Politics of Caste in West Bengal

THE POLITICS OF CASTE IN
WEST BENGAL
“This important volume imaginatively unravels heterogeneous, sub-

terranean, and formative histories . . . of caste.”
– Saurabh Dube, Center for Asian and African Studies,
El Colegio de México
“This timely, provocative, and scholarly book . . . [wi]th its rich eth-
nography and fine-tuned political sense . . . challenges the comfort-
able assumptions of Bengal as a society in which class relations have
trumped traditional inequities and hierarchies . . . [T]he authors put
caste centrally on the agenda of social theory in South Asia.”
– Dilip Menon, Centre for Indian Studies in Africa,
University of Witwatersrand
This volume offers – for the first time – a comprehensive and in-depth
analysis of the making and maintenance of a modern caste society in
colonial and postcolonial West Bengal in India. Drawing on cutting-
edge multidisciplinary scholarship, it explains why caste continues to
be neglected in the politics of and scholarship on West Bengal, and how
caste relations have permeated the politics of the region until today. The
chapters presented here dispel the myth that caste does not matter in
Bengali society and politics and make possible meaningful comparisons
and contrasts with other regions in South Asia.
The work will interest scholars and researchers in sociology, social
anthropology, politics, modern Indian history, and cultural studies.
Uday Chandra is postdoctoral research fellow at the Max Planck

Institute for the Study of Religious and Ethnic Diversity, Göttingen,
Germany.
Geir Heierstad is research director at the Department of International
Studies, Norwegian Institute for Urban and Regional Research, Oslo,
Norway.
Kenneth Bo Nielsen is postdoctoral research fellow at the Depart-

ment of Sociology, University of Bergen, Norway.
Contributors: Sarbani Bandyopadhyay/Sekhar Bandyopadhyay/Anasua

Basu Ray Chaudhury/Kenneth Bo Nielsen/Uday Chandra/Partha Chat-
terjee/Geir Heierstad/Indrajit Roy/Arild Engelsen Ruud/Dwaipayan
Sen/Moumita Sen/Praskanva Sinharay.
EXPLORING THE POLITICAL IN SOUTH ASIA
Series Editor: Mukulika Banerjee
Associate Professor, Department of Anthropology,
London School of Economics and Political Science
Exploring the Political in South Asia is devoted to the publication

of research on the political cultures of the region. The books in this
series present qualitative and quantitative analyses grounded in field
research, and explore the cultures of democracies in their everyday
local settings, specifically the workings of modern political institu-
tions, practices of political mobilisation, manoeuvres of high politics,
structures of popular beliefs, content of political ideologies and styles
of political leadership, among others. Through fine-grained descrip-
tions of particular settings in South Asia, the studies presented in this
series inform, and have implications for, general discussions of democ-
racy and politics elsewhere in the world.
Also in this Series
THE VERNACULARISATION OF DEMOCRACY

Politics, Caste and Religion in India
Lucia Michelutti
978-0-415-46732-2
RISE OF THE PLEBEIANS? THE CHANGING FACE OF

THE INDIAN LEGISLATIVE ASSEMBLIES
Editors: Christophe Jaffrelot and Sanjay Kumar
978-0-415-46092-7
BROADENING AND DEEPENING DEMOCRACY

Political Innovation in Karnataka
E. Raghavan and James Manor
978-0-415-54454-2
RETRO-MODERN INDIA
Forging the Low-caste Self
Manuela Ciotti
978-0-415-56311-6
POWER AND INFLUENCE IN INDIA

Bosses, Lords and Captains
Editors: Pamela Price and Arild Engelsen Ruud
978-0-415-58595-8
DALITS IN NEOLIBERAL INDIA

Mobility or Marginalisation?
Editor: Clarinda Still
978-1-138-02024-5
WHY INDIA VOTES?

Mukulika Banerjee
978-1-138-01971-3
CRIMINAL CAPITAL
Violence, Corruption and Class in Industrial India
Andrew Sanchez
978-1-138-92196-2
THE POLITICS OF CASTE IN WEST BENGAL

Edited by Uday Chandra, Geir Heierstad, and Kenneth Bo Nielsen
978-1-138-92148-1
POLITICS, LANDLORDS AND ISLAM IN PAKISTAN

Nicolas Martin
978-1-138-82188-0
THE POLITICS OF CASTE
IN WEST BENGAL
Edited by Uday Chandra,

Geir Heierstad, and Kenneth
Bo Nielsen
First published 2016
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2016 Uday Chandra, Geir Heierstad, and Kenneth Bo Nielsen
The right of Uday Chandra, Geir Heierstad, and Kenneth Bo Nielsen to
be identified as the authors of the editorial material, and of the authors
for their individual chapters, has been asserted in accordance with
sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic, mechanical, or
other means, now known or hereafter invented, including photocopying
and recording, or in any information storage or retrieval system,
without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Catalog record for this book has been requested
ISBN: 978-1-138-92148-1 (hbk)

ISBN: 978-1-315-68631-8 (ebk)
Typeset in Sabon
by Apex CoVantage LLC
CONTENTS
List of figures ix
List of tables x
Foreword xi
Acknowledgements xiv
List of abbreviations xv
Introduction 1
UDAY CHANDRA, GEIR HEIERSTAD, AND KENNETH BO NIELSEN
1 Kol, coolie, colonial subject: a hidden history of caste

and the making of modern Bengal 19
UDAY CHANDRA
2 Another history: Bhadralok responses to Dalit political

assertion in colonial Bengal 35
SARBANI BANDYOPADHYAY
3 Partition, displacement, and the decline of

the Scheduled Caste movement in West Bengal 60
SEKHAR BANDYOPADHYAY AND ANASUA BASU RAY CHAUDHURY
4 Partition and the mysterious disappearance of

caste in Bengal 83
PARTHA CHATTERJEE
5 An absent-minded casteism? 103

DWAIPAYAN SEN
vii
CONTENTS
6 The politics of caste and class in Singur’s anti-land

acquisition struggle 125
KENNETH BO NIELSEN
7 Building up the Harichand-Guruchand movement:

the politics of the Matua Mahasangha 147
PRASKANVA SINHARAY
8 Transformative politics: the imaginary of the Mulnibasi

in West Bengal 169
INDRAJIT ROY
9 From client to supporter: economic change and

the slow change of social identity in rural West Bengal 193
ARILD ENGELSEN RUUD
10 Craft, identity, hierarchy: the Kumbhakars of Bengal 216

MOUMITA SEN
11 The commodification of caste and politics in

Kolkata’s Kumartuli 240
GEIR HEIERSTAD
Glossary 263
Index 267
viii
FIGURES
10.1 Pratima of Goddess Kali being finished at

Kumortuli, 2012. 218
10.2 Tarit Pal’s Studio, Krishnanagar, 2013. 219
10.3 Dilip Pal making a portrait. 230
10.4 Two signs showing the names of the mritshilpis on
pandals. 234
ix
TABLES
4.1 Distribution of caste and religious groups as

percentage of total district population, 1931. 86
4.2 Literacy and occupation of selected castes as
percentage of the total population of the various
castes in Bengal, 1931. 87
4.3 Proportion of Scheduled Castes, Scheduled Tribes,
and Muslims as percentage of total district
population, 2001. 87
4.4 Religion/caste and literacy in Calcutta, 1931. 89
6.1 Frequency of household assets per 100 households,
calculated as the total number of an asset relative to
the total number of households. 137
7.1 Organisational structure of the Matua Mahasangha. 157
9.1 Landownership in percentage of total village land
in 1957. 197
9.2 Landownership by jati in percentage of total land
1957, 1981, 1993, and 2011. 202
9.3 Landownership groups by jati in number of
households 2011. 202
x
FOREWORD
‘Caste is not important in Bengal’ is an often-repeated phrase in discus-

sions of contemporary India. It is used when comparing West Bengal
with other states, where caste is considered a more determining fac-
tor in many issues. What is, of course, meant by this remark is that
the democratic upsurge that India has witnessed in large parts of the
country since the 1990s, which mobilised lower castes into active politi-
cal participation and gave them an unprecedented visibility in electoral
politics, did not happen in Bengal in the same way. This argument of
the relative lack of importance of caste in Bengal is further supported by
the lack of any of Bengal’s major political parties being associated with
any particular caste or cluster of castes; it has been dominated first by
the Congress, and then by the Left Front for most of its postcolo-
nial history and neither party has advocated a caste-based identity or
interest.
But this leads to the crucial question whether the absence of caste
as a mobilisation strategy in electoral politics justifies its dismissal in
shaping the politics of Bengal? Or indeed, does the very dismissal indi-
cate a nascent upper-caste bias in how the narrative of the politics of
Bengal is told? Given the importance of Bengal’s intellectual and cul-
tural leadership in shaping India’s social renaissance and the national-
ist movement, this question has salience not just for an understanding
of Bengal, but that of India as a whole.
This volume addresses this important issue head-on and does it with
academic thoroughness. The editors have been driven by the urgency
of this question through observing the politics of contemporary Ben-
gal, which each of them do in their individual capacity. But they also
recognise that it is impossible (like so many other issues in understand-
ing the political in South Asia) to gain any understanding of it with-
out going back at least 200 years in the region. Thus, in this volume,
xi
FOREWORD
historians first provide fine-grained case studies of particular castes

and place them within the wider canvas of the hugely innovative and
exciting politics of 19th-century Bengal. Their sociologist and anthro-
pologist colleagues are then able to do the same with more contempo-
rary examples by placing particular incidents of caste mobilisations,
displacements, and changing hierarchies within the wider politics of
Communism and its challengers.
This volume shows that, whatever the time period examined or the
disciplinary lens through which it is examined, the puzzle of how a tiny
majority of upper castes (about 10 per cent in the early 20th century)
were able to exercise a remarkable hegemonic hold over the rest of
the population remains. There have been no bahujan political parties
in West Bengal, as in north India, nor has there been an anti-Brahmin
movement, as in south India. It begs the question as to why Bengal is so
different and, predictably, the answer is complex. It has to be first recog-
nised that the distinguishing feature of the Bengal elite, the bhadralok –
a group that had otherwise caste ramifications within it – is their col-
lective complete disavowal of manual work. The aspiration for any of
the lower castes to become a member of the bhadralok required, there-
fore, first and foremost, their ability to give up manual work. But as
the accounts in this book show, the elites deployed a number of strate-
gies to keep this from happening. A wide variety of governance strate-
gies such as the stereotyping of lower castes and indigenous peoples,
determining the settlement destinations of refugee populations from
the east during and just after Partition, and their mastery of electoral
strategies in independent India maintained their dominance. Also,
as the scholars in this volume and beyond have pointed out, factors
such as Partition also made the sort of alliance between Muslims and
lower-caste peasants, which we have seen in other parts of India, dif-
ficult. Further, the lower castes themselves – Rajbanshi, Namshudra,
Bagdi, and Paindra – do not share many commonalities to unite under
a single banner. Thus, the combination of post-Partition dynamics, the
advent of electoral politics, and the lack of a single mobilising caste
identity together created conditions under which lower-caste politics
of the sort we have seen elsewhere in India is well-nigh impossible in
Bengal.
In addition, the dominance first of Congress politics and then
the alliance of the Left Front parties for much of postcolonial Ben-
gal has been no small factor. Caste was not the determining fac-
tor in the electoral strategy for either political formation and the
34 year period of communist rule gave rise to the notion that ‘class’ was
xii
FOREWORD
the more important variable for determining electoral fortunes than

‘caste’. Furthermore, vote share figures at each election gave rise to
the notion of the ‘Muslim block vote’, similar to those elsewhere
in India, and one which the winning party got. The lack of preci-
sion of these ideas is no more evident than in the villages, where my
own research has been based for the past 15 years. There, among the
Muslims who form the majority of paddy cultivators, there are four
castes – Syeds, Shekhs, Mughals, and Pathans. The Syeds own the
land, the rest work as sharecroppers or daily wage labourers. The land
reforms of the 1980s changed this scenario and led to new dynamics
of ownership, income, and mutuality. The local village comrade, as
perfect reflection of the membership of the CPI(M)’s politburo, is an
upper-caste Syed who formed his fiefdom by mobilising the labouring
castes. However, his machinations also alienated most of the other
upper-caste Syeds, but not all. In addition, his uneven dealings had also
alienated some of the Shekhs and Pathans. During elections, people
tended to vote on the basis of which party they felt most attachment
for and on the basis of kinship links. Thus, both the Congress and Left
Front vote contained many upper- and lower-caste votes, each formed
according to a complex algorithm of loyalty, caste, and class identities.
This changed further with the rise of Trinamul Congress. Thus, the
dynamics of caste in a Muslim setting – and a quarter of West Bengal’s
populations is Muslim – is far from predictable, certainly not replaced
by class and upper-caste hegemony has been enhanced by the long
years of the Left Front.
And yet, as some of the studies in this volume show, the hegemony
of the upper castes as well as the case for West Bengal ‘exceptional-
ism’ is less secure than is supposed. Challenges by organised caste
organisations, the defiance of bhadralok norms by the grassroots Con-
gress leader, Mamata Banerjee, herself an upper caste by birth but one
who questioned the patrician manners of communist leaders, and the
eastward aspirations of Hindutva politics are all contributing factors.
The next 10 years in West Bengal will be very unlike the previous 10.
A volume such as this had been long overdue. But at least it has
arrived in good time.
Mukulika Banerjee
London
December 2014
xiii
ACKNOWLEDGEMENTS
This anthology grew out of a scholarly exchange on the political life of

caste in West Bengal in the pages of Economic and Political Weekly in
the latter half of 2012. Praskanva Sinharay, who is also a contributor to
this volume, started the debate when he published a short piece on the
politics of the Matua Mahasangha in contemporary West Bengal. Sin-
haray’s article elicited a critical commentary from Uday Chandra and
Kenneth Bo Nielsen, and later, from Sarbani Bandyopadhyay and Par-
tha Chatterjee. Chandra and Nielsen subsequently received an email
from Sekhar Bandyopadhyay, the pioneering historian of caste politics
in 20th-century Bengal, suggesting that we organise a workshop on
this neglected topic, which, all of us felt, was in need of greater and
more substantial scholarly attention. Several of the contributors to
this volume then met in the Norwegian capital, Oslo in September
2013 for a day-long discussion on how to make sense of the role of
caste in social and political life in West Bengal. This volume is based
on many of the papers presented at the Oslo workshop, the spirit of
which we seek to convey in the pages that follow.
The editors would like to thank the Nordic Forum for South Asia;
the Centre for Development and the Environment and the Depart-
ment of Culture Studies and Oriental Languages, University of Oslo;
and the University Grants Commission (India), for their kind sup-
port. Thanks also to Amalie Meling Vikse and Jin Kathrine Fosli, who
worked hard to make the Oslo workshop a success. Special thanks
are due to those authors who joined us later in the process, enabling
us to secure a broader thematic coverage. The editors would also like
to thank series editor Mukulika Banerjee for her encouragement and
Routledge India for opening its press for us.
xiv
ABBREVIATIONS
AISCF All India Scheduled Castes Federation

