two main political parties, the Awami League and the BNP, are the mother organisations for most mastans. Both have a range of wing organisations (student and youth wings, women’s and labourers’ wings, farmers’ and fishermen’s wings, among others) with which mastans are primarily affiliated and through which they often join politics. Many students, most of whom come from outside Dhaka, initially join student politics because they need contacts to get a ‘seat’ in a hall of residence or simply to make friends. The story of Ruby’s entry into politics through BNP’s student wing, Chhatro Dol, is typical. He said: ‘When I came to Dhaka [university], I had a “brother” [bhai, also “cousin” or “friend”] from my home area

The political bully in Bangladesh 309

here, and I started spending time with his friends. They were with the Chhatro Dol’. They did ‘good work’, he said, meaning they were friendly and helpful, and so he stuck to them. Ruby became a BNP man almost by accident. Friends of friends and friends of relatives lead to student life in the big city. As networks of friends and family tend to follow party colours, students often end up in one or the other camp. Alamgir’s story was similar. It had the same element of chance, but it also showed that there is more than chance to accidents; they are part of an intricate real-life web through which young men like him enter particular lines of work. Alamgir was born in a village into a poor labourer’s family and he never went to school. At 15 he followed his father to Dhaka, where he worked on construction sites, and took up the same jobs as his father, first as a help to others, and later as an increasingly skilled labourer. His father was involved with a criminal gang in Dhaka, which Alamgir joined too. Alamgir is not tall, but he is powerfully built. ‘I was very strong even when I was young’, he once told me, ‘When I was 15, I was as strong as my father. When I was 18, I was stronger’. The leader of their gang was one Selim-bhai. Although he played a large part in Alamgir’s mastan career, I never understood who exactly Selim-bhai was or what position he occupied, except that he was associated both with small construction companies and with the local leaders of BNP. ‘We were the BNP’, said Alamgir, ‘There were many good leaders in the BNP. Not like the Awami League. We used to hang out in the [local] party office. And we demonstrated. The leaders were always very respectful to us because we were in Selim-bhai’s group’. If you were with Selim-bhai, nothing could harm you. Selimbhai was ‘very powerful; he always had money but he always spent money. He had friends everywhere. Even in the police’. Alamgir’s entry into politics was purposeful, and somehow very human. He was not drawn to the party by force of ideological conviction, but joined it through his father’s associations and Selim-bhai, who gave him protection and respect. How Babul joined the Awami League was slightly different. His story goes a long way to show how rural Bangladesh was politicised, how families came to identify with this or that political party, how individuals were recruited to politics and what kept them there. His story also helps to understand the nature of politics in Bangladesh and how Bangladeshi political parties function as patrons. Babul’s tale

310  Arild Engelsen Ruud

begins with a quarrel with a neighbouring family, a large family with five sons who insisted that Babul’s sister should marry one of them. Babul’s family resisted their advances, making itself the subject of malicious rumours and even stone pelting. When Babul’s sister married another man, the wedding party was attacked and forced to seek refuge in a police station. The acrimony carried on for several years and, as Babul put it, ‘it became necessary to avenge the family’. Quietly, they started to cultivate relations with a locally influential ‘Awami League family’, into which another one of Babul’s sisters eventually married. Babul himself was introduced to the Awami League-affiliated Chhatro Dol at his college, and his family slowly acquired the reputation of an ‘Awami League family’. Then the neighbours let up. This plotline is quite complex. Several families and individuals embroiled in this conflict sought external protection, and a number of party workers, activists and mastans stepped in to help, all the while capitalising on the dispute to expand their parties’ followings. Association with the Awami League increased Babul’s family’s influence in the village and others kept respectfully away. The party had tinted yet another family with its colours; the family was now bound to the party by an accumulation of debt which it could only repay with political loyalty. These changes contributed to the dismantling of the old patronage system dominated by landowners, who provided protection in villages and who were eventually replaced by various new patrons, ranging from school teachers and NGO workers to business people and political activists (Westergaard and Hossain 2005). Many local ‘social workers’ or ‘political activists’ share the qualities of the ‘new leaders’ (naya netas) in India, described by Krishna (2007) and Alm (2010). They are intelligent operators, often without the wealth, kin or ritual status which traditionally secured patronal influence.

Reciprocity, ‘contacts’ and loyalties When I first met Babul and Ruby, Babul described the relation between mastans and politicians in this way: People in the underworld maintain a relationship with political leaders, whether the Awami League or the BNP. When the BNP is in power, they will use the BNP MP [Member of Parliament], whoever he is. If there is an

The political bully in Bangladesh 311

administrative problem or some problem inside the police station, you tell the MP and the problem goes away. The MP can provide an exemption [from prosecution]. In return, if there is some Awami League leader whose house has to be bombed, then he will use them [mastans].

The political patron may require violent means to enact his patronly duty. In exchange, he protects the bully. Ruby’s story offers another illustration. Soon after coming to Dhaka and the university, he established a certain reputation for being a ‘tough’. He was young, ambitious, keen on getting some ‘action’ and probably also in need of money. When he was still a junior ‘activist’ associated with the BNP student wing, he was asked by some student leaders and more senior party members to gather ‘some information’. He never told me what it was or even what it was about, but he said: ‘They needed me to get that information’. At the time, he lived near the national parliament and hinted that the people he knew and worked for had contacts far beyond the university, possibly even among senior party leaders. When he went to gather this information, which required some force to extract, he brought a gun, which he ended up using in an unfortunate sequence of events. Although Ruby did not kill anyone, he fired some shots and a First Information Report was lodged against him with the police. He went to the party leaders, reporting what happened, and said: ‘Bhai, what should I do now?’ They told me to stay low and out of town for some 10–12 days. ‘We will send you news’. Some senior leaders then made the case vanish. Then they sent news to me and I returned to Dhaka. One of them was one of the 4–5 top leaders of the Chhatro Dol. I can’t tell you his name. He belongs both to the university and to the central organisation now.

The BNP was in power then, and people from the party’s top echelons intervened on Ruby’s behalf, making him feel protected and perhaps even invincible. This marked the beginning of a lasting collaboration between Ruby and the party leaders, an exchange of his services for their protection. In the early years of his career, Ruby managed to use his ‘credit’ with the party to help his father in an old struggle against neighbours in their village. The tale has a familiar opening.

312  Arild Engelsen Ruud

Our neighbours were all in the Awami League. I was doing the BNP. My father was a freedom fighter [entailing a special social standing and respect], but he was alone [as the only grown man in the family]. Our neighbours were 4–5 brothers and all in all there were 20 or 25 boys and men in that family. Because that clan had a lot of clout, they could harass us in many ways without repercussions. The problem was that they occupied some of our land. It was our land, but when my father went to them and complained, they harassed him and shouted at him.

The quarrel went on for a few years and Ruby’s father was on the verge of losing. At one point, when Ruby was back home from the university, he went to speak to the neighbours to ‘explain’ his family’s position and asked them to return the land. At that time Ruby’s growing reputation as a political activist in the capital had not yet reached the village, and the neighbours refused to negotiate. This is when Ruby turned to his political contacts in Dhaka. I was in Dhaka then. I had a friend there whose uncle was an officer-in-charge at the police station to which our village belonged. I phoned the uncle and I said that there is this quarrel over our land. ‘Please look into the matter. Do some investigation’. And they did. After the police had made their inquiries, the neighbours gave in. They were afraid. They did not have a legal case. And so they came and told my father: ‘There will be no more such errors. We are giving up the land’.

By this time, Ruby’s contacts in the city must have had considerable weight. Otherwise his plea would not have made an impression on the police station’s chief, and certainly not through a casual phone call. In all likelihood, local party leaders already knew about Ruby’s exploits in the city, and the officer in charge of the station sounded out Ruby’s request with them before approaching the neighbours. In Bangladesh, as in most of South Asia, what local administration does and does not do is substantially decided by the local ruling party’s leaders. This is perhaps more so in Bangladesh than elsewhere in the region. The 2006 BRAC report on the state of governance in the country characterised it as a ‘partyarchy’ in which party leaders wield enormous power over bureaucrats

The political bully in Bangladesh 313

and local society. With the police force and a powerful figure in the ruling party stacked up against them, the neighbours finally caved in. By this point Ruby had warmed to the story, elaborating on its colourful finale, which underscored the potency of his contacts: The people in the village came to my father and said to him: ‘Please don’t lodge a case against your neighbours. If you do, they won’t be able to stay in their home. They are going to suffer in so many ways’. So I said: ‘Ok, if they give up the land, we will have no further conflict with them’. And the problem was solved.

Ruby, who was then still a minor activist-mastan away in the capital, seemed important enough to the party’s local leadership to care to influence the police in his favour. The point was not lost on his neighbours or other villagers, who suddenly made peace with him and his family. Notably, his story involves no open violence. The initial shootout which established Ruby as a party mastan took place in the capital and, in any case, it was hushed up and spoken about only obliquely back home. Ruby’s tale was not about violence, but about its threat which was ever present in the way that he carried himself. His erratic movements were punctuated with sudden bursts of anger and energy, which alarmed anyone who spent time with him, including his fellow villagers. There were vague rumours about the shootout in Dhaka, but they alone were not enough to terrify anyone. Muscular bearing requires political contacts of proven worth and Ruby did not have to use violence to convince his neighbours of just how wrong they were to quarrel with him. The frightening thing about him was not muscle, but his connections and what they entailed. After the incident, people knew that he could make the police initiate investigations. His contacts and the power to put people in prison, or even deprive them of their homes, were just a phone call away. Ruby’s exploits in Dhaka had generated ‘capital’. As the party’s client, he could ‘cash in’ his due not only in Dhaka, but also in his village, tied into the hawala network that constitutes political party membership in Bangladesh. The patronage in which he was engaged was thus not a dyadic bond with a single patron, but a much bigger network of loyalties and ‘connections’.

314  Arild Engelsen Ruud

If mastans enjoy near-total protection when their parties are in power, the costs to them are high when their parties step down. Awami League ‘activists’ with whom I spoke during the BNP reign in 2001–2006 experienced serious deprivations. One man, Siddique, was only relatively safe in a very small area. Saleem, another Awami League activist I met at a tea stall, kept looking over his shoulder, fearing the police may spot him as the tea stall was in an area where he was well known. He generally avoided sleeping at home and stayed with different friends every night. A third activist, Chandanbhai, moved around only with friends, who after only half an hour of conversation with me in a coffee shop grew fidgety and got up to leave. We parted company outside and they left through the back door. Under ‘their government’ these mastans had been among the most influential ‘activists’, but now they were ‘criminals’ in hiding. Yet, however difficult life may be for the mastans when their party is in opposition, they rarely switch sides. Party loyalties, in other words, are in most cases for life.

Action and violence All three mastans, and particularly Ruby and Alamgir, loved ‘action’ and took pride in the tales of their exploits. Alamgir described an incident in which he and his gang had opposed another and said to them: ‘you should leave now’ (tomra chole jao na). The deliberately casual turn of phrase referred to a dangerous shootout and a little later Alamgir rolled up his sleeve and showed me a scar on his upper arm, where a bullet had gone right through the muscle. He dismissed the episode, but there was more than a hint of pride in his memento of war. Alamgir’s incident was relatively innocent in comparison to Ruby’s experiences. Ruby told me that he was once involved in an arms trade case, which led to more stories about extensive use of violence. The first was about his student days in Dhaka and it had the familiar features of political student life: an almost continuously high level of tension, violent clashes and beatings. Ruby: I did not have arms [in English], but there were arms, on campus. AER: Did you see firearms yourself? When you participated in the beating, what did you carry then? Ruby: We had clubs (lathis), hockey sticks, Chinese axes. But those are not arms.

The political bully in Bangladesh 315

AER: How are these weapons used in politics? Ruby: If you want to beat someone, it is easier if you have a big thing with you. You see, there is also the chance that I will be beaten. AER: Have you ever been beaten? Ruby: No, no. Never. When I was in college, during Hasina’s government [1996–2001], say some boy [chhele, meaning student] from the Chhatro Dol [opposition student wing] would not listen… There was a storage room near the games room in our college where we kept some arms; we would take that boy there, beat him and then send him to hospital.

We may recall that Ruby himself was with the Chhatro Dol. The story, then, was about maintaining discipline within the ranks of one’s own party, when it was out of power. It may have also been part of factional squabbles. Babul too was once involved in beating a student, after which he was accused of attempted murder by the police. The fight broke out under unexpected circumstances. The opposition had prevented the local wing of the ruling party from organising a procession. But Babul told me about this incident in order to explain how a handful of determined people could manage to resist the dominant political force. The opposition controlled this particular college and gave an order against demonstrating, threatening to thrash anyone who was caught doing so. ‘For beating they would select anyone, randomly. It could be here [on campus], it could be elsewhere’. It is not uncommon for bullies and political activists to be attacked by their opponents, to be surprised as they walk in the park or sip tea at a stall and be beaten up. To loosen their hold on this particular campus, Babul and his associates intervened: they grabbed a boy from the opposition and thrashed him sufficiently to frighten others—enough for him to need hospital care. In the end, Babul and Ruby acknowledged that they used violence ‘when necessary’. And there were, and are, deaths. Babul lost one of his friends to wounds from bullets fired some years earlier during a demonstration. In Ruby’s, Alamgir’s and Babul’s stories violence was a routine, accepted aspect of politics, and they describe it as an unavoidable, almost casual, part of the game. But violence and brutality can also be exaggerated. Violence is frequently a display of threat and a performance of a certain style

316  Arild Engelsen Ruud

of public authority, which involves both ruthlessness and generosity, and which may help to defend as well as define a neighbourhood (Spencer 2007, 31; Berenschot 2011, 266). It can also work the other way around. In his study of musclemen in Hyderabad, Sudhir Kakar showed that violence can generate a tough or a criminal reputation. The goonda-king of Hyderabad explained to Kakar that young men who are wrongly booked by the police on murder charges tend to see themselves as killers after they are released. They emerge from the station ‘swaggering, as if they really are killers, and what better credentials can they have?’ (1997, 57). Fear is central to how mastans are perceived, and how they present themselves, and Ruby and Alamgir knew that they frightened others. Throughout South Asia, the reputation of a gangster is a valuable asset. If it is taken seriously, it often does not perpetuate violence, but conversely makes it unnecessary. Ruby’s tough reputation, and his political contacts, helped to resolve his family’s land dispute without the use of force. Then again, the threat of violence is only effective if it is backed by rumours about instances of actual brutality. The presence of knives, cudgels and firearms, the risks of being arrested, beaten, have your arms broken or tendons cut (the Islamists’ preferred attack method), with hospital visits and bullet wounds, generate fear only if they are occasionally instantiated.

Politics as morality In spite of a certain pride in mementos of war and the acknowledgement of ubiquitous violence, all three mastans told me that violence was ancillary to their ethical aims. Their main motivation for engaging in politics, they insisted, was moral; and the desire to do what was good and just legitimised brutality. They claimed that all their activities benefitted their fellow-men, and in particular the ‘ordinary man’ (shadharon manush). They worked for the poor, the vulnerable and the weak. This world was deeply unfair, and the under dogs were ever in danger of falling prey to predators or just becoming misfortune’s victims, they held. So the use of violence and exposure to danger were sacrifices necessitated by moral duty. Alamgir and Ruby insisted that they always helped people from their home area. ‘When they come to Dhaka’, said Ruby, ‘I look after them. When you come from my home district to Dhaka, you get off at Kalyanpur, and I meet you there. But they do not only come for

The political bully in Bangladesh 317

political reasons. There are many reasons. They know me and love me, and I try to help them in many ways’. Every instance of violence, or its threat, described by Ruby was morally justified. When he spoke of his family’s quarrel with their neighbours, he insisted that his neighbours had no ‘legal claim’ to the disputed land. He fought back not only to defend his father but also to uphold the law. I asked him whether he thought he frightened people, and he said: ‘a little’, quickly adding that ‘ordinary people’ loved him: ‘90% love me, and 10% are afraid. The bad people are afraid’. Alamgir was more candid: ‘the bad people in the neighbourhood were afraid of us. We were also bad [bodmaishi]. There was violence [khun, suggesting ugly violence, torture]’. At first he only admitted to having heard about such incidents, but then he described the fights, shootouts and kidnappings (mar, guli khawa, kidnyap) in which he had taken part. None of these, however, involved ‘ordinary people’, only the ‘baddies’. ‘Ordinary people loved us. We helped them’. Alamgir: Selim-bhai always helped the poor. The workers loved us because we always protected them. Once there was a raid on their quarters. We ran over and told the attackers to leave. We always helped the workers. AER: Were the workers from Dhaka? Alamgir: No, they were from other places. Many of them were from my district, but also from other districts. They all knew us. When we went there, to their quarters, they would make us sit, give us tea and cigarettes. We could go there and they gave us food to eat. Any time.

This is a Robin Hood tale: a story of bandits resisting violence and injustice. It is a story of force used to right what is wrong and help the poor and the oppressed. This moral vision is central to how bullies in Bangladesh and other parts of South Asia present themselves. The gang leader in Hyderabad described by Kakar was known for helping both ‘his community’ and ‘the people’ at large. In one story retold by Kakar, he helped to implement court orders, which could not have otherwise been carried out by the police (1997, 79). While he was not averse to charging a fee for his efforts—‘the minimum fee is a hundred thousand rupees’—in his own eyes, and perhaps in the eyes of the many he helped, he remained a respected fighter for ‘the people’ whose ‘love’ he enjoyed.

318  Arild Engelsen Ruud

In Bangladesh ‘action’ (a word used in English), which figured prominently in the bullies’ narratives, was almost invariably described as engagement on behalf of ‘the poor’, an argument that legitimised even the most brutal aspects of their work. In their own descriptions, the role they play is essentially virtuous, made more so by their courage and self-sacrifice. The virtue of bravery came through in bold decisions taken on the spur of the moment. Notably, in their stories of courage the mastans never said they were acting as representatives of political parties. Instead, they presented themselves as independent guardians of a very high moral set. These carefully framed moral self-portraits were airbrushed even more than many political activists’ presentations of self. Here is Babul’s account of why he once ‘went into action’: At about 10 at night I heard that some mastans were planning to kidnap [a local businessman]. Some people from his area told me, they were with me that night. There were others too. Together we decided to go there to prevent the kidnapping. People like these criminals often have arms, and those who have arms are often involved in this sort of work. AER: Did they belong to another political party? Babul: Criminals have no party affiliations. AER: Who was it they planned to kidnap? Babul: A man in our neighbourhood. To get money. They would take him and hold him, bind his feet and hands to keep him still and then extort money from his relatives. First they inform the family about the kidnapping. When they get the money, they let him go. In this case I more or less knew who the kidnappers were, and after a few rounds of bullets and bombs they really understood that they had to leave the place. AER: They could not kidnap him? Babul: No. AER: What was your gain, then, when you resisted them successfully? Babul: That people will love me.

The political bully in Bangladesh 319

A similar sentiment emerges in Alamgir’s stories of his early adventures, in which he describes himself even more vividly as a brave and selfless defender of the poor and vulnerable, in this case a woman: Alamgir: A shop owner from that neighbourhood died. He had two sons. They were small then. A gang went to his wife. They said: ‘your husband owed us money. 300,000 taka. He is dead, so you give us’. They went to her and demanded money. She was a poor woman, her husband was dead, she had two small children. How can you do something like that? So she came to Selim-bhai. She said: ‘Selim-bhai, you are a good man. I am poor. What can I do?’ So we went there [to confront the gang]. There was some shooting. One man died. Another man was hit [by a bullet], and they took him to hospital. AER: You did this because you wanted to help the widow? Alamgir: Yes. AER: Were you frightened? Alamgir: No, no! We are in the hands of God.

Musclemen in Bangladesh, no less than elsewhere in South Asia, appeal to moral frameworks to justify what they do and turn what may be seen as an ‘offence’ against society into the defence of moral duty, usually in response to injustice done unto others or unto themselves. Indian musclemen involved in communal riots also present their actions as morally correct. Indeed, the more objectionable are the bullies’ actions, the more important it is to turn them into noble heroics.

Free-range clients in the underworld Patron–client relations contain a contradiction between the reciprocal demands and the selfless affection and respect meant to sustain that relationship (see the Introduction and Chapter 5 in this volume). For the political bully who tries to style himself a patron, the conundrum lies in the uneasy relation between the fear which he must instil to maintain his position and the ‘love’ which he hopes to generate. The finely woven Robin Hood rhetoric unravels when we realise that much of what mastans actually do lacks morals or

320  Arild Engelsen Ruud

ideology, something audible in their (often implicit) admissions that they are much more interested in extortion and making deals than in helping the poor. Once upon a time, Ruby, Babul and Alamgir were innocent youngsters who received an education in violence when they became involved in politics. In the course of their careers, crime and moneymaking became aims in their own right. One day, when conversation turned to money, Alamgir told me about the extravagances of the mastan world, and about ways to get hold of money to fund these expenses. I had an elder brother, Dulal. Everybody loved him. He was good looking, and he had a beautiful singing voice. He was getting married to a girl from his village. She was from a good family. But Dulal’s father had no money. He drank. So we had to help with money for the feast. 100,000 taka or 200,000 taka, something like that [$1300–2600]. Siddik-uncle [kaka] then said that there is this businessman in Kolabagan [a neighbourhood in Dhaka]. So we got money from there. Not enough, but something. We did not kill him.

Perhaps his motives were honourable, to help a friend pay for a wedding feast. But what he described was a robbery, a brutal raid on a businessman not directly guilty of his friend’s lack of funds. Their associations with the BNP did not prevent Alamgir and his friends from operating on their own, and sometimes very violently. Ruby and Babul, and certainly Alamgir, were more involved in the ‘underworld’ than they had first admitted and than would have been common among ‘activists’ and party workers. As he wrapped up his story of preventing the kidnapping, Ruby surprised me by saying that he knew the kidnappers ‘personally’ (amar porichito) and that he did not think they were ‘bad people’. ‘People are not born bad. It is circumstances that compel them to be bad’, he reflected. Nonetheless, he prevented them from kidnapping a man with the use of ‘a few rounds of bullets and bombs’. During this time Babul and Ruby were registered as university students, and both lived among ordinary students. They had a wide circle of friends and acquaintances (including my research assistant), and they spent much of their time on campus. Both had been ‘student activists’ for several years and they knew senior student leaders, and probably senior party bosses

The political bully in Bangladesh 321

as well. Their lives straddled the criminal ‘underworld’ and the worlds of higher education and street politics. The extent of Babul’s and Ruby’s involvement in the underworld emerged in a story they told me about a brawl over a fish tank. Ruby: Next to the Meherpur police station [thana] there is a large tank, called the Rajbari Beel. The fish caught and sold from this tank is worth some 30–35 lakhs [lakh = 100,000] of taka every year. But it was impossible for honest people to lease the tank because of a gang of ‘miscreants’ [in English]. So the lease was given to the Chairman [of the Union Council, a locally powerful figure] who cooperated with criminals. Nowadays the Chairman holds the lease. Every year he gets 4–5 lakhs out of it, and they keep 30 lakhs. Babul: As long as the Chairman did not agree, no one could lease the tank. Now, say you let me lease the tank. I am involved in the underworld. If someone wants to protest, I will tackle them. All you have to do is say: ‘Yes, I gave it to him’. I will take care of the rest. A lot of clashes have taken place in the last 10 years over matters like these. Say, I am the Chairman and I tell someone: ‘You lease the tank’. And then someone else, in a higher position, says the same thing to another group. Then there would be war between these two groups. The group that is able to occupy the spot, they win. AER: How do they enforce their possession? Babul: I was coming to that. Imagine that I had taken the lease of that tank, and that he wants it [pointing to Ruby]. To take the tank away from me, he would first have to see how much force I have. And then he will try to outdo me. And I will try to muster as much force as I can to keep it. If he brings a pistol, I will try to counter that, and then there may be some shooting. This is just an example. If something like this happened and two of my people died, but I still retained possession of the tank, it means that he lost. It slipped out of his hands. Ruby: The leaders need those who have power [khomota]; and those who do not have power, political leaders do not need them.

322  Arild Engelsen Ruud

AER: What do you mean by khomota? Ruby: Arms and people. Whoever has more is more powerful [khomotaban].