BAMCEF All India Backward and Minority Communities Employee
Federation
BJP Bharatiya Janata Party
BJS Bangiya Jana Sangha
BPL Below the Poverty Line
BPSCF Bengal Provincial Scheduled Castes Federation
CPI Communist Party of India
CPI(M) Communist Party of India (Marxist)
EIRC East India Refugee Council
INR Indian Rupee
KMC Kolkata Municipal Corporation
KMDA Kolkata Metropolitan Development Authority
KRRS Karnataka Rajya Raitha Sangha
LF Left Front
MM Matua Mahasangha
NREGA National Rural Employment Guarantee Act
OBC Other Backward Classes
PSP Praja Socialist Party
RSS Rashtriya Swayamsevak Sangh
SBBS Sara Bangla Bastuhara Samiti
SC Scheduled Caste
SKJRC The Committee to Save the Farmland of Singur
ST Scheduled Tribe
TMC Trinamul Congress
UCRC United Central Refugee Council
xv
This page intentionally left blank
INTRODUCTION
Uday Chandra, Geir Heierstad, and

Kenneth Bo Nielsen
This anthology explores a much-neglected theme in South Asian his-

tory and politics – namely, the politics of caste in colonial and post-
colonial West Bengal. Caste in West Bengal has been notoriously
understudied for at least three reasons. First, the political culture of
postcolonial West Bengal has tended to make all talk of caste a taboo.
Second, West Bengal, unlike many other regions in India, has not
experienced major caste-based social movements since 1947. Third,
the ruling elites of this eastern Indian region, the bhadralok, though
internally differentiated along many axes, have exercised a virtually
uncontested social dominance that is rather unique, even in a country
where the reproduction of power relations is anything but uncommon.
In The Politics of Caste in West Bengal, we address each of these intel-
lectual and political concerns by taking an interdisciplinary perspec-
tive that draws on scholarship in the fields of history, anthropology,
sociology, political science, and cultural studies. Our purpose is to
interrogate not only why caste continues to be neglected in the politics of
and scholarship on West Bengal but also how caste relations have, in
fact, permeated the politics of the region in the colonial and postcolo-
nial eras.
Caste in West Bengal, as our contributors show, has overt and
covert aspects. On the one hand, there is the obvious issue of upper-
caste dominance in the domain of formal politics despite the sway of
communism for more than three decades. On the other hand, there
are hidden, even insidious, ways in which a modern caste society has
flourished since colonial times and shaped academic, journalistic and
popular understandings of Bengali society, culture, history, and poli-
tics. This volume explores both the overt and covert workings of caste
as it interrogates the ‘long-held political myth’ (Roy 2012: 948) about
its irrelevance in West Bengal. Our endeavour in this anthology is not
1
U D AY C H A N D R A E T A L .
to offer the final word on the politics of caste in West Bengal, nor do
we purport to offer an exhaustive account of the career of caste in
West Bengal across the colonial–postcolonial divide. Instead, we aim
to provide the reader with a collection of stimulating chapters that
identify key events, processes and issues so as to sustain an intellectual
conversation that is both timely and relevant for those interested in
understanding the nature of politics in contemporary South Asia. In
this brief introduction, we situate the complex question of caste in West
Bengal in a broader context and provide an overview of the topics that
the book engages with. The individual chapters engage with topics
ranging from caste and the colonial encounter; Dalit political assertion
in the colonial context; the consequences of Partition; the construction
of bhadralok hegemony; the impact of the class-based politics of the
Left parties; the effect of commodification and economic transforma-
tion; and the changing dynamics of caste in contemporary popular
politics.
Situating West Bengal exceptionalism

Caste used to be India, and vice versa, as an India dominated by ‘caste
hierarchy fulfils the need for a single and powerful organising image
which enables people in the west to think about a particular non-
western society’ (Searle-Chatterjee and Sharma 1994: 2). Moreover,
caste ‘capture(s) internal realities in terms that serve the discursive
needs of general theory’ (Appadurai 1992: 45). As such, caste, or its
traits, has appeared both perpetual and ephemeral, as founded in
structure and practice, providing both emancipation and subjugation.
Debating the origins of caste as a system, or a set of systems, its ideo-
logical foundation or how it is embedded in practice has a long history
beyond the politics of West Bengal (Lewis 1958; Beidelman 1959;
Pocock 1962; Nicholas 1965; Dumont 1980; Dirks 1987, 2001;
Gupta 2004) that the contributors to this volume are well aware of.
However, in this volume, we seek not to engage directly in defining
what caste (jati, varna) really is or used to be. Instead, we recognise
that caste implies different things to different people at different times,
and that caste is used politically in numerous ways. Inelegantly speak-
ing, the contributors to this volume approach caste as something
(groups of) people in India – and in West Bengal, in particular – do to
each other in a more or less systematic manner.
The impetus for putting together this volume came in the wake
of the publication of a short paper by Praskanva Sinharay in
2
INTRODUCTION
the pages of Economic and Political Weekly in 2012. Sinharay, who

is also a contributor to this volume, wrote that the politics of West
Bengal had historically been ‘truly unique’ when compared to other
Indian states with regard to the caste question because in that state,
caste had never been a relevant political category. Now, however,
massive changes were underway, spearheaded by the assertive voice
of the Namasudra, as channelled through the Matua Mahasangha.
The emergence of the strong, organised voice of the Matuas, Sinharay
argued, signalled the coming of an entirely new politics of caste in a
state otherwise inhospitable to what Rajni Kothari (1970: 5) long ago
called ‘the politicisation of caste’.
Sinharay’s (2012: 26) claim that West Bengal is (or was until
recently) a ‘truly unique’ case when set in the broader Indian context is
indicative of what is often labelled ‘West Bengal exceptionalism’. Sev-
eral chapters in this volume detail the historical trajectory of the mak-
ing and consolidation of this ‘exceptionalism’, and we refer the reader
to the contributions by Sekhar Bandyopadhyay and Anasua Basu Ray
Chaudhury and Partha Chatterjee, in particular, for a more detailed
overview and analysis. But, put briefly, the impact of Partition; the
general mainstreaming of caste groups and religious minorities around
large, secular parties (first the Congress, later, the CPI(M)) in the post-
colonial period; the ideological subsumption of caste by class under
prolonged Left Front hegemony; the almost complete dominance of
rural Bengal’s ‘party society’ (Bhattacharya 2011) by the Left par-
ties, ostensibly underpinned by loyalties forged ‘across divisions of
caste and community’ (Chatterjee 1997: 69); and the conspicuous – if
often under rug swept – sociopolitical dominance of the self-professed
‘casteless’ upper-caste bhadralok (Lama-Rewal 2009) have combined
to render the language of caste illegitimate in political discourse (Ban-
dyopadhyay 2012). As a result, the politics of caste in West Bengal
has rarely been foregrounded in scholarly work as being of any par-
ticular significance in shaping struggles over power, influence and the
distribution of resources, whether material or symbolic. This has been
most clearly the case when such work has focussed on the aggregate
level of state politics, where none of the major parties champion the
cause of any singular caste group or conglomerate of caste groups, and
where the said parties attract voters from all castes and communities,
albeit to varying degrees. This scenario, of course, contrasts starkly
with the recent rapid ‘rise of the plebeians’ (Jaffrelot and Kumar 2009) –
the SCs and OBCs – to power and influence across north India (see,
e.g. Jaffrelot 2003; Michelutti 2008). In Uttar Pradesh, for instance,
3
the Yadav-dominated Samajwadi Party competes with the Chamar-

dominated bahujan/sarvajan alliance of the Bahujan Samaj Party for
power and influence in a patronage democracy (Chandra 2004); while
in Bihar, the Yadav-dominated Rashtriya Lok Dal of Lalu Prasad has,
until very recently, battled it out against the Kurmi-Mahadalit vote
bank of Nitish Kumar. And on the sidelines, both the Jats of Uttar
Pradesh, led by Ajit Singh, and the Dusadhs of Bihar, led by Ram Vilas
Paswan, have caste-based ‘parties of their own’. In West Bengal, by
contrast, organised and sustained Dalit assertion has been negligible;
no dominant OBC has emerged as a political force post-Mandal; the
electorate has not been polarised along communal lines; and attempts
at forming a broader bahujan political identity out of a conglomerate
of castes and communities have been largely unsuccessful.
Sinharay’s article elicited a critical commentary from Uday Chandra
and Kenneth Bo Nielsen (2012), and later, from Sarbani Bandyopad-
hyay (2012) and Partha Chatterjee (2012), all of whom are contribu-
tors to this volume. While sympathetic to Sinharay’s analysis of the
‘new caste politics’ of the Matua Mahasangha, the commentaries
pointed out the need for situating emerging forms of caste politics
in a broader ethnographic and historical context. From such a con-
textualised perspective, one would be led to question both the onto-
logical standing of ‘West Bengal exceptionalism’ and Sinharay’s claim
to ‘newness’ on behalf of the Matua mobilisation. Both Bandyopad-
hyay and Chatterjee shared the trained historian’s scepticism towards
such claims to ‘newness’, speculating instead that ‘the structures of
bhadralok dominance are too well-fortified for one assembly election
to bring about a dent in those structures’ (Bandyopadhyay 2012: 73).
Yet, rather than simply debunking the ‘exceptionalism’ of the Ben-
gal situation as pure myth, the critiques encouraged us to scrutinise
it more carefully – an argument recently echoed by Ranabir Samad-
dar (2013). All the chapters in the present volume have taken this
task seriously as they grapple with the constitution, reproduction and
transformation of specific forms of politics and contestation in which
caste is operative. Importantly, our authors are sensitive to the fact
that whether caste emerges as a relevant category in the politics of
West Bengal depends crucially on how one defines ‘politics’ and how
one studies it. As Partha Chatterjee (1997: 83, 86) rightly notes, in the
‘apparently uninstitutionalised world of what may be called politics
among the people’, caste categories have continued to provide many
of the basic signifying terms through which collective identities and
social relations are still perceived. This is not so different from other
4
INTRODUCTION
states such as Tamil Nadu, Maharashtra or Bihar, where political par-

ties have coalesced diverse communities along caste lines, and where
the impact of caste on organised politics is obvious (see, e.g. Gorringe
2005; Waghmore 2013; Witsoe 2013). In states such as West Bengal,
where the caste question does not formally dominate party politics,
we may be mistaken to conclude that caste loyalties have disappeared
from popular consciousness (Chatterjee 1997: 84). There may, indeed,
be an exceptional contrast between the politics of caste in West Ben-
gal and in other north Indian states, which arises primarily from the
dominance of the upper-caste Hindu middle classes, the bhadralok.
But this contrast does not necessarily extend to the level of popular
ideology or consciousness (Chatterjee 1997: 86).
Critically interrogating the standard narrative about West Bengal
exceptionalism, the historically informed contributions by Uday Chan-
dra and Sarbani Bandyopadhyay bring out the importance of caste
as a crucial political category in undivided colonial Bengal. Chandra
scrutinises the remaking of caste relations in an emerging capitalist
political economy over the 19th century, while Bandyopadhyay analy-
ses caste as a centre of gravity for collective political action in the first
half of the 20th century. That the politics of caste was obviously cen-
tral to the political life of the province at the time can be gleaned from
how, from as early as the 1880s, subordinated caste groups such as the
Namasudras had organised themselves in ritual and economic spheres
against the upper-caste bhadralok (Bandyopadhyay 2011: 35–48). The
conscious materialism of the Matua cult, a phenomenon addressed
in this volume by Praskanva Sinharay, Sekhar Bandyopadhyay and
Anasua Basu Ray Chaudhury, contrasted starkly with Ramakrishna’s
other-worldly exhortations against work (kaaj) and wealth (kanchan)
(Sarkar 1992). Sekhar Bandyopadhyay (2011: xi), in his magisterial
history of social protest by the Namasudras of Bengal, thus attacked
‘the powerful political myth that caste did not matter in this part of the
subcontinent’. Even during the Swadeshi and nationalist movements
in late colonial Bengal, lower-caste and Adivasi groups did not make
common cause with the bhadralok. Hence, Aloysius (1998: 69) writes
that the perception that ‘the whole of Bengal Presidency supported the
bhadralok-sponsored renaissance and the subsequent phenomenon of
Swadeshi nationalism is a myth’, one that has been perpetrated by
a great many writers at that. In this context, Sarbani Bandyopadhy-
ay’s article on Dalit assertion is a timely and detailed reminder of the
‘hyper-visibility of caste’ (Bandyopadhyay 2012: 71) in the domain of
institutionalised politics in the late colonial period.
5
If caste in West Bengal was thus ‘hyper-visible’ right up to the time

of Independence, how and why did it disappear so rapidly from popu-
lar political discourse and action? The articles by Bandyopadhyay and
Ray Chaudhury, and Chatterjee offer two stimulating answers to the
puzzle of this strange disappearance of the caste question that centre
on the long history of Partition and its aftermath. The spatial displace-
ment, massive migration and sometimes violent intimidation associ-
ated with Partition, they argue, diluted in many respects the politics of
Dalit assertion in the postcolonial context. Since migration patterns,
the degree of spatial rupture, and the modalities of postcolonial gover-
nance were refracted through the prism of caste, certain caste groups
were favoured over others. For complex and contested reasons, this
effectively undermined the potential for the formation of strong ‘Dalit
counter publics’ (Hardtmann 2009) and a concomitant caste-based
Dalit counter-politics in the postcolonial era. Yet, while Bandyopad-
hyay and Ray Chaudhury, and Chatterjee, in effect, offer a perspec-
tive of the longue durée, centred on the event of Partition, to explain
the decline of Dalit assertion and the consolidation of bhadralok hege-
mony in the postcolonial context, Dwaipayan Sen’s refreshing chapter
cautions against an overly structuralist reading of the aftermath of
Partition. Sen argues that the ‘silencing’ of caste was also, albeit not
exclusively, the outcome of the exercise of social agency on the part
of the bhadralok. Rather than deciding which of these two interpreta-
tions is more accurate, we find it fruitful to keep the inherent produc-
tive tension of this exchange alive throughout this volume. As Uditi
Sen (2013) has shown recently, careful study of the structural trans-
formations wrought by Partition-enforced migration to West Bengal
is hardly incompatible with a fine-grained understanding of how
bhadralok domination came to be re-asserted in the postcolonial era.
‘Post-Communism’ and the normalisation of

West Bengal politics?
Anthropologists with fieldwork experience in rural West Bengal have,
of course, for long been well aware of the persistence of caste in popu-
lar consciousness and social and political life, both under Congress
rule and the prolonged hegemony of the Left Front. Moving beneath
the radar screen of aggregate state-level electoral politics, Dayabati
Roy’s (2012; 2014) recent fieldwork finds caste hierarchies widespread
in village society and demonstrates the entrenchment of a caste con-
sciousness among the upper- and middle-caste leaders and cronies of
6
INTRODUCTION
the CPI(M). Similarly, Mukulika Banerjee’s (2010) case-study of the

CPI(M)’s comrades in Birbhum shows how a local party boss from the
dominant Syed caste in the village commands the loyalty of lower-
caste Muslims, such as Sheikhs and Pathans. This may not come across
as very surprising, given the social origins of bhadralok or madhy-
abitta Marxism in early-20th-century Bengal (Dasgupta 2005). Since
the 1930s, the politics of bhadralok Marxism was, as much as that of
the right-wing Hindu Mahasabha, an upper-caste Hindu alternative to
a weak, declining Congress in the province (Gallagher 1973). In effect,
the politics of caste tended to remain a key organising principle of
everyday rural life.
Anthropologists of rural West Bengal have similarly underscored the
limited social transformation wrought by the CPI(M) in a society where
caste remained an everyday reality. Arild Engelsen Ruud’s pioneering
ethnographic work has, for example, examined in great detail how
the Marxist penetration of rural Bengal did not lead to any deeper
revolution in local perceptions of power and influence as in, say, post
colonial Bihar (Kunnath 2012). Ruud (1994) suggests that, although
the Marxist movement may have mobilised the masses, particularly
lower-caste groups, it nonetheless behaved and was perceived as a tra-
ditional patron (albeit a more just and potent one than older patrons).
And elsewhere, Ruud (2003: 146) has demonstrated that local rural
responses to the Marxist message were generally influenced by local
histories and experiences, inflected by caste relations and stereo-
types. Hence, the dominant ideology of village society in West Ben-
gal remained one of inequality, hierarchy and rank, separateness and
distinction, as Marvin Davis’s (1983) ethnographic work also brings
out. While Ruud’s contribution to this volume suggests that caste hier-
archies in rural Bengal may now, in fact, be in the process of partly
withering away, Kenneth Bo Nielsen’s chapter brings to light the con-
tinued ability of caste hierarchies to structure political action, even
within contexts in which caste is upheld as irrelevant. That the politics
of caste in rural West Bengal has thus led a healthy life, casts consider-
able doubt on just how deep West Bengal exceptionalism ran in the
first place.
Yet, in spite of these deeper continuities, there has been a growing
sense in recent years that something radically new is underway with
regard to the role of caste in the politics of the state. This growing
common sense has been spawned by a series of interrelated develop-
ments that have combined to propel communal and caste questions
to the top of the political agenda. The cumulative dissatisfaction with
7
Left Front rule; the CPI(M)’s heavy-handed land acquisitions in Singur

and Nandigram (Sarkar and Sarkar 2008; Jones 2009; Nielsen 2010);
its alienation of Muslim voters (Nielsen 2011); and the unexpected
ability of Mamata Banerjee to project herself as the new champion of
the peasantry and the poor and downtrodden, finally dislodged the
Left Front from power in the state in 2011. And in contrast to her
‘secular’ communist counterparts, Mamata Banerjee has no scruples
about appealing overtly to caste and communal sentiments. Her ‘post-
bhadralok’ (Gupta 2012: 132) style of politics has, in turn, opened up
new spaces for articulating a plurality of communal identities in politi-
cal forums – something which, for instance, the Muslim electorate has
made good use of (Nielsen 2011; 2012). So, too, have the numeri-
cally strong Matuas, as Sinharay demonstrates in this volume; and
there have even been attempts at forging a more inclusive Mulnibasi
political identity among marginalised groups, as Indrajit Roy’s chapter
shows. At the same time, Hindu nationalist sentiments promulgated
by the Sangh Parivar – most notably, the Rashtriya Swayamsevak
Sangh (RSS) and the Bharatiya Janata Party (BJP) – have now gained a
foothold in a state where they were, for long, considered a non-entity.
At the recently concluded 2014 Lok Sabha elections, the BJP polled
a full 17 per cent of the popular vote, whereas the once undefeatable
electoral machinery of the CPI(M) managed just over 22 per cent. The
surge of the BJP has drawn nourishment from long-standing concerns
among Bengali Hindu voters over illegal migration of Muslims from
Bangladesh, whom the BJP calls ‘Muslim infiltrators’, and the discon-
tent with what the BJP projects as Mamata Banerjee’s pro-Muslim pol-
itics. Yet, apart from thus seeking to consolidate and add to whatever
support the party has traditionally had among upper-caste Hindus,
the BJP, in fact, appears to have expanded the most in the southern
SC, ST and OBC-dominated areas of the state. The party has success-
fully organised SC and ST communities against Muslim infiltrators
in some of the border areas; and more generally, the fact that the BJP
has focussed on the living conditions of backward Hindu communi-
ties means that nearly three out of four of the party’s local leaders hail
from the backward castes. The irony that a party that is, for very good
reasons, conventionally seen as a bastion of conservative and patri-
archal upper-caste values may be capable of denting bhadralok hege-
mony by effecting a genuine transformation in the social composition
of political leadership in the state is, perhaps, obvious.
At another level, these ‘post-communist’ transformations in the
state’s political culture have, in many respects, brought West Bengal
8
INTRODUCTION
closer to the general north Indian pattern, where caste and communal
identities are politically salient, competitive populism is deeply embed-
ded in electoral politics and cycles, and a considerable level of politi-
cal violence is the order of the day (Nielsen 2014). At the same time,
it is, as Ranabir Samaddar (2013) has reiterated, surely too early to
proclaim the dawn of an entirely new politics of caste in West Bengal:
Mamata Banerjee’s Trinamul Congress may yet assume its place as the
new king of ‘party society’, thereby closing off competing channels of
mediation and transactions, and thus, once again gloss over the opera-
tions of the politics of caste in everyday social and political life.
The chapters
The chapters in The Politics of Caste in West Bengal are structured
around four key topics: (1) caste and colonialism; (2) Partition and the
making of a modern caste society; (3) caste and popular politics; and
(4) caste, stratification and the economy.
Uday Chandra’s opening chapter probes the limits of recent aca-
demic writing that treats modern articulations of caste as a direct or
indirect consequence of colonial governmentality. The example of the
Kols of Chotanagpur in the westernmost fringe of 19th-century Ben-
gal, he explains, points to an alternative theoretical perspective on how
caste relations were transformed as the social relations of production
under colonial conditions. The Kols migrated from their homes to
work as construction workers and sweepers in colonial Calcutta, as
forest-clearers in the Sundarbans and Assam, and as plantation labour-
ers in the tea gardens of North Bengal and Assam. Despite being clas-
sified in late-19th-century ethnographic accounts as a ‘tribe’, the Kols
became the labouring caste par excellence in the political economy
of colonial eastern India. The hard and dirty labour they performed
placed the Kols as the lowest of the low at the bottom of the emerging
social hierarchies of modern Bengal. The winners in the new capitalist
economy were not only European plantation owners and representa-
tives of the Raj but also the Bengali landowners, managers and better
off labouring groups in the same social field. Power and domination
were thus diffused across economy and society. The historical processes
that led to this reordering of Bengali society have been obscured to a
great extent, and this chapter goes some distance towards unravelling
these processes.
Sarbani Bandyopadhyay’s contribution is an examination of the
dialectics between Dalit political assertion and bhadralok responses
9
in the first half of the 20th century. In this period, the alienation of a
large section of the Bengali population from mainstream bhadralok
society became visible when the Hindu bhadralok failed to draw the
‘untouchable’ and some ‘lower castes’ in adequate numbers into the
Swadeshi movement that began in 1905. These castes opposed it, and
most allied with the Muslims and the colonial government to thwart
the Swadeshi programme. To the lowest castes, Swadeshi was little
more than a conspiracy to further bhadralok dominance. To con-
tain this and similar forms of Dalit assertion, influential sections of
bhadralok society actively sought to appropriate this assertion to serve
nationalist goals. They did so, Bandyopadhyay shows, by embarking
on a programme of building ‘Hindu unity’ through caste reforms –
a process which also entailed the communalisation of identities. By
drawing on consolidated and influential networks among caste Hin-
dus, and by discursively subsuming caste into the larger question of
class, independent forms of Dalit mobility and mobilisation were
effectively blocked.
Sekhar Bandyopadhyay and Anasua Basu Ray Chaudhury’s chapter
proceed from the pertinent observation that the Scheduled Caste
movement before 1947 was most powerful in east and north Ben-
gal: for both the Namasudras and Rajbansis, their close geographical
location in these areas offered them a crucial spatial capacity for social
mobilisation. Therefore, the loss of that spatial anchorage as a result
of Partition and the consequent physical displacement and dispersal
of a large section of the Dalit peasant population of Bengal had an
adverse impact on their social and political movements, which were,
from then on, overshadowed by their struggle for resettlement. From
1950, Scheduled Caste peasants migrated on a large scale from East
to West Bengal. The presence of these Dalit peasant refugees changed
the texture of politics in postcolonial West Bengal as the displaced
Dalits acquired the new identification of being ‘refugees’ – the only
publicly identifiable oppressed group in a new post-Partition discourse
of victimhood. As refugees defined by the experience of migration and
the camp life, they faced a different kind of struggle – the struggle for
resettlement. While the refugees were never a homogenous category,
in the interest of a united struggle, their left-liberal and predominantly
high-caste leadership deliberately purged the vocabulary of caste from
their language of protest, which could then be more easily appropri-
ated into the modern tropes of social justice deployed by the main-
stream political parties and the state. This did not imply that the caste
10
INTRODUCTION
question was resolved; it only meant that caste became less conspicu-
ous, though not non-existent, in the public discourse of social justice
and political protest.
Partha Chatterjee’s chapter similarly grapples with the conse-
quences of Partition for the politics of caste in postcolonial West
Bengal. Chatterjee is in agreement with Sarbani Bandyopadhyay’s
argument that, as in other regions of India, the initial dominance of
upper-caste Hindus in middle-class occupations during the colonial
period came under severe challenge in Bengal in the last two decades
before independence. The rise of a new educated middle class from
among the superior peasantry and popular political mobilisation led
to an assault on the institutions of upper-caste privilege. But the con-
sequence of independence and the partition of the province was that
the erstwhile dominance of upper castes was re-established in West
Bengal. The reversal happened during the lifetime of a single genera-
tion without anyone talking about it. To Chatterjee, this is nothing
less than ‘a social counter-revolution’ that took place behind a veil of
silence. Breaking this counter-revolution down to its constituent parts,
Chatterjee offers an historical account of eight features of ‘the new
middle-class formation’ that successfully constituted itself as a domi-
nant culture that was, in Gramsci’s sense, hegemonic. This dominance
is not – even today – in any serious danger, Chatterjee suggests, not
least because of the immensely superior control exercised by the upper
castes over the mechanisms of electoral democracy through their dom-
inance of the party system, from left to right.
Dwaipayan Sen’s chapter explores the analytically vexed problem
of agency with respect to the following anomaly: the domination
of West Bengal’s political, social and cultural domains by the upper
castes, even as it was surely proclaimed that caste did not matter – the
perpetuation of caste inequality by those who disavowed the salience
of caste, as it were. But who, or what, was the agent of this domi-
nation? Sen asks. This question requires consideration because, Sen
argues, the resumption of upper-caste domination and concomitant
decline of the political visibility of caste have been explained primarily
as a consequence of social structure in the first case, and acquiescence
and accommodation in the latter. In contrast, Sen encourages us to
consider upper-caste domination as far more willed and coercive than
we are usually given to believe. The disproportionate influence com-
manded by the upper castes of West Bengal and the related ‘silence’
about the caste question, Sen argues, was also the outcome of their
11
exercise of social agency. Sen’s chapter also focuses on contemporary

activist discourse about upper-caste domination and Dalit inequality,
and the prospects for alliances between Dalit and Muslim communi-
ties. The excursion into those fields leads Sen to argue – and in contrast
to what Chatterjee’s chapter suggests – that the presumed bhadralok
hegemony is far less secure than we might think.
Kenneth Bo Nielsen’s chapter portrays one of the most talked-about
rural movements in West Bengal in recent years – namely, that against
the Tata Motors car factory in Singur. The movement of Singur’s
unwilling farmers to resist forced land acquisition was instrumental in
turning the rural mood against the incumbent Left Front government,
and pried open new political spaces for Mamata Banerjee’s assertive
populism (Banerjee 2004). While much has been written on the impli-
cations of the Singur movement for political change and development
in West Bengal, the role of caste in it has, as Nielsen (forthcoming) has
argued elsewhere, remained a neglected issue. While the scant atten-
tion generally accorded to the politics of caste in West Bengal may
form part of the explanation for why this should be the case, another
part of the explanation is that support for the Singur movement (or
lack thereof) on the ground did not neatly follow caste lines. Yet, as
Nielsen’s detailed ethnography shows, the everyday politics of caste
was effectively operative in shaping both the local organisation of the
Singur movement and the articulation of its agenda. Nielsen’s con-
tribution is, thus, further evidence of the continued salience of caste
hierarchies in forms of popular politics that operate beneath the level
of the aggregate.
Praskanva Sinharay’s contribution is a study of the political mobili-
sation of the Matuas, constituted almost entirely by the lower-caste
Namasudras on the grounds of caste loyalty, under the banner of
Matua Mahasangha. Over the past few elections held in the state – the
panchayat polls in 2008; the Lok Sabha elections in 2009 and 2014;
and the Legislative Assembly elections in 2011 – all political parties
engaged in a tug-of-war to ensure en bloc electoral support from the
Matuas in the around 35 state assembly constituencies, where their
votes are believed to be decisive. By detailing three crucial aspects of
the politics of Matua Mahasangha – its organisational history and
contemporary structure, its role in the Dalit literary movement and
its political activism as the community organisation of the Matuas –
Sinharay argues that the politics of Matua Mahasangha has introduced
‘a new politics of caste’ in West Bengal. By carving out an autonomous
12
INTRODUCTION
political position outside the ambit of bhadralokism, and by declaring

itself the vanguard of the Dalits and the underprivileged people, the
Matua Mahasangha is, Sinharay suggests, in the process of creating
a counter-discourse that can stand outside and at the same time effi-
ciently challenge bhadralok dominance.
Indrajit Roy’s contribution analyses the work and writings of
the Mulnibasi Samiti, the West Bengal State body of the Rashtriya
Mulnivasi Sangh (National Mulnibasi Organisation), which, again,
functions as the cultural wing of the All India Backward and Minor-
ity Communities Employee Federation (BAMCEF). BAMCEF was
founded in 1983 by the charismatic Kanshi Ram, and BAMCEF and
the National Mulnibasi Organisation share the same political slogans:
‘Jai Mulnivasi’, or victory to Mulnibasi. In its imagination, the fig-
ure of the Mulnibasi refers to the autochthonous inhabitants of India,
that is, the members of the Scheduled Castes, Scheduled Tribes, Other
Backward Classes, and religious minorities. Its ‘Other’ is the Ary-
ans: the Brahmins, the Kshatriyas and the Vaishyas, with Eurasian
antecedents. To the supporters of the Mulnibasi Samiti, the histori-
cal experience of West Bengal had been uniquely unfortunate. Despite
the language of modernity permeating the state’s cultural and politi-
cal spaces for nearly two centuries, caste remained a central idiom
of a social life characterised by relations of exploitation and inequal-
ity between the Mulnibasis and their tormentors. In a telling quota-
tion from one of Roy’s informants, the oppressed in West Bengal are
upheld as being worse off than ‘our people in Bihar’, who at least
‘enjoyed some respect’. To Roy, the negotiations, conversations and
agonisms attending to transforming identifications towards the figure
of the Mulnibasi are indicative of the transformative politics being
advanced by a new political generation in rural West Bengal. Just how
far this project is able to go remains to be seen.
Arild Engelsen Ruud’s chapter is a historical study of the trans-
formations occurring in caste and the economy in a Bengali village.
Although West Bengal is often portrayed as somewhat of a laggard state
when compared to, say, Gujarat or the Punjab, the economic transforma-
tions that have occurred in the Bengali countryside over the past five
decades have been nothing short of spectacular. The stark inequali-
ties and agrarian impasse that had characterised rural West Bengal
for a very long time (Bhattacharyya and Bhattacharyya 2007) gave
way, from the 1970s, to pro-poor land reforms and a period of sus-
tained and high agricultural growth. Living standards improved
13
considerably and later, many rural Bengalis – almost from across the
caste spectrum – have successfully diversified out of agriculture and
now derive their livelihoods from a variety of sources. As a result,
the very hierarchical land-based patronage relations of old between
upper-caste landlords and lower-caste dependents have crumbled.
Today, the core of patron–client relationships is neither caste, jati nor
labour, but rather ‘politics’ – that is, the distribution of protection
and access to state resources and programmes, mediated by politi-
cal parties. In this radically changed economic and political context,
the sociopolitical salience of caste identities is both transformed and
increasingly withering away.
Moumita Sen’s contribution is a historical study of how the Kumb-
hakar, the potter caste of West Bengal, has negotiated with questions
of identity, hierarchy, power, and status through the practice of its
craft. Tracing the evolution of different art styles over time, Sen anal-
yses the construction of hierarchies within the caste-based artisanal
community centred on their capacity to appropriate different ways of
moulding reality. By tying together shifting forms of patronage and
alternating regimes of art, aesthetics and taste, Sen presents a compel-
ling account of how the changing nature of a craft can lead to signifi-
cant changes in individual and collective status and identity.
Geir Heierstad’s contribution builds on the chapters by Ruud and
Sen. Like Sen, Heierstad too writes about the Kumars of Kolkata’s
Kumartuli, the last of the city’s larger caste-based neighbourhoods.
And like Ruud, he examines the impact of economic transformations on
the meaning and practice of caste. Caste identity among the Kumars
has, Heierstad demonstrates, increasingly been turned into an emblem
under which a craft is practised, products sold, markets monopolised,
and political battles fought. This, to Heierstad, amounts to a ‘com-
modification of caste’. Today, the Kumars see themselves, and are seen
by society at large, as bearers of Bengali tradition and history, who
have, at the same time, skilfully adapted to the demands of contem-
porary consumers and clients. This ‘commodified’ caste identity may,
in turn, be leveraged as a political resource to bring development and
other benefits to Kumartuli. By thus analytically integrating changing
self-understandings and art practices among the Kumars with broader
patterns of socio-economic change, Heierstad brings to light crucial
aspects of the historical transformations in the politics of caste among
this particular caste group.
In combination, the 11 chapters in The Politics of Caste in West Ben-
gal offer rich and stimulating accounts of the making and remaking
14
INTRODUCTION
of caste relations and identities in a diversity of sociopolitical settings.