Ruby and Babul described a business deal between the Chairman and the gang. The Chairman had outsourced control of the tank in exchange for a regular income, agreeing an annual lease contract of 4–5 lakhs taka and giving the gang full freedom to assert control over the tank and to earn from it. Deals like these are struck all over the country, where political leaders supply income opportunities and protection to thugs in exchange for a cut of the profit and the muscle which they may occasionally require. Ruby and Babul carried on with their story. It turned out that both had been involved in an intense struggle over this tank and that they were actually close associates, although Ruby was with the BNP and Babul with the Awami League. AER: Have you seen such incidents yourself? Ruby: That tank I was talking about. I have eaten fish from that tank. Babul: There was a fight over that tank. We even used submachine guns and AK-47s. We put a lot of effort into that fight. AER: How did you get the arms? Ruby: The tank is close to the Indian border. And we got money from local businessmen.

Then followed a long and rather complicated story about the struggle to control the tank, in the course of which Babul’s cousin was kidnapped by the opposing party and kept in confinement in a neighbouring village. In the end he was released when Ruby, who knew the kidnapper, intervened. In fact, Ruby was friendly with the kidnapper and agreed to stage the victim’s ‘escape’, rather than a willing release, which would have undermined the kidnapper’s reputation. The whole affair seemed fairly civilised, until Ruby mentioned, almost as an afterthought, that one of the kidnappers, a local man and owner of the house where the firing occurred died in a crossfire.

The political bully in Bangladesh 323

Although in the capital the two mastans belonged to opposing camps, in a remote village they had fought on the same side, defending the Chairman’s racket. I never learned which party the Chairman belonged to, and it probably did not matter. He offered a good source of income and the rest was left to the mastans’ own devices, which included shootouts, AK-47s, kidnappings and death. Relations between gangsters and politicians are often highly unsentimental, based on monetary considerations rather than ideological sympathies or personal bonds, and they often involve agreements about the division of labour and spoils. Babul and Ruby moved quite freely between politically sponsored crime and violence used for non-political, self-supporting purposes. Similarly, for all his loyalty to the BNP, Alamgir was quick to loot for a friend in need. Another one of Ruby’s stories suggests that ties to parties may occasionally be used as support for the mastans’ freelance activities, in which party bosses have no direct interest. Here is Ruby: This is an incident from two years ago. It was in the winter. It was early morning. It happened in the morning fog. Some cloth merchants had hired a bus and were returning from a cloth market called Poradoho. Then they were suddenly stopped and robbed by a criminal gang. The local MP took 40% of the loot, and the gang kept the rest. There was no police case opened against those who had taken part in the raid [dakati]. Some local people had witnessed the whole raid, but still there was no case lodged against the criminals. Some others from that area, people who are opposed to the  MP, people from his party or from another party, they wrote a list [of names] and filed it with the police. [But the police took no action.] As long as the MP is in their [the gang’s] hands, the administration is also in their hands.

The details of his description—the morning fog, the name of the cloth market, the local witnesses—were all very intimate and I asked whether he was part of the gang. In response he only smiled. If initially Babul, Ruby and their associates required the involvement of party bosses to ‘get their work done’, as their careers matured, they came to generate income and to exercise influence over countryside affairs—often through force and bloodshed—quite independently of political patrons. Once they found their footing, they might have remained ‘party activists’, but they also acquired considerable liberty to act as they wished. Alamgir, who entered party politics by a less

324  Arild Engelsen Ruud

direct route, considered himself a ‘party activist’ through his adult life, and saw his involvement in violence as an extension of his political work. In 2006 he depended for his safe ‘retirement’ on the patronage of the political leader through whom he and I initially met. Yet he could not link most of his stories to party politics. In fact, in accounts provided by all three bullies political leaders hardly figured at all. Of the many stories they told, few were painted in party colours, and in some cases their violence was indeed aimed at their own party folk. All three were no doubt familiar with the idea of ‘clean’ politics and were fairly well versed in party rhetoric. But formal politics did not offer the opportunities they lived for. Much of their work focused on moneymaking deals like fish tank racketeering or looting poor cloth merchants on their way home on jointly hired buses. Their business nevertheless needed political protection and moral legitimation to operate.

Conclusion The modern South Asian state has created political spaces and opportunities for crime, and the presence of criminals in politics is thought to be on the rise, even if the ‘criminal’ role remains poorly understood. Some analysts have suggested that the processes of populist democratic mobilisation in poor, uneducated and postcolonial societies ‘deinstitutionalised’ the state, making it vulnerable to criminal elements. Alternatively, others have argued that the inability of the modern state to reach its poorer citizens has created openings for criminals, allowing them to fill a structural space of mediation; or simply that criminals are recruited for specific purposes by opportunistic politicians. The general sentiment is that the two spheres of activity are analytically distinct from each other, that their personnel is not overlapping, and that ‘politics’ has somehow allowed ‘criminals’ to enter its fold. I suggest that in the context which I describe crime cannot be understood as an integral part or an extension of politics, but as a set of associated activities. The relationship between the two sets of activities, or the ‘fields’ of crime and politics, in Bangladesh and South Asia at large is intimate, although complicated. Politics is a springboard that can propel youngsters who possess the right balance of ambition, recklessness and brutality into a life of crime.

The political bully in Bangladesh 325

It provides the young mastan with protection, friends, stature and moral legitimacy. For the more mature political bully, the party as a political patron provides a wide network of contacts in cities and villages, among the high and the low. This network may be called upon and it calls upon mastans in return. But economically the mature bully has to fend for himself. The affiliation of political bullies to a party lasts a lifetime, or until they manage to retire gracefully, due at least in part to the brutality they impose on their opponents, but also due to the fact that political bullies are recruited from the ranks of aspiring politicians. They are, then, both personally acquainted with people who later become political leaders, but more importantly they are versed in the logic and manners of political competition—at least on the level of the street.

The dark side of patronage in the Pakistani Punjab


Nicolas Martin

Introduction Look, we get elected because we are ba asr log [effective people] in our area. People vote for me because they perceive me as someone who can help them. And what help do they seek from me? Somebody’s brother has committed a murder and he comes to me and I protect him from the authorities. Somebody’s son is a matric fail and I get him a job as a teacher or a government servant. Somebody’s nephew had been caught thieving and I protect him. This sort of thing. That is my power. This is what they perceive as power. You know, somebody has not paid up their loan and I try to have the payment delayed, etc. That means that I get elected because I am doing all the wrong things…. My skill is that laws don’t mean anything to me, and that I can cut right across them and help people whether they are in the right or in the wrong. If somebody’s son is first class, he’s not coming to me to get him a job. If somebody has merit they very rarely come to me—occasionally they come to me. But it’s the real wrongdoers who come to me (anonymous politician quoted by Wilder 1999, 204).

According to Partha Chatterjee (2004), the sphere of civil society— where the public gathers to hold their government to account and to press legislative demands—is bourgeois, and excludes the majority of India’s population.1 He argues that the rule of law, through which civil society operates, protects the interests of the propertied classes and tends to further the interests of corporate capital. In Central India, it is after all under the rule of law that the state hands over 1

A different version of this chapter was published in the Journal of Agrarian Change (Martin 2013). I would like to thank the European and British Research Councils for funding the preparation of this chapter for publication.

The dark side of patronage in the Pakistani Punjab 327

land occupied by indigenous communities to mining companies.2 On the other hand, he argues that what he calls ‘political society’—a sphere of patronage that bypasses ‘bourgeois law’ and bureaucratic procedure—helps ordinary people extend their ‘freedoms using means that are not available to them in civil society’ (2004, 67; 2011). Politicians operating within ‘political society’ circumvent and even break the law to appease their constituents, making allegedly rigid and unresponsive post-colonial states more amenable to ordinary people’s needs. Chatterjee cites the example of slum dwellers who avoid eviction by appealing to patrons who create legal and bureaucratic exceptions which allow them to stay where they are. Although such patronage is neither secure nor permanent, writes Chatterjee, at least it prevents slum dwellers from being evicted as they would have been under ‘bourgeois’ state law. Here I argue that the social costs of patron-client relations outweigh their benefits. This is so because clientelistic exchanges which constitute Chatterjee’s political society are unequal and they benefit the political and economic elites more than they help most ordinary people. I show that patronage undermines universal public service provision, and therefore that its dark side significantly overshadows the bright. I describe affairs in a village in the district of Sargodha in the Pakistani Punjab to show how the de facto privatisation of governmental services via patronage bonds reproduces traditional clan- and class-based patterns of dominance and how, in the process, it prevents the expansion of rights of citizenship to the rural poor.3 My ethnography shows how a government without a stable framework of rights, laws and bureaucratic procedures can serve elite interests just as much as—if not more than—the rule of law and bureaucratic procedure. As Matthew Nelson (2011) has shown, while Pakistani state law grants to female heirs and sharecroppers certain 2

Many of these land acquisitions are facilitated by India’s colonial ‘laws of eminent domain’, which allow the state to make purchases of land in the name of the public good. 3 These findings agree with those of Craig Jeffrey (2001) and Jonathan Pattenden (2011) who show that elites in rural Uttar Pradesh and Karnataka, respectively, maintain their positions of power by colonising the state and appropriating its resources. These conclusions are also opposed to those of South Asia’s development optimists, like Oliver Mendelsohn (1993), who believe that political and economic development has so much diminished the power of landed elites that it no longer makes sense to speak of ‘dominant castes’ (à la Srinivas 1987).

328  Nicolas Martin

rights to land, politicians deprive them of these rights by ignoring and subverting the law to gain political support among kin and clients. This chapter similarly shows that in Pakistan it is not the implementation of state laws, policies and bureaucratic procedures that oppresses the poor but the powerful members of landowning clans (biraderi) who undermine them. While Nelson (ibid.) and Lieven (2011) contend that clan-based politics render the Pakistani state ineffective, I argue that the state’s ability to deliver basic services is impaired by classes who possess disproportionate social, political and economic power that allows them to subvert bureaucratic procedure, the implementation of policies and the rule of law. Because landlords can evict their tenants—despite legal occupancy rights—they can extract from them free labour (begar) and votes. Moreover, because landlords use their influence to flout bureaucratic procedure to advance their own interests, and those of their supporters, they deprive the majority of rural Pakistanis of access to state resources. As a result of landlords’ patronage, absentee and unqualified doctors and teachers retain their positions, depriving villagers of even the most basic medical and schooling facilities. The landlord-patrons also protect criminals, whether petty thieves or heroin traffickers, making them immune to policing and placing villagers’ property and security at perennial risk. In short, I show that patronage benefits the well connected few at great cost to the rest. Moreover, patronage prevents political unity and class mobilisation among subordinate classes: instead of uniting to challenge their subordinate standing, subalterns align themselves individually to powerful patrons in search of particular benefits. As a result, subordinate classes remain weak, fragmented and easily exploited. Contrary to Lieven (ibid.) and Lyon (2004) who claim, like Chatterjee (2004), that patronage makes state resources more accessible to the common man, I argue that patronage bolsters the power of the elites by redirecting to them state resources and in the process undermining the political unity of subordinate classes and public service delivery, and consequently the majority population’s economic and physical security, freedom, education and health.

Changing patterns of dominance The dominant Jat clan known as Gondal in the village of Bek Sagrana belonged to the class of middle to large landholders who gained

The dark side of patronage in the Pakistani Punjab 329

political prominence in the 1960s and 1970s, following the Green Revolution and the advent of popular politics during the premiership of Benazir Bhutto (Zaidi 2005, 5174).4 From colonial times, the compact Jat clan structure (based on endogamy: see Alavi 1972a), their numerical preponderance, land ownership and role as mediators between villagers and the state has made Gondals the dominant clan around Bek Sagrana. Yet until the 1970s their political influence had not extended far beyond the village. Prior to this, aristocratic (ashrafi) Makhdooms, or landlords, controlled the provincial and national assembly seats representing the area. Their political influence derived largely from their ownership of more than 1000 acres of land and from their contingent ability to control the livelihoods of a large number of people. But changes in the rural economy, including the decline in village crafts and agricultural mechanisation, loosened this control. In Bek Sagrana villagers said that the Makhdooms were unable, and often unwilling, to reassert their control because they had grown ‘soft’ from their excessively luxurious and indolent lives. Although these claims contained an element of truth, it was also true that, unlike the Gondals, the Makhdooms could afford to leave the local political scene. Some of them had graduated from prestigious universities in Britain and the US and could pursue lucrative careers in banks or multinational corporations, escaping the cutthroat world of Pakistani politics. Other Makhdooms obtained high-ranking bureaucratic positions, from which many moved into international organisations like the United Nations, the World Bank and the Asian Development Bank. Although not all aristocratic landlords in Pakistan have abandoned rural politics, many followed a path like the Makhdooms’. In Bek Sagrana their exit opened the doors to the many forceful Gondals. In Bek Sagrana proper, the Gondals comprise 13 of the 123 households in the village, and constitute 74% of the village’s total landowning population. In clusters of settlements (deras, literally ‘encampments’) that surround the farmhouses of the wealthiest Gondals, the distribution of land was even more skewed: Gondals constituted more than 95% of the landowning population. Only two Gondal households in the village did not own any land and worked as overseers (munshis) for their wealthier 4

Hamza Alavi (1973) classes farmers who have 0-5 acres of land as ‘small farmers’, those with 5-25 acres as ‘medium farmers’ and those with more than 25 acres as ‘large farmers’.

330  Nicolas Martin

clanmates. The average size of landholding among the remaining 11 Gondals in the village was 25 acres, with the poorest Gondals owning three and the wealthiest 100 acres. However, in the 1980s the wealthiest Gondals moved out of the village centre to live in spacious, newly built compounds amidst their citrus orchards. The wealthiest among them owned over 400 acres of land and the second wealthiest around 300; both acquired their landholdings, and a variety of urban businesses and properties, after entering provincial and national politics. By 2004 all except the two poorest landowning Gondals in the village spent most of their time in Sargodha, the district capital, where their children were being educated and where they had government jobs and businesses. The wealthiest lived in Lahore, where they also educated their children.5 Most returned to the village on a weekly or fortnightly basis and spent extended periods of time in the village only during the citrus and wheat harvesting seasons, and during elections. Citrus orchards, which were introduced to the area in the 1970s and 1980s and which now cover most of the canal-irrigated land in the village, bring high profits and do not require sharecroppers or much supervisory work. The Gondals could shift to the cultivation of citrus fruit because most of them owned well-irrigated lands near a major canal.6 Some still employed sharecroppers (even though tractorisation and citrus orchards had drastically reduced their demand), but since the 1970s they have increasingly leased out land not used in citrus cultivation for cash. In 2005 an acre of citrus orchard generated an annual income of 50,000–60,000 rupees ($1000–1250), so that an owner of 25 acres could earn roughly 1.5 million rupees ($31,000), untaxed, from citrus alone every year. Although wheat provided some additional income, it was grown mainly for domestic consumption, to pay farm and domestic servants, and to compensate village menials and artisans (kammis) for their services. A number of Gondals also held government jobs in the departments of health, education and agriculture, set up small shops in town and 5

Only the poorest Gondals in the village, two siblings who each owned three acres of land, lived permanently in the village, but sent their children to live with their youngest brother who lived in rented accommodation in Sargodha. 6 To produce plentiful fruit, most Gondals in Bek Sagrana used more than their share of irrigation water, thereby depriving those who lived downstream of their share. People at the tail end of the irrigation canals in nearby villages complained that they could not grow citrus because the Gondals ‘stole all their water’.

The dark side of patronage in the Pakistani Punjab 331

bought small commercial properties for let. Most of them obtained their government jobs via links to Gondal patron-politicians. Many supplemented their incomes by working as contractors (thekedars) to build government-funded infrastructure when leading village Gondals were in power and could arrange their contracts. Some Gondals significantly supplemented their incomes by trafficking stolen buffalos, cars and even narcotics with the protection of Gondal politicians and their allies. More recently, younger Gondals have been securing jobs in banks and corporations and, since 2008, migrating to Europe, in addition to occupying government jobs. The rest of the households in Bek Sagrana are now occupied by smallholders, sharecroppers7 and landless artisans (kammis) who remain socially subordinate to the Gondals, even though their livelihoods no longer depend exclusively on the landlords. Subordination to the Gondals is now less direct and is based increasingly on intermittent wage labour. Because Gondals are now largely absent from villages, they no longer control their labourers day and night. Relations between Gondals and their servants are now primarily contractual, short-term and increasingly devoid of mutual social obligations. In Breman’s (1974) terms, patronage has turned into exploitation. The power of Gondals is now largely derived from their successful capture of state resources. They are among South Asia’s post-Green Revolution rural elites who, according to Herring (1984), invest in supra-local politics to retain their dominance and bolster their capacity to exploit labour. Kammis constitute 35% of the village population; this proportion is much higher in the deras around Gondal farmsteads where all houses, aside from the landlords’, belong to the kammis.8 Traditionally, kammis rendered their services to groups of Gondal landlords in exchange for a fixed portion of harvested wheat and access to fodder and firewood on their lands. These arrangements, called seypi, resembled the traditional jajmani relations originally described in north India by William Wiser (1936).9 In the past, before mass production replaced most local manufacture and before people started commuting


In the area surrounding Bek Sagrana, the main landowning (zamindar) clans were the Lurkas and the Sagranas, who were involved with Gondals in sharecropping and constituted 15% of the village population (19 out of 123 households). 8 In the dera I lived in, 16 out of 18 households belonged to the kammis. 9 See Kessinger (1974) for a description of these arrangements in a village in East Punjab.

332  Nicolas Martin

to work,10 all kammis who lived in the village were employed by the Gondals and by one another.11 This system was supported by customary law and codified in British statute, which required kammis to provide services to the village as a proprietary body. Kammis who did not wish to render their services on this basis (and owed no debts to the Gondals) had to leave the village; that is, their duties towards the Gondals were based on residence rather than personal liability.12 Arif Hassan has argued that economic change ‘freed the kammis from servitude and because the kammis had marketable skills they improved their social and economic standing’ (2009 [2002], xvi). He further argued that one significant shift that liberated the kammis was the growth of market (mandi) towns resulting from expanding demand for agricultural input and machinery during the Green Revolution. Market towns provided an escape from exploitative, ‘feudal’ power structures, where kammis with sufficient skills and capital could set up businesses, often using their hereditary skills. Work opportunities in cities and better educational opportunities in larger towns gave further impetus to the kammis’ migration out of villages.13


The district centre, Sargodha, is 30 km away from the village. Well into the 1970s buses were not readily available for commuting to the market or work. 11 Kammis included carpenters (Tarkhans), potters (Kumhars), blacksmiths (Lohars), cobblers (Mochis), bakers (Machis), weavers ( Julahas), barbers (Nais/Hajaams), bards (Mirasis), drummers (Pirhains) and sweepers (Mussallis), among others. Barbers, for example, visit their patrons’ households once or twice a week to cut hair and shave the men. They also deliver invitations to weddings and funerals and perform circumcisions. The Mussallis’ principal role is in their patrons’ cattle shed, courtyard and street in front of the house; other duties include preparing mud for the roofs of traditional houses, plastering with mud the chaff and straw collected during the wheat harvest, manufacturing ropes used for making string cots (charpais), and making cooking fires during weddings and funerals. The Tarkhans principally make string cots, fit door frames, make certain agricultural implements and dig graves for funerals. In the traditional village economy all other kammis similarly have a principal occupation and a variety of supplementary roles. Occasionally kammis also owe Gondals unremunerated labour (begar). In addition to serving the Gondals, kammis exchange their services with one another. A Mussalli might help a Lohar to mend his roof in exchange for repairs to his sickles. 12 Rose writes that ‘The village abadi [population] belongs to the proprietary body of the village and the custom assumes that nonproprietors have settled under grants from that body’ (Rose 1911, 217). 13 Muhammad Azam Chaudhary (1999) traces the movement of kammis in Misalpur (near the textile producing city of Faisalabad) from traditional village work to textile factories, the establishment of their own power-looms, and (via education) office work.

The dark side of patronage in the Pakistani Punjab 333

In Bek Sagrana only blacksmith and carpenter kammis managed to turn their traditional skills into successful businesses; among them, only one carpenter household was planning to move permanently to a nearby market town to escape from extortion by Gondals. Most were wage ‘hunters and gatherers’ (Breman 1996) who moved from job to job as daily wage labourers, principally as construction and agricultural labourers, but also as carriers, cleaners and security guards. Mussallis (sweepers)—the largest kammi group with 15 households in the village—as well as Mirasis, Machis and Pirhains, all of whom did not have especially marketable skills, derived most of their incomes from wage labour, which some of them supplemented with earnings from the seypi arrangements.14 These kammis comprised the overwhelming bulk of labourers during the citrus and wheat harvests; they were also most likely to work as the Gondals’ domestic and farm servants, and to become indebted to them (although other kammis and former sharecroppers sometimes did so as well). The main causes of indebtedness were weddings and medical expenses, which they could not afford, because poorly paid and intermittent wage labour made it difficult to accumulate savings. Indebted labourers with able-bodied siblings who could help repay their debts within a couple of years, or even within a few months, but those without siblings were often condemned to exploitative and abusive work conditions for their entire lives (Martin 2009). Even when village kammis prospered, they failed to shed the shackles of subservience to the Gondals. Because the Gondals had successfully captured state power, even relatively prosperous kammis continued to depend on them for access to basic state services. Many Gondals employed armed toughs (goondas) who threatened kammis with violence and eviction from their homes, keeping them subservient and forced to provide Gondals with free labour.15 Thus, some carpenters who had a thriving furniture business and could afford to buy several buffalos and a motorcycle 14

Their seypi arrangements no longer included the duty to sweep people’s houses; when they did this, it was in their capacity as privately employed servants. Villagers were aware of the fact that under seypi arrangments Mussallis had once been the sweepers, but pointed out that this was the case some time before the 1947 partition. 15 This was the case despite Bhutto’s homestead reforms, which granted people ownership over their village houses. See Rouse (1983) and Gazdar and Mallah (2011) for accounts of the effectiveness of the homestead reforms.

334  Nicolas Martin

still had to fix the doors and electrical wiring in the house of a local Gondal landlord without payment. It was in order to escape such duties that they hoped to set up their businesses and move to the nearby market town.

Popular politics and the dominance of landlords Chowdri Nawaz Ali Gondal, the eldest of 5 siblings, owned 70 acres of fine, canal-irrigated land when he became the first Gondal in parliamentary politics. He was the second Gondal, after his father, to obtain further education and to qualify as a lawyer. In a situation where clients most commonly approached their leaders to help them with land disputes and cases involving police stations (thanas) and local courts (kacheri), law provided an ideal entry into politics (Nelson 2011). Having acquired a reputation for being effective, forceful and astute in his handling of legal cases, Chowdri Nawaz Ali was eventually noticed by local leaders of the Pakistan People’s Party (PPP). When I first met him in 2004, Chowdri Nawaz Ali claimed to be a ‘socialist’, even though it had been nearly three decades since he abandoned the PPP for Nawaz Sharif’s conservative, pro-business party. When I asked him how a big feudal landlord like himself could be a socialist, he corrected me and said that he was a member of the middle class (darmiyan tabka), not a feudal lord. He argued that although there were still plenty of ‘feudal’ politicians who owned hundreds, if not thousands, of acres of land, obtaining political office no longer simply depended on status and land ownership. Now politicians had to do ‘people’s work’ (logon ka kaam) by delivering patronage, jobs and infrastructure. Despite what Chowdri Nawaz Ali claimed, it would be misleading to assume that the rise of popular politics meant that political and economic development had made the Pakistani state more responsive to pressure from below. The Gondals largely failed to redistribute the spoils of their power to the poorest members of society, and what little was redistributed did not compensate for the loss of access to essential state services. The main beneficiaries of patronage were the Gondals themselves and other Jat landlords who supported Gondals politically by pledging to them their votes and the votes of their dependants. After Zia-ul-Haq’s military takeover, the Gondals joined the new ruling coalition with a socially and economically conservative

The dark side of patronage in the Pakistani Punjab 335

ideology, making it plain that their commitment to ‘socialism’ was little more than lip service. Over the following years, this shift of allegiance proved highly beneficial to the political and economic fortunes of leading Gondals. After 1985 both Chowdri Nawaz Ali and his younger brother Chowdri Mazhar Ali benefitted greatly from the unprecedented levels of patronage that characterised Nawaz Sharif’s several terms in power, first as the Chief Minister of Punjab and later as the country’s Prime Minister.16 Their personal gain is dramatically illustrated by the fact that by 1999 Chowdri Mazhar Ali owned almost 400 acres of land and several Pakistan State Oil (PSO) petrol stations throughout Sargodha district. When he began his political career in 1985, he owned no more than 70 acres of land. Chowdri Abdullah Gondal, the son of the lambardar 17Ahmed Rasool Gondal, who shared a paternal great-great-grandfather with Chowdri Mazhar Ali, increased the amount of land he owned from 20 acres to more than 400. He did this by obtaining a sand extraction licence from Chowdri Mazhar Ali, trafficking heroin and grabbing land from the onceinfluential Makhdooms and poor peasants. During his three terms in office, Mazhar Ali Gondal obtained a variety of lucrative contracts and a steady flow of development funds for his factional constituency and his biraderi. During his last term in office (1997–1999), Chowdri Mazhar Ali obtained contracts to construct stretches of the motorway from Lahore to Islamabad, engineered by the South Korean Company Daewoo. He took advantage of these contracts himself and granted contracts to his relatives and other supporters who became extensively involved in the project. His younger brother even managed to place one of his domestic servants—a kammi—onto the payroll of Daewoo without the servant ever having to actually work for the company. According to the servant in question, this arrangement allowed his master to


Wilder reports that Nawaz Sharif was able to fill thousands of government jobs with his supporters during his time in office: ‘…he appointed hundreds of loyalist police officers, particularly into the lucrative positions of Assistant Sub-Inspectors and Station Head Officers. This was especially significant as the police play a central political role in Pakistan because of their ability to selectively apply laws in order to harass opponents or to turn a blind eye to the misdeeds of political allies’ (1999, 139). 17 Lambardars were village officials traditionally in charge of collecting revenue and mediating between villagers and the state.