While we have aimed for a broad thematic coverage, we do not aspire
to offer the final word on the politics of caste in West Bengal. Indeed,
there are several crucial areas of discussion that the book does not
address. This includes, crucially, the politics of caste among Chris-
tian and Muslim communities, and the complex intersections between
caste and gender. We, therefore, hope the reader will see this book as
an invitation to further comparative research on different dimensions
of caste in West Bengal.
Bibliography
Aloysius, G. 1998. Nationalism without a Nation in India. New Delhi: Oxford
University Press.
Appadurai, Arjun. 1992. ‘Putting Hierarchy in Its Place’, in George E. Marcus
(ed), Rereading Cultural Anthropology, pp. 34–47. Durham: Duke University
Press.
Bandyopadhyay, Sarbani. 2012. ‘Caste and Politics in Bengal’, Economic and
Political Weekly, 47(50): 71–73.
Bandyopadhyay, Sekhar. 2011. Caste, Protest and Identity in Colonial India:
The Namasudras of Bengal, 1872–1947. New Delhi: Oxford University
Press.
Banerjee, Mukulika. 2004. ‘Populist Leadership in West Bengal and Tamil
Nadu: Mamata Banerjee and Jayalalithaa Compared’, in Rob Jenkins (ed),
Regional Reflections: Comparing Politics across India’s States, pp. 285–308.
New Delhi: Oxford University Press.
Banerjee, Mukulika. 2010. ‘Leadership and Political Work’, in Pamela Price
and Arild Engelsen Ruud (eds), Power and Influence in India: Bosses, Lords
and Captains, pp. 20–43. New Delhi: Routledge.
Beidelman, Thomas O. 1959. A Comparative Analysis of the Jajmani System.
New York: J. J. Augustin.
Bhattacharyya, Dwaipayan. 2011. ‘Party Society, Its Consolidation and
Crisis: Understanding Political Change in West Bengal’, in Anjan Ghosh,
Tapati Guha-Thakurta and Janaki Nair (eds), Theorizing the Present:
Essays for Partha Chatterjee, pp. 226–50. New Delhi: Oxford University
Press.
Bhattacharyya, Maumiti and Sudipta Bhattacharyya. 2007. ‘Agrarian Impasse
in West Bengal in the Liberalisation Era’, Economic and Political Weekly,
42(52): 65–71.
Chandra, Kanchan. 2004. Why Ethnic Parties Succeed: Patronage and Ethnic
Head Counts in India. Cambridge: Cambridge University Press.
Chandra, Uday and Kenneth Bo Nielsen. 2012. ‘The Importance of Caste in
Bengal’, Economic and Political Weekly, 47(44): 59–61.
15
Chatterjee, Partha. 1997. The Present History of West Bengal: Essays in Political
Criticism. New Delhi: Oxford University Press.
Chatterjee, Partha. 2012. ‘Historicising Caste in Bengal Politics’, Economic
and Political Weekly, 47(50): 69–70.
Cohn, Bernard S. 2004. ‘The Census, Social Structure and Objectification in
South Asia’, in An Anthropologist among the Historians and Other Essays,
pp. 224–54. New Delhi: Oxford University Press.
Dasgupta, Rajarshi. 2005. ‘Rhyming Revolution: Marxism and Culture in
Colonial Bengal’, Studies in History, 21(1): 79–98.
Davis, Marvin. 1983. Rank and Rivalry: The Politics of Inequality in Rural
West Bengal. Cambridge: Cambridge University Press.
Dirks, Nicholas B. 1987. The Hollow Crown: Ethnohistory of an Indian King-
dom. Cambridge: Cambridge University Press.
Dirks, Nicholas B. 2001. Castes of Mind: Colonialism and the Making of
Modern India. Princeton: Princeton University Press.
Dumont, Louis. 1980 (1970). Homo Hierarchicus. The Caste System and Its
Implications. Oxford: Oxford University Press.
Gallagher, John. 1973. ‘Congress in Decline: Bengal, 1930 to 1939’, Modern
Asian Studies, 7(3): 589–645.
Gorringe, Hugo. 2005. Untouchable Citizens: Dalit Movements and Demo-
cratization in Tamil Nadu. New Delhi: Sage Publications.
Gupta, Dipankar (ed). 2004. Caste in Question: Identity or Hierarchy?
New Delhi: Sage Publications.
Gupta, Monobina. 2012. Didi: A Political Biography. Noida: HarperCollins.
Hardtmann, Eva-Maria. 2009. The Dalit Movement in India: Local Practices,
Global Connections. New Delhi: Oxford University Press.
Jaffrelot, Christophe. 2003. India’s Silent Revolution: The Rise of the Lower
Castes in North India. London: C. Hurst & Co.
Jaffrelot, Christophe and Sanjay Kumar (eds). 2009. Rise of the Plebeians?
The Changing Face of Indian Legislative Assemblies. New Delhi: Routledge.
Jones, Jonathan D. 2009. ‘Negotiating Development: A Study of the Grass-
roots Resistance to India’s 2005 Special Economic Zones Act’, unpublished
Ph.D. dissertation, University of Florida.
Kothari, Rajni. 1970. ‘Introduction’, in Rajni Kothari (ed), Caste in Indian
Politics, pp. 3–26. Hyderabad: Orient Longman.
Kunnath, George J. 2012. Rebels from the Mud Houses: Dalits and the
Making of the Maoist Revolution in Bihar. New Delhi: Social Science
Press.
Lama-Rewal, Stephanie Tawa. 2009. ‘The Resilient Bhadralok: A Profile of
the West Bengal MLAs’, in Christophe Jaffrelot and Sanjay Kumar (eds),
Rise of the Plebeians? The Changing Face of Indian Legislative Assemblies,
pp. 361–92. New Delhi: Routledge.
Lewis, Oscar. 1958. Village Life in Northern India: Studies in a Delhi Village.
New York: Vintage.
16
INTRODUCTION
Michelutti, Lucia. 2008. The Vernacularisation of Democracy: Politics, Caste

and Religion in India. New Delhi: Routledge.
Nicholas, Ralph W. 1965. ‘Factions: A Comparative Analysis’, in Max Gluck-
man and Fred Eggen (eds), Political Systems and the Distribution of Power,
pp. 21–61. London: Tavistock.
Nielsen, Kenneth Bo. 2010. ‘Contesting India’s Development? Industrialisa-
tion, Land Acquisition and Protest in West Bengal’, Forum for Development
Studies, 37(2): 145–70.
Nielsen, Kenneth Bo. 2011. ‘In Search of Development: Muslims and Electoral
Politics in an Indian State’, Forum for Development Studies, 38(3): 345–70.
Nielsen, Kenneth Bo. 2012. ‘Campaigning for Inclusion: Muslims and Social
Exclusion in Contemporary West Bengal’, in V. Rama Krishna, R. Shashid-
har and M. Muniraju (eds), Social Exclusion – Inclusion Continuum: A
Paradigm Shift, pp. 119–35. Bangalore: Niruta Publications.
Nielsen, Kenneth Bo. 2014. ‘Mamata Banerjee: Redefining Female Leadership
in India’s Democracy’, in Arild Engelsen Ruud and Geir Heierstad (eds),
India’s Democracies: Diversity, Co-optation, Resistance, pp. 80–111. New
Delhi: Primus Books.
Nielsen, Kenneth Bo. Forthcoming. ‘Managing “Communities” of Resistance:
Negotiating Caste and Class in an Anti-Land Acquisition Movement in
West Bengal’, in Uday Chandra (eds), Staking Claims: The Politics of Social
Movements in Contemporary Rural India. New Delhi: Oxford University
Press.
Pocock, David F. 1962. ‘Notes on Jajmani Relationships’, Contributions to
Indian Sociology, 6(1): 78–95.
Roy, Dayabati. 2012. ‘Caste and Power: An Ethnography in West Bengal,
India’, Modern Asian Studies, 46(4): 947–74.
Roy, Dayabati. 2014. Rural Politics in India: Political Stratification and Gov-
ernance in West Bengal. Delhi: Cambridge University Press.
Ruud, Arild Engelsen. 1994. ‘Land and Power: The Marxist Conquest of
Rural Bengal’, Modern Asian Studies, 28(2): 357–80.
Ruud, Arild Engelsen. 2003. Poetry of Village Politics: The Making of West
Bengal’s Rural Communism. New Delhi: Oxford University Press.
Samaddar, Ranabir. 2013. ‘Whatever Has Happened to Caste in West Bengal?’
Economic and Political Weekly, 48(36): 77–79.
Sarkar, Sumit. 1992. ‘ “Kaliyuga”, “Chakri” and “Bhakti”: Ramakrishna and
His Times’, Economic and Political Weekly, 27(29): 1543–66.
Sarkar, Sumit and Tanika Sarkar. 2008. ‘A Place Called Nandigram’, in Gau-
tam Ray (ed), Nandigram and Beyond, pp. 19–53. Kolkata: Gangchil.
Searle-Chatterjee, Mary and Ursula Sharma. 1994. ‘Introduction’, in Mary
Searle-Chatterjee and Ursula Sharma (eds), Contextualising Caste: Post-
Dumontian Approaches, pp. 1–24. Oxford: Blackwell.
Sen, Uditi. 2013. ‘The Myths Refugees Live By: Memory and History in the
Making of Bengali Refugee Identity’, Modern Asian Studies, 48(1): 37–76.
17
Sinharay, Praskanva. 2012. ‘A New Politics of Caste’, Economic and Political

Weekly, 47(34): 26–27.
Waghmore, Suryakant. 2013. Civility against Caste: Dalit Politics and Citizen-
ship in Western India. New Delhi: Sage Publications.
Witsoe, Jeffrey. 2013. Democracy against Development: Lower-Caste Politics
and Political Modernity in Postcolonial India. Chicago: University of Chicago
Press.
18
1
KOL, COOLIE,
COLONIAL SUBJECT
A hidden history of caste and the
making of modern Bengal
Uday Chandra
Historical anthropologists of modern India, such as Bernard Cohn

(1996), Arjun Appadurai (1993), and Nicholas Dirks (2001), have
argued forcefully that caste, as a modern social institution, came to be
revived and reproduced by the colonial state via its classificatory and
enumerative policies. Yet, this colonialism-centred perspective, though
useful in many senses, obscures the everyday sociocultural and political-
economic processes by which the colonised organised themselves
under colonial overlordship. Insofar as caste is a system of organising
labour on the basis of a hierarchical social logic, it is important to
understand how distinctive ‘regional modernities’ (Sivaramakrishnan
and Agrawal 2003) were built, quite literally, on the backs of labour-
ing groups assigned the lowest ritual and socio-economic status in
these new regions.
This chapter uncovers a ‘hidden’ history of one such labouring
group in 19th-century Bengal, who appear in the colonial archives as
‘Kols’, despised in caste terms by the Hindu bhadralok, yet categorised
subsequently via ethnological accounts as ‘tribes’. The Kols, some-
times known as Dhangars, appear in the colonial records from the
time they helped build the imperial capital of Calcutta from the neigh-
bouring forest highlands of Chotanagpur in the first decade of the
19th century. Decried as dirty or impure but valued for their ability to
perform hard physical labour, the Kols served as construction work-
ers as well as sweepers and cleaners in Calcutta. By the middle of the
century, colonial archives suggest that the Kols had turned into coolies
for the indigo and tea plantations of modern Bengal. In the plantation
economy, the lowly Kols – men and women alike – performed hard
19
U D AY C H A N D R A
agricultural labour that other caste groups were deemed incapable of.
Subsequently, as land had to be reclaimed and forests cleared in the
Sunderban delta, the Kols were called upon to alter the natural and
human ecology of the area. Even as they were classified as ‘tribes’
by anthropologist-administrators in Chotanagpur, the Kols became
the labouring caste par excellence in modern Bengal. The sociocul-
tural and political-economic processes by which this occurred have,
nonetheless, been hidden from the gaze of later historians raised on
the venerable caste/tribe dichotomy in Indian sociology. This chapter
offers a preliminary sketch of this hidden history of labour, caste and
subjecthood on which Bengali regional modernity came to rest by the
end of the 19th century and which continues to pervade the postcolo-
nial present.
‘Dirty Swines’ in the imperial capital

In Die Gossnersche Mission Unter den Kols (1874 – ‘The Gossner
Mission among the Kols’), the Lutheran pastor Dr Alfred Nottrott
describes the Kols as ‘the wanderers of Calcutta’. These ‘mountain-
ous black children’, he proceeded to describe how he first met them,
‘were engaged in mean works like sweeping the roads and carrying
the goods etc. in this world town’. ‘In that age’, he added, ‘it [Kols]
meant “Dirty Swines” ’ (cited in Mahto 1971: 21). Some of these
Kols, as they were known throughout Bengal in the early 19th cen-
tury, appear with shovels as the ‘scavengers of Calcutta’ in the litho-
graphs of Colesworthy Grant1 (1846) or the many paintings by
Company School artists depicting the construction of Fort William in
colonial Calcutta. Yet, we know exceedingly little about these men
and women from Chotanagpur: what they did in the city, their every-
day pursuits and their shifting position between the rural and urban
worlds.
Our earliest encounters with the Kols or ‘Coles’ in the Bengal Presi-
dency are in colonial records that view these labouring groups as dif-
ferent from other inhabitants of Bengal by virtue of their strength and
tempermant, shaped apparently by the rugged environs in which they
were raised. In the words of Major J. Sutherland, the Kols were ‘one
family’, ‘wild’, ‘savages’, but ‘as free and independent as any people
1 See, for example, BL/IOC/APAC/P2553.
20
KOL, COOLIE, COLONIAL SUBJECT
on the Earth’.2 The Kols were also, in his opinion, ‘an industrious
people, possessing a beauty and mostly a highly cultivated country’
in the highlands of Chotanagpur. Yet, colonial officials were unsure
how to situate the Kols in their sociological understanding of Indian
society. One military officer described them as:
A race distinct from the great Hindoo family both in manners,

language, religion and appearance inferior in some respects to
the common inhabitants of the hills in point of civilization,
but superior to them in courage and industry, and possess-
ing large and flourishing villages with extensive tracts of well
cultivated land.3
A civilian official, however, saw the Kols as ‘the lowest kind of Hin-
doos’.4 Without the caste/tribe dichotomy that has dominated Indian
sociology since the mid-19th century, such confusion in the colonial
records over the Kols is understandable and, in fact, rather revealing.
Despite their confusion, British officialdom soldiered on and divided
the Kols into two groups, ‘Lurka Coles’ and ‘Dhanger Coles’, better
known from later colonial ethnological works as Hos and Mundas,
respectively. The Lurka Kols, so called for their reputation as fearless
fighters (lurka literally means ‘fighter’), resided on the southern edge of
the Ranchi plateau in Singhbhum. Major Edward Roughsedge described
them during his military expedition in Singhbhum in 1820 as follows:
Not having any of the feelings of veneration for Bramins

Cows which pervade Hindoos of every description they make
no scruple of putting to death any man of respectable caste
who presumes to enter their Territory, nor is there . . . a single
Bramin Rajpoot or Mussulman in any one of the numerous
and well inhabited villages, they possess. A traveller would as
soon think of visiting into a Tiger’s den, as of traversing any
part of Lurka Cole.
2 ‘Note by Major J. Sutherland, Private Secretary to the Hon’ble the Vice President’,
Fort William Judicial Consultations No. 44 of 17 April 1832 and Charles Metcalfe,
‘Vice President’s Minute’, Fort William Judicial Consultations No. 16 of 17 April
1832, IOR/F/4/1363/54227.
3 Extract Political Letter from Bengal, dated 9 May 1823, IOR/F/4/800/21438.
4 S. T. Cuthbert, Magistrate of Ramghur, to Mr Secretary Shakespear, Bengal Judicial
Consultations No. 53 of 14 June 1827, letter dated 21 April 1827, BL/IOR/E/4/731.
21
To compare the Lurkas to the Dhangers, Roughsedge wrote, ‘they [the

Lurkas] are as much superior in size and form to the tame Danghers,
if I may use the expression of Chota Nagpore, though of one common
origin, as wild Buffaloes are to the village Herds’.5 In Major Sutherland’s
view, the Dhanger Coles of the Chotanagpur plateau were ‘a remarkably
industrious and peaceable people and who have a character for truth
and honesty beyond that of many of the people of India’. The Dhangers
were hitherto subjects of the Nagbanshi rajas of Chotangpur in ‘nearly
4,000 inhabited villages’ in the five Parganas (‘Paanch Pergunnah’) of
Rai, Bundu, Silli, Tori, and Tamar.6 As the Collector S. T. Cuthbert noted
after his extensive tour of the Chotanagpur countryside in 1826–27, the
Dhangers were seen to ‘emigrate in great number annually during the
agricultural off-season in search of employment’ to Calcutta as well as
other districts in rural Bengal. ‘They are’, he wrote, ‘generally preferred to
the labourers of other parts of the Country on account of their perform-
ing more work and at a lower rate’. That meant, typically, that ‘in a fam-
ily consisting of four or five persons, two are left at home to take care of
the family affairs and cultivation and the rest go abroad to seek service’.7
This is how the Kols came to be regarded in early colonial Bengal
as labourers par excellence. Deemed to be a lowly caste by the British
and their upper-caste collaborators, the Kols went about building Cal-
cutta’s new ‘white town’, keeping its streets clean and drains unclogged,
and digging the banks of canals. It is here that Christian missionaries
‘discovered’ them. In November 1845, when the first batch of German
Lutheran evangelists arrived in Calcutta, ‘strolling one morning in the
narrow streets . . . by the bank of the river Hooghly . . . they saw some
natives with dark skin’. Curious, they asked the wife of Anglican Bishop
Hoeberlin, ‘Who are these people we saw, so low and so degraded?’ Mrs
Hoeberlin replied, ‘They are Cols from West Bengal’. From these Kols,
the missionaries learned that they were migrants from Chotanagpur,
located (in those days) 15 days west of Calcutta. Their home, the Kol
migrants reported, ‘was full of green forests, high mountains and a large
number of big and small rivers flowing under the clear blue sky’. They
had come to ‘the din and bustle of a metropolitan city like Calcutta in
order to earn some “Paise” [cash] which could improve their material
5 Extract Bengal Political Consultations No. 38 of 3 June 1820, IOR/F/4/800/21438.

6 ‘Note by Major J. Sutherland, Private Secretary to the Hon’ble the Vice President’,
Fort William Judicial Consultations No. 44 of 17 April 1832, IOR/F/4/1363/54227.
7 S. T. Cuthbert, magistrate of Ramghur, to Mr Secretary Shakespear, Bengal Judicial
Consultations No. 53 of 14 June 1827, letter dated 21 April 1827, BL/IOR/E/4/731.
22
conditions at home’ (cited in Mahto 1971: 19–22). In a similar vein,

Eyre Chatterton (1901: 5), an Anglican priest who spent five decades in
Chotanagpur, recalled in his memoirs how, as ‘young missionaries’, they
‘were at once struck by these dark-skinned, bright, merry-faced people’
labouring on the streets of imperial Calcutta. It was in the same circum-
stances that the Belgian Jesuit priest Constant Lievens, too, encountered
‘these people [who] are rather dark, but not negroes . . . [with] thick
lips, a flat nose, a round face, long black hair and are almost beardless’,
proceeding thereafter to the Chotanagpur ‘Mission in the West’ (cited
in Clarysse 1985: 68–72, 128). It is at the margins of missionary narra-
tives, therefore, that we learn of the common sight of Kol labourers on
the streets of the imperial capital of Calcutta.
Anthropological theories of caste and tribe do not explain very well
where the labouring Kols of Chotanagpur fit into the overall sociology
of South Asia. If we follow the dominant wisdom since colonial times,
the Kols are simply ‘tribes’, and hence, the colonial and missionary
records cited above merely misrecognise this fact by referring to them
as a lowly labouring caste. The problem with this reading is that it
relies on a colonial ideology of ‘primitivism’ (Chandra 2013a), which
ascribes a permanent ontological reality to a racialised notion of tribe
(Fried 1975; Béteille 1986). Even today, when the colonial notion of
tribe is under attack, even from many of those placed in the ‘savage
slot’ (Trouillot 2003; Chandra 2013b), colonial ideas of the primitive
Other continue to haunt the postcolonial present (Kuper 2005; Chat-
terjee 2013). So, we turn then to the other side of the caste/tribe binary,
namely, caste. Since it is no longer held that caste is entirely a ritual
matter of purity and pollution in the Dumontian sense, it has become
axiomatic among South Asianists that the modern caste order is a func-
tion of post-1858 colonial governmentality in British India (Cohn 1987;
Appadurai 1993; Dirks 2001). But if caste is taken to be a by-product of
colonial state-making processes, then it must be pointed out that colo-
nial censuses, surveys and ethnographic experiments did not produce
the Kols out of thin air. We thus face a conundrum: the conventional
academic wisdom on the much-vaunted caste/tribe dichotomy fails us
here. The next section seeks to unravel this conundrum.
Kols as coolies
It would be wrong to conclude from the evidence presented so far that
only the new British capital of Calcutta relied on Kol labour to func-
tion. The Bengal countryside relied on Kol labour too. Indeed, the
23
Kols were treated by higher caste landowners in the south-western

frontier of Bengal as forest-clearers, above all. The zamindars on this
jungli frontier, as in other forested regions in India, typically enlarged
their estates since at least the 15th century by sending out bands of
forest-clearers deeper and deeper into the most deciduous forests of
the subcontinent.8 Accordingly, the American anthropologist Richard
Fox (1969) labelled these forest-clearing bands ‘professional primi-
tives’ to rescue them from the colonial discourses of primitivism in
which they were enmeshed. In exchange for their labour in clearing
forests and expanding the arable frontier in Bengal and beyond, Kols
and other professional primitives received either rent-free lands or
lands at nominal quit rents. Early modern state formation in South
Asia, as Sumit Guha (1996) has shown so brilliantly, relied precisely
on such forest-clearing labour.
The onset of colonialism, as I have argued elsewhere,9 deepened and
hastened the early modern processes of regional state-making and the
development of land markets in eastern India. The progressive break-
down of social order in jungle zamindaris from the 1780s onwards,
owing to increasing subinfeudation and rent burdens on those previ-
ously paying little or no rents, released massive flows of labour into the
rest of early colonial Bengal, including Calcutta. As social structures
were reorganised across Bengal, the Kols of Chotanagpur came to occupy
their lowly status at the bottom of new hierarchies of life and labour.
The term ‘Kol’ was, as the opening quotation from Father Reverend
Nottrott about ‘dirty swines’ suggested earlier, a common ‘epithet of
abuse, applied by the Brahminical race’ (Dalton 1866: 144) or those
claiming superior caste status in early 19th-century Bengal. This usage,
steadily adopted by colonial officialdom too, referred to dark-skinned
migrants from forest zamindaris on the western frontier of colonial
Bengal, who performed degrading and demanding physical labour in
rural and urban settings. Without the later ethnological names given
by anthropologist-administrators to newly discovered ‘tribes’ such as
Munda, Oraon or Santal, the term ‘Kol’ acted as a catch-all term for
migrants from the forest highlands of Chotanagpur.
The Kols went to every corner of the Bengal Presidency: the indigo
farms in the plains of Bihar and Bengal proper, the deltaic swamps of
8 For south-western Bengal, see Sivaramakrishnan (1999); for examples from Bastar
and the Nigiris, see Sundar (1997) and Hockings (2013).
9 See chapters 2 and 3 of my doctoral dissertation (Chandra 2013c).
24
the Sunderbans and the tea plantations in Assam and sub-Himalayan

Bengal. Consider the indigo farms that came up from the 1830s.
Charles Metcalfe wrote in a Minute in 1832 that the ‘quiet inoffen-
sive Character’ of the Kols ‘and their industrious habits cannot but
be known to . . . the numerous European Gentlemen in the Lower
Provinces, who have been accustomed annually to employ large bod-
ies of these people in the manufacture of Indigo’.10 These Kol labour-
ers were recruited by men of higher caste rank sent by the indigo
plantation owners in the chief growing areas in colonial Bengal. For
instance, we hear of a certain ‘Kumul Sing a Servant of an Indigo
Planter who had been sent by his Master into the country to hire
Dhangers’11 from the area that eventually became Ranchi district.
Upper-caste labour recruiters, such as Kumul Sing, were as impor-
tant to the indigo set-up as the Kols themselves. They embodied,
in a sense, the new hierarchy of labour that was emerging under
colonial overlordship and capital. Jacques Pouchepedass (1999) has
shown how the most arduous tasks in the indigo production pro-
cess, for example, physically entering a tank full of water and indigo
to stomp out impurities, were carried out by the so-called jungli
Kols, because no one else would do it and the Kols were seen as
ideally suited for such tasks. So, in the caste hierarchy of the indigo
plantation, Kols effectively stood below the raiyats or rent-paying
peasants who were forced to cultivate indigo on their lands, the
landless farm workers who were dependents of zamindars or indigo
contractors, the upwardly mobile labour recruiters, the zamindars
who leased out their lands to European contractors, and the con-
tractors themselves. It is not surprising, therefore, that, when the
famous Santal Hul broke out in July 1855, district officials from
Bhagalpur, Aurangabad and Rajmahal regularly reported threats to
and attacks on European contractors, their indigo factories, as well
as their native subordinates in indigo-cultivating villages.12 Equally
10 The Vice President’s Minute, Fort William Judicial Consultations No. 46 of 17 April
1832, BL/IOR/F/4/1363/54227.
11 S. T. Cuthbert and T. Wilkinson, Joint Commissioners, to James Thomason, Deputy
Secretary to Government in the Judicial Department, Fort William Judicial Consul-
tations No. 59, dated 12 February 1832, BL/IOR/F/4/1363/54227.
12 See, for example, G. J. Brown, Commissioner of Circuit for the Bhagulpore Divi-
sion, to W. Grey, Secretary to the Government of Bengal, letter dated 11 July 1855;
W. Grey to the Secretary, Government of India, Military Department, letter dated
21 July 1855, BL/IOR/P/145/14.
25
unsurprising is the fact that the fierce mid-19th-century debates over

the rights of the indigo-cultivating raiyat in colonial Bengal (Sartori
2011) completely sidestepped the Kols’ circumstances at the bottom
of the labour hierarchy. Later historians, including those of a Subal-
ternist persuasion, have followed suit (see, for example, Guha 1974).
Sidestepping the Kols was not so easy, however, in the tea planta-
tions of Assam and northern Bengal, where they were, in fact, the
principal labour force as ‘coolies’.13 Until the passage of the Trans-
portation of Native Labourers Act (1863), which encouraged so-
called free emigration of coolies to Assam and Cachar, Kols from
Chotanagpur were taken primarily to Mauritius and the East Indies.14
Until then, the colonial government had ‘failed in inducing any of the
Coles or Dangurs, to proceed to Assam, where their labour in the
manipulation of tea [was] so required’.15 The Kols in the tea planta-
tions were in more or less the same circumstances as those in the
indigo plantations. The only difference lay in the proximity of the
latter to their rural homes in comparison with Assam, which the Kol
coolies often took to be ‘the end of the world’ (Sharma 2011: 73).
As in the indigo plantations, there was in Assam an elaborate hierar-
chy of labour: headed by European planters, upper-caste Assamese or
Bengali managers, upwardly mobile arkattis from central and eastern
13 Many etymological origins have been suggested for the word ‘coolie’, but Kol
remains a strong contender.
14 Lieutenant Colonel E. T. Dalton, Commissioner of Chotanagpur, to the Hon’ble
Ashley Eden, Secretary to the Government of Bengal, No. 1171, letter dated 28 June
1865, General (Emigration) Proceedings B62, July 1865, WBSA. The historian Jag-
dish Chandra Jha (1986: 9–11) gives us rare details of the first batch of 34 coolies
from Chotanagpur shipped to Mauritius in the aftermath after the Kol Insurrection
of 1831–32. According to Marina Carter (1995: 104), roughly a third of the 7,000
indentured coolies who arrived in Mauritius in 1837–38 were dhangars (‘Kols’)
from Chotanagpur. Widespread condemnation of the awful living conditions of
these early coolies en route to and in Mauritius briefly stopped emigration between
1838 and 1842, but it resumed thereafter under a more ‘managed’ system of inden-
tured labour. By the 1840s, however, coolies from Chotanagpur avoided Mauritius
in favour of the emerging tea plantations in Assam and north Bengal.
15 J. R. Ouseley, Governor-General’s Agent in the South West Frontier, to F. J. Halliday,
Secretary to the Government of Bengal, letter dated 8 June 1839, Home (Revenue –
Agriculture) Proceedings 21–22, 24 June 1839, WBSA. I am grateful to Andy Liu for
pointing me to the contents of this file. A similar lament appears as late as 1857 in a
letter that unsuccessfully seeks Kol coolies to help complete the highway connecting
the Dhaka and Arakkan Divisions: see India Public Works (Bengal), letter dated
14 July 1857, BL/IOR/E/4/845.
26
India who scouted for and recruited coolies, the local labourers who
were characterised as lazy and averse to physical labour, and finally,
the Kols and other coolies imported into Assam. Later, garden sir-
dars, chosen from among the Kol coolies, were sent to recruit more
coolies from rural Chotanagpur, independent of the arkattis.16 In her
study of the cultural worlds of the Assam tea plantations, Jayeeta
Sharma (2011: 74) writes:
By the end of the nineteenth century, Chotanagpur labourers