336  Nicolas Martin

employ him at no cost.18 Chowdri Mazhar Ali’s younger brother, nephews, cousins and other allied Jats also received several local road construction projects. The state of these roads only a few years after their completion suggests that the contractors spent far less on their construction than they had invoiced. By 2004, the metal road running alongside a canal near Bek Sagrana, built only 7 years earlier, was already thoroughly pitted with potholes and its large parts had collapsed into the canal. Chowdri Mazhar Ali was also granted several franchises to build the PSI petrol stations along the main roads leading to Sargodha. During Chowdri Mazhar Ali’s terms in office, the village of Bek Sagrana received a large number of development projects, the building contracts for which were all allocated to Gondals. The village acquired a post office, a water tower, a basic health unit (BHU), a new school and a community centre. Neither the water tower nor the post-office was ever used and the BHU was not operational. The water tower was part of a scheme originally meant to connect households to a water supply system in areas with water scarcity and salinity problems. Bek Sagrana did not suffer from any of those problems, and most people got freshwater from hand pumps in the courtyards of their houses. No one in the village opted for connections to the new water system because they saw no reason to pay monthly bills. The post office housed one of Mazhar Ali’s former employees. The BHU, like many other health units I visited throughout the district, included a hospital building with two operating rooms for minor surgeries like appendicitis and childbirth, and a medical dispensary. In addition, it had housing for hospital staff and a large two-story residence for the doctor. The doctor, a wealthy Gondal landlord who managed to get a position in the BHU of his own village through connections to Chowdri Mazhar Ali, never attended to his duties, only paying it occasional visits to falsify the attendance register.19 The BHU was occupied by his maternal cousin’s family, who was also Chowdri Mazhar Ali’s nephew. A few rooms meant for the hospital staff were occupied by Chowdri Mazhar Ali’s servants, one of whom used a room to keep broiler chickens, which he 18

The servant even told me that his salary from Daewoo was quite high and that his employer kept part of it for himself. 19 Chowdri Nawaz Ali was married to the doctor’s maternal aunt.

The dark side of patronage in the Pakistani Punjab 337

sold in the village. The clerk in charge of the medical dispensary allegedly made money by selling medical supplies from the health unit. As a result, villagers continued to rely almost exclusively on local healers (hakeems) and dubious practitioners of allopathic medicine.20 In cases of serious illness, when poor villagers needed medical treatment in expensive private clinics in town, they sought loans from Gondals. In general, only servants who maintained close service-ties with the Gondals were ever likely to receive medical care at no cost. Poor voters could also occasionally obtain medical treatment, cash, a Gondal’s favourable intercession in a court case or even a government job, but such brokerage did not compensate for a nearcomplete lack of public access to state resources. The appropriation of the spoils of power by politicians did not make state resources more accessible to people, as Lieven (2011) and Lyon (2004) had claimed. Instead of access to healthcare, communal spaces, sound infrastructure, education and justice, poor villagers received cash payments—money that rarely lasted more than a week—in exchange for their votes. Access to public resources became the charity of local power-brokers. The lucky ones on the ‘bright side’ of patronage could obtain low-level government jobs. But they did this in ways that undermined the delivery of public service to most of the population. Although some landless villagers did obtain government jobs through their Gondal patrons, generally, Gondals prevented villagers from empowerment through jobs, improved access to education and other public resources. Villagers repeatedly told me that Gondals did not want them to get good jobs and improve their lot in life because they would no longer be able to use young men as their gunmen.21 Time and again they told me that, whenever the Gondals noticed an ordinary villager wearing nice clothes or using a new mobile phone, they grew annoyed as they felt they were losing their ‘slaves’ (ghulaams) and responded by mocking and humiliating the villagers. The Gondals, in turn, claimed that nowadays kammis want to be kings (badshahs) and that they no longer wish to do respectable, 20

Most BHUs I have visited in rural areas were not fully operational, and doctors employed to work in them came to collect their salaries once a month. 21 Whenever the Gondals engaged in land disputes, or when they contested elections and quarrelled with rivals, they demanded that their tenants and labourers pick up arms and fight for them.

338  Nicolas Martin

hard work, but instead want to lounge about in ‘flashy clothes’ and make money ‘the easy way’, by running shops or by stealing and  bootlegging. The only sweeper (Mussalli) in the village to have completed high school education told me that the Gondals could not bear the fact: one had mockingly asked whether he thought he was a ‘big man’ (baraa aadmi), now that he spoke English with me. Several Gondals told me that education only planted unrealistic career expectations in the kammis’ minds and made it more difficult for Gondals to obtain labourers.22 In any case, they did not believe that kammis could reap the benefits of education because they were inherently incapable of reason and hard work. They  complained that nowadays the only way they could lure labourers was by providing them with small consumption loans or larger loans to cover weddings and medical expenses (Martin 2009). While giving little support to secular education, the Gondals focused on proselytising Islamic education meant to morally ‘uplift’ the poor and raise the Gondals’ status and spiritual merit (savaab). Instead of sending their child servants, working to repay their parents’ debts, to a secular school, they gave them basic home schooling and taught them how to read the Qu’ran. They built mosques and paid preachers to teach village children about Islam, but made no effort to improve the village primary school. One village schoolteacher told me that because the Gondals increasingly educated their children in town, teachers had little incentive to teach in villages. Previous generations of Gondals had received their primary schooling in villages, and their parents ensured that teachers turned up to schools. This is no longer the case. Teachers are now often absent and regularly get away with falsifying attendance registers. Even when they are around, teachers spend significant amounts of time chatting to each other, drinking tea and performing prayers in the village mosque. During class, they often send children on errands to buy biscuits or cigarettes for them. When the village school master asked Gondal landlords to impose absentee fines on the teachers, they made an initial show of goodwill, but nothing came of it because Gondal power brokers were more interested in teachers’ electoral loyalty than in village education. Teachers tend to belong to the landed, rural middle class, and unlike 22

A wealthy landlord and hereditary saint (pir) from a district in South Punjab once boasted to me that he did not let schools operate in his region because they created unrealistically high career and earning expectations among labourers.

The dark side of patronage in the Pakistani Punjab 339

poor villagers their votes are more difficult to purchase or force. They expect politicians to help them obtain and retain their jobs, prevent their transfers to undesirable locations, obtain promotions and even relieve them of having to work at all. In one instance, Chowdri Mazhar Ali helped to avoid disciplinary action against a Gondal schoolteacher who often did not turn up to school because he ran his own school in a nearby village. Absenteeism and careless teachers were in large part to blame for the 18.5% literacy rate in Bek Sagrana reported in the 1998 census. The Gondals were likewise averse to helping the few kammis with the requisite educational qualifications to obtain low-level jobs as clerks or guards. Chowdri Mazhar Ali, who was thought to be more forward-thinking than other Gondals, managed to obtain petty government jobs as guards for 4 smallholders and 3 kammis during his time as a Member of the Provincial Assembly; he had nevertheless faced resistance from his relatives anxious that this would deprive them of cheap labour. In one widely discussed case, Chowdri Mazhar Ali obtained a ‘class 4’ job for a blacksmith (Lohar) in the Ministry of Agriculture, but because his nephew wanted the blacksmith to remain a labour supervisor in his brick kiln, he withdrew the offer. In half of the cases where Chowdri Mazhar Ali obtained government jobs for poorer villagers he had done so to acquire free servants for himself and his close kin: the villagers in question were on the payroll of the state but worked for Gondal landlords as domestic servants, labourers and overseers. One Rajput smallholder and a kammi bread maker (Machi) were on the payroll of the Ministry of Agriculture but actually worked as overseers on Chowdri Mazhar Ali’s estate. Two smallholders and one kammi employed as watchmen on the premises of the BHU spent a great deal of their time serving one of Chowdri Mazhar Ali’s nephews who used the bungalow meant for the doctor as a weekend retreat.23 As I mentioned earlier, a landless member of the Lurka biraderi had been on the payroll of Daewoo for several years while actually working as a driver and farm servant for Chowdri Mazhar Ali’s younger brother, Chowdri Arif.


The bungalow was a ‘bachelor pad’ where he regularly hosted friends over drinks, and occasionally dancing girls from Sargodha.

340  Nicolas Martin

In addition to undermining public service delivery, patronage fostered criminality. Gondal politicians frequently intervened in the running of courts and the police on behalf of criminals they employed as toughs. Because they protected buffalo thieves, people had to sleep next to their livestock at night. Because they protected heroin traffickers, the village became a converging point for heroin addicts who stole anything that came their way. Throughout my stay, bicycles, clothes and agricultural tools were being constantly stolen by heroin addicts. The involvement of patrons in dispute resolution undermined impartiality in the delivery of justice in both formal and informal forums (Chaudhary 1999) and meant that people with the ‘right connections’ could get away with a wide range of illegal and socially unacceptable behaviour. In one instance, a Lurka who was widely known to beat his wife when drunk, escaped prosecution because he had good relations with the head of one of the village factions. His brother-in-law spent a great deal of time and money in order to file a First Information Report (FIR) against him with the police, but was forced to withdraw it after being threatened with eviction by landlords allied with Chowdri Abdullah, on whose land he had built his house. Like many other villagers, he complained that Gondals were only concerned with consolidating their power and had no interest in justice (insaaf). Finally, patronage generated division and mistrust among villagers because it forced people to expend great efforts on ingratiating themselves with the landlords and snitching on one another. Whenever a theft took place in the village, there was never a shortage of those willing to convey ‘information’ about its perpetrators to the Gondals. This gave rise to widespread mistrust and suspicion among the village poor and to claims that Bek Sagrana was a bad place because it was full of snitches (chugal khor). From the moment of my arrival, I was repeatedly told that villagers could not be trusted and that they were rotten (kharaab) and fraudulent (do nambar).

Why elections did not improve political accountability At the time of fieldwork, the majority of villagers did not think their votes  to be effective bargaining tools for obtaining the patronage of Gondal politicians. Aside from electoral rigging by authoritarian governments (Waseem 2006, 24), severe political and economic

The dark side of patronage in the Pakistani Punjab 341

inequalities and the absence of political parties that genuinely represent popular interests meant that elections in rural Pakistani Punjab could hardly be described as competitive and as occasions when poor voters could press their demands and hold politicians to account. Poor voters’ right to political representation was curtailed by the fact that the only people who could contest elections were those with money, political influence and muscle power. This was— and remains—partly because political parties relied on the private fortunes of their candidates to finance electoral campaigns. When (in 2005) I asked kammis why none of them considered contesting elections for the seat of the Union Council Nazim (the lowest seat in the electoral hierarchy), they thought my question ridiculous. While these elections were ostensibly meant to empower the poor, with General Musharraf declaring that they would cause an unprecedented transfer of power ‘from the elites to the vast majority’ (ICG 2004, 5) of poor villagers, it was obvious that in order to contest elections candidates needed influence to broker access to state resources, money to fund expensive election campaigns (in which votes were purchased with cash and minor infrastructure projects) and muscle power to fight rivals.24 Kammis have none of these, and there is no coherently structured, well-funded political party that may support their bid to contest elections.25 Poor people’s access to political representation was even more directly blocked by the fact that landlords instructed them on how to vote—and that these instructions were backed by threat of eviction, unemployment and even physical harassment. Customary practices whereby kammis and villagers owed their services and political allegiance to the dominant Gondal biraderi, remained in place despite Bhutto’s attempts to reform land ownership in the 1970s. When a kammi settled in the village or in a nearby dera, he generally did so on the understanding that he would have to provide the local landlord with labour and votes. During the 2005 elections to local council, Gondal landlords gathered all of their kammis and instructed them on how to vote. The threat of eviction meant that they complied, in part because they knew that 24

In 2005 I was told that an election campaign for a provincial assembly seat cost around 600,000 rupees and that a campaign for the seat on the Union Council Nazim cost around 80,000 rupees. Among other things, money is used to purchase votes, build minor infrastructure projects and buy fuel for campaigners’ motorcades, billboards, posters and leaflets. 25 Political parties tend to rely on the candidates’ private fortunes.

342  Nicolas Martin

the Gondals used spies to find out people’s voting intentions and to look over people’s shoulders as they cast their ballots inside polling booths. Eviction was a real threat. Following the 2005 local council elections, a Mussalli family failed to vote according to the instructions of the Gondal landlord on whose land they had built their house and was consequently evicted. Similarly, some smallholders who used a government-built wedding hall as a petrol station were also evicted  when they were found to have voted against the dictates of the landlord on whose land the wedding hall had been built. Finally, the types of activity that made the electoral process ineffective included various forms of pre-poll and polling day rigging. Candidates used their influence and muscle to hassle voters, opponents and even polling officers. They used connections to place their own polling officers and policemen ready to overlook irregularities such as multiple voting, and violations of the secrecy of the ballot box. Although a great deal of the rigging reported to have taken place during the Union Council Elections in 2005 was instigated by the central government to benefit pro-government candidates, powerful opposition candidates also meddled with elections, using their influence, political connections and muscle to intimidate rivals, voters and government officers. Chowdri Mazhar Ali Gondal—who remained loyal to Nawaz Sharif’s opposition Pakistan Muslim League Noon (PML-N) and was jailed for it following General Musharraf’s military coup in 1999—used his elder brother’s connections in the judiciary to put in place a polling officer willing to overlook multiple voting by his Gondal supporters in the village. Where he was unable to obtain favourable polling officers, his   heavily armed supporters managed to intimidate presiding officers into overlooking irregularities in the polling process.26 To prevent electoral irregularities, Musharraf’s government is reported to have attempted to position the polling stations in territories controlled by pro-government politicians, or at least in places where 26

In Bek Sagrana the police officer in charge of the two polling stations did not wish to antagonise either Chowdri Mazhar Ali or Chowdri Abdullah and their armed supporters; during elections he spent most of the time eating and drinking inside a tent some distance away from the booth. After the 2008 election to the provincial assembly, I heard that in a neighbouring village government officers were so frightened of the heavily armed supporters of PPP, that they allowed them to intimidate voters and cast multiple ballots.

The dark side of patronage in the Pakistani Punjab 343

pro-government and opposition supporters were equally represented, so that neither side could capture the polling booth (see ICG 2005). The dominance of landlords among electoral candidates combined with pre-poll day and polling day rigging meant that elections failed to empower the electorate and help substantially redistribute state resources.

Conclusions In Pakistan clientelistic exchanges reinforce existing power structures and undermine popular freedoms. The dark side of political society is here much larger than Chatterjee (2004) concedes. Instead of protecting the masses from dispossession, clientelistic transactions are integral to the processes of dispossession, as authors examining links between capitalism, corruption and criminality have already shown.27 Thus, while Chatterjee is in part right to see clientelism as a response to popular need, he fails to acknowledge adequately the ways in which it exploits the citizenry and the electorate to reproduce elite power. He also fails to acknowledge that patronage undermines public service delivery. By endorsing political society as the principal means through which popular freedoms can be extended in post-colonial states, he undermines the contribution  to  democratisation of grassroots and rights based movements which emphasise institution-building and legislation, and which have played a particularly salient role in India’s democratic process. It may be said, in defence of Chatterjee, that while such political exploitation may persist in ‘feudal’ and authoritarian Pakistan, it has been vastly reduced in democratic India where politicians have been forced to become more accountable through consistently and frequently held elections. Illustrating this point, Steven Wilkinson (2011) describes the splits within Congress after the 1960s and the emergence of new parties (representing businesses, landlords, farmers, regional linguistic groups and communists), which intensified political competition and forced politicians scrambling for votes to increase the supply of patronage to their constituents. In villages such competition, sometimes combined with rising living standards, resulted in political parties’ increasing provision of goods like clean 27

See Arlacchi (1983; 1986), Gambetta (1993), Handelman (1994), Kang (2001), Varese (2001), Volkov (2002) and Herzfeld (2009).

344  Nicolas Martin

drinking water, housing plots, credit and subsistence wages for their voters. These achievements may give reason to celebrate political society. I would argue nevertheless that Indian democracy has deepened despite political society, not because of it. If India’s subalterns have benefitted from democracy, it is mainly because ‘elections occur regularly, democratic institutions are strong, people are free to protest and civil rights are guaranteed. The media is lively and social movements are tolerated by the state’ (Corbridge, Harris and Jeffrey 2013, 157). The success of Indian democracy also arguably relies on the fact that many of the directive principles contained in the Indian constitution, including the right to work, education and healthcare, have been upgraded to fundamental rights by Supreme Court judgements (Birchfield and Corsi 2010, 713). What  has  impeded  democratisation by undermining the effectiveness of public institutions is the ad-hoc nature of political society embodied in ‘the expansive network of patronage founded on caste and class inequalities’ (Corbridge, Harris and Jeffrey 2013, 157). In Pakistan patronage networks headed by powerful political brokers divert state resources away from the general public and make people’s legal entitlements contingent on political loyalties. Patronage networks further fragment the subaltern classes by co-opting their different segments and thereby preventing the emergence of strong opposition to dominant classes. As Alavi (1971; 1972b) and Mohmand (2008) have shown, the landless seek to attach themselves to powerful patrons instead of organising themselves politically to defend their interests. This means that there is little independent associational activity among them. The presence of powerful political brokers also limits meaningful political participation, leaving the poor with fewer opportunities to complain about government services or become involved in politics even at the local level. In Pakistan it is the alliance between successive military regimes and the landed class that has perpetuated the rule of political society and prevented the emergence of rights-based movements which have aided democratisation in India. General Zia-ul-Haq, for example, played an important role in forestalling the emergence of popular movements which could have challenged landed power (Waseem 1994; Jalal 1995). And, as Javid Hassan (2011) has shown, landed

The dark side of patronage in the Pakistani Punjab 345

power in Pakistan has remained intact largely because military and authoritarian governments have repeatedly allied with factions of the landed political class.28 The rule of ‘political society’ in Pakistan is a product of this unholy alliance—which perpetuates inequalities and diminishes politicians’ accountability—not a sign of deepening democratisation.


Alavi (1972b) argued that the Pakistani post-colonial state acted as a mediator between the often contradictory interests of capitalist and feudal factions within the ruling class. Hassan also recently argued that over time the relationship between state and ‘the landed class grew to become one of mutual structural dependence’ (2011, 358).

Patronage and printing innovation in 15th-century Tibet


Hildegard Diemberger


any of those who write about patronage in modern politics assume that it is a backward, pre-modern, traditional, hierarchical and fundamentally conservative institution that thwarts innovation and change.1 This chapter gives a case of the opposite. Focusing on the introduction of printing in 15th-century Tibet, it shows how relations between lay rulers and their spiritual preceptors, as described by Seyfort Ruegg (2004; Chapter 2 in this volume), contributed to what may be called a ‘Tibetan printing revolution’, or the rapid diffusion of block printing in the 15th century, which had far-reaching social and cultural consequences (Diemberger 2012, 18–39). I argue that ‘patronage’, articulated in the widespread IndoTibetan formulation as the bond between lay donor and monastic donee, was at the heart of innovation, even in a society usually seen as subject to the moral authority of the past and ancient traditions. In the context of book-related patronage and technological innovation in Tibet, this was as true in the 15th century as it continues to be in the 21st. The relation between lay donors and monastic donees is central to the advancement of Buddhist doctrine, for which the former provides material means. In Tibet this relation takes on a distinctive shape and plays a special political and religious role when it involves politically powerful donors. Introducing a volume on patronage in Tibet, Ruegg observed that a close association between the religious and the secular— the spiritual and the temporal—has characterized polities [in Tibet and Inner Asia]. And a special kind of very close 1

This paper is based on research carried out in the context of the project ‘Transforming Technologies and Buddhist Book Culture: the Introduction of Printing and Digital Text Reproduction in Tibetan Societies’, funded by the Arts and Humanities Research Council UK (RG 55631).

Patronage and printing innovation in 15th-century Tibet 347

co-ordination, or dyarchy, of orders and functions—one exercised together by a lay ruler-donor (yon bdag) and his religious counsellor-donee and preceptor-officiant (mchod gnas, Skt. daks. in.-1ya)—has thus been a regularly recurring feature of these polities. This traditional association of a religious preceptor-officiant and a ruler by (religious) law (chos rgyal = dharmara-ja) is expressed in Tibetan by the copulative compounds mchod yon and yon mchod (2004, 9).

While it is clear that the two parties involved in this relation are mutually dependent, the difference in status between them is not set in stone. Although in Buddhist principle spiritual preceptors stand above their lay donors, in the practice of life, they are part of political hierarchies headed by royal or imperial donors. Strategic manipulations of this ambiguity by both sides had far-reaching consequences, most strikingly obvious in the contrasting readings of the history of Tibet and its imperial rulers (Sperling 2004). In times of political fragmentation, the negotiability of status built into this relationship gave scholars and artists much space for manipulating their circumstances and for making significant technological, aesthetic and literary advances in what could otherwise be perceived as a context dominated by re-enactment of antiquity.

Figure 15.1 Map of southwestern Tibet.

348  Hildegard Diemberger

Books as relics and merit-makers In Buddhist narratives the idea that books embody the Buddha’s speech, and are therefore sacred objects, is born at the very moment of  Gautama Buddha’s demise. The Maha- parinibbannasutta, the early Buddhist scripture describing the end of the Buddha’s life, relates: Then the Bhagvan [Lord Buddha] addressed the venerable - nanda [his favourite student]: ‘It may be, A - nanda, that A some of you will think, “The word of the Teacher is a things of the past; we have no Teacher”. The Doctrine - nanda, which I have taught and and the Discipline, A enjoined upon you is to be your teacher when I am gone’ (cited in Warren 1985 [1896], 107).

The Perfection of Wisdom literature elaborated on this theme, promoting what over time grew into a true cult of the book.2 This finds expression in the following early verses: … when, through the Tatha-gata’s sustaining power, it has been well written, in very distinct letters, in a great book, one should honour, revere, adore and worship it, with flowers, incense, scents, wreaths, unguents, aromatic powders, strips of cloth, parasols, banners, bells, flags and with rows of lamps all round, and with manifold kinds of worship…One should know that those beings are living in the presence of the Tatha-gata who will hear this perfection of wisdom, take it up, study, spread, repeat and write it, and who will honour, revere, adore and worship it (Prajña-pa-ramita-, Conze 1973, 299–300).

In fact, in the Maha-ya-na literature, including centrally the Perfection of Wisdom and related works, the cult of relics and the cult of the book were virtually indistinguishable. Texts could be treated as relics, worshipped and placed inside Buddhist reliquaries (stu- pas). The relevant acts could be multiplied, re-enacting, when performed by rulers, the virtuous deeds of the first Buddhist Virtuous Ruler (cakravartin) As´oka, who was known for distributing relics and 2

Emerging in the 1st and 2nd centuries CE, the Perfection of Wisdom literature became central to Maha-ya-na Buddhism and remains the source of some of the most widely reproduced Buddhist scriptures.