acquired the highest rank among Assam coolies. They were
known as ‘Class I junglies’ in the planter’s lexicon. In the
recruitment market, [they] were the most prised and the most
expensive: Planters ranked them high in terms of resilience,
labouring ability, and resistance to disease.
According to a newspaper report in The Times, between 700,000

and 750,000 ‘tea coolies’ came to Assam, of which roughly a third,
a quarter of a million coolies, were Kols from Chotanagpur (cited in
Cotton 1911: 264).
While postcolonial scholars of the Assam tea plantations have cer-
tainly not neglected Kol coolies from Chotanagpur, many have mis-
taken the cart for the horse in positing a ‘market for aboriginality’
(Ghosh 1999) at work there. Jayeeta Sharma (2011: 72), for instance,
cites Brian Hodgson and George Campbell, the latter a prominent
lieutenant governor of Bengal, to argue that the British ‘employed the
tenets of race science’ to justify the use of Kol coolies over local tribes
from Cachar. However, such an explanation presumes, without suffi-
cient warrant, that racial notions associated with ‘tribes’ caused planta-
tion owners and colonial administrators to seek Kol coolies to perform
certain difficult and degrading labour. Cultural-historical explanations
of this kind fail to see that the Kols were brought as coolies to Assam
because of their prior reputation as low-cost, hardworking manual
16 See detailed descriptions of the labour recruitment mechanisms in the following two
files: V. T. Taylor, Commissioner of the Chota Nagpore Division, to the Secretary to
the Government of Bengal, Judicial Department, No. 1658, letter dated 28 June
1877, Emigration Department, Financial Branch Proceedings No. 13–14, August
1877, WBSA; J. Ware Edgar, Officiating Commissioner of the Chota Nagpore Divi-
sion, to the Secretary to the Government of Bengal, Judicial Department, No. 479CR,
letter dated 3 November 1882, General (Emigration) Proceedings No. 11, December
1882, WBSA.
27
labourers in western Bengal, though as pointed out earlier, it was not

easy to induce more than a handful to undertake the arduous jour-
ney to Assam before the coercive indentured system took shape in the
1860s. Additionally, as Prabhu Mohapatra and Rana Behal (1992:
146) argue, ‘there is no doubt that the high wages demanded by local
labourers was an important consideration in the resort to long distance
recruitment’. George Campbell (1866: 34) himself admitted to ‘cheap-
ness of labour’ being a significant factor in the long-distance recruit-
ment of Kol coolies to Assam. In the words of a memorandum from
the Indian Tea Districts Association to London in 1880, the ‘future of
the tea industry hinged on the maintenance of an adequate supply of
coolie labour at a cost calculated to leave a fair margin of profit on
the capital invested’ (cited in Mohapatra and Behal 1992: 147). Lastly,
the Kol coolies, unlike local labourers, were subject to the provisions
of the 1863 Act, by which plantation owners enjoyed special privileges
that permitted them to catch and imprison any coolies who escaped.
Racial stereotypes of Kols, and later, individual tribal groups such as
Mundas and Oraons, did develop on the plantations (see, for example,
Tea Districts Labour Association 1924; 1925), but these were post-hoc
constructions that situated the plantations into a wider all-India ideol-
ogy of colonial primitivism. As such, we should be wary of reinscrib-
ing ‘primitivism’ onto the past by privileging post-hoc justifications for
recruiting Kol coolies over contemporaneous explanations rooted in a
simple socio-economic logic.
A labour theory of caste domination?

What was this socio-economic logic at work in Calcutta as well as the
indigo and tea plantations in colonial Bengal? Having ruled out
racialised notions of tribal or primitive labour and popular notions of
caste animated by principles of ritual purity/pollution or colonial gov-
ernmentality, I suggest we think in terms of caste as embodying the
social relations of production in modern India. The evidence presented
in this chapter from 19th-century Bengal reinterprets the Kols’ situation
at the bottom of the socio-economic ladder. The Kols were not merely a
subaltern class in itself, but the lowest of the low in the reworked caste
order of modern Bengal. At this point, the sceptical reader may justly
inquire whether I am not conflating ‘caste’ with ‘class’. Indeed, I am:
what I am proposing effectively is that, viewed from the bottom of the
social ladder, caste in modern India is, in fact, ‘the specifically Indian
28
form of material relations at the base, with its own historical dynamic’
(Chatterjee 1989: 175).17
This is not entirely a novel argument, of course. The likes of D. D.
Kosambi (1944), Dipankar Gupta (1980), Gail Omvedt (1982), Kum-
kum Roy (2008), and Anand Teltumbde (2012) have all been here
before. Indeed, as Teltumbde (2012) points out, Ambedkar himself
‘practised class politics, albeit not in the Marxian sense [insofar as]
he always used “class” even for describing the untouchables’. Where
I differ from these luminaries is in my singular focus on labour and
its relationship to both the production of value and the social hierar-
chies that are built on the backs of the labouring multitudes. In sum,
a labour theory of caste domination. The extraction of surplus value
in capitalist production processes should be clear enough, but the spe-
cifically Indian character of class relations qua caste lies in its visible
manifestations as ‘discrimination’, ‘ritual purity’, and/or ‘voting pat-
terns’. However, to take these manifestations of caste to be the same
as their underlying basis is the fundamental empiricist error that unites
academic and lay commentators on caste.
If we are to take the suggestion of a labour theory of caste domina-
tion seriously, then labouring groups such as the Kols in colonial Ben-
gal must be understood in terms of their position at the bottom of
the caste pyramid – assigned the hardest and most degrading physi-
cal labour imaginable. Regional modernities in Bengal and elsewhere
were built on the backs of labouring groups such as the Kols of
Chotanagpur. Histories of such groups are, however, ‘hidden’ by the
colonial ideology of primitivism that has seeped into the postcolo-
nial present. It is true that conditions of colonial capitalism produced
these ‘hidden’ histories of caste, but it would be wrong to see these as
merely a function of colonial governmentality. After all, Kols cleared
forests for jungle zamindars and rajas well before the onset of colo-
nial modernity. The late medieval Chandimangala of Mukundaram
Chakrabarti, for instance, refers extensively to ‘Beruniyas’,18 a Persian
17 Partha Chatterjee’s argument in this well-known essay on caste and subaltern con-
sciousness is, of course, not the same as mine here. Indeed, Chatterjee is criticising
the position that I am proposing here (or at least the versions of it that he had found
a quarter century ago). Due to constraints of space, I am unable to discuss Chat-
terjee’s argument and its relationship to mine in this chapter.
18 I am grateful to Professor Ralph Nicholas for directing me to the ‘Beruniyas’ in the
Chandimangala.
29
term for casual labour in zamindari estates who cleared forests and
built dams (Raychaudhuri 1969: 199). Moreover, the lines between
the everyday state and society are blurred in modern India and beyond
(Fuller and Bénéï 2001), and when writing about the colonial period,
it is useful to remember that ‘the raj was part of the same social field
as its subjects’ (Washbrook 1981: 713). So, if the British administra-
tors and capitalists were the ‘ruling caste’ (Gilmour 2006), it is worth
contemplating how ‘brown sahibs and white sahibs sought to escape
their fears about the instability of social hierarchy . . . covering extant
hierarchies [of caste] with the mantle of the natural and the primor-
dial’ (Guha 1998: 438). And, when Ranajit Guha (1983), the found-
ing father of Subaltern Studies, turned to study the ‘elementary forms
of peasant rebellion in colonial India’, it is unsurprising that he put
himself in a longer genealogy of bhadralok writers going back at least
to Sanjeeb Chandra Chattopadhyay (Banerjee 2006), who delighted
in romanticising the misfortune of those whom they and their fore-
fathers took great pains to keep at the bottom of the caste hierarchy.
Little wonder, then, that caste is almost completely absent from the
early volumes of Subaltern Studies (with the honourable exception of
Chatterjee (1989)).
Yet, caste domination is an inescapable reality in postcolonial West
Bengal as it is in the rest of India. To understand why, we would do
well to consider the sociologist Charles Tilly’s classic work Durable
Inequality (1999), which outlined a set of social mechanisms that
place productive resources in some hands at the expense of oth-
ers. Because haves and have-nots are subsequently locked in a vari-
ety of everyday transactions, categories that sustain socio-economic
inequality and power differentials in society arise. Much like class,
race, and gender, caste is also sustained in this manner as a principle
of categorising different sections of society. David Mosse (2010) has
recently expanded on Tilly’s thesis to offer a multidimensional ‘rela-
tional’ explanation of ‘durable poverty, inequality and power’ among
Dalits and Adivasis in modern India. By ‘relational’, Mosse means
that those who are ranked at the bottom of Indian society today are
poor not because they lack any intrinsic qualities that others possess,
but because of the power others enjoy over them under conditions
of modern capitalism. To the extent that caste embodies the social
relations of production in India, a relational theory of durable pov-
erty, inequality, and power leads us to appreciate how those occupying
the lowest rungs of caste society are subject to the most exploitative
labour regimes even as the bhadralok or ‘middle class’ is defined by
30
an aversion to physical labour. This state of affairs persists because

economy and society are happily in sync with each other. Max Weber’s
(1978: 933–34) perceptive analysis of how status groups harden into
castes is worth recounting here:
A status segregation grown into a ‘caste’ differs in its struc-

ture from a more ‘ethnic’ segregation: The caste structure
transforms the horizontal and unconnected coexistences of
ethnically segregated groups into a vertical social system of
super- and subordination.
As Weber rightly recognised, caste is a matter of political economy,

above and beyond the symbolic difference markers that are most
apparent to observers (Guha 2013: 9–11). For our purposes, the social
reproduction of vertically ordered strata of different kinds of labour
holds the key to understanding how caste domination persists as well
as the ways in which it is cunningly obscured in popular and academic
discourses. This is why uncovering hidden histories of caste in colonial
Bengal today is as much an exercise in reconstructing the past as it is
about making sense of the present.
Postscript
For the benefit of the reader, I want to recount my inspiration for this
chapter. One monsoon evening in 2010, after a day at the West Bengal
State Archives on Bhawani Dutta Lane, I boarded a Kolkata taxi en
route to Park Street. In the lengthy traffic jam that is typical during the
monsoons in Kolkata, the taxi driver and I started chatting in Bengali
about our respective places of origin. I learned then that Madan, as I
shall call him here, hailed from a Dom family in a village in the Sunder-
bans. Soon, the conversation turned to my upcoming field trip to Khunti
(Jharkhand). On hearing the word ‘Khunti’, Madan gasped. ‘Do you
know Longa gram?’ he asked. ‘Of course, I do’, I replied. I had been
there several times as an NGO worker and was planning to do so again
as a researcher. ‘Are there any Salupurti or Masapurti families there?’
Indeed, I said, there are many. ‘Those are my brothers-sisters, aunts-
uncles’, he exclaimed, with a smile of relief on his face. Over a hundred
years ago, Madan’s ‘grandfather’s grandfather’ (thakurdadar thakurd-
ada) had gone to the Sunderbans to clear forests before eventually mar-
rying a local woman and settling there. Those were days of intense
agrarian disputes in Chotanagpur, and Madan’s great-great-grandfather
31
was hardly alone in leaving the region in search of a better future. Appar-
ently, census surveyors had declared one of Madan’s ancestors a Dom,
and he now, officially, belonged to a scheduled caste (SC), unlike his
Munda (ST) extended family in Khunti. Caste histories of the kind that I
discuss in this chapter are thus ‘hidden’ in another sense too: beyond the
grasp of census officials, tax collectors, and historians who rely so heavily
on paper trails. I cannot say I have done more than to scratch the surface
ever so slightly.
Bibliography
Appadurai, Arjun. 1993. ‘Number in the Colonial Imagination’, in Carol
A. Breckenridge and Peter van der Veer (eds), Orientalism and the Post-
Colonial Predicament, pp. 314–40. Philadelphia: University of Pennsylvania
Press.
Banerjee, P. 2006. Politics of Time: ‘Primitives’ and History-Writing in a Colo-
nial Society. New Delhi: Oxford University Press.
Béteille, Andre. 1986. ‘The Concept of Tribe with Special Reference to India’,
European Journal of Sociology, 27(2): 296–318.
Campbell, George. 1866. ‘The Ethnology of India’, Journal of the Asiatic Soci-
ety of Bengal, 35(2): 1–150.
Carter, Marina. 1995. Servants, Sirdars and Settlers: Indians in Mauritius,
1834–1874. New Delhi: Oxford University Press.
Chandra, Uday. 2013a. ‘Liberalism and Its Other: The Politics of Primitivism
in Colonial and Postcolonial Indian Law’, Law & Society Review, 47(1):
135–68.
Chandra, Uday. 2013b. ‘Beyond Subalternity: Land, Community, and the State
in Contemporary Jharkhand’, Contemporary South Asia, 21(1): 52–61.
Chandra, Uday. 2013c. ‘Negotiating Leviathan: Statemaking and Resistance
in the Margins of Modern India’, unpublished Ph.D. dissertation, Yale Uni-
versity.
Chatterjee, Indrani. 2013. Forgotten Friends: Monks, Marriages, and Memo-
ries of Northeast India. New Delhi: Oxford University Press.
Chatterjee, Partha. 1989. ‘Caste and Subaltern Consciousness’, in Ranajit
Guha (ed), Subaltern Studies VI, pp. 169–209. New Delhi: Oxford Univer-
sity Press.
Chatterton, E. 1901. The Story of Fifty Years’ Mission Work in Chhota Nag-
pur. London: Society for Promoting Christian Knowledge.
Clarysse, L. 1985. Father Constant Lievens, S. J. Ranchi: Satya Bharati.
Cohn, Bernard S. 1996. Colonialism and Its Forms of Knowledge: The British
in India. Princeton: Princeton University Press.
Cotton, H. J. S. 1911. Indian and Home Memories. London: T. F. Unwin.
Dalton, E. T. 1866. ‘The “Kols” of Chota-Nagpore’, Journal of Royal Asiatic
Society of Bengal, 35(2): 153–98.
32