Patronage and printing innovation in 15th-century Tibet 349

inscribing pillars with sacred texts. The proliferation of books was seen as a merit-generating activity: the accumulation of meritorious deeds that leads to a better rebirth and ultimately to enlightenment. The understanding of text as a relic and Buddhist book-related meritgenerating practices were thus also linked to the development of early printing in China, Korea and Japan (cf. Strong 2004; Schopen 2005, 309–311). In his book, The Woman Who Discovered Printing, Tim Barrett (2008) described how the controversial Chinese Empress Wu, at the end of a life in which she had much to atone for, ordered 100,000 copies of the ritual text, Great Spell of Unsullied Pure Light, to be printed and distributed throughout the empire. Barrett argues that the empress’s distribution of books in a bid to generate merit and enact the role of the righteous ruler, as she worked to establish a dynasty of her own, led to the discovery of printing as early as the end of the 7th century. The same treatment of books as relics, artefacts and ritual objects,3 I argue, underpinned the production of the earliest Tibetan prints. Printing was popular among Buddhists along the Silk Road, where printed editions of Buddhist texts were used as gifts in diplomatic exchanges between polities (e.g., Dunnell 1996). The earliest extant Tibetan prints do not come from Tibet proper. In the 12th century Tibetan texts (the so-called Hor prints) were produced by the Tanguts, and in the 13th and 14th centuries in Beijing under Khubilai Khan (and his successors) with the guidance of Tibetan curators.4 Some of these prints were later taken to Central Tibet, and an increasing number of them has been discovered recently by Tibetan scholars (Shes rab bzang po 2009, 41–50). Although we find mentions of a print in eastern Tibet as early as 1207 (van der Kujip 2010, 441ff), so far the text printed in 1407 at Shel dkar5 remains the earliest extant print, which was produced in Tibet and which by a few years predates the printed works of the great Buddhist teacher Tsong kha pa (1357–1419), which were previously regarded as earliest (Ehrhard 2000, 11; Jackson 1990, 107ff.). 3

The conception of sacred texts as relics is also reflected in the way that Tibetan books are treated as holy and honoured persons. This is expressed in the vocabulary and honorific terms used to address them: they are traditionally wrapped in robes (na bza’), tied with belts (sku rag) and invited (spyan ‘dren) to events and places. 4 Kurtis Schaeffer (2009) mentions a 1153 Tangut print found at Kharakhoto. For a discussion of Tibetan prints produced by the Tanguts, see also Shen (2010, 337ff). 5 This was found last year by a Tibetan colleague, Gyutog Dawa, among the Bo dong pa texts which had been taken from Shel dkar to sNye mo in the 17th century.

350  Hildegard Diemberger

13th and 14th-century Tibetan prints, such as those said to have been made by U rgyan pa (1230–1309) after he had been involved in the printing of Tibetan texts at the Yuan court, may emerge eventually.6 In any case, these would have been much rarer than those brought from Beijing, which is probably why none have surfaced as yet. By the 15th  century, the popularity of printing spread with remarkable speed. Within a few decades of the introduction of printing technology, the printing of editions of Buddhist texts became an important meritmaking activity for rulers seeking to enhance and consolidate their profile, as well as for lay and monastic networks.

A witness to the Tibetan printing revolution In 1407 the printing of Haribhadra’s commentary on the Abhisamaya. alanka-ra (Ornament of Clear Realisation[s]) was sponsored  to commemorate the death (in 1402) of Si tu Chos kyi rin chen, ruler of Southern La stod,7 by his son Si tu lHa btsan skyabs.8 This popular text, which elaborated on the Perfection of Wisdom and was authored by an 8th-century disciple of S´a-ntaraks.ita, the Indian scholar celebrated for introducing Buddhism to Tibet, was and still is today an important part of the standard monastic curriculum in Tibet (Dreyfus 2003, 174–182). It is in the format of a pothi with 90  folios (some now missing) and many of the physical features of a manuscript.9 In the centre of the text, there are two circles (see  Figure  15.2) imitating the rings on older paper manuscripts, which mimicked the palm leaf Buddhist manuscripts originally brought to Tibet from India and tied  together with strings passing through two decoratively encircled holes.


Grub thob U rgyan pa (1230–1309) was a great adept of the bKa’ brgyud school of Tibetan Buddhism, particularly renowned for his knowledge of the Ka-lacakra Tantra. He was born in an area located between Southern La stod and Gung thang. He was the student of the second Karma pa and recognised the third Karmapa as his teacher’s reincarnation. He travelled widely and visited the Yuan court where he was involved in the printing of Tibetan texts. 7 Southern La stod (also La stod lho or lHo phyogs) in south-western Tibet was one of the administrative units, called myriarchies (khri skor), during and after the Sa skya-Yuan period (1271–1368). It corresponded to present-day Dingri County in the Shigatse Prefecture, just north of Mount Everest (see map in Figure 15.1 above). 8 This text reproduces The Commentary That Clarifies Meaning (‘Grel ba don gsal ba). 9 A pothi is a bundle of elongated, loose sheets of paper held together between two wooden covers.

Patronage and printing innovation in 15th-century Tibet 351

Figure 15.2 The 1407 print of the sBas gdon gsal ba.

Figure 15.3 Block printing at the Sera monastery in Tibet. The text includes an extensive colophon celebrating the land and the ruler of Southern La stod and gives some details of the production of the print edition. After a lengthy description of the celebrations,

352  Hildegard Diemberger

which took place on the work’s completion, the colophon concludes with the following passage: Auspicious garlands of flowers were raining [from the sky]. Beautiful girls were singing melodiously: There is no obstacle in the happiness of the myriarchy (khri skor)!10 Glory was rising amidst the feast; higher than the mountain of pride was the purest among the human rulers. This had become the true navel of  the land of wealth. Likewise, the printing that spread 100,000s of texts like rays of light dwarfed the sun. By whatever power the doctrines of the Buddha could uplift from dullness, such was the power of the ruler of the Southern Regions, endowed with the deeds of the ancestral kings of the past. From the palace of Shel dkar this Commentary that Clarifies Meaning like the sun and the moon, [a text] said to convert everybody, was printed in the year of the fire female hog (1407), in the month of S´ravana (gro zhun, July-August) thanks to the endeavour of the official Jo bo rgya ‘dar. It shall spread in every direction, at all times and on all occasions.11

A short manuscript called the Outline of the Great Situ [lHa btsan skyabs]’s print edition of the Short Commentary that Explains Meaning12 reproduces the same text with an added remark that this outline was part of the collected works of the great master Jigs ‘bral, Victorious in All Directions.13 This great master was the young 10

A myriarchy (khri skor) is a Mongol–Tibetan administrative unit. During the Sa skyaYuan period, Tibet was subdivided into 13 myriarchies, literally ‘units of 10,000’, following the Mongolian Model. 11 bkra shis me tog char ‘phreng ‘bab pa’i tshe//bu mo mdzas ma’i mgrin snyan sdebs pa las//khri skor bde skyid bar chad med do zhes//dga’ ston dbus su snyan pa ‘phel bar ‘rgyur//nga rgyal ri bo mtho ba’i mi bdag dag; nor ‘dzin sa’i lte ba mad gyur kyang//’bum phrag gzhung gi ‘od zer kun bgyed pa’i spar byed rta ljang rol ba’i nyi ‘dra nyung//stobs chen gang gis sangs rgyas bstan pa dag//’ngam du bying ‘dra gyen du rab ‘degs pa// mes dbon sngon rgyal po’i rnam thar can//lho phyogs rgyal po’i dbang phyug gang ‘di yin// dpal shel dkar gyi pho brang nas// nyi zla ltar grags pa’i sgrel ba gdon gsal ba ‘di / nang gnyer jo bo rgya ‘dar gyis/gzo bo’i zhabs tog brtson bar bsgrubs/thams cad ‘dul zhes bya ba/ me mo phag gi lo; gro zhun gyi zla ba la spar du bsgrubs pa ‘dis; phyogs dang dus dang gnas skabs kun du kyab par gyur cig/ (folio 90r) 12 Ta si tu lha btsan skyabs kyi ‘grel chung don gsal par du bsgrubs pa’i dkar chag tshigs bcad la. 13 This great master was most commonly known by the name of Bo dong phyogs las rnam rgyal (the founder of the Bo dong pa tradition, literally ‘The Victorious

Patronage and printing innovation in 15th-century Tibet 353

abbot of the Shel dkar monastery and the spiritual preceptor of the ruling family of Southern La stod, and was apparently extensively involved in the production of the print edition. In the following decades, he would become one of the most prolific polymaths in Tibetan history. The production of the print edition described in the ‘mission statement’ follows the prescription contained in the final lines of the Perfection of Wisdom: to honour, revere, adore and worship the book. The new technology allowed this to be followed in an unprecedentedly effective way. More strikingly, however, this print is a departure from the traditional production by Tibetan rulers of lavishly decorated copies of the Perfection of Wisdom and other popular canonical texts as ritual objects. In this case, although the printing operations were eventually eulogised in ritual terms (including the sacred number 100,000), it is also clear that numerous people involved in the production engaged with the content and used the text for teaching and learning purposes. But first let me briefly describe the protagonists.

A devout innovator in pursuit of legitimacy and power In the early 14th century Tibet was ruled by the combined powers of the Sa skya Buddhist sect and the Chinese Yuan Dynasty. Marrying Mongolian imperial power and Tibetan spiritual guidance, this was the most famous donor-donee, mchod-yon, relationship in the history of Tibet. Political control in Tibet was then exercised by a local authority called Sa skya dPon chen, after its seat in Sa skya in Southern Tibet. Si tu Chos kyi rin chen (c. 1330–1402) was the son (possibly illegitimate) of dPon chen Don yod dpal bzang who had died in the 1330s on his way from Tibet to China, just after establishing a new capital for his myriarchy of Southern La stod, in a place called Rin chen spo (Everding 2006, 96–101). The shift of the capital from Ding ri sGang dkar14 to the area a few miles north of the hill (where the Shel dkar fortress was later built) [master] in All Directions from Bo dong’). The Bo dong E monastery, established in 1049 west of Shigatse, in central Tibet, was considered the main seat of the Bo dong pa tradition, rooted in the corpus of scriptures compiled by him and known as the dPal de kho nan yid ‘dus pa, or the ‘compendium of reality’ or even as an ‘encyclopaedia tibetica’. 14 Ding ri sGang dkar is a town in the Ding ri area currently also called Old Dingri. It is some 100 km west of Shel dkar, currently known as New Ding ri.

354  Hildegard Diemberger

is described in The History of Shel dkar (Shel dkar chos ‘byung)15 as necessitated by the need for its protection from the incursions of northern neighbours and by its proximity to Sa skya and other areas on which the king levied taxes. The move is also said to have followed a drought and was attended by an intense construction of irrigation systems and the creation of new fields.16 With the demise of the Sa skya-Yuan rule in 1368, the Phag mo gru pa took loose control of Tibet, which became an arena of competing polities, each wrestling to control the others. The confederacy of the 13 Tibetan myriarchies, which was consolidated under the Sa skya-Yuan regime, became the playing field for a new power game (Everding 2006, 93). Northern and Southern La stod were two of these myriarchies, which evolved into autonomous polities whose relations with each other (and their neighbours) vacillated between warfare and marriage alliances. The more powerful Northern La stod vied, with intermittent success, to control Southern La stod, which in turn struggled to maintain its autonomy. Despite the controversial circumstances of his birth, and perhaps because his character might have been strengthened by this challenge, Si tu Chos kyi rin chen brought to Southern La stod a period of relative autonomy and prosperity, and became a celebrated hero in the dedicated narrative, The Thirteen Deeds of the Situ.17 Defeated and taken prisoner by the army of the Northern La stod’s ruler, he miraculously escaped death, married the enemy ruler’s daughter and eventually returned home, triumphant. Here he established a new capital on a steep hill which became known as Shel dkar. The new castle gave Southern La stod better protection from its northerly neighbour (Figure 15.4). And by expanding its nomadic areas and improving trade, the polity entered a new era of prosperity which was eventually reflected in the establishment and expansion of Buddhist sites (Everding 2006, 102). The most important of these was the Shel dkar chos sde monastery established in 1385 under the guidance of 15

The Shel dkar chos ‘byung is a religious history of the area corresponding to Southern La stod (present-day Dingri County, Shigatse Prefecture, Tibet Autonomous Region). The text was composed in 1732 by Ngag dbang skal ldan rgya mtsho, who collated earlier historical sources. For a facsimile edition and a translation of the document, see Wangdu and Diemberger (1996). For a discussion of this source, see also Everding (2006). 16 According to Sinha et al. (2010, 1–16), a mega-drought with weak monsoon affected South Asia over most of the 14th and the early-15th centuries. 17 This 15th-century narrative is reported in the Shel dkar chos ‘byung (folio 8v–34v).

Patronage and printing innovation in 15th-century Tibet 355

Figure 15.4 The Shel dkar (Shekar) fortress, the capital of La stod lho (Southern Lato). the great translator Grags pa rgyal mtshan (1352–1405). Shortly afterwards this translator established a school of philosophy (bshad grwa) where the monastic community was to be trained. Providing full support to the monks was a greater material commitment than supporting the numerous temples he had already established, and the Shel dkar chos ‘byung details the food and supplies that had to be provided on a regular basis. Enormous expense required by the project is also highlighted in the chronicle (folio 33v): ‘One third of the revenue of Shel dkar chos rdzong [castle] was offered for the provision of services and honours to sustain the monks for the summer and winter religious festivities in the great monastery of Shel dkar and to support the worship of the Jewel [of Buddhism] in the temples in general’.18 All this was achieved through rule according to both religious and civil laws: ‘The above-mentioned deeds are of a religious nature, but a lord (mi rje) who rules according to both [religious and civil] laws (lugs gnyis) requires perfect majesty (mnga’ thang). For this reason, the outer majesty [made of] horses, mules, mdzo [a hybrid of yak and domestic cattle], yaks, goats, and sheep 18

The Three Jewels of Buddhism, or the three articles of Buddhist faith in which they ‘take refuge’, are the Buddha, the Buddha’s teaching and the monastic community.

356  Hildegard Diemberger

had to be increased’ (folio 33r). Accordingly, Si tu Chos kyi rin chen gave tax breaks to people who owned many animals; this worked well as an incentive and the kingdom’s pastoral wealth significantly increased. Thriving agriculture, supported by the irrigation system, expanded the granaries and passes, and pathways were opened in all directions, encouraging thriving trade. The idea of an interlinked inner and outer ‘majesty’, of spiritual and worldly force, builds on the ancient Tibetan political notion of ‘majesty’—mnga’ thang—which could also be glossed as ‘power’, ‘might’, ‘dominion’ or ‘authority’, and which can already be found in 6th to 9th-century dynastic sources describing ancestral and territorial cults. Overlaid by the Buddhist ideal of the Virtuous Ruler, the cakravartin, and the political relevance of merit-generating activities, the notion of ‘majesty’ was reframed in light of the rule of ‘the two laws’—the spiritual and the worldly—and of the donorpreceptor relationship which reflected that. Adding to, and sometimes competing with, the memory of ancient Tibetan emperors, Khubilai Khan (1215–1294) was the model Righteous King (dharmara-ja) who ruled according to the two laws, inspiring local rulers to emulate him long after his death. The Mongol emperor and his family had harnessed and developed the many legacies of the people they integrated into their empire, including Tibetan Buddhism and the culture of print. Thus, in the colophon of the 1407 print, Situ Chos kyi rin chen was celebrated as having come as ‘the jewel in the crown for all’, endowed with the qualities (mtshan ‘dzin pa) of the Righteous King Khubilai Khan.

The princely patron of print Si tu lHa btsan skyabs ascended the throne of Southern La stod after the death of his father in 1402. The Shel dkar chos ‘byung provides little information about him, as the author relied on a narrative concerned primarily with the father and not the son. But biographies of spiritual masters from this period tell us that Si tu lHa btsan skyabs was a highly charismatic figure and a very devout Buddhist who acted as patron not only to Bo dong Phyogs las rnam rgyal but also to Thang stong rgyal po, another great innovator famous for his ironchain bridges over the gTsang po river and the architectural feat of the gCung Ri bo che Stu-pa (Stearns 2007). As his father’s successor, Si tu lHa bstan skyabs had a great legacy to preserve. The printing operations that took place in 1407 certainly

Patronage and printing innovation in 15th-century Tibet 357

commemorated his father’s death and celebrated his country. But they also crucially generated the patron’s merit. A passage in the colophon states: In order to complete the spiritual aspirations of the Dharmara-ja Chos kyi rin chen with a spiritual aspiration of great merit, king lHa btsan [skyabs] who is like the king of the gods, had the printing of an excellent scripture made in his honour […] the 100,000s printed texts like rays of light were profusely transferred, shining in the lotus groves of the dharma-schools thus endowed with the wings of scriptural authority and reasoning… (folio 89v)19

Merit-making for the deceased was not new to Buddhist society, but the way in which it was done on this occasion was probably quite new, at least in the local context. Aside from the absence of earlier extant prints, literature seems to confirm that these were early days for printing in Tibet. When between around 1488 and 1495 gTsang smyon Heruka in Southern La stod decided to print his famous biography of Milarepa, he was deliberately embarking on a new project, considered the pioneering enterprise in early Tibetan printing (Ehrhard 2000, 17).20

The lama who oversaw the printing process Bo dong Phyogs las rnam rgyal (Figure 15.5) was born in 1376 in a pastoral area northwest of Shel dkar into a prominent spiritual lineage in the region.21 His birthplace, like that of his predecessors, was not far from the home of U rgyan pa, the Lama who was instrumental to the production of 13th-century prints at the Yuan court and who is likely to have left important legacies for the Bo dong pa tradition.22 19

Lha dbang dang mtshungs lha btsan rgyal po des//chos kyi rgyal po chos kyi rin chen gyi thugs dgongs rdzogs phyir thugs dgongs rab dge bas// bzang po’i zhabs kyi gzhung bzang spar du bsgrubs /[…]/ spar ‘bum phrag ‘od zer rab ‘phos te// chos gra so so’i pad tshal rnams su bkra// de’i tshe lung dang rigs pa’i ‘dab bshog can// 20 As reported in his gTsang smyon Heruka’s biography written by his disciple Lha btsun Rin chen rnam rgyal, 7 years was considered an exceptionally long time to spend on the print of one work (Ehrhard 2000, 17). 21 This was a nephew-lineage (dbon rgyud) stemming from dPang lo tsa ba and continuing with Byang chub rtse mo, Grags pa rgyal mtshan. They were all born in Zur tsho, which is very close to the birth place of Urgyan pa, and is an area that in the 16th century acquired particular significance for printing operations (Ehrhard 2010, 129ff; Sernesi 2011, 180ff). 22 One of these Yuan prints is included in Bo dong Phyogs las rnam rgyal’s Collected Works (Encyclopedia Tibetica, vol. 116).

358  Hildegard Diemberger

Figure 15.5 A statue of Bo dong Phyogs las rnam rgyal at the Porong Pemo Choding monastery in Kathmandu. According to the Shel dkar chos ‘byung, just before handing over the abbot’s seat to his nephew in 1403, Bo dong Phyogs las rnam rgyal’s uncle Grags pa rgyal mtshan initiated proceedings in honour of the deceased Si tu Chos kyi rin chen, and these included the printing of Haribhadra’s commentary. Whether initiated by Grags pa rgyal mtshan or Bo dong Phyogs las rnam rgyal, or both, the departed ruler was honoured with a new and bigger ambition: the promotion of a larger-scale and deeper engagement with the Buddhist doctrine through its enhanced study by the monastic community. Taking his ancestral legacy to new heights, Bo dong Phyogs las rnam rgyal had a real passion for the arts and the crafts and was no doubt aware of the printing operations in China. He could

Patronage and printing innovation in 15th-century Tibet 359

also count on the wood carving skills of a growing number of Newari artisans (and their Tibetan students) who had come to the region in increasing numbers from the 13th century. Both Grags pa rgyal mtshan and Bo dong Phyogs las rnam rgyal were in a position to manipulate the relationship to the ruler to support the newly established school of philosophy and expand the monastic curriculum. The chosen text strongly suggests itself as a tool for religious training. We know about a few other texts that were dedicated to the study of Tantra and that were also printed at Shel dkar under Bo dong Phyogs las rnam rgyal’s aegis around 1410.23 Further insight into Bo dong Phyogs las rnam rgyal and the way his relations with worldly rulers spurred innovation can be found in the neighbouring kingdom of Gung thang.

The history of Gung Thang The Royal Genealogy of Gung Thang (Gung thang rgyal rabs)24 lists the rulers of the Gung thang kingdom and describes their deeds and the production of sacred objects they patronised. Here the Buddhist making of volumes often appears alongside the making of statues, paintings, temples and shrines as merit-making activities, especially after the passing of a family member. Until the reign of Khri Lha dbang rgyal mtshan (r. 1419–1464), these were manuscript copies, sometimes lavishly illuminated and written in gold on deep blue paper. However, with the arrival of Khri Lha dbang rgyal mtshan, the printing of Buddhist texts appears for the first time in the narrative as a crucial merit-generating activity. The texts he had printed were the teachings of Bo dong Phyogs las rnam rgyal, which he gave in his public teaching sessions, and whose memorialisation in print was supervised by Khri lha dbang rgyal mtshan’s famous daughter and the master’s disciple, Chos kyi sgron ma (1422–1455; see Figure 15.6).25 23

Extant copies have just been discovered by Tibetan colleagues (see the Bod kyi shing spar lag rtsal gyi byung rim mdor bsdus lag rtsal gyi byung rim mdor bsdus collection published in 2013 by the Paltsek Research Institute). 24 The Gung thang rgyal rabs is an 18th-century source compiled by the Tibetan historian Rig ‘dzin Tshe dbang nor bu (1698–1755) on the basis of pre-existing sources. It gives an account of the foundation of this kingdom established by descendants of the ancient Tibetan imperial lineage until its end, brought about by the occupation in 1620 by the gTsang King. For a translation into German and a discussion of this source, see Everding (2000). 25 Chos kyi sgron ma’s biography, folio 62v. Chos kyi sgron ma was the first of the bSam sding rdo rje phag mo, Tibet’s most famous female lineage of reincarnation (Tsering 1993; Diemberger 2007).

360  Hildegard Diemberger

After a short and unhappy marriage in the ruling family of Southern La stod (as the daughter-in-law of Situ lHa btsan skyabs), Chos kyi sgron ma became a nun, and during her peripatetic life she made a point of teaching nuns how to read. More generally, this was the beginning of a period in which more popular, grassroots forms of religious practice developed through increased access to printed texts (Ehrhard 2004, 149). In the kaleidoscopic landscape of constantly shifting alliances, Bo dong Phyogs las rnam rgyal was able to manipulate his actual and potential patrons without fully falling under the control of any one of them. What

Figure 15.6 A depiction of Chos kyi sgron ma.

Patronage and printing innovation in 15th-century Tibet 361

he did was by no means a repetition of established norms, but a pursuit of a bold and innovative vision of reviving Buddhist civilisation. He could pursue it because he was not under the strict control of any single ruler or institution, but could call upon and seek help from several patrons at a time. Sometimes he made daring choices. Printing was only one of his many innovative initiatives which included new forms of painting and metallurgy, the staging of ritual dances, the participation of women (on Chos kyi sgron ma’s instigation) in ritual dances as well as fuller forms of female ordination (Diemberger 2007). Khri lha dbang rgyal mtshan and Chos kyi sgron ma’s printing initiatives were followed by those of his son and further descendants on the Gung thang throne, leading eventually to the production of a whole host of print editions, including the biography and songs of Milarepa (perhaps the most famous work of Tibetan literature) mentioned above. At the same time the printing of works such as the Man.i bka’ ‘bum,26 Testament of the First Tibetan Emperor Srong btsan sgam po, contributed to the spreading of Tibetan imperial legacy and a sense of proto-national consciousness (Dreyfuss 1994). Activities in Southern La stod and Gung thang were not unique. In other parts of Tibet, rulers such as the lords of rGyal rtse, gYa’ bzang and Phag mo gru engaged in similar operations, and by the turn of the 16th century, Tibet was abuzz with printing houses.

Printing for change in Tibet With the collapse of the Sakya-Yuan rule in Tibet, as polities fragmented, competing fiercely against one another, local rulers emulated the dharmara-jas of the erstwhile Sa skya-Yuan imperial period, rivalling one another in the generation of merit as well as in cultural and artistic innovation. In the spirit of this competition, the Shel dkar chos ‘byung ends its enumeration of the great deeds of Situ Chos kyi rin chen with comparison to another great contemporary ruler who also engaged in early printing (Ehrhard 2000, 12): The great deeds of [Tibetan] kings and ministers are told in annals, religious histories, and so on. In an analogous way the great deeds of Rab brtan kun bzang ‘phag, the lord of rGyal rtse, are recorded in the so-called ‘Eighteen 26

For a discussion of the early prints made in Gung thang, see Ehrhard (2000, 15); a remarkable early 16th-century version of the text produced by the Gung thang royal house is kept in the Cambridge University Library.

362  Hildegard Diemberger

perfect deeds’. The above-mentioned great deeds of Situ Chos kyi rin chen together with those of his ancestors and descendants can compete with them (Sheld dkar chos ‘byung 33v, in Wangdu and Diemberger 1996, 61).

The proliferation of printing was further supported by the memory of the archetypal Righteous King, Kubilai Khan, who was a great patron of printing. This printing revolution was carried further forward by a period of economic prosperity that followed the opening of new trading routes and the communicative infrastructure established by the Yuan Empire. There was more wealth and more potential patrons and readers. Printing did not only remain the prerogative of devout and ambitious rulers, but also quickly became  the pursuit of informal lay and monastic networks.27 Patronage offered by both rulers and ordinary people were manifestations of the same principle—merit-making through donation—and books became powerful agents which connected people via sacred objects.

In the 21st century If in the 15th century the meritorious reproduction of books shaped forms of patronage and technological innovation, and more than 500  years later, it still does. Today Buddhist books mobilise many Tibetans (and not only Tibetans) engaged in the recovery of textual collections, which may have survived the Cultural Revolution in monastic libraries, shrines, caves and in private collections. Their activities are often supported by sponsors who see in the rescue and reproduction of these texts an important source of merit.28 To handle this task, they also support the use and development of information technologies, such as the digital reproduction of texts. There are clear parallels between these two moments of technological innovation in Tibetan book culture (Baron, Lindquist and Shevlin 2007, 9). Patronage, devoutness and technological innovation still go hand in hand and mutually reinforce one another.


The colophons of their print editions include long lists of donors’ names and donations. 28 Many sponsors are Tibetan traders, but some are also wealthy patrons from mainland China and the West.

Patronage and printing innovation in 15th-century Tibet 363

Figure 15.7 A monk at the Paltsek Research Institute in Lhasa. Tibetan politics in the 21st century is markedly different: the Chinese state apparatus defines a dominant secular modernity in contrast to the alleged backwardness of Tibetan Buddhist religious practices and beliefs. Nevertheless, informal networks of scholars, monks, traders, students and even farmers and nomads remain at the forefront of technological innovation, mobilising the material and intellectual resources to develop and customise whatever is needed to achieve their goals. In their feelings about books as sacred objects and in their readiness to embrace innovation while respecting the moral authority of Buddhist antiquity, new forms of Buddhist entrepreneurship (as well as some state-sponsored cataloguing and digitising projects that may liaise with them) echo the ancient relations which once linked princely donors to their spiritual preceptors.

An ending thought When I first started researching 15th-century Tibet, I  was struck by how much it resembled in its political and artistic culture Italy during the Renaissance: its efflorescence of great artistic and cultural achievement, rampant political factionalism and multiple

364  Hildegard Diemberger

political rivalries which morphed into and were conducted through competitive patronage of the arts. However historically frivolous, this parallel highlights something important about the relation between patronage and innovation. If innovation in art, literature or technology needs political support, the similarities between approximately contemporary Italy and Tibet press us to rethink the assumption that political decentralisation, fragmentation and patronage thwart the forward movement of creative life beyond politics. Here we see the opposite of that. In both cases patrons were keen to show off, as it were, with creations which were distinctly their own and which would be new but shaped within the bounds of accepted tradition. At the same time, as those whom they patronised—whether sculptors or spiritual preceptors—enjoyed special standing in society, the innovators had the space necessary to experiment and innovate as no power had full control over them, and they could move between competing patrons. While Michelangelo and Bo dong Phyogs las rnam rgyal produced works of great novelty, their worldly patrons were pursuing age-old political aims: the prestige and grandeur that wins them authority among those who call them kings. In the words of historian Elisabeth Eisenstein, patronage on the Mediterranean and in the Himalayas ‘broke new paths in the very act of seeking to achieve old goals’ (1979, 693).

The (im)morality of mediation and patronage in south India and the Gulf


Filippo Osella


atrons, big-men, brokers and fixers who mediate access to the state or the market inhabit complex, and often contradictory, moral spaces that awkwardly straddle reciprocity and unabashed self-interest, munificence and greed.1 These morally ambiguous figures have had a chequered history in studies of everyday life in South Asia, and beyond.2 In this literature, their penchant for turning social relations into a means to accumulate status, political power and capital often overshadows their ability to extend help or represent the interests of their clients in the face of market forces or the quagmires of state bureaucracies. Their know-how and knowwho, many have argued, contribute to the (re)production of social hierarchies and inequalities, thus coming in the way of democratic participation and free market exchange. This chapter traces the ways in which modes of connectedness in Kerala—from the extended family to networks held together by the well-known figure of the broker or go-between—articulate with styles of networking in the Gulf, and especially with the Arab system of wasta, or ‘advantage via social connection’.3 By considering the role 1

The first draft of this paper was written jointly with Caroline Osella and a different version of it was published as Osella and Osella (2012). I thank Geert De Neve, Tom Widger, Kaveri Qureshi, Vekkal J. Varghese and Anastasia Piliavsky for their comments on drafts. Over the years, my research has been supported by the British Research Councils, the Wenner-Gren Foundation and the Nuffield Foundation, to which bodies I am thankful. 2 To cite a few: Wolf (1956), Geertz (1960), Boissevain (1974), Davis (1977), Gellner and Waterbury (1977), Eisenstadt (1984), Reddy and Haragopal (1985) and James (2011). 3 I have been working in Kerala, South India, since 1989 and since 1996 in various Gulf States among migrants from Kerala (Malayalis). In Kerala, as in India on the whole, Hindus are the majority, but Kerala also has substantial Muslim and Christian populations, each of which are highly concentrated and dominant in

366  Filippo Osella

of connectedness and mediation in the everyday life of migrants and would-be migrants to the Gulf Cooperation Council (GCC) countries of West Asia—for long a popular destination for hundreds of thousands of Keralan migrants—I aim to unsettle abstract moral evaluations which often inform distinctions drawn between the figures of patron and broker.4 Although the actual work that both do substantially overlaps, whilst the patron’s largesse and apparent generosity are normally associated with forms of mutual dependency and hierarchical reciprocity characteristic of a ‘moral economy’, brokers are deemed to operate in the amoral, or altogether immoral, realm of calculated self-interest that ostensibly defines market exchange (see Introduction, 23–24). The notion of ‘moral economy’ needs clarification insofar as in academic writing it subsumes two related meanings—its economic aspect and its moral content (Fassin 2009). In the first, moral economy refers to the social embeddedness of economic practice, or the degree to which economic activity is shaped by and carried out amidst social and cultural relations.5 In the second, moral economy draws attention to relations between social actors based on mutual obligations and expectations of reciprocity—a pre-modern social contract, that is—that might provide subaltern groups with the means to engage critically with, or resist the penetration of, capitalist relations and state power (Thompson 1971; Scott 1976; Taussig 1977).6 From this perspective, geographically specific zones. I have worked in two main field sites. Throughout the 1990s I did research in an inland rural area (pseudonym Valiyagramam), roughly equally split between Hindus and Christians and formerly dependent upon rice cultivation, where ties to the Gulf developed only after the 1970s. Since 2000 I worked in an urban coastal town known as Calicut or Kozhikode in the zone of greatest Muslim dominance, where recorded ties to the Gulf stretch back to the tenth century. 4 For a classic discussion of these typologies, see Reddy and Haragopal (1985). See also Boissevain (1969). 5 For example, Mauss (2002 [1925]), Polanyi (1944), Gregory (1982), Granovetter (1985) and Carrier (1995). Although embeddedness has oft been associated solely with ‘traditional’ or pre-capitalist economies (Polanyi 1944), research by economic sociologists, historians and anthropologists has underscored that the lifeworld of contemporary capitalism—even in its more abstract expressions, such as financial markets—is deeply imbued with moral and ethical normative orientations (Parry and Bloch 1989; Sen 1989; Hirschman 1997; Callon 1998; Mitchell 2005; Thrift 2005; Fourcade and Healy 2007; MacKenzie et al. 2007; Muniesa, Millo and Callon 2007; Ho 2009). 6 In the context of migration studies, for instance, the chain migration (Werbner 1990; Gardner 1995; see Vertovec 2009 for an overview) and the distribution of remittances (Schmalzbauer 2004; Sánchez-Carretero 2005; Carling 2008) have been explained with

The (im)morality of mediation and patronage 367

then, market exchange under capitalism might constitute a moral economy, but not a full-fledged moral economy. Moral understandings of the relation between self and Other are thus reduced to a number of mutually reinforcing oppositions, pitting against each other different modes of production or exchange (Scott 1976; Taussig 1977; Gudeman and Rivera 1990), economic cycles of individual and community reproduction (Bloch and Parry 1989), tradition and modernity (Mills 1999), the local and the global (Comaroff and Comaroff 2000; cf. Tsing 2000).7 What brings together these diverse analyses is not just the benign consistency attributed to moral economies, but the fact that the conditions of their production and rhetorical deployment in everyday life are either taken for granted or remain hidden (Ewing 1990; Roitman 2000; Schielke 2009). With a sleight of hand, the moral economy is eventually subsumed under ethical or moral universals that might be closer to the anthropologists’ own (political) sensibilities than to those of their respondents (see, for instance, Browne and Milgram 2008). I have argued elsewhere that in Kerala accusations of corruption against patrons and big-men do not concern the exchange of money for favours, but the failure to fulfil promises made to clients (Osella and reference to the fulfilment of reciprocal obligations binding communities in complex webs of social relations that undermine the individualising logic of accumulation under capitalism. Most commonly, the apparent moral ambiguity routinely attributed in popular discourse to migrants and migration has been related to inevitable tensions engendered by the encounter of discrete moral worlds, brought into contact by transnational flows. The morality of communities— informed by orientations towards mutual obligations and various degrees of reciprocity—confronts the (a/im)morality of a world beyond the local, normally associated to capitalist production and market exchange, profit and individual accumulation (Gardner 1995; Mills 1999; Gamburd 2000; Rudnyckyj 2004). 7 In studies of South Asia, morality, as a set of norms or discourses seeking to regulate behaviours and exchanges between people or social groups, has been discussed, either implicitly or explicitly, in the context of debates concerning ideologies sustaining caste/community identities (Dumont 1980; Daniel 1984), notions of personhood (Östör, Fruzzetti and Barnett 1982; Carrithers, Collins and Lukes 1985), patron-client relations (Breman 1993; Mines 1994; de Neve 2000), the regulation and normalisation of sexualities and gender (Uberoi 1996; Patel 2006; Devika 2007; Osella 2012), the constitution of national identities/the nation in opposition to/dialogue with colonialism (Chatterjee 1993; Roy 1998), subaltern politics and solidarities (Hardiman 1987) and, more recently, consumption, censorship, as well as the cultivation of ethical dispositions/qualities (Pandian 2009; Pandian and Ali 2010). With a few exceptions (notably analyses of gender/sexuality and ethics/qualities), these debates share the shortcomings indicated here.

368  Filippo Osella

Osella 2000a). The brokers and middle men I discuss in this chapter are judged to be immoral when they cheat would-be-migrants by taking money for non-existent visas or jobs, not because they put a price on connections that should be provided at no cost as part of wider obligations between kin and friends. Likewise, the relational obligations of a big-man might be less effective—but no less instrumental— in providing connections and employment than the self-interested practices of a businessman. Many Gulf-based Malayali entrepreneurs mobilise their business skills to sustain community ‘upliftment’ back in Kerala, and many help by privileging the recruitment of migrant labour from among ‘their own people’ (Osella and Osella 2007; 2009). The accumulation of capital through market exchange does not thwart, but enables the objectification of moral dispositions. At the same time, whilst not all patrons or big-men wish to become brokers, they might use their connections to make a fast buck on their clients’ backs, or extend patronage via strategic gift-giving to their close circle, portraying kinship obligations as the voluntary, open-hearted patron’s largesse (Osella and Osella 1996). Although brokers might not have the standing and clout of big-men, they frequently seek to transform purely economic transactions into long-term relationships of patronage and thus claim corresponding status (Osella and Osella 2000b). Extending one’s connections to provide a passage to and job in the Gulf is in any case shrouded in ambiguity: a broker who regularly receives money for his services would rather present himself and be known as a benevolent patron eager to fulfil responsibilities towards clients and kin. A would-be-migrant, on the contrary, might prefer to see the help received from a ‘patron’ or ‘big-man’ as a purely commercial transaction—a return for unpaid services rendered in the past—that does not create long-term obligations. Malayalis, then, are well aware that the politics of connectedness and mediation inevitably entail a degree of calculation and return, and that exchanges embedded in or expressed through the idiom of kinship are at best ambiguous and contradictory. Trust might be expected, but seldom can be taken for granted, even within the close circle of kin and friends.8 That is, in Kerala patrons, big-men, brokers 8

Over the years Kerala has witnessed the proliferation of structures (temples, churches, shrines, ashrams, hospitals) and practitioners (astrologers, healers, imams, priests, poojaris, psychologists, counsellors) who specialise not only in smoothing the path to successful migration, but also in helping families to keep a hold on their migrant members, so that remittances are sent regularly and savings are not wasted abroad.

The (im)morality of mediation and patronage 369

and fixers occupy the same moral space, where degrees of trust and morality are determined by the context and outcome of mediation processes. It is the scrutiny and evaluation of the relative efficacy of forms of connectedness and the success of mediation that constitute an ‘ethical scene’ (Cohen 2011) through which people—individually and collectively, as a public—apprehend themselves as moral subjects against an imaginary abject (immoral, corrupt, rapacious) Other. Along the way, discussions of the moral faults of patrons and brokers—instances of blatant corruption or cheating connected to Gulf migration which are regularly taken up by the media—produce the effect of an immanent morality actually already there, suggesting that there is a substantial reality in what is alluded to in debates (Butler 1993). In exploring the moral ambiguities and unpredictable outcomes of mediation, then, I heed Kate Meagher’s warning against the essentialism and cultural determinism of ‘abstract models of solidarity and connectivity’ and focus instead ‘on the specificities of how particular types of networks operate’ (2005, 226).

State, migration and the politics of mediation The moral perils of mediation underpin ongoing debates about policy interventions into regulating migration from India to the Gulf. A consensus between the governments of labour-sending and labourreceiving countries, international labour organisations and NGOs, and market actors/employers has started to emerge. It suggests that an adequately regulated and formally mediated migration would produce a more efficient and reliable labour market, simplify the policing and governance of migration, and at the same time substantially reduce the exploitation of migrant labour by recruiters and employers alike. This consensus has led to plans proposing the phasing out or the redefinition of the current ‘sponsorship’ (kafala) system9 and the delegation of recruitment (in the country of origin) and employment (in the place of destination) of all migrant labour to governmentapproved agencies meant to guarantee contractual fairness and transparency to migrants and employers alike (Breeding 2012; Gardner et al. 2013). Agreement about the need for a new migration regime places substantial limits on informal networks of migration 9

To live and work in the Gulf countries, all foreigners require a sponsor as a legal guarantor. For a detailed discussion of the kafala system, see Beaugé (1986) and Gardner (2010). See also Frantz (2011).

370  Filippo Osella

brokerage, judged to be inherently exploitative and corrupt, if not altogether illicit and criminal. Middlemen and fixers who facilitate migration to the GCC countries (often for a hefty fee) are identified as the problem: unscrupulous, rapacious individuals who take advantage of migrant labour, raising the costs of an already expensive migration process and condemning many migrants’ families to long-term debt. Attempts to regulate and formalise recruitment of migrant labour to date have had, at the best, mixed results. Regardless of formal contracts with government-approved recruitment agencies, most migrants have to renegotiate wages or terms of employment on arrival in the GCC countries; wages are often lower than what had been agreed, ‘donations’ are still necessary for smooth migration, employers have not ceased to retain migrants’ passports and so on (Rahman 2011; Breeding 2012; Gardner et al. 2013). In other words, the regularisation of informal channels of migration and brokerage—whereby the recruitment, training and employment of migrant labour are standardised and guaranteed by the state (Rudnyckyj 2004; Lindquist 2010; Frantz 2011; Rahman 2011)—has not lived up to its promise. Whilst in practice formal and informal migration networks might overlap (c.f. Mollona 2005), migrant labour continues to rely on fixers and middlemen who enable South Asian emigration to the Gulf. Whilst the GCC countries have attracted a great deal of scholarly and activist attention for implementing policies which, by liberalising the circulation of capital, significantly curtail labour rights (Longva 1997; Gardner 2010; Mahdavi 2011), far less has been said about the governance of emigration in India. Under the guise of offering state protection to the most vulnerable of migrants, the 1967 Indian Passport Act and the 1983 Indian Emigration Act create substantial bureaucratic hurdles for low-skill and unskilled migrant labourers, and particularly for women (Irudaya, Varghese and Jayakumar 2010; Kodoth and Varghese 2012). Whether we turn our gaze to one shore of the Indian Ocean or to the other, we find migrants regularly confronting complex regulations, hostile bureaucracies, police  controls  and stringent residency and employment laws.  It  is perhaps all too obvious that navigating the maze of migration bureaucracy—from acquiring a passport and obtaining an emigration clearance certificate, to securing a visa and getting a job—requires competences and contacts which may not be available in equal measure to all. This means that the relationship between migrants and state bureaucracies normally necessitates various degrees of mediation.

The (im)morality of mediation and patronage 371

In India, as much as in the Gulf, the protection and representation of migrants, especially unskilled male labourers and female domestic workers, is increasingly undertaken by or delegated to a plethora of (state-sponsored) non-governmental organisations. In Kerala, for instance, the Overseas Development and Employment Promotion Consultants Limited (ODEPC) and the Non-Resident Keralites Affairs Department (NORKA) are the government-sponsored initiatives explicitly aimed at regulating informal networks, with the view to  shifting mediation entirely to the non-governmental sector. This move is part of much wider-reaching management of transnational circulation of labour, which anthropologist Johan Lindquist relates to broader political and economic contradictions ‘between the “upward” concentration of capital and the “downward” outsourcing of labour’ and ‘between the dispersion and fragmentation of labour management and the centralisation of migration control’. These contradictions create spaces for brokerage and mediation where ‘the bodies of migrants are quickly transformed into valuable commodities to be controlled and protected’ (2010, 118).10 But unlike Indonesian or Sri Lankan migrants, Malayalis routinely shy away from government-licensed migration agencies and continue to rely instead on less expensive and more secure informal connections to support their migration projects.11 Informal networks and interpersonal relationships, then, constitute the main means through which Malayali migrants both engage with state bureaucracies and the labour market, and negotiate their everyday lives in the GCC countries. Informal networks offer Gulf migrants the first recourse through the migration process, from the first steps in search of employment and acquisition of visas, to arrival and settlement into accommodation, and dealing with post-arrival problems or the desire to change their contract. Personal connections and bonds achieve what official institutions cannot. 10

See also Rudnyckyj (2004), Peck, Theodore and Ward (2005), Coe, Jennings and Ward (2007), Biao (2008), Rogaly (2008) and Wills (2009). 11 On Indonesia, see Rudnyckyj (2004), Lindquist (2010); see also Yeoh and Huang (2010), Constable (2007). On Sri Lanka, see Gamburd (2000), Frantz (2011). Rajan et al. (2010, 26ff) suggest that in Kerala only 25% of migrants use formal channels of migration, such as NGOs or CONGOs (Conferences of Non-Governmental Organizations). The reasons they give include greater efficiency, lower costs, as well as friendlier and more secure service. Recently, Gardner et al. (2013) have shown that only 56% of the labour migrants surveyed in Qatar were recruited through agent-brokers; the rest effected this through kith and kin.

372  Filippo Osella

Neha Vora (2008) notes that professional and middle-class migrants are very vocal in their criticism of the UAE for its failure to live up to the neoliberal and capitalist ideals it claims to espouse—no freedom of movement, labour, or hiring and no market-led salary structure. Networks mitigate some of these restrictions. Another type of migrant criticism takes the far more concrete form of simply trying to evade as far as possible severe controls; here informal networks become critical. In other words, it is the state and the restrictions it puts on migrants’ lives (including limits on migrants’ representation, control of the media and the enforcement of highly restrictive residence and employment regulations) that simultaneously provide the grounds for  the flourishing of informal networks and restrict or criminalise the  working of the latter (Gardner 2010).12 The criminalisation of informal migration networks is indeed central to the working of contemporary global capitalism. To use Di Giorgi’s words, it ‘work[s] symbiotically toward the reproduction of a vulnerable labour force, suitable for the most exploitative sectors of the post-Fordist economy’, as is the case with migrant labour in the Gulf (2010, 153). In practice, then, new migration regimes proposed across the Indian Ocean (re)produce forms of ‘organised informality’ in migration and employment that deprive labour migrants of personal networks of protection and support through which they navigate the everyday predicaments of migration (Gooptu 2013).

Connectedness across the Indian Ocean In India networking practices flourish; brokers, fixers and go-betweens are ubiquitous. They negotiate arrangements from marriages to business deals and school admissions, even down to mundane matters like buying an air ticket. Two major factors contribute to mediation. The first is the vast gap between an arcane and time-consuming postcolonial bureaucracy, and a relatively uneducated and often illiterate population, which has been socialised to believe that official matters and things requiring ‘paperwork’ are difficult to navigate alone. Among the lower classes of a highly hierarchical society, there is also a widespread tendency to think of oneself as less skilled or less capable 12

Meagher notes that informal networks ‘are not defined by their autonomy from the state, but are critically shaped by the nature of their relationship with the state’ (2005, 226). See also Varghese and Rajan (2010).

The (im)morality of mediation and patronage 373

in technical or bureaucratic matters than ‘experts’. But in Kerala, which boasts a fully literate population and a high competence in expressions of modernity, these are not adequate explanations. Just as in the rest of India, in Kerala most can well apply for a driving licence or passport, pursue a job, buy an air ticket or rent a house, directly. But people routinely turn to others for advice and help, and in many circumstances, they approach various agents, brokers, fixers or ‘uncles’ who make their living by mediation (Tarlo 2003; Gandhi 2010). South Asianist scholarship has underscored the pervasiveness of various forms of mediation or brokerage and the iconic figures of patron, big-man, sardar, kangani, jobber, maistri or dalaal, whose influence was expanded by unfolding colonial and post-colonial political and economic relations.13 Contrary to early predictions of informal mediation withering away once both capital and democracy reached ‘maturity’,14 recent research has shown that in post-liberalisation India’s political and labour brokerage has only flourished.15 Whilst this apparent ‘informalisation’ of political and economic life is undoubtedly related to the requirements of global capital—producing and demanding flexibility at every level of social life—another factor plays an important role. Many Indians are brought into modes of subjectivation that produce forms of highly socially connected selves, intersecting with bourgeois notions of selfrealisation, entrepreneurship and productivity. A non-autonomous self who works happily around and within family, marriage and lifedecision making (study, job), which do not place all the burden of decision or action upon an isolated individual is what works best, and is widely cultivated.


For example, Srinivas (1976), Chakrabarty (1989), Gupta (1995), Hansen (2001), Ruud (2000), Chatterjee (2006), Chandavarkar (1994, 2008), Roy (2008), Breman, Guérin and Prakash (2009), Picherit (2009), Prakash (2009), Roesch (2009) and Sen (2010). 14 During the early post-colonial era, when villages were generally conceptualised as closed corporate groups, anthropologists Eric Wolf (1956) and Clifford Geertz (1960) described brokerage and patronage as ‘vertical’ mediation between village and  state. For Geertz, the broker who bridged the divide (either in cultural or economic terms) between the village and the city was bound to disappear as the new nations developed. See also Boissevain (1974) and Davis (1977). 15 See Picherit (2009), Roesh (2009) and James (2011).