Fried, M. H. 1975. The Notion of Tribe. Menlo Park, CA: Cummings Publish-
ing Company.
Fox, Richard G. 1969. ‘ “Professional Primitives”: Hunters and Gatherers of
Nuclear South Asia’, Man in India, 49: 139–60.
Fuller, Chris J. and Veronique Bénéï (eds). 2001. The Everyday State and
Society in Modern India. London: C. Hurst & Co.
Ghosh, K. 1999. ‘A Market for Aboriginality: Primitivism and Race Classifica-
tion in the Indentured Labor Market of Colonial India’, in Gautam Bhadra,
Gyan Prakash and Susie Tharu (eds), Subaltern Studies X, pp. 8–48. New
Delhi: Oxford University Press.
Gilmour, David. 2006. The Ruling Caste: Imperial Lives in the Victorian Raj.
New York: Farrar, Straus and Giroux.
Grant, Colesworthy. 1846. Sketches of Oriental Heads: Being a Series of
Lithographic Portraits Drawn from Life, Intended to illustrate the Physiog-
nomic Characteristics of the Various People and Tribes of India. Calcutta:
Thacker & Spink.
Guha, Ranajit. 1983. Elementary Aspects of Peasant Insurgency in Colonial
India. New Delhi: Oxford University Press.
Guha, Ranajit. 1974. ‘Neel Darpan: The Image of a Peasant Revolt in a Liberal
Mirror’, Journal of Peasant Studies, 2(1): 1–46.
Guha, Sumit. 1996. ‘Forest Polities and Agrarian Empires: The Khandesh
Bhils, c.1700–1850’, The Indian Economic and Social History Review,
33(2): 133–53.
Guha, Sumit. 1998. ‘Lower Strata, Older Races, and Aboriginal Peoples:
Racial Anthropology and Mythical History Past and Present’, Journal of
Asian Studies, 57(2): 423–41.
Guha, Sumit. 2013. Beyond Caste: Identity and Power in South Asia, Past and
Present. Leiden: E. J. Brill.
Gupta, Dipankar. 1980. ‘From Varna to Jati: The Indian Caste System, from
the Asiatic to the Feudal Mode of Production’, Journal of Contemporary
Asia, 10(3): 249–71.
Jha, Jagdish Chandra. 1986. ‘Early Indian Immigration into Mauritius (1834–
1842)’, in U. Bissoondoyal and S. B. C. Servansing (eds), Indian Labour Immi-
gration, pp. 9–19. Moka, Mauritus: Mahatma Gandhi Institute.
Kosambi, D. D. 1944. ‘Caste and Class in India’, Science & Society, 8(3):
243–49.
Kuper, Adam. 2005. The Reinvention of Primitive Society: Transformations of
a Myth. London: Routledge.
Mahto, S. 1971. Hundred Years of Christian Missions in Chotanagpur since
1845. Ranchi: Chotanagpur Christian Publishing House.
Mohapatra, Prabhu M. and Rana P. Behal. 1992. ‘ “Tea and Money versus
Human Life”: The Rise and Fall of the Indenture System in the Assam Tea
Plantations 1840–1908’, Journal of Peasant Studies, 19(3): 142–72.
33
Mosse, David. 2001. ‘A Relational Approach to Durable Poverty, Inequality

and Power’, Journal of Development Studies, 46(7): 1156–78.
Omvedt, Gail (ed). 1982. Land, Caste and Politics in Indian States. Delhi:
Guild Publications.
Pouchepedass, Jacques. 1999. Champaran and Gandhi: Planters, Peasants
and Gandhian Politics. New Delhi: Oxford University Press.
Raychaudhuri, Tapan. 1969. Bengal under Akbar and Jahangir: An Introduc-
tory Study in Social History. New Delhi: Munshiram Manoharlal.
Roy, Kumkum. 2008. ‘Kosambi and Questions of Caste’, Economic and Polit-
ical Weekly, 43(30): 78–84.
Sartori, Andrew. 2011. ‘Indigo and Independence in Nineteenth-Century
Bengal’, paper presented at the British Historical Studies Colloquium, Yale
University, 15 December.
Sharma, Jayeeta. 2011. Empire’s Garden: Assam and the Making of India.
Durham: Duke University Press.
Sivaramakrishnan, K. 1999. Modern Forests: Statemaking and Environmental
Change in Colonial Eastern India. Stanford: Stanford University Press.
Sivaramakrishnan, K. and A. Agrawal (eds). 2003. Regional Modernities: The
Cultural Politics of Development in India. Palo Alto: Stanford University
Press.
Tea Districts Labour Association. 1924. Hand-Book of Castes and Tribes
Employed on Tea Estates in North-East India. Calcutta: Catholic Orphan
Press.
Tea Districts Labour Association. 1925. Language Hand-Book: Mundari. Cal-
cutta: Catholic Orphan Press.
Teltumbde, Anand. 2012. ‘It’s Not Red vs Blue’. Outlook, 20 August, http://
www.outlookindia.com/article.aspx?281944 (accessed 10 January 2014).
Tilly, Charles. 1999. Durable Inequality. Berkeley: University of California
Press.
Trouillot, Michel-Rolph. 2003. Global Transformations: Anthropology and
the Modern World. New York: Palgrave Macmillan.
Washbrook, David A. 1981. ‘Law, State and Agrarian Society in Colonial
Weber, Max. 1978. Economy and Society (2 vols). G. Roth and C. Wittich
(eds). Berkeley: University of California Press.
34
2
ANOTHER HISTORY
Bhadralok responses to Dalit political
assertion in colonial Bengal
Sarbani Bandyopadhyay
A glance at the few works on the caste question in Bengal indicates that
caste in colonial Bengal was an intensely debated issue that gave rise to
different and competing articulations.1 These works (Bandyopadhyay
1990; 2011; Sen 2012; 2014) point to the significant role caste played
not merely in the everyday life and politics of Bengali society, but increas-
ingly, in the domain of institutionalised politics and through attempts to
bring about reforms in society and introduce radical change in modes of
governance. In this chapter, I attempt a study of the implications of Dalit2
1 I thank Rowena Robinson for her conceptual insights and help with editing a draft of
this chapter, Kushal Deb, my Doctoral Committee at IIT Bombay and the editors of this
volume for their comments; Anand Chakravarti for creating my interest in studies of
caste; Anjan Ghosh and Sweta Ghosh for encouraging me to study caste in Bengal; my
parents, sister, mesho and mashi for being my finest teachers; my mother Alo Banejee
especially has a strange faith in me which has turned out to be my greatest source of
inspiration and I dedicate this chapter to her memory; Jagadish Mandal, Ekushe,
Asimda, Rajat and my participants for their help and Sayantani Mitra for many things.
2 This term usually refers to the former ‘untouchable’ castes. But some such castes in Bengal
refused to accept Depressed Class status since they believed that would seriously jeop-
ardise their claims to high-caste Hindu status. While the Bangiya Jana Sangha was formed
by leaders of many ‘untouchable’ and some middle-ranking castes, some of the latter, like
the Mahishya, do not seem to have remained with this endeavour and similarly refused
Depressed Class status. Sardar (2012) mentions how Mahishyas played the role of an
oppressor caste when ‘lower placed’ castes claimed high-caste status. This is an aspect
that was visible in other parts of India too. Mandal (1922) used the term ‘undeveloped’
(anunnoto) to refer to a situation of backwardness enforced upon the ‘lower’ castes by
Bengali caste society. He further described them as dalito, ‘exploited’ and ‘trampled
upon’. None of these terms indicate complete homogeneity and reproduction of caste
hierarchy acted against the development of solidarity among them. I use the term ‘Dalit’
to refer to castes who later entered the governmental categories of Depressed Classes and
Scheduled Castes. In general I use the term ‘marginalised caste’ instead of ‘lower caste’.
35
S A R B A N I B A N D Y O P A D H Y AY
political assertions and the bhadralok3 responses they evoked: one in the
form of building Hindu sangathan (organised society) through caste
reforms, the communalisation of identities and ‘networking’ among
caste Hindus to prevent independent forms of Dalit mobility and mobili-
sations; another, the subsuming of caste into the larger question of class.
Both are, I argue, variants of bhadralok politics.
Bhadralok (or the genteel folk) was seen as a status category consist-
ing of upper/high-caste men who pursued a modern education, were
engaged in the secular professions, had links with land in the form of
rentier interests and were disdainful of manual labour. ‘High culture’
and ‘high’-caste orientation were essential attributes of the bhadralok
(Mukherjee 1975; Sarkar 1998). However, available (mainstream)
literature on the bhadralok has, misleadingly, seen it as a generally
achieved – rather than ascribed – category. From the early 20th century,
castes placed lower in the hierarchy started making efforts to enter the
fields of the Bengali (Hindu) bhadralok. These were the domains of
modern education, professions, the field of culture and politics – all
that distinguished the bhadra from the abhadra. Although these caste
movements began as ones claiming higher ritual status – and thereby,
higher social rank – soon, they started petitioning the government for
social and economic justice and demanded proportional representa-
tion in the fields that had become the new markers of Hindu bhadralok
selfhood. The eligibility of caste Hindus to occupy and colonise the
new modern institutions and opportunities was largely a function of
what I call ‘caste capital’.4 Anjan Ghosh (2001) has shown how in
early colonial Bengal upper castes were able to convert themselves into
the new middle classes. This conversion, quick and secure, was made
possible by the cultural, social, economic, and political resources they
enjoyed from precolonial times. These kinds of resources were them-
selves products of the caste system. Hence, ‘caste capital’ may help us
better capture, in this context, the complex of what is generally known
as ‘cultural and social capital’ (Srinivas and Béteille 1964; Bourdieu
2002). This caste capital originated in the pre-modern institution of
caste, but made itself remarkably malleable to suit modern institutions
and interests that made it grow further. Colonial middle-class capital
3 This is not to suggest that practices of caste, domination and resistance were confined
to bhadralok society; in fact, such layered practices existed beyond bhadralok space;
but this falls outside the scope of this chapter.
4 It must be stated here, that caste functioned as capital for the upper castes and not for
the marginalised castes.
36
A N O T H E R H I S T O RY
was caste capital in modified forms, access to such capital was, and
still is, largely governed by caste.
These very privileges the caste structures once guaranteed were,
from the early 20th century, being questioned by the marginalised
castes. Colonial society became, as we shall see, that structure of
multiple ‘fields’ (Bourdieu 2002), that is, networks of relations and
practices, in which antagonistic groups articulated, reproduced and
changed their competing dispositions and vied for control over all
kinds of resources at stake.
The alienation of a large section of the population from mainstream
society became visible when the Hindu bhadralok failed to draw the
marginalised castes in adequate numbers into the Swadeshi movement
(launched against Curzon’s plan to partition Bengal), which began in
1905. These castes opposed it and most allied with the Muslim ‘Other’ and
the colonial government to thwart the Swadeshi programme. To the Dalit
castes, Swadeshi was akin to a conspiracy to further bhadralok dominance
(Sarkar 1987; Chatterjee 2002; Bandyopadhyay 2011). Non-cooperation
of marginalised castes was sought to be countered with caste reforms
aimed at removing untouchability and the jalchal-ajalchal divide.5 By that
time, however, such ritual disabilities alone were not the only concern
of the marginalised castes. Their fight against these disabilities became
intertwined with that of seeking material improvement, and these could
only be achieved through a strong marginalised caste organisation. Soon,
their demands included legitimate share in the political resources of the
colonial state as well. Their attempts at forming a separate caste organisa-
tion that would, if required, follow a separatist agenda made clear their
political position, and more importantly, it pointed out the possibility of
developing their bargaining power. To these castes, colonial rule was like
a blessing, for the nature of this rule made unstable the structural basis
of caste society. A host of factors from the 20th century onwards, not
least their numerical strength, made these castes indispensable to nation-
alist mobilisations. The Muslims were already a majority in Bengal while
nearly 50 per cent of the Hindu population was constituted by ‘untouch-
able’ castes. Breaking the alliance between Muslims and ‘untouchable’
castes, therefore, became important for the nationalist movement.6
5 Jalchal castes were those from whom Brahmins could accept water.
6 This alliance, however, was for the most part tenuous and conflicts between the
‘untouchable’ castes and Muslims were not uncommon, although they did not become
communalised till before the 1920s (Bandyopadhyay 2011). This conflict continued
even later, when Scheduled Caste legislators joined League ministries (Mandal 2003).
37
The Bangiya Jana Sangha: attempting

a forbidden move
From the early 20th century, the marginalised castes were mobilising
for proportional representation in all educational, administrative and
political bodies of the province. The Bangiya Jana Sangha (BJS – Bengal
People’s Association), formed in 1922, was the first organised attempt
of these castes to storm the stage of institutionalised politics. Succes-
sive organisations, such as the All Bengal Depressed Classes Associa-
tion, formed in 1926, and later, the Bengal Depressed Classes League
and the Bengal Provincial Scheduled Castes Federation, all kept push-
ing the limits of political imagination of these castes.
The primary aim of the BJS was to bring all marginalised castes
together under one organisation. The initiative was taken by a well-
known senior, Paundra-Kshatriya7 leader, Manindranath Mandal. All
the appeals, proposal and communication exchanges show that this
organisation had no intent of seeking support from the upper castes.
An exclusively marginalised castes’ organisation was deemed neces-
sary to make these castes self-reliant so that they could live with dig-
nity and self-respect. The main agenda were (a) to bring about unity
among these castes, (b) to act as a pressure group on the colonial
regime and Bengali society and (c) to consequently fight for their col-
lective social and political rights (Mandal 1922). However, excessive
dependence on the state for distributive justice often led to infighting
between these castes and weakened their solidarity.
Addressing two kinds of audience, the oppressed and the oppres-
sor castes, Mandal cleverly appropriated two upper-caste icons,
Vivekananda and Gandhi, to legitimise the politics of the BJS. The
BJS claimed it was seeking swaraj for all – true swaraj could only be
attained through self-reliance. In order to further this goal, which the
BJS scandalously insisted was in ‘conformity’ with Vivekananda’s
and Gandhi’s ideals, they required a separate organisation of the mar-
ginalised castes and argued that no true nationalist could be averse to
the idea of the oppressed emancipating themselves (Mandal 1922).
Mandal also made explicit a politically blasphemous agenda: if its
demands were not met, it would follow in the steps of the Muslim
League and pursue a separatist politics, away from the Hindu (and,
by definition) ‘national’ fold. The BJS argued their separatist politics
7 An ‘untouchable’ caste claiming Kshatriya status.
38
was necessary for ushering in true unity in Hindu society. In that social
and political context, bhadralok society could not afford the further
alienation of these castes as their alliance with Muslims was consid-
ered ominous for bhadralok domination. Although the BJS survived
for barely two years, the legacy it created endured until its abrupt end
in August 1947.
Hindu bhadralok response: Hindu Sangathan

and Bharat Sevashram Sangha
Post-Swadeshi, we see strenuous efforts by the bhadralok at building
and institutionalising Hindu sangathan on a sustained large-scale
manner through Hindu Mahasabha-affiliated organisations,8 such as
the Bharat Sevashram Sangha founded by Pranabananda. I focus here
on the Sangha not only because of the radical social, political and
organisational role it played in sangathan work but also because its
activities came up in many of the Scheduled Caste life narratives I col-
lected in course of my research. Sangh activities were primarily con-
centrated in eastern and northern Bengal, which had the highest
concentration of Dalit castes. For the Sangha, it was necessary to build
a Hindu equivalent of the mosque, the Hindu Milan Mandir (HMM)
by creating an overarching Hindu identity that would supersede all
smaller, conflicting, contradictory identities based on caste, sect,
region, or language along with this, it tried to channel margin-
alised caste hatred of upper castes in ways that would be productive
to the project of building the Hindu dharma rashtra: the Rakshi Dal,
formed in 1935, was to perform that function. Its political function
was cloaked by a welfare agenda of protecting Hindus from Muslim
violence (Juktananda 2007). It was supposed to deal effectively with
the hurdles the Communal Award created for the realisation of ‘Hindu
unity’. By the time it set about its task of building the Milan Mandir
and the larger Hindu Samaj Samanvyay Andolan in 1934 (Atmananda
2004), the Sangha had considerable material resources under its con-
trol and had engaged itself in relief work and in reorganising village
economy; as a cognate of the Hindu Mahasabha, it established pro-
ductive networking with the bhadralok society. It was, thus, well-
placed to take the challenges of separatist Dalit politics head-on.
8 While removal of caste disabilities was a central agenda, their programme was more
informed by the larger goal of sangathan work than by concerns for social justice.
39
By 1926, the Sangha established nine ashrams in eastern and north-