374  Filippo Osella

Far from positing worn-out orientalist exceptionalism,16 I argue that the cross-cultural prevalence of and value attributed to social connectedness (see footnote 2) suggests that the autonomy and isolation normatively demanded of the contemporary Northern European or American is nearing pathological in the degree of pressure it places upon a single person. At the same time, research by anthropologists and economic sociologists has underscored not only the sociality inherent to the production and reproduction of markets, but also recovers the economy of affects, values, dispositions and habituations underpinning and enabling everyday market practices (e.g., Ho 2009; LiPuma and Lee 2012). Ethnographies of stock markets and financial institutions further reveal that even the virtualisation of market exchange through sophisticated technical devices is undergirded by personal connections, life-style commitments, corporate fashions, creative intuitions, blind faith and plain luck, showing that the teleology of disembeddedness staked in mainstream social theory as a defining feature of the modern is simply implausible.17 It is not then a judgement, but a plain observation that in much of South Asia18—and beyond—family socialisation styles and wider realms of action encourage a group-centred, connected self. The iconic South Indian ‘big-man’, for instance, is a wealthy and well-connected patron who balances the desire for personal fulfilment with obligations towards relatives and neighbours, whilst attracting clients and dependants through strategic deployment of unreciprocated largesse.19 Orientation towards social connectedness means that it is quite normal and expected for people to turn to others for opinion and help and to respond in return to demands for support and advice. Brokerage, then, is one logical outgrowth of the connected self. The groundwork is then laid for a class of ‘Mr Fixits’ who, with a little more education, worldly wisdom, connectedness and relevant experience, can offer to help, for a fee. India is a society in which mature and functioning economic and political institutions are criss-crossed by networks of mediation 16

See Chandavarkar’s criticism of Chakrabarty’s orientalist understandings of labour brokerage in colonial India (2008). 17 See Janine Wedel’s (2009) excellent study of neo-con networks in the United States and her notion of ‘flex-power’. 18 For example, Roland (1988), Obeyesekere (1990), Trawick (1990), Kurtz (1992) and the essays in Pandian and Ali (2010). 19 Srinivas (1976), Mines (1994; also Chapter 1 in this volume) and Osella and Osella (2000a).

The (im)morality of mediation and patronage 375

and practices (as well as aesthetics) of informality that simultaneously respond to and inflect the workings of political and economic life.20 Research on migration has often focused on the importance of personal connections for making migration happen. Yet analysts of chain migration have concentrated predominantly on the role of family and friends in passing information and offering practical support to would-be migrants, leaving the role of migration brokers somewhat overlooked (Massey et al. 1998, 190).21 Networks that expand beyond friends and family are indeed central to various instances of transnational social formations (cf. Vertovec 2009). In Kerala, where the economy relies heavily on remittances and Gulf connections,22 where Dubai is more familiar than Delhi, and where the Gulf is considered to be part of Kerala, it has long been clear that access to information, capital, contacts and hospitality has been critical to people’s migration strategies. Anxiety about obtaining, maintaining, improving and expanding one’s connections is correspondingly high, so that no chance may be missed or lost out to somebody else, and energetic networking around issues of access to the Gulf is a central part of Kerala life. In a terrain of economic liberalisation and limited employment chances, Malayali would-be migrants are not mere surplus labour waiting to be called upon. They are actively gaining permissions, connections and information needed for success, in a highly entrepreneurial manner. The centrality of information for negotiating the ever-shifting and fast-moving migration scene in the Gulf makes networking crucial, in the way that it is among Malaysian entrepreneurs or in socialist societies where access to scarce resources and passage through labyrinthine regulations is mediated by the right ‘connections’ (Yang 1994; Sloane-White 1999). Here I do not take a naïve view of the migrants’ networks, as instances of demotic resistance to state and capitalist markets, or to formal, bourgeois associationalism (as does Chatterjee 2004; for a critique see Meagher 2005; Roy 2009). Literature on patron-client relations in India, for instance, has produced extensive accounts of power relations within informal networks (e.g., Breman 1993; 1996; 20

See Roy (2009). For an exception, see Lindquist (2010; 2012) and Lindquist, Xiang and Yeoh (2012). 22 While in Kerala direct official remittances have dropped from 25% of the GDP in 1998 to 18% in 2008, the secondary economic development in the form of real estate, sales of consumer goods and so on is hard to estimate, but is clearly vast (Rajan, Varghese and Jayakumar 2010). 21

376  Filippo Osella

also Bourdieu 1990), even as recent research highlights the leverage available to clients in negotiating or resisting patrons’ demands.23 Clearly, favours have to be repaid, in cash or otherwise. But these informal networks, through which information, visas, jobs and accommodation prospects circulate among migrants and would-be migrants, while entailing various degrees of power inequality and exploitation, simultaneously offer a source of necessary connections and social capital. Ambivalence defines relationships within informal migrant networks, moral evaluations and assessments of their licitness or illicitness depending on context, specific experiences, social position and, indeed, necessity. The role of brokers or mediators in these networks is equally complex in that the same person can help out or cheat someone with equal ease, depending on opportunities and social relations engendered within (and outside) these networks (see also Chandavarkar 2008).

Expanding one’s networks I have observed over the years a range of different styles of social networking in operation. Some are based on pre-existing connections, via kinship, friendship and neighbourliness back home; others are more expansive and, as we shall see, often built out of chance encounters. Sitting on the Kuwait City seafront, I share take away food with a few Malayali friends. It is a mixed group—young migrants and old hands, kin and unrelated professionals and house servants— who meet regularly because they are all Muslims from the same neighbourhood in Kozhikode. There are two new arrivals: Faizal, who has recently moved from Muscat, where he had to abandon his small haulage business as the result of ‘Omanisation’ policies that restrict certain occupations or businesses to Omani nationals; and Gafoor, a young man who, after some years in Dammam, could no longer stand Saudi restrictions on ‘bachelor’ use of public space. The conversation soon turns to serious matters. Faizal is staying with his sister’s husband, an accountant who has put him in contact with a number of shops which are now hiring him to transport goods with the van he bought. He considers himself lucky, although he claims to ‘have lost a fortune’ in the move and complains that he is 23

Appadurai (1990), Osella and Osella (1996), De Neve (2000) and Staples (2003).

The (im)morality of mediation and patronage 377

hardly making a living. Faizal’s brother-in-law, who had arranged the work permit, now negotiates low van hiring rates from Faizal as a favour to some of his own clients. The group of friends listen carefully, but no one has harsh words for the brother-in-law, simply reminding each other that ‘it is always hard when you arrive in a new place’. Concerned, though, about Faizal’s reduced income, and by what a protracted reduction of remittances to his family would do to his reputation back home, a few discreetly offer to lend him money. Gafoor has not yet found a steady job and is helping out in a supermarket belonging to one of his father’s friends ‘for time passing’. After spending all his working life in Dammam, his father has now retired back to Kerala. Concerned with Gafoor’s unemployment, he is setting up a shoe shop for him in Kozhikode. A long exchange about how much the Gulf has changed ensues: it has become much harder to find a job, salaries have gone down, working conditions are not what they used to be and, in general, there are far fewer ‘chances’ than there were before. Yes, with the right connections it is still possible to make money, but no one has any illusions about the Gulf anymore. Newcomers, argue my friends, have to learn fast that the streets of the Gulf are not paved with gold, as they once were. The oldest migrants in this group arrived in the late 1960s, brought by Kuwaiti traders who, after many years in Kozhikode, had closed their businesses in the wake of the oil boom and returned to the Gulf. Trading partnerships across the Indian Ocean, built over many years, sometimes over several generations, allowed Malayalis to find government employment in post offices or the police. As the Kuwaiti economy developed, young professionals— engineers, accountants and medical doctors—used longstanding trade connections to join private and public companies. In the meantime, Kuwaiti and Malayali trading partners used the last boats (dhows) plying the moribund Indian Ocean to ‘smuggle’ migrants— mostly labourers—to the Kuwaiti coast. ‘They were dropped outside the city, on the beach’, my friends recounted. ‘There was a Malayali “hotel” which gave shelter to these men, and found them jobs’. Over time, as migration became a steady stream, networks diversified. On the one hand, often overlapping familial, friendship and business relations provided reliable information, contacts and support for the would-be-migrants. On the other, established migrants doubled their income by moonlighting as middlemen

378  Filippo Osella

between Arab sponsors and Malayali ‘agents’, offering their services to those with no direct connections in the Gulf. Back in Kozhikode, in a Muslim neighbourhood where most households have at least one member working in the Gulf and where many families now hold three generations of migrants, information about job opportunities, salaries and so on is readily available and discussed on a daily basis.24 Here would-be-migrants seldom rely on ‘agents’, and young men have a good sense of what awaits them in Dammam, Sharjah or Doha: the embellished tales of holidaying migrants are quickly subdued by direct and honest accounts of relatives and family friends. People decide on whether or not to migrate judiciously. Men may first go and stay with friends and relatives in the Gulf on a tourist visa, visiting different places to find out first-hand what opportunities they may have. At the same time, many long-term migrants have grown weary of extending help to all those who approach them, either directly or via network chains, for help. The reputation of a ‘fixer’ or patron certainly increases one’s status and income, but it is also troublesome and time-consuming. If you ‘arrange’ a job for the son of a family acquaintance and the employer turns out to be a cheat, paying a much lower salary than he promised, it is then your duty to mediate and find a satisfactory solution, or to arrange alternative employment. There is a significant difference between migrant networks in urban Muslim Kozhikode and rural Hindu/Christian Valiyagramam, both in the way that information circulates and in the manner in which help is extended to support aspiring migrants. In Kozhikode, which is tied to the Arabian Peninsula by historical trade, religious and heavy migration flows, locals never hang on the migrants’ words, taking as true fantastic stories about glamorous life and easy money, which first-hand experience proves wrong. In Kozhikode, familiarity with the Gulf is so deep and the transnational information and gossip networks are so extensive that masculine boasting about one’s good fortune or cunning can be easily disproved and ridiculed (Osella and Osella 2000a). In Kozhikode, everyone knows current salaries for different jobs in the Gulf, in which states or cities life is cheaper and easier, how migration laws are changing and where it might be easier 24

Women play a crucial part in flows of information and assessments of chance. At Keralan weddings and dinners, in school lobbies, beauty parlours and other places where women socialise, ‘stay-behind wives’ routinely share information about their husbands’ work abroad.

The (im)morality of mediation and patronage 379

to circumvent them. In Valiyagramam this is the privileged knowledge of a few longer-term migrants, who cash in on this disclosure and create brokerage jobs. At the same time, there is also a difference between Muslim, Christian and Hindu networks. Among Muslims, reliable connections extend well beyond close family to include a wider circle of friends and thus limit the extent to which mutuality can be commoditised into a service offered for a fee, as is often the case among Hindus and Christians (Osella and Osella 2009).25 Many migrants who shared with me their stories were first taken to the Gulf by a relative, lived there with a relative and had a job arranged by a relative. This often results in a safe situation, quite unlike the labour bondage described in many studies of migration to the Gulf.26 Mujib’s father’s brother, Aziz, is an accountant in a supermarket in Abu Dhabi. Aziz managed to get his young nephew a job as a sales assistant in the electrical department and took him into his apartment to live. This gave Mujib the time to find his footing and get ready to move on to look for a better contract or chance. People often agree: the important thing is to get to the Gulf. Only once you are there can you begin to meet the people you need to know and gather the information you need in order to make your migration successful. Even Malayalis who come ‘alone’ will have two or three names and phone numbers of direct or indirect friends and relatives, and will be able within a week of arrival to insert themselves into multiple networks. While most come already armed with some connection, many more contacts are forged in the Gulf. This can happen via expansion along the Indian line of working to familiarise the unfamiliar and give oneself a sense of belonging, through contacts with friends of friends of friends. These networks tend to be narrow, reaching back to roots in the homeland and hence usually to specific communities or kin groups. In central Dubai there is an apartment building where a handful of original residents (some related) found comfort and convenience in co-residence, and attracted others from the same neighbourhood in Kozhikode who also appreciated the shared cooking of favourite ‘home foods’, the ability to share and discuss news 25

Non-Muslims sometimes carp at this ethos of mutual help and express envy at the wider circles of trust which they observe among Muslims. They do not pause to reflect that this may have in the first place developed in response to the socio-economic hostilities endured by Muslims in India (see Hansen, 2007). 26 For a counterargument, see Mahdavi (2011).

380  Filippo Osella

from home as well as constant and easy access to communication.27 Those who work outside the city or in ‘labour camps’ use this address for correspondence or as a resting place on the rare days off.28 Wellknown both back home and around the Gulf, this block is the hub of a tightly knit network that, although extremely effective in extending support to a relatively small circle of people, does not promote the expansion of migrants’ connections. Going to church or mosque affords other opportunities for meeting co-nationals and making new contacts. An apartment in downtown Karama, Dubai, was shared by a group of South Indian Christians who attended an underground evangelical fellowship and had become good friends through the church. This bachelor flat highlighted the possibilities for expanding networks beyond hometown, state and even linguistic boundaries through shared religious affiliation. Half of the men worked in a handbag distribution warehouse, where they had been recruited by a fellow church member, and were happy with their working and living conditions. Some were officially on visas and contracts for different jobs and were thus ‘illegals’, but they were not unduly anxious, in part because they envisaged making money and then returning to India, in part because they felt that the warehouse owner had the wasta (social connections) to ensure that no raids or deportations would take place at the workplace (Cunningham and Sarayrah 1993). Reggie has a little old house on the outskirts of Ras al Khaimah, where he works as a manager in a mall. His spends his days off moving energetically around the stretch of the Emirates, seeing people whom he knew back home and who are now settled here, as well as new people from his home area who have been introduced to him through the friend-of-a-friend networks. He may only stay with each host for half an hour, but he tries to maintain regular contact with as many hometowners as he can. Sometimes, on a Friday night he and a few of his friends go out to one of the discreet Indian drinking clubs. There, he may meet new friends-of-friends who have been invited to relax in the company of the regular group of six or so men. As the evening progresses, drinking buddies share new Indian songs, movie 27

Migrants usually avoid the Indian postal system and send letters, cash and small gifts with other migrants travelling home. Living among hometown folk offers them an uninterrupted and easy line of transit. 28 Large companies normally accommodate migrant labourers in specially built compounds outside the main cities, known as labour camps (Gardner 2010).

The (im)morality of mediation and patronage 381

clips and pornographic mini-videos on their mobile phones, and finish, phone numbers swapped, with new contacts, and everyone’s networks strengthened and widened. Like churches and mosques, bars are spaces where people can make connections and widen networks beyond hometown circles or other direct links. Settling into a place can also happen through one of the many ambiguous ‘self interested’ new friends one may meet, and such networks are far wider and cross any boundary, so long as the price is right or the mutual advantage good enough. Such connections in the Gulf cross the boundaries of religion and caste, even gender and, to some degree, class, which are normally observed in India. Varghese, an old friend from Valiyagramam, once took me to meet his ‘very good friend’, Abu Awais, in a rundown little office in downtown Ras-al-Khaimah. Abu, a hugely fat and jovial Somali man, sitting behind a desk littered with three phones and many piles of paperwork, was introduced as one of the long-term Gulfies, someone who had been there since the 1960s. Varghese pointed out that Abu knew hundreds of people, understood all the regulations and—more importantly—knew how to get round them. He had many Arab friends and he spoke, read and wrote Arabic. Ostensibly employed as a scribe and document translator, Abu drew his real income by liaising between Indian agents seeking papers or jobs for their Indian clients and locals looking to make money out of sponsoring visas, offering fake employment letters, or acting as sitting partners in businesses (genuine or fraudulent). One triangle, composed of a local Arab, a Malayali and an East African, registered a window frame supply business in Ajman. The enterprise was based in a yard, where there was some basic equipment, materials, and a few half-made window frames. But their real business was in the sale of work visas with employment at the company. Men who paid for these visas back in India were not cheated. They knew that the offer of employment was designed to get them a visa and into the UAE, where they could then seek employment. A series of officials and the combined wasta of the three partners had been involved in obtaining the clearances, premises and other things needed to set up the business, and would continue to steer clear of being detected as a ‘ghost’ company. The Malayali partner in the triangle, Shajan, was also involved in many other small networks and moneymaking initiatives; he did his best never to turn anyone down, to connect people and build his reputation as ‘a man who can’ (Osella and Osella 2000a).

382  Filippo Osella

One evening, while driving his car around Dubai, Shajan took a call from somebody reporting that the management of a labour camp in Al Quoz was unhappy with the price they were paying their food contractors. Moving back and forth between Hindi and Malayalam, he asked for the menu and details of prices, told the caller that he knew other contractors who could supply similar menus at a lower cost and said that he would get back to him within a day or two. He then rang up somebody, passed on the details and asked them to offer him a competitive price. Soon after, he took another call from somebody looking for second-hand sanitary fittings, which would be extracted from bathrooms in hotels under refurbishment in UAE and then shipped to Kuwait. He promised to locate the party that could supply these. A third call came from somebody working for an Arab household and looking for a recommendation for somebody offering Ayurvedic treatments for diabetes and courses in medicinal oil massages. Shajan immediately named two establishments, giving the name of the head masseur, and agreeing to ‘adjust’ the caller (give commission) for this business when they next met. Shajan himself took commission from the Ayurvedic clinics which got the business. This is small-time stuff: ‘lunch money’, ‘drinks money’ or ‘tea money’ passes hands in many of these transactions. For Gulf citizens these sums of cash are laughable, but in the hands of a careful Indian broker they can add up to help scrape together a liveable income. It is the extent, size and heterogeneity of a broker’s network that makes for his success. Shajan receives non-stop calls and manages to make several small deals every day because of his reputation as an ultra-connected man. New arrivals to the Emirates are likely to end up with his number in their phone, on recommendation, because he is a useful person. Many migrants in the Gulf, like Shajan, actually have as their main business (whatever their official business or job may be) the work of a broker who arranges visas, jobs, contracts, certificates, supplies and whatever else that might be needed. Acting as the hubs of many overlapping networks, these energetic entrepreneurs find out about a variety of needs and connect—for a favour or fee—two or more parties. Some of them open their houses as temporary stopovers for migrants who are in-between work contracts or looking for connections. Obviously, this is not an open service, and one needs to be personally known or strongly recommended and tightly connected to one of the broker’s own close contacts to be invited to stay in a broker’s home.

The (im)morality of mediation and patronage 383

Men like Shajan embody the qualities of a person known for being able to get things done by operating and establishing connections across different social worlds. They are the prototypical brokers who use their wherewithal to mediate all sorts of deals, with an eye to making a profit, small as it might be. Their everyday strategies are complex and fraught with ambiguities. They might offer their services for free to a complete stranger in the hope of cementing a long-term and mutually advantageous partnership, or charge a fee for a deal struck with a friend or kin in the knowledge that in the long term ‘everything will be adjusted’. Their business of mediation requires both the self-interested shrewdness of brokers and the largesse of patrons or big-men.

Modalities of connectedness and mediation Indians who arrive in the Gulf find a great deal of continuity between the ways things are done locally and the ways they are done back home. They are skilled and accustomed to negotiating relations, and through networks around which favours, benefits, profit and information circulate frantically. Some Indians in the Gulf find continuity between older forms of patronage and networking with which they were familiar back home, and the new ones operative under the logic of wasta. Back in Kerala, Meena’s father, Babuchan, had been the president of a village council (panchayat) and a big-man of the village and its immediate neighbourhood. Although well educated, he never took up formal employment, preferring to retain the status of ‘cultivator’ of his many hectares of paddy land and to become an energetic member of Kerala’s Communist Party. His home was always full of people and he was forever out meeting some block office bureaucrat or Member of Legislative Assembly. He helped people to sort out problems, liaised with government offices on people’s behalf, connected people for business, marriage and whatever else that they needed, and was invariably a reliable source of news on what was going on both in the village and immediately beyond. A couple of years after his daughter arrived in the Gulf, she was already working as a broker. She had her fingers in just as many pies as her father, finding it natural to connect people for the sake of business. Once, juggling two of her mobile phones, through which she was breaking a deal between a ‘buyer’ and a ‘seller’, she smiled at us and said, with a wink: ‘don’t look so surprised. What do you expect? I am Babuchan’s daughter!’

384  Filippo Osella

Undoubtedly, Meena was an excellent networker and a maker of deals, skills she has developed and used for many years to support her family. The tricks of her trade flourished in the Gulf, but were learnt at home, at her father’s knee. There is a continuity of skill, a set  of habits and, in Malayali idiom, a ‘social-mindedness’ which Meena’s father put to good use in local politics and ‘social work’, and which Meena was able to turn into income in the Gulf. As Meena’s case suggests, there is a happy articulation of expectation and practice between Kerala forms of brokerage or mediation and the Arab wasta. Malayalis are used to operating through go-betweens and to the idea that the best way to get things done is through connections. The quality of ‘connections’—invariably entailing various degrees of obligation and a trade in favours—ranges from openly commercial transactions in which brokers claiming better access or knowledge are paid to facilitate a process, to more commercially and morally ambiguous long-standing friendships and patronage ties (Osella and Osella 1996; 2000a). But the means of connection—gifts and hospitality—are consistent. Many explained the success of Malayali entrepreneur, Yusuf Ali, leader of the Emke group which owns the LuLu department stores ubiquitous in the Gulf (and the fourth richest Indian in the Gulf), as a function of his ability to combine good business sense with wasta, to develop and cultivate mutually advantageous personalised relationships. They typically remarked: ‘He is very good at knowing what the Arabs want’. I have explored elsewhere the frankness and clarity with which Indian entrepreneurs speak about their success and their vision of contemporary entrepreneurship as relying on a winning combination of technocratic ‘know-how’ and the older skills of ‘know-who’ (Osella and Osella 2011). This stands in sharp contrast to Northern Europe or North America, where we cling to the fictions of impersonal institutions, meritocracy, transparency and open competition. There are of course differences between Indian and Arab practices of connectedness. To Malayali migrants, Indian forms of mediation or brokerage appear much more democratic than wasta in their openness: the favour-seeker can be a tenuous and hardly known friend of a friend of a friend, or somebody only just met with no depth of relation, or even someone openly paying a fee. Whether the relationship with a  broker or ‘fixer’ slides from patronage to a trade in favours, the passing of gifts or open monetary exchange is carefully judged and negotiated when parties meet. There is such a large population of

The (im)morality of mediation and patronage 385

South Asians who rely on regular income from brokerage and act as hub points for connections and deals in the Gulf, that one can always find several paths towards the resolution of one’s problem, and choose the most suitable. Even when help is dressed in the rhetoric of friendship, with no fee or obvious material benefit accruing to the favour-giver, the hierarchy within the relationship is always well understood by both parties. The ‘friend’ receiving a favour knows that at one point, payback will be called in. Sometimes ‘payback’ consists of no more than joining the patron’s entourage or being part of a group that reflects a big-man’s glory, for the patron’s own satisfaction and for the maintenance of his reputation as someone who matters and can get things done. Plenty of men seek to augment their (masculine) status, stand at the centre of many network hubs and be recognised as one of the hegemonic males (de Neve 2004; Osella and Osella 2006a). Despite restrictions of access, Indians are well used to dealing with different forms and styles of mediation and brokerage, and find wasta easy to understand and navigate. Shabith, who had worked in the administration of a shipping office told me happily how, every Thursday, his boss had bought biryani for all the workers and sat to eat with them, sharing food off the same plate.29 Shabith thought of this as a pretence at friendliness and equality, a performance, because of course the boss was still boss—and an Arab to boot. But Shabith also understood that the boss was trying to extend his relationship with his workers beyond contract, that he was buying their loyalty and silence about certain accounting and recording practices, and that he had, through the Thursday night meals, become a valuable source of wasta, somebody to whom Shabith could turn in time of trouble. On Thursdays Shabith would have preferred to leave as soon as his precious weekend began, but he stayed behind to share food with his boss and co-workers, knowing well the value of this weekly meal. And he never interpreted the invite as an innocent offer of hospitality.

Morality and legitimacy in the informal life Some networks are benign groupings of people who offer help to each  other, but many more are much more ambivalent. Cheating, hustling, lying and hiding are commonplace among migrants in 29

For non-Muslim Malayalis, whose normative food practices include avoidance of food sharing beyond family intimates, this is an extraordinary gesture of equality and intimacy (Osella and Osella 2008).

386  Filippo Osella

the  Gulf: the  stakes for financial gain and ruin are high, financial demands and prestige pressures back home are enormous, and people act accordingly. More generally, though, in Kerala it is recognised that personal advancement and successful manhood necessarily entail, and justify, a degree of rule-bending and chicanery in business as much as in social obligations to kin and friends (Osella and Osella 2000b; Osella and Osella 2009). Moral evaluations of the actions of patrons, middlemen, brokers and mediators, then, are contingent upon the fulfilment of expectations and promises, and are evaluated over a long time. The difference between types of connection does not map neatly onto a continuum between ‘trusted familiars’ and ‘risky strangers’. Time and again I have observed that one is as likely to be cheated by one’s own brother as by a stranger-hustler to whom one has paid cash in hope of a contract or permit. ‘Cheating’ can range from plain swindling (typically when an ‘agent’ cashes in his fees, but fails to deliver a visa or job) to failure to deliver on promises of quality jobs, salaries or accommodation. In Valiyagramam, Baby John had the reputation of a successful businessman, having built, after some years away, a thriving water bottling plant in Sharjah. He readily offered help to relatives, neighbours and friends, recruiting them to work in his plant. Soon, however, many of those who had joined him returned to the village, complaining that ‘office positions’ turned out to be labouring jobs. What’s more is that Baby John treated them harshly, and even refused to pay salaries, arguing that it had cost him a lot to bring them there. Hussein, belonging to a reputed Kozhikode trading family, had decided to try his luck in Dubai, accepting some old family friends’ offer to become a partner in their tyre dealership. Although he contributed several hundreds of thousands of rupees to the business, after a few months in Dubai he realised that he was being used as a shop assistant and received no salary, because he was not an employee, but a partner, under the pretext that they had to teach him the trade. Eventually, he left, losing all of the money that he invested. Over the years, I have heard hundreds of similar stories, which revealed not only the economic aggressiveness of life in the Gulf, but also the precariousness of relationships which involve migrants, especially with people whose trust one would hope to assume. Ties of reciprocity and mutuality, expected of kinship and friendship, can easily turn into deception and naked exploitation. Migrants are no

The (im)morality of mediation and patronage 387

less vulnerable back home, once they seek to invest remittances and savings. Ahmedkutty had for a long time worked as an accountant in a bank in rural Saudi Arabia. This was a good job and he accumulated substantial capital. As he approached retirement, he joined a couple of old friends in buying a brick factory back in Kerala. As the three partners were still in the Gulf, they left the deal in the hands of a relative who had proposed and brokered the investment. Money was paid, papers were signed, but Ahmedkutty and his friends soon learned that their partner simply ran off with the cash, and Ahmedkutty had to postpone his return for several years30. We cannot simply identify and deplore the sharp practitioners in these tales. For the man who cheats you may at the same time be helping somebody else. One can never identify agents clearly as either ‘social workers’ or ‘exploitative hustlers’: the same people play both roles at different times. It is the nature of the moment, the other agents and the particular assemblage of circumstances and actors involved that flavour the event. Personal relations offer a shifting ground, and the highly ‘immoral’ can also be helpful and work ‘for the good’. A great deal of ambivalence prevails in evaluations of each transaction. You may now think that the man who lured you out to the Gulf on the false promise of a job was a cheat, but in 10 years’ time you may remember him as a benefactor who first brought you out of the village and helped you to get a foothold in the Gulf. The ambivalent nature of ‘networked connections’ and personal relations highlights the unhelpfulness of abstract notions of mutuality, obligation or reciprocity which often inform debates about ‘moral economies’ (e.g., Browne and Milgram 2008). Riju, whom I have known since he was an adolescent and who has spent the last 20 years in Saudi Arabia and then in Dubai, recounted: ‘When I lost 30

Questions of where to invest savings, whom to trust back home and how to avoid being cheated are issues of great and constant concern for most migrants, especially Muslims, who are prevented from putting money in the bank and taking interest (riba), which is forbidden (haram) and considered a major sin. Malayali Hindu and Christian migrants have long circumvented the problem of cheating by investing their savings in life insurances and various bank saving schemes, a fact deplored by Kerala state government anxious to see Gulf savings used in ‘productive’ investment and development back home (Osella and Osella 2000a; 2006b). Eventually, for returnees in all communities the most common (and perceived as least risky) form of long-term investment are flats in apartment blocks which are being built across Kerala in a property speculation boom.

388  Filippo Osella

my job in Saudi, I decided to return to Valiyagramam and set up a small business, a photocopy and computer shop. But it didn’t work out… my partner, my cousin-brother [father’s brother son] cheated me and I lost everything, so I had to go back [to the Gulf]’. Indeed, there is a substantial gap in Kerala between public moral rhetoric and actual practice in which even bonds of kinship do not ensure probity. Moreover, the moral criteria deployed to evaluate the actions of those operating within or connected to various networks may not meet standard or formal expectations, or conform to normative liberal ideas of ‘morality’. One broker-hustler told me happily about having finally found a good Gulf Arab national who offered phony letters of employment to be sold to would-be-migrants, for an honest price. For the majority of Gulf migrants an illegal labour broker who finds them a job or an agent who sells fake school certificates and phoney job contracts is often a godsend. There is morality here, make no mistake, but it is a broader morality of opportunity for those who struggle to eke out a living in the interstices of the Gulf’s opulence. It would be tempting to argue that migrant networking practices and their moral frameworks are born of a mix of desperation and a desire to do well—that they are conditioned, as Andrew Gardner (2010) argued, by the instrumental demands of overarching ‘orchestrating’ structures.31 They might correspond to what Asef Bayat (2010) has identified as the wider imperative upon the marginalised and the economically dispossessed to live an ‘informal’ life (see also Mollona 2005). Yet observations made in other societies where networking is intense, complicates analysis. Examples are legion around the globe and across class: middle-class sociality in Malaysia described by Sloane-White (1999); Chinese guanx chi (hierarchical networks of influence and patronage) dynamics which bring people together through gifts, favours, feasting and hosting (Yang 1994, 90–99); extralegal ‘informality’ that undergirds land use at both ends of the social spectrum in India (Roy 2009); the morals of networking with the global financial and political ‘superclass’ or ‘shadow elite’ (Rothkopf 2008; Wedel 2009). Ubiquitous instances of mutually advantageous networking and brokerage draw attention to the centrality of personal bonds and networks (closed and secretive in some cases), which they


See also De Certeau (1984), Hobbs (1988), Bourdieu (1990), Wacquant (2007), Gardner (2010) and Mahdavi (2011).

The (im)morality of mediation and patronage 389

comprise and which are part and parcel of global capital and its power base. Aside from India’s business elite, where top entrepreneurs play at very high levels, mingling with senior Indian and Arab politicians and acting as the hubs of major contracts and agreements between states, most Indian networking is small-scale and clandestine. Much networking, favours and mediation are done surreptitiously, often inside people’s homes. A contractor may treat a Public Works Department official to a lavish and alcohol-fuelled lunch in his home. A broker may receive visitors seeking favours and help them in the evenings from his veranda or car. If the elite—Kerala’s Gulf entrepreneurs, British cabinet ministers and Indian politicians, Gulf sheikhs and North American industrialists—can all increase their share of renown, respect, advantage and profit through energetic networking, then why and with what reason should migrant labourers be denied similar rights? If rich entrepreneurs and businessmen can obtain multimillion dollar contracts or political honours through elite networking practices, or by cutting bureaucratic corners for advantage, why should ordinary migrants be denied respect, reputation and material gain for doing no more and no less than that?

Conclusion Networks where friendship, support and instrumental gain are entwined and mutually reinforcing run right across the spectrum of social life. In the GCC countries there are many wealthy Indian business people and professionals (Vora 2008). There are also plenty of lower-middle-class Arab nationals who are struggling to keep up with the economic boom (Bristol-Rhys 2010). At issue is in no way a simple relation between hosts and outsiders with advantage and disadvantage divided cleanly between two sides of the divide. Networking among Indian billionaires, rich Arab families and Indian politicians is part of the same social process that involves smalltime brokers selling fake SSLC certificates or bogus offers of work to Indian migrants. We should be wary, then, of positing a split between the practices of labour and capital acquisition. As I have argued elsewhere, in the Gulf, much as in the rest of the world, technocratic know-how informing the discourse of modern business management remains indissoluble from longstanding social skills of know-who (Osella and Osella 2009). Social networks are by no means an unusual

390  Filippo Osella

or an imperfect aspect of the neo-liberal project, but a foundational feature of global capitalism.32 What differentiates the Gulf countries or India from northern Europe or the United States, then, is not the degree to which social networks inform business relations, but whether or not these networks and attendant practices of sociality are readily named and acknowledged as a legitimate feature of economic life (Kahn and Formosa 2002). In this light, the distinction often drawn between ‘patronage’/‘brokerage’ and ‘networking’/‘lobbying’ as opposed associational modes begins to dissolve. In the literature, this distinction usually appears in the contrast between ‘illicit’ and ‘primordial’ mediation and ‘licit’ and ‘modern’ forms of social connectivity. In practice, however, both work in analogous ways by connecting people through mutually advantageous relations, thus underscoring the social and affective embeddedness of economic relations under capitalism. Migrants’ networks connecting South India and the Gulf, and often spanning out to Europe, United States and South-East Asia, and personalised forms of mediation by fixers are neither marginal practices nor demotic alternatives to formal channels of migration and employment. The success of the Gulf economy relies substantially and directly on relations between migrants, local informal networks and brokers, most of whom stand outside the law. It relies on these relations to acquire and maintain a cheap, docile and flexible workforce via the kafala system and widespread subcontracting (Gardner 2010). These practices, which are as common in the Gulf as they are in the United States or the United Kingdom, bring together multinational companies and a multitude of small employers, who recruit labour through informal migrant networks, in which they often play central roles (Mollona 2005). Notwithstanding recent shifts towards extending limited rights of citizenship to certain categories of migrants and reforming the kafala system, transnationalism remains as much an expression of cross-continental relations as a condition of life forced upon migrant labour by global capital. By criminalising informal brokerage, proposed policy shifts towards regularising and formalising the recruitment of migrant labour objectifies the neo-liberal notion of socially disembedded individual entrepreneurship as a condition for participating in the global labour market (McKeown  2012). That is, the formalisation of recruitment 32

See for instance Granovetter (1985), Thrift (2005), Harvey (2007) and Wedel (2009).

The (im)morality of mediation and patronage 391

and employment opens the way, via a structure of sub-contracting, to a substantial further informalisation of labour regimes in which migrant labourers lose the (already limited) means to negotiate the insecurity of precarious contracts or to resist demands for increased flexibility. Gardner (2010) has argued that the migrants’ ‘illicit’ practices, which circumvent or diminish the governments of the GCC countries, emerge from the deep-rooted tradition (harking back to maritime trade links) of offering carefully limited ‘hospitality towards strangers’, who are neither expected to become part of society nor to expect rights on par with their hosts. Whether or not we can read in the present the legacy of time-honoured cultural practices, the contemporary Gulf does produce its own illicitness by means of its regulations, control of institutions, media, its anxieties about free movement and its determination to hold on to ethnic privilege such that non-citizens cannot hope for a fair treatment in court or state. What Bayat (2010) calls the ‘informal life’ is encouraged by the prevalence of poverty, inequality and debt bondage with no hope of gain or redress by formal means. In a wider ideological sense, Gulf economies also sanction the flourishing of the illicit in as much as they promulgate and support (by means of targeted and energetic structural and material incentives set within a wider ethos of risk and entrepreneurship) the values of lavishness and glorification of wealth, placing an overwhelming stress upon the material. In analysing guang-xi, Yang (1994) follows feminist arguments made by Gilligan (1982) to characterise this type of non-disembedded, connected, personalised, horizontal integration via relationship as ‘feminine’. We can pick this move apart and instead of accepting the (common) ‘gendering’ of connectedness, note the studies that trace the contours of connectedness as a value in many social arenas. We can then move on to ask ourselves: how, in Western social and political theory, did disembeddedness, abstraction and disconnectedness all come to be gendered as masculine, forced into a teleology of progress and finally touted as superior as at once associated with the gender deemed superior and evaluated as more socially and politically ‘mature’? Turning away from Western social and political theory’s teleological progress narrative which privileges disembeddedness and abstraction over the material and affective politics of everyday personalised connectedness (Gilligan 1982; Yang 1994), of which mediation is inevitably part, one could say that the problem is not the system

392  Filippo Osella

itself, but the fact that in the Gulf not everyone has access to wasta, which continues to operate in strictly controlled circles. The problem  then  parallels many others—citizenship, accommodation, healthcare, education—and is better framed as one of unequal access. Just as Vora’s (2008) informants claimed rights of belonging in Dubai, while drawing her attention to the novelty and unfairness of racial discrimination there, and to the gap between the rhetoric of free and open market in the Gulf, and its restrictive hiring and working practices, the issue can be understood from a user’s point of view as one of unfair monopoly by citizens on a form of capital which is a basic necessity. If we do not start from neo-liberal or even classic liberal standpoints and their (imagined) values,33 but from a more neutral admission that wasta exists, and is undoubtedly a resource in Gulf society, then we may see that from the perspective of migrant labourers more openness, more wasta-building opportunities, would be a good thing. To be able, as in India or China, to rely upon no more than a gift, or to be able to marshal endless connections, would give migrants better access to social hubs that actually matter and which can make things move. Even in India, where huge numbers of labour recruiters and Mr Fixits roam, the questions of the scale of gifts required to forge a connection or to adjust a situation leads us to the issue of resources and cash. In Kerala, a 500 rupee note to ward attendants is enough for anybody to help make sure their loved one has a smooth and pleasant stay in a local government hospital; but to secure admission for one’s child to a top engineering college will cost hundreds of thousands of rupees. Even the widest and most open of networks, like the Indian system of mediation, is always inflected by class inequality. The latter is a relatively open system relying on gifts, favours, tenuous connections and money; from the perspective of migrant labour, wasta is a relatively closed and hierarchical/vertical system relying heavily on the cultural capital of lineage, name and substantial connection. Both Indian wasta and Indian brokerage channel and restrict access to resources, just like the state institutions, 33

There is no space here to treat this argument fully: as with other classic liberal desirable goals—modernity and secularism, for example—the projected image is far from the reality. For overviews and critiques of classical modernity theory, see Gaonkar (1999), Mitchell (2000), Kahn (2001) and Osella and Osella (2006b). On Western secularism, see Asad (2003) and Mahmood (2006).

The (im)morality of mediation and patronage 393

which place legal restrictions on citizenship and related benefits. Are we then better off re-thinking these networks in terms of their specific histories and operation, where we are better able to perceive that they are neither separate from the state and the market, nor are they actually subverting them, but acting as a form of ‘soft power’ which eventually works towards the same goals as both market and state: the maintenance and protection of privilege for a minority under a global capitalist regime?


Anastasia Piliavsky is a social anthropologist and Research Fellow of King’s College, Cambridge. She holds degrees from Boston University and from Oxford, where she read anthropology as a Rhodes Scholar. She has written about Indian politics, crime and secrecy for Comparative Studies in Society and History, Modern Asian Studies, Cambridge Anthropology, and other journals. She is co-Investigator on a collaborative study of democratic cultures and ‘muscular’ politics in India, Pakistan and Bangladesh, funded by the European and British Research Councils. Ward Berenschot is a researcher at the Royal Netherlands Institute for Southeast Asian and Caribbean studies (KITLV) where he is currently working on a comparative research project on political clientelism in Indonesia. He is author of Riot Politics: Hindu-Muslim Violence and the Indian State (2011) and of several articles on ethnic violence, public service delivery and politics in western India. He has worked with development agencies on legal aid and access to justice and is currently coordinating a collaborative research programme on citizenship and democratization in Indonesia. Lisa Björkman is a political ethnographer and Fellow of the Max Planck Institute for the Study of Religious and Ethnic Diversity in Göttingen. She holds a PhD in Politics from the New School for Social Research. Her forthcoming book, Pipe Politics: Mumbai’s Contested Waters, which explores the politics of water access in Mumbai, was awarded the 2014 Joseph W. Elder Prize in the Indian Social Sciences from the American Institute of Indian Studies. Lisa’s most recent work has been concerned with the relations of the institutions of electoral democracy and everyday politics in urban India. Hildegard Diemberger is a Fellow of Pembroke College, Cambridge, and director of the Mongolia and Inner Asia Studies Unit at the University of Cambridge. She has written widely on Tibet and the Himalayas and has also translated several historical texts from

396  Contributors

Tibetan. She is the author of When a Woman Becomes a Religious Dynasty (2007) and co-editor of Feast of Miracles: The Life and the Tradition of Bodeng Chole Namgyal (with Pasang Wangdu, Marlies Kornfeld and Christian Jahoda, 1997), Territory and Identity in Tibet and Himalayas (with Katia Buffetrille, 2002) and the Mongolia-Tibet Interface (with Uradyn E. Bulag, 2007). David Gilmartin is professor of History at the North Carolina State University. He is the author of Empire and Islam: Punjab and the Making of Pakistan (1988) and co-editor of Beyond Turk and Hindu: Rethinking Religious Identities in Islamicate South Asia (with Bruce B. Lawrence, 2000) and Pakistan at the Millennium (with Charles H. Kennedy, Kathleen McNeil and Carl Ernst, 2003). Sumit Guha is the Frances Higginbotham Nalle Centennial Professor in History at the University of Texas at Austin. He is the author of The Agrarian Economy of the Bombay Deccan, 1818–1941 (1985), Environment and Ethnicity in India, 1200–1991 (1999), Health and Population in South Asia (2001), and Beyond Caste: Identity and Power in South Asia, Past and Present (2013), among other books. His awards include a senior fellowship at the American Council of Learned Societies, received in 2003–2004, and a John Simon Guggenheim Memorial Fellowship, received in 2008–2009. Beatrice Jauregui is assistant professor of Criminology and Sociolegal Studies at the University of Toronto. She is co-editor of Anthropology and Global Counterinsurgency (2010) and author of articles published in American Ethnologist, Law and Social Inquiry, Journal of South Asian Studies, and Asian Policing. Her forthcoming book, provisionally titled Police, Power and Public Order in Postcolonial India, explores everyday policing practices in Uttar Pradesh and the intersection of power and morals in contemporary northern India. Nicolas Martin is a research associate at the Department of Social Anthropology, University College London. He holds a PhD from the London School of Economics, where he also taught anthropology. He is the author of Despotic Politics: Landlords, Clients and Islam in the Pakistani Punjab (2014) and is now conducting research on rural politics in the Indian Punjab as part of a research project on democratic

Contributors 397

cultures and ‘muscular’ politics in South Asia, funded by the European and British Research Councils. Lucia Michelutti is lecturer in Anthropology at the University College London. Her major research interests include popular politics, religion, law and order, as well as violence in South Asia and Latin America. She is the author of The Vernacularisation of Democracy (2008) and several articles on caste, race, leadership, ‘muscular politics’ and political experimentation. Her next monograph is entitled In the Name of the People: Gods and Revolutionary Politics in a Venezuelan Village (2014). Diane P. Mines is professor and chair of Anthropology at Appalachian State University. She is the author of Caste in India (2009) and Fierce Gods: Inequality, Ritual and the Politics of Dignity in a South Indian Village (2005). She is also co-editor, with Nicolas Yazgi, of Do Villages Matter? (2010) and Everyday Life in South Asia (with Sarah Lamb, 2010). Mattison Mines is professor emeritus in the Department of Anthropology, University of California Santa Barbara, and affiliate professor in South Asian Studies at the Jackson School of International Studies, University of Washington, Seattle. He has done field research in Tamil Nadu among Tamil-speaking Muslims, Kaikkoolar weavers, and in Chennai, among the Beeri Chetti and Komati Chetti merchants. He is the author of Muslim Merchants (1972), The Warrior Merchants (1984) and Public Faces, Private Voices (1994), and co-author with Vijayalaksmi Gourishankar of ‘Leadership and Individuality in South Asia: The Case of the South Indian Big-Man’, Journal of Asian Studies 49(4): 761–86. Filippo Osella is reader in Anthropology at the University of Sussex. He is co-author of Men and Masculinities in South India (with Caroline Osella, 2006), Social Mobility in Kerala (with Caroline Osella, 2000); co-editor of Islam, Politics, Anthropology (with Ben Soares, 2010), Migration, Modernity and Social Transformation in South Asia (with Katy Gardner, 2004); South Asian Masculinities (with Caroline Osella and Radhika Chopra, 2004) and Islamic Reform in South Asia (with Caroline Osella, 2012). His current research concerns relations between economic and religious practice amongst

398  Contributors

South Indian Muslims, and he is co-Investigator on an ESRC/DfIDfunded research project on contemporary charity and philanthropy in Sri Lanka. Pamela Gwynne Price is professor emerita of History at the University of Oslo and the author of Kingship and Political Practice in Colonial India (1996) and State, Politics, and Cultures in Modern South India (2013). She is also co-editor of Power and Influence in India: Bosses, Lords and Captains (with Arild Engelsen Ruud, 2010). D. Seyfort Ruegg is professor emeritus of Sanskrit and Tibetan Studies at the School of Oriental and African Studies and the author of The Literature of the Madhyakama School of Philosophy in India (1981), Buddha-Nature, Mind and the Problem of Gradualism in a Comparative Perspective (1989), Three Studies in the History of Indian and Tibetan Madhyamaka Philosophy (2000), and The Buddhist Philosophy of the Middle (2010). Arild Engelsen Ruud is professor of South Asian Studies at the University of Oslo and the author of Poetics of Village Politics (2003). He is also co-editor of Power and Influence in India: Bosses, Lords and Captains (with Pamela Gwynne Price, 2010) and India’s Democracies (with Geir Heierstad, 2014). Dusi Srinivas is a PhD candidate at the Centre for Historical Studies at the Jawaharlal Nehru University in Delhi and Lecturer in History at the J.V.R. Government College in Sathupally, Andhra Pradesh. His forthcoming dissertation is titled ‘Tribes, Forests and Livelihoods: A Study of Northern Andhra, 1839–1972’. Steven I. Wilkinson is professor of Indian and South Asian Studies and Political Science and International Affairs at Yale and author of Army and Nation: The Military and India’s Democracy since Independence (2014), Votes and Violence: Electoral Competition and Ethnic Riots in India (2006), and Religious Politics and Communal Violence (2008). He is also co-editor of Patrons, Clients, and Policies: Patterns of Democratic Accountability and Political Competition (with Herbert Kitschelt, 2007).


Aggrawal, Ankita. 2011. ‘The PDS in Rural Orissa: Against the Grain?’ Economic and Political Weekly 66(36):21–23. Aiyar, A. N., ed. 1953. Election Law Reports, vol. 3. Delhi: Election Commission of India, Manager of Publications. Aiyar, A. N., ed. 1959. Election Law Reports, vol. 18. Delhi: Election Commission of India, Manager of Publications. Alavi, Hamza. 1971. ‘The Politics of Dependence: A Village in West Punjab.’ South Asian Review 4(2):111–128. Alavi, Hamza. 1972a. ‘Kinship in West Punjab Villages.’ Contributions to Indian Sociology 6(1):1–27. Alavi, Hamza. 1972b. ‘The State in Post-Colonial Societies: Pakistan and Bangladesh.’ New Left Review 1(74):59–81. Alavi, Hamza. 1973. ‘Peasant Classes and Primordial Loyalties.’ Journal of Peasant Studies 1(1):23–62. Alesina, Alberto, and Eliana Ferrara. 2000. ‘Participation in Heterogenous Communities.’ Quarterly Journal of Economics 115:847–904. Alesina, Alberto, and Eliana Ferrara. 2000. ‘Who Trusts Others?’ Journal of Public Economics 85:207–234. Alesina, Alberto, Reza Baqir, and William Easterly. 1999. ‘Public Goods and Ethnic Divisions.’ Quarterly Journal of Economics 114(4):1243–1284. Ali, Imran. 1988. The Punjab under Imperialism, 1885–1947. Princeton: Princeton University Press. Ali, S. Mahmud. 2010. Understanding Bangladesh. New Delhi: Foundation Books. Alm, Björn. 2010. ‘Creating Followers, Gaining Patrons: Leadership Strategies in a Tamil Nadu Village.’ In Power and Influence in India: Bosses, Lords and Captains, edited by Pamela Price, and Arild Engelsen Ruud, 1–19. London: Routledge. Anjaria, Jonathan S. 2011. ‘Ordinary States: Everyday Corruption and the Politics of Space in Mumbai.’ American Ethnologist 38(1):58–72.