ern Bengal, to which were attached schools, charitable dispensaries,
cottage industries and weaving institutions. The Sangha was estab-
lished as full-fledged organisation with its head office in Calcutta in
1922. By 1940, it had founded 500 HMMs and 30,000 armed volun-
teers of the Rakshi Dal (File 75/47). The Sangha would rescue villagers
affected by natural calamities and communal violence and take them
to common shelters where there was no scope for caste discrimina-
tion. Like other reformers before him, Pranabananda argued that the
powerlessness of Hindus stemmed from evils such as untouchability
that kept a large part of Hindus ‘outside’ Hindu society. Pranabanda
campaigned among marginalised castes and insisted that they were to
form the backbone of Hindu regeneration.
Well aware of the problems that affected the villages, Pranabananda
embarked on a comprehensive plan to deal with the caste problem by
merging the religious, social, political and economic concerns with
the central problematic of Hindu sangathan. He combined his relief
work with the setting-up of small-scale cottage industries and weav-
ing institutions; local machinery for husking paddy were also made
available in each village to groups of Hindu families so as to make
the economic conditions of the villages and marginalised castes better.
Thus the Sangha’s upliftment programmes took off. Handsome dona-
tions to the Sangha allowed for all their activities to be carried out
without financial hurdles (File 531/26; Bedananda 1996; Smart 2010).
Sangha documents show that within just a few years, their ashrams
were established in almost all districts of eastern and northern Bengal.
Pranabananda made it clear that he did not want a social revolution
that would thoroughly transform the existing caste society; he saw the
caste system as an essentially positive, co operative and accommoda-
tive system. The Sangha only wanted to cure the society of its ‘ills’ in
a way that would not entail a structural alteration. He appealed to
the upper castes to ‘embrace the untouchables as their brethren’; the
latter, however, ‘must leave behind their hatred for and jealousy of the
“upper” castes and be part of Hindu fraternity’ (BSS 2006). The sys-
temic nature of upper-caste oppression is never alluded to in Sangha
literature, and upper castes are absolved of their roles in sustaining the
structures of oppression and domination against which there had been
organised protest.
Bengal later saw the formation of the successor to the BJS, the Bengal
Depressed Classes Association. Its agitational politics merged it with
the All India Depressed Classes Association. In 1930, in its Executive
40
Committee meeting, the Bengal representative, Birat Chandra Mandal

condemned the Civil Disobedience movement and proposed to orga-
nise loyalist movements against it. It also welcomed the Round Table
Conference. Further, in the Bengal Legislative Council, the Depressed
Classes representatives voted against the resolution recommending
full dominion status for India. In 1932, in the midst of Gandhi’s fast
against the Communal Award that granted separate electorates to the
Depressed Classes, leaders of the Depressed Classes here clearly stated
that they were not ready to accept anything less than what the Com-
munal Award granted. At an emergency meeting on 26 September,
these leaders criticised Ambedkar for ‘assuming the role of a dictator’
and sacrificing the real cause of the depressed classes by signing the
Poona Pact. They made it clear that they accepted the Pact ‘out of
necessity’ and not ‘out of choice’ (Bandyopadhyay 1990).
The project of building Hindu unity could only take place if the
objectives of marginalised caste organisations could be appropriated
by caste Hindu politics. Being ostensibly a socio-religious organisation
with its purported central focus on service to the ‘downtrodden’ and
‘neglected’ sections of Hindu society, the Sangha’s relief work was wel-
comed by sections of those otherwise at the receiving end of caste soci-
ety. Its programmes for village welfare yielded results. The energetic
relief work had also drawn these sections towards the Sangha. Yet,
the demand for separate electorates and the support it drew from the
Depressed Classes caused great anxiety to the project of Hindu nation-
alism. Hindu nationalist politics required foot soldiers; the castes that
specialised in defence activities were also among the ‘lowest’ in the
hierarchy, and therefore, these sections were required to be integrated
into this project. Greater organisational effort was hence required to
attain this objective. The main aim was to withdraw these sections
from separatist political goals and redirect them towards the goal of
Hindu nationalism.
From the 1920s, the Sangha actively participated in the Shuddhi
(purification and reconversion) movement, which gave rise to increased
communal violence, which was converted to fertile grounds of the
Hindu nationalist project. However, this movement created problems
for caste society since there was a lack of consensus on where the
re converts to Hinduism would be placed in the varna hierarchy
(Ghoshbarma 1926). This difficulty was sought to be tided over by
making the ‘Hindu’ identity primary: it was argued that this Hindu
identity made every member ‘touchable’. The new mlechhas (untouch-
able) were those non-Hindus, especially the immediate ‘other’ of the
41
Hindu self: the Muslim. This displacement of hatred and the concerted
attempts for building a singular Hindu identity were the only pos-
sible ways in which the now politically disruptive question of the caste
order could be handled. One could both avoid meaningfully engaging
with the problem of caste (beyond questions of untouchability) and
create an all-encompassing Hindu identity that could be effectively
deployed against anti-Hindu forces. This required the crystallisation
of identities and conflicts along communal lines.
Marginalisation and attempted appropriation of

Dalit politics
The Hindu Samaj Samanvyay Andolan, through the HMM and Rakshi
Dal was the direct, open and critical attempt at building the Hindu
community by challenging divisive politics of all kinds. The HMM
was that radical institution that had: (i) an all-caste prayer hall for
routine prayers, rituals, festivals and reading out of sacred texts; (ii) a
sabhamandap for discussion of issues affecting Hindu society and
their solutions, issues of education, morals and so on; (iii) a vidyaman-
dir (free school) for imparting general education to all Hindu children
along with religious education. The young would be taught history,
geography, general and health science, and be educated in Hindu reli-
gion and culture to grow up as ‘good’ Hindus; (iv) a school for the
elderly where shastras would be taught in a simple language and prob-
lems of Hindu society would be discussed; (v) a library for all and on
all subjects, including on Hinduism and the Hindu way of life;9 (vi) a
martial arts training and body building centre where all Hindus could
learn the art of self-defence; (vii) Rakshi Dal – armed core volunteer
groups who would collectively defend Hindus against their enemies;
(viii) charitable dispensary and a service force where qualified doctors
would serve the needy; and (ix) pracharak force that would prevent con-
version to other religions and promote Hindu unity. The HMMs would
take charge of village festivals, which would be open to members of all
castes. The function of such ritual gatherings and the highly publicised
conversion ceremonies was to create a religious fervour in which caste
barriers were supposed to melt.
9 This library set up branches across Bengal to build an All Bengal Circulating Library,
whose main objective was to ‘awaken the political and national consciousness among
the masses to whom patriotic writings are to be read out’ (File 531/26).
42
The Calcutta head office directed that at least 10 well-known indi-

viduals would be selected to establish the HMMs in each village.
Available evidence suggests they were mostly bhadralok, but included
a few well-known members of local marginalised castes (Juktananda
2007; File 279-J/25). This helped in the reorientation of programmes
to build inroads into such communities as it did not make their activi-
ties appear as mere bhadralok activities. HMMs were required to
organise an all-caste (Hindu) public meeting to publicise the Mandir,
its principles, aims and methods of work. Further, the HMM volun-
teers were to collect from each Hindu family, a handful of rice and a
monthly subscription for the HMM treasury. In times of need such
as riots, famines and so on, these would be used to aid the Hindus.
The HMM became the place for the resolution of all problems affect-
ing the members of the Hindu community. The Rakshi Dal also was to
be formed in each village (Juktananda 2008).
All these were of immense significance for sangathan work and for
contesting and appropriating Dalit politics. Unlike some other well-
known missionary institutions, the Sangha had a few important func-
tionaries who did not rank too high in the caste hierarchy; Bedananda,
the successor of Pranabananda was a non-upper-caste person. This
possibly impacted on marginalised caste society: it could be seen as an
avenue for social mobility. The common fund collected from all Hin-
dus, irrespective of caste, and the lack of provision for separate caste
kitchens also helped develop among these castes a sense of belonging.
Although the Sangha was not the only organisation working in this
field, it was – as archival records, Sangha literature, and Chatterjee
(2002) show – unmatched in terms of resources, branches and net-
works. It was the Sangha which effectively took up the task of village
reorganisation and helped better the conditions of the marginalised
castes. Mobilisational literature of these castes of the late 19th and early
20th centuries show how these castes sought to bridge the gap between
their economically well-off and poor sections through practices such
as charity work and forming a common granary. Through very simi-
lar practices, the Sangha helped forge some degree of closeness with
these castes. Moreover, since some sections of marginalised castes had
claimed an ‘authentic’ Hindu identity in their search for higher social
status, the friction and conflict between marginalised caste peasantry
and the Muslim peasantry, which did not necessarily always spring
from religious differences when channelled along communal lines,
proved instrumental in this context. Communalism was an important
and often successful way of dealing with separatist challenges from the
43
marginalised castes. Class differentiation that occurred with economic

development among these castes (Bandyopadhyay 2011; Sardar 2012)
also helped in the communalisation of agrarian relations.
For the vast majority of the marginalised castes, attaining educa-
tion was almost impossible; it was expensive and often beyond their
means; schools were too far-off and not compatible with the timings of
a vast population who spent much of their daytime in the agricultural
fields. The free education offered by the Sangha appealed to those who
aspired for education. This further helped in slowly weaning away a
section of these castes from their English-educated articulate sections.
The loss of Hindu masculinity (pourush) is mourned in all Sangathan
literature. While Gandhi’s work among ‘untouchables’ was acknowl-
edged, his non-violence was heavily condemned; the failure of the
Khilafat/Non-Cooperation movements and the Bengal Pact10 and the
importance of the mofussil in provincial politics after 1935 (Chatter-
jee 2002) were seen as potentially threatening the dominance of the
Hindu bhadralok. From 1937, in the Bengal Provincial cabinets, Dalit
representatives tried to make their presence felt as autonomous politi-
cal beings through their legislative activities (Mandal 1999; 2003). The
non-participation of the Depressed Classes and backward Muslims in
the Swadeshi as well as the Khilafat/Non-Cooperation movements, the
organised attempts of the Depressed Classes to push for the implemen-
tation of the Communal Award and their criticism of the Poona Pact,
all drove home the point that the bhadralok needed to assimilate these
castes into the Hindu fold. The census played its due role in raising the
alarm in bhadralok quarters: it was necessary to prevent the rise of
Muslims to prominence; but it was of a greater necessity to reverse the
alliances between marginalised castes and Muslims against bhadralok
interests. Communalisation of social structures and massive drives at
the ‘integration’ of these sections into the Hindu fold were two of the
strategies adopted by the Hindu bhadralok to counter threats to its
political death. Large-scale organised Sangathan work that began full-
scale from 1912 especially that of the all-pervading programme of the
Sangha was able to make a dent in emerging Dalit politics.
The Rakshi Dal was constituted mainly of Namasudras, Paundra-
Kshatriyas and Bagdis in eastern, western and south Bengal, and of
Rajbanshis in northern Bengal, all known for their physical strength
10 An arrangement made by C. R. Das of Congress that granted ‘concessions’ to Muslims

to secure their support against colonial rule.
44
and valour. Some castes, such as the Namasudras, were also the tradi-
tional village defenders. From being the outcastes from which (caste)
Hindu society had to protect and preserve itself, they were now
becoming the defenders of that very society from external and internal
threats, as if they were the very sections on which the survival and res-
urrection of Hindu society depended. This had the prospect of draw-
ing members of these castes into Sangathan work. The leaders of the
dals were mostly upper-caste bhadraloks who were associated with
revolutionary parties and organisations such as Anushilan Samity and
Jugantar Party. All professed violent Hindu nationalist politics.
The Hindu nationalist project necessitated the displacement and deflec-
tion of the violence constitutive of the caste system onto some other group
external to the Hindu community. Along with ‘upliftment’ programmes
this process and ritual of displaced violence could create in some measure
‘community consciousness’ and ‘solidarity’ with a Hindu identity when
quotidian structures of caste stood as a barrier against such ‘solidarity’.
Although a substantial section did act as foot soldiers of Hindu national-
ism, particularly since the 1940s, a much larger section was still suspi-
cious of its agenda. This was one reason behind what Sen (2012) calls the
‘nationalist resolution of the caste question’: the Partition of Bengal. Sev-
eral Depressed Classes associations emerged in the late 1930s and 1940s,
some owing allegiance to the Congress, some to the Hindu Mahasabha
and some to Ambedkar’s All India Scheduled Castes’ Federation (AISCF).
The Bengal Provincial Scheduled Castes Federation (henceforth
Federation), formed in 1943, had been the main opponent of Hindu
Sangathan and had been advocating unity between Scheduled Castes
and Muslims against caste Hindu domination. With almost half the
Bengal Muslim population being converts from marginalised Hindu
castes this was presumed to create considerable scope for alliances
between Scheduled Castes and ‘low-caste’ Muslims (Kotewal 1944).
Some lack of sympathy on the part of non-Ashraf Muslims with the
League demand for Pakistan (File 388A/40; Kotewal 1944; Mandal
1999) seemed to brighten the chances of such alliances. These alli-
ances were also a reality against which the Sangha and other Hindu
organisations had to wage their battles. With some powerful Dalit
organisations ready to openly contest sangathan work, it is perhaps
not surprising that each place Pranabananda visited and set up ash-
rams in the course of time witnessed communal riots.11
11 Sangha literature is replete with these cases.
45
The quarterly reports on the Sangha’s activities published in its

organ, Pranab, abound with descriptions and news of the ways in
which the presence of Rakshi Dal helped marginalised caste peas-
ants and labourers to work on the fields without attacks from the
Muslims; the training in martial arts imparted by Namasudra and
Paundra-Kshatriya caste people; relief activities in riot-torn areas and
the festivals and sabhas it organised, which were attended in large
numbers by members of marginalised castes. Possibly in response to
the Rakshi Dal, in 1943, the Muslims founded the All Bengal Lathial
Samity, whose objectives were similar to the Sangha’s: to train Muslim
youth in martial arts for self-defence (File 83/47). Communal violence
increasingly held out the best prospects in this situation. Chatterjee
(2002) shows how caste Hindus were equally prepared for the Cal-
cutta Killings of 1946; the caste Hindu campaign for Partition of
Bengal also gained political mileage from this carnage (ibid.). This
massacre also made Dalits extremely suspicious of Muslim intent; and
the news of the Killings led to riots in eastern Bengal, which made the
cry for Partition more rabid.
Contestations, caste networks and Dalit politics

From contesting the existing caste structures by claiming high caste
status, Dalit politics moved towards what it considered to be the more
important domains of political, economic and educational resources.
The methods they increasingly adopted were agitational and separat-
ist. With the formation of the Federation in 1943, this mode of politics
became one of the critical markers of Dalit politics in Bengal. Earlier,
in the 1930s, nascent Dalit politics had made its presence felt with its
critique