400  Bibliography

Appadurai, Arjun. 1981. Worship and Conflict under Colonial Rule: A South Indian Case. Cambridge: Cambridge University Press. Appadurai, Arjun. 1985. ‘Gratitude as a Social Mode in India.’ Ethos 3:236–245. Appadurai, Arjun. 1986. ‘Is Homo Hierarchicus?’ American Ethnologist 13(4):745–761. Appadurai, Arjun. 1990. ‘Topographies of the Self: Praise and Emotion in Hindu India.’ In Affecting Discourse: Anthropological Essays on Emotions and Social Life, edited by Catherine Lutz and Leela AbuLughod, 92–112. New York: Cambridge University Press. Appadurai, Arjun. 1993. ‘Numbers in the Colonial Imagination.’ In Orientalism and the Post-Colonial Predicament, edited by Carol A. Breckenridge and Peter van der Veer, 314–340. Philadelphia: University Of Pennsylvania Press. Arlacchi, Pino. 1983. Mafia, Peasants and Great Estates: Society in Traditional Calabria. Cambridge: Cambridge University Press. Arlacchi, Pino. 1986. Mafia Business: The Mafia Ethic and the Spirit of Capitalism. London: Verso. Asad, Talal. 2003. Formations of the Secular: Christianity, Islam, Modernity. Stanford: Stanford University Press. Aspinall, Arthur. 1931. Cornwallis in Bengal: The Administrative and Judicial Reforms of Lord Cornwallis. Manchester: Manchester University Press. Auyero, Javier. 1999. ‘“From the Client’s Point of View”: How Poor People Perceive and Evaluate Political Clientelism.’ Theory and Society 28(2):297–334. Auyero, Javier. 2001. Poor People’s Politics: Peronist Survival Networks and the Legacy of Evita. Durham: Duke University Press. Babb, Lawrence A. 1975. The Divine Hierarchy: Popular Hinduism in Central India. New York: Columbia University Press. Bailey, Fredrik G. 1963. Politics and Social Change: Orissa in 1959. Berkeley: University of California Press. Balzani, Marzia. 2003. Modern Indian Kingship: Tradition, Legitimacy and Power in Jodhpur. Oxford: James Currey. Banerjee, Mukulika. 2007. ‘Sacred Elections.’ Economic and Political Weekly. 42(17):1556–1562. Banerjee, Mukulika. 2008. ‘Democracy, Sacred and Everyday: An Ethnographic Case from India.’ In Democracy: Anthropological Perspectives, edited by Julia Paley, 63–95. Santa Fe: Advanced Research Press.

Bibliography 401

Banerjee, Mukulika. 2010. ‘Foreword’ to Power and Influence in India: Bosses, Lords and Captains, edited by Pamela Price and Arild E. Ruud, xv–xvii. London: Routledge. Banerjee, Sumanta. 2011. ‘Anna Hazare, Civil Society and the State.’ Economic and Political Weekly 66(36):12–14. Banfield, Edward, and James Q. Wilson. 1963. City Politics. Cambridge, MA: Harvard University Press. Banfield, Edward C. 1958. The Moral Basis of a Backward Society. Glencoe, IL: Free Press. Banfield, Edward C. 1969. Urban Development: A Reader in Administration and Politics. London: Collier-Macmillan Limited. Banks, Nicola. 2006. ‘A Tale of Two Wards: Political Participation and the Urban Poor in Dhaka City.’ Centre for Governance Studies, BRAC University, working paper 2. Bardhan, Pranab, and Dilip Mookherjee. 2012. ‘Political Clientelism and Capture: Theory and Evidence from West Bengal, India.’ United Nations University WIDER Working Paper 2012/97. ISBN: 978-92-9230-561-1. Barnett, Steven. 1975. ‘Approaches to Changes in Caste Ideology in South India.’ In Essays on South India, edited by Burton Stein, 149–180. Hawaii: University Press of Hawaii. Barth, Fredrik. 1959. Political Leadership among Swat Pathans. New Brunswick, NJ: Athlone Press. Baruah, Sanjib. 2010. ‘Regionalism and Secessionism.’ In The Oxford Companion to Politics in India, edited by Niraja Gopal and Pratap Bhanu Mehta, 181–91. New Delhi: Oxford University Press. Bate, Bernard. 2000. ‘Metaittamil: Oratory and Democratic Practice in Tamilnadu.’ PhD diss., Department of Anthropology, University of Chicago. Bate, Bernard. 2009. Tamil Oratory and the Dravidian Aesthetic: Democratic Practice in South India. New York: Columbia University Press. Baud, Isa, and Joop de Wit, eds. 2008. New Forms of Urban Governance in India: Shifts, Models, Networks and Contestations. New Delhi: Sage. Baxi, Upendra. 1982. The Crisis of the Indian Legal System: Alternatives in Development Series: Law. New Delhi: Vikas Publishing House. Bayart, Jean-François. 2009. The State in Africa: The Politics of the Belly, second edition. London: Polity Press.

402  Bibliography

Bayat, Asef. 2010. Life as Politics: How Ordinary People Change the Middle East. Stanford: Stanford University Press. Bayly, Chris A. 1973. ‘Patrons and Politics in Northern India.’ Modern Asian Studies 7(3):349–388. Bayly, Chris A. 1983. Rulers, Townsmen and Bazaars: North Indian Society in the Age of British Expansion, 1770–1870. Cambridge, MA: Cambridge University Press. Bayly, Chris A. 1996. Empire and Information: Intelligence Gathering and Social Communication in India, 1780–1870. Cambridge: Cambridge University Press. Bayly, Susan. 1999. Caste, Society and Politics in India: From the 18th Century to the Modern Age. Cambridge: Cambridge University Press. Beaugé, Gilbert. 1986. ‘La Kafala: Un Système de Gestion Transitoire de la Main-d’œuvre et du Capital dans les Pays du Golfe.’ Revue Européenne de Migrations Internationales 2(1):109–122. Beidelman, Thomas O. 1959. A Comparative Analysis of the Jajmani System. Locust Valley, NY: J.J. Augustin. Belur, Jyoti. 2010. Permission to Shoot? Police Use of Deadly Force in Democracies. London: Springer. Benjamin, Solomon. 2007. ‘Occupancy Urbanism: Ten Theses.’ Sarai Reader 07: Frontiers 538 (New Delhi). Accessed July 4, 2008. http:// a7ad6edfa9704962dce48e91bc99367d. Bentinck, William C. 1977. The Correspondence of Lord William Cavendish Bentinck (in two volumes), edited by Cyril H. Philips, Oxford: Oxford University Press. Berenschot, Ward. 2010. ‘Everyday Mediation: The Politics of Public Service Delivery in Gujarat, India.’ Development and Change 41(5):883–905. Berenschot, Ward. 2011a. ‘On the Usefulness of Goondas in Indian Politics.’ South Asia 34(2):255–275. Berenschot, Ward. 2011b. ‘Political Fixers and the Rise of Hindu Nationalism in Gujarat, India: Lubricating a Patronage Democracy.’ Journal of South Asian Studies 34(3):382–401. Berenschot, Ward. 2012. Riot Politics: Hindu-Muslim Violence and the Indian State. London: Hurst & Co. Béteille, André. 1965. Caste, Class, and Power; Changing Patterns of Stratification in a Tanjore Village. Berkeley: University of California Press.

Bibliography 403

Béteille, André. 1991. ‘Distributive Justice and Institutional WellBeing.’ Economic and Political Weekly 26(11/12):591–600. Béteille, André. 1994. ‘Secularism and Intellectuals.’ Economic and Political Weekly 29(10):559–566. Bhalla, Angad, and Siddharta Luther. 2007. ‘Writings on the Wall.’ Time of Day Productions, New Day Films. Toronto: Ontario Arts Council. Biao, Xiang. 2008. ‘Transplanting Labour in East Asia.’ In Transnational Migration in East Asia: Japan in a Comparative Focus, edited by Yamashita Shinji, Makito Minami, David Haines, and Jeremy Edes, 175–186. Osaka: National Museum of Ethnology. Biography of Bo dong Phyogs las rnam rgyal = Amoghasiddhi ’Jigs med ’bangs, Bo dong pan chen gyi rnam thar. Lhasa: Bod ljongs bod yig dpe rnying dpe skrun khang, 1990. Biography of Chos kyi sgron ma = dPal ’chi med grub pa (?) Ye shes mkha’ ’gro bsod names ’dren gyi sku skyes gsum pa rje btsun ma chos kyi sgron ma’i rnam thar. Incomplete manuscript of 144 folios. Biography of Thang stong rgyal po = ’Gyur med bde chen, Thang rgyal rnam thar. Chengdu: Si khron mi rigs dpe skrun khang, 1982. Birchfield, Lauren, and Jessica Corsi. 2010. ‘Between Starvation and Globalization: Realizing the Right to Food in India.’ Michigan Journal of International Law 31(1):691–764. Björkman, Lisa. 2014. ‘Becoming a Slum: Criminalizing Infrastructure in Liberalisation-Era Mumbai.’ Forthcoming in International Journal of Urban and Regional Research 38(1):36–59. Blackburn, Stuart. 1985. ‘Death and Deification: Folk Cults in Hinduism.’ History of Religion 3:255–274. Bloch, Maurice, and Jonathan Parry, eds. 1989. Money and the Morality of Exchange. Cambridge: Cambridge University Press. Blok, Anton. 1974. The Mafia of a Sicilian Village, 1860–1960: A Study of Violent Peasant Entrepreneurs. Oxford: Basil Blackwell. Bo dong chos ’byung = ‘Chi med ’od zer dPal de kho na nyid dus pa las bo dong chos ’byung gsal byed sgron me zhes bya ba dPal thams cad mkyen pa ’Chi med ’od zer gyis mdzad pa. Manuscript of 35 folios, kept at Bodong E monastery. Bo dong Phyogs las rnam rgyal’s Collected Works = dPal de kho na nyid 'dus pa/Encyclopaedia Tibetica. Edition of the collected works of Bodong Chogle Namgyal, 119 vols. 1969–1981. New Delhi: Tibet House Library.

404  Bibliography

Bod kyi shing spar lag rtsal gyi byung rim mdor bsdus lag rtsal gyi byung rim mdor bsdus = Collection of Tibetan early prints published by the Paltsek Research Institute, Lhasa, 2013. Bodeman‚ Michael. 1988. ‘Relations of Production and Class Rule: The Hidden Basis of Patron-Clientage.’ In Social Structures: A Network Approach, edited by Barry Wellman and Stephen D. Berkowitz, 198–220. Cambridge: Cambridge University Press. Boissevain, Jeremy. 1969. ‘Patrons as Brokers.’ Sociologische Gids 16(6):379–386. Boissevain, Jeremy. 1974. Friends of Friends: Networks, Manipulators and Coalitions. Oxford: Basil Blackwell. Bouglé, Celestin. 1992. ‘The Essence and Reality of the Caste System.’ In Social Stratification, edited by Dipankar Gupta, 64–73. Delhi: Oxford University Press. Bourdieu, Pierre. 1990. The Logic of Practice. Cambridge: Polity Press. BRAC. 2006. State of Governance Report. Dhaka. Brass, Paul. 1965. Factional Politics in an Indian State. Berkeley: University of California Press. Brass, Paul R. 1984. ‘National Power and Local Politics in India: A Twenty-Year Perspective.’ Modern Asian Studies 18(1):89–118. Brass, Paul. 1990. The Politics of India since Independence. Cambridge: Cambridge University Press. Brass, Paul R. 1997. Theft of an Idol. Princeton, NJ: Princeton University Press. Brass, Paul R. 2003. The Production of Hindu-Muslim Violence in Contemporary India. Delhi: Oxford University Press. Breckenridge, Carol. 1977. ‘From Protector to Litigant: Changing Relations between Hindu Temples and the Raja of Ramnad.’ The Indian Economic and Social History Review 14(1):75–106. Breeding, Mary. 2011. ‘The Micro-Politics of Vote Banks in Karnataka.’ Economic and Political Weekly 46(14):71–77. Breeding, Mary. 2012. ‘India-Persian Gulf Migration: Corruption and Capacity in Regulating Recruitment Agencies.’ In Migrant Labor in the Persian Gulf, edited by Kamrava Mehran, and Zahra Babar, 137-154. New York: Columbia University Press. Breman, Jan. 1974. Patronage and Exploitation: Changing Agrarian Relations in South Gujarat, India. Berkeley: University of California Press. Breman, Jan. 1993. Beyond Patronage and Exploitation: Changing Agrar­ ian Relations in South Gujarat. New Delhi: Oxford University Press.

Bibliography 405

Breman, Jan. 1996. Footloose Labour: Working in India’s Informal Economy. Cambridge: Cambridge University Press. Breman, Jan. 2004. The Making and Unmaking of an Industrial Working Class: Sliding Down the Labour Hierarchy in Ahmedabad, India. New Delhi: Oxford University Press. Breman, Jan, and Isabelle Guérin. 2009. ‘Introduction: On Bondage: Old and New.’ In India’s Unfree Workforce: Of Bondage Old and New, edited by Jan Breman, Isabelle Guérin, and Aseem Prakash, 1–17. New Delhi: Oxford University Press. Breman Jan, Isabelle Guérin and Aseem Prakash, eds. 2009 India’s Unfree Workforce: Of Bondage Old and New. New Delhi: Oxford University Press. Briquet, Jean-Louis, and Frédéric Sawicki, eds. 1998. Le Clientélisme Politique dans les Sociétés Contemporaines. Paris: Presses universitaires de France. Bristol-Rhys, Jane. 2010. Emirati Women: Generations of Change. London: Hurst. Bronger, Dirk. 1975. ‘Jajmani System in Southern India.’ Journal of the Indian Anthropological Society 10:1–38. Broomfield, John H. 1968. Elite Conflict in a Plural Society: TwentiethCentury Bengal. Berkeley: University of California Press. Browne, Katherine E., and B. Lynne Milgram, eds. 2008. Economics and Morality: Anthropological Approaches. Lanham, MD: Rowman Altamira. Bunsha, Dionne. 2006. Scarred: Experiments with Violence in Gujarat. New Delhi: Penguin Books. Butler, Judith. 1993. Bodies That Matter: On the Discursive Limits of ‘Sex.’ London: Routledge. Caldwell, Bruce. 1991. The Jajmani System: An Investigation. Delhi: Hindustan Publishing Corp. Callon, Michael. 1998. ‘Introduction: The Embeddedness of Economic Markets in Economics.’ In The Laws of the Market, edited by Michael Callon, 1–57. Oxford: Blackwell. Campbell, John K. 1964. Honour, Family, and Patronage: A Study in Institutions and Moral Values in a Greek Mountain Community. Oxford: Clarendon Press. Carling, Jørgen. 2008. ‘The Human Dynamics of Migrant Transnation alism.’ Ethnic and Racial Studies 31(8):1452–1477. Carrier, James G. 1995. Gifts and Commodities: Exchange and Western Capitalism Since 1700. London: Routledge.

406  Bibliography

Carrithers, Michael, Steven Collins, and Steven Lukes, eds. 1985. The Category of the Person: Anthropology, Philosophy, History. Cambridge: Cambridge University Press. Centenary Celebration Souvenir. 18 January 1993. Dharmamurthi Rao Bahadur Calavala Cunnan Chetty Charities Hindu Higher Secondary School Thiruvallur-602 001. Chakrabarty, Bidyut. 2008. Indian Politics and Society since Independence: Events, Processes and Ideology. London: Routledge. Chakrabarty, Dipesh. 1989. Rethinking Working-Class History: Bengal 1890–1940. Princeton: Princeton University Press. Chakrabarty, Dipesh. 2000. Provincializing Europe: Postcolonial Thought and Historical Difference. Princeton: Princeton University Press. Chambers, William Nisbet, Walter Dean Burnham, and Frank J. Sorauf, eds. 1967. The American Party Systems: Stages of Political Development. New York: Oxford University Press. Chandavarkar, Rajnarayan. 1994. The Origins of Industrial Capitalism in India: Business Strategies and the Working Classes in Bombay, 1900–1940. Cambridge: Cambridge University Press. Chandavarkar, Rajnarayan. 2008. ‘The Decline and Fall of the Jobber System in the Bombay Textile Industry, 1870–1955.’ Modern Asian Studies 42(1):117–210. Chandra, Kanchan. 2004. Why Ethnic Parties Succeed: Patronage and Ethnic Headcounts in India. Cambridge: Cambridge University Press. Chatterjee, Partha. 1993. The Nation and Its Fragments: Colonial and Postcolonial Histories. Princeton: Princeton University Press. Chatterjee, Partha. 1998. ‘Community in the East.’ Economic and Political Weekly 33(6):277–282. Chatterjee, Partha. 2004. The Politics of the Governed: Reflections on Popular Politics in Most of the World. New York: Columbia University Press. Chatterjee, Partha. 2011. Lineages of Political Society: Studies in Postcolonial Democracy. New York: Columbia University Press. Chaudhary, Muhammad A. 1999. Justice in Practice: Legal Ethnography of a Pakistani Punjabi Village. Oxford: Oxford University Press. Clark-Decès, Isabelle. 2011. A Companion to the Anthropology of India. Malden, MA: Wiley-Blackwell.

Bibliography 407

Clemente, Michela. 2007. ‘Colophons as Sources: Historical Information from Some Brag dkar rta so Xylographies.’ Rivista di Studi Sudasiatici 2:121–160. Coccari, Diane. 1989. ‘The Bir Babas of Benares and the Deified Dead.’ In Criminal Gods and Demon Devotees: Essays on the Guardians of Popular Hinduism, edited by Alt Hiltebeitel, 251–269. Albany: State University of New York Press. Code of Criminal Procedure (CrPC). 2005 [1973]. Lucknow: Eastern Book Company. Coe, Neil, Jennifer Jennings, and Kevin Ward. 2007. ‘Mapping the Globalisation of the Temporary Staffing Industry.’ The Professional Geographer 59(4):503–520. Cohen, Lawrence. 2011. ‘Ethical Publicity: On Transplant Victims, Wounded Communities, and the Moral Demands of Dreaming.’ In Ethical Life in South Asia, edited by Anand Pandian and Daud Ali, 253–274. Bloomington: Indiana University Press. Cohn, Bernard. 1956. ‘Some Notes on Law and Change in North India.’ Economic Development and Cultural Change 8(1): 79–93. Cohn, Bernard. 1965. ‘Anthropological Notes on Law and Disputes in North India.’ American Anthropologist (n.s.) 67(6/2):82–122. Cohn, Bernard. 1996. Colonialism and Its Forms of Knowledge. Princeton: Princeton University Press. Comaroff, Jean, and John L. Comaroff. 2000. ‘Millennial Capitalism: First Thoughts on a Second Coming.’ Public Culture 12(2): 291–343. Commander, Simon. 1983. ‘The Jajmani System in North India: An Examination of Its Logic and Status across Two Centuries.’ Modern Asian Studies 17(2):283–311. Conlon, Frank F. 1977. A Caste in a Changing World: The Chitrapur Saraswat Brahmans, 1700–1935. Berkeley: University of California Press. Constable, Nicole. 2007. Maid to Order in Hong Kong: Stories of Migrant Workers. Ithaca: Cornell University Press. Constas, Helen. 1958. ‘Max Weber’s Two Conceptions of Bureaucracy.’ American Journal of Sociology 62(4):400–409. Conze, Edward. 1973. The Perfection of Wisdom in Eight Thousand Lines and Its Verse Summary. Delhi: Sri Satguru Publications. Cooper, Randolf G.S. 2003. The Anglo-Maratha Campaigns and the Contest for India. Cambridge: Cambridge University Press.

408  Bibliography

Copeman, Jacob. 2011. ‘The Gift and Its Forms of Life in Contemporary India.’ Modern Asian Studies 45(5):1051–1094. Corbridge, Stuart, Glyn Williams, Manoj Srivastava and Reneé Véron, eds. 2005. Seeing the State: Governance and Governmentality in India. New York: Cambridge University Press. Corbridge, Stuart, John Harris and Craig Jeffrey, eds. 2013. India Today: Economy, Politics and Society. Cambridge: Polity Press. Council Proceedings of the Bengal Legislative Council, 45th session 1935—11th–28th February 1935. 1935. Alipore: Superintendent, Government Press. Crook, Malcolm, and Tom Crook. 2007. ‘The Advent of the Secret Ballot in Britain and France, 1789–1914: From Public Assembly to Private Compartment.’ History 92(308):449–471. Crook, Tom. 2011. ‘Secrecy and Liberal Modernity in Victorian and Edwardian England.’ In The Peculiarities of Liberal Modernity in Imperial Britain, edited by Simon Gunn and James Vernon, 72–90. Berkeley: University of California Press. Cunningham, Robert, and Yasin K. Sarayrah. 1993. Wasta: The Hidden Force in Middle Eastern Society. Westport, CT: Praeger. Cüppers, Christoph, ed. 2004. The Relationship Between Religion and State in Traditional Tibet. Lumbini: Lumbini International Research Institute. Daniel, E. Valentine. 1984. Fluid Signs: Being a Person the Tamil Way. Berkeley: University of California Press. Daniel, E. Valentine. 2010. ‘The Lost Ur.’ In Village Matters: Relocating Villages in the Contemporary Anthropology of India, edited by Diane P. Mines and Nicolas Yazgi, 317–339. New Delhi: Oxford University Press. Das, Sisir K. 2001. Public Office, Private Interest: Bureaucracy and Corruption in India. New Delhi: Oxford University Press. Das, Veena. 2004. ‘The Signature of the State: The Paradox of Illegibility.’ In Anthropology in the Margins of the State, edited by Veena Das and Deborah Poole, 225–252. Oxford: Oxford University Press. Davis, John. 1977. People of the Mediterranean: An Essay in Compar­ ative Social Anthropology. London: Routledge and Kegan Paul. Davis, Marvin. 1983. Rank and Rivalry: The Politics of Inequality in Rural West Bengal. Cambridge: Cambridge University Press. Dayal, Rajeshwar. 1998. A Life of Our Times. Delhi: Orient Longman. de Certeau, Michael. 1984. The Practice of Everyday Life. Berkeley: University of California Press.

Bibliography 409

de Giorgi, Alessandro. 2010. ‘Immigration Control, Post-Fordism, and Less Eligibility: A Materialist Critique of the Criminalization of Immigration across Europe.’ Punishment and Society 12(2): 147–167. de Heusch, Luc. 1997. ‘The Symbolic Mechanisms of Sacred Kingship: Rediscovering Frazer.’ Journal of the Royal Anthropological Institute 3(2):213–232. de Neve, Geert. 2000. ‘Patronage and “Community”: The Role of a Tamil “Village” Festival in the Integration of a Town.’ Journal of the Royal Anthropological Institute (n.s.) 6(3):501–519. de Neve, Geert. 2004. ‘The Workplace and the Neighbourhood: Locating Masculinities in the South Indian Textile Industry.’ In South Asian Masculinities, edited by Radhika Chopra, Caroline Osella and Filippo Osella, 60–95. New Delhi: Kali for Women and Women Unlimited. de Wit, Joop. 1996. Poverty, Policy and Politics in Madras Slums: Dynamics of Survival, Gender and Leadership. London: Sage. de Wit, Joop. 2013. ‘The Mumbai Municipal Elections of February 2012: Money, Muscle Power and Machine Politics.’ Unpublished Paper. de Wit, Joop, and Erhard Berner. 2009. ‘Progressive Patronage? Municipalities, NGOs, CBOs and the Limits to Slum Dwellers’ Empowerment.’ Development and Change 40(5):927–947. de Zwart, Frank. 1994. The Bureaucratic Merry-Go-Round. Amsterdam: Amsterdam University Press. Desai, Ashok. 1999. ‘The Economics and Politics of a Transition to Open Market Economy: India.’ OECD Technical Paper 155. Paris: OECD. Devika, Jayakumari. 2007. En-Gendering Individuals: The Language of Re-forming in Early Twentieth Century Kerala. Hyderabad: Orient Longman. Dhuru, Arundhati, and Sandeep Pandey. 4 April 2007. ‘Mulayam Faces Uphill Battle in UP.’ Rediff News, Accessed on June 13, 2013 http:// Diemberger, Hildegard. 2007. When a Woman Becomes a Religious Dynasty. New York: Columbia University Press. Diemberger, Hildegard. 2012. ‘Quand le Livre Devient Relique—Les Textes Tibetains Entre Culture Bouddhique et Transformations Tecnologiques.’ Terrain: Revue d'Ethnologie Européenne 59:18–39.

410  Bibliography

Dirks, Nicholas B. 1979. ‘The Structure and Meaning of Political Relations in a South Indian Little Kingdom.’ Contributions to Indian Sociology 13(2):169–206. Dirks, Nicholas B. 1987. The Hollow Crown: Ethnohistory of an Indian Kingdom. Cambridge: Cambridge University Press. Dogan, Mattei, and Dominique Pleassy. 1984. How to Compare Nations. Chatham, NJ: Chatham House Publishers. Dreyfus, Georges. 1994. ‘Proto-Nationalism in Tibet.’ In Tibetan Studies, edited by Per Kvaerne. Oslo: Institute for Comparative Research. Dreyfus, Georges. 2003. The Sound of Two Hands Clapping: The Education of a Tibetan Buddhist Monk. Berkeley: University of California Press. Dube, Saurabh. 1998. Untouchable Pasts: Religion, Identity, and Power among a Central Indian Community, 1780-1950. Albany: SUNY Press. Dumont, Louis. 1980. Homo Hierarchicus: The Caste System and Its Implications. Chicago: University of Chicago Press. Dumont, Louis. 1986. Essays on Individualism: Modern Ideology in Anthropological Perspective. Chicago: University of Chicago Press. Dundas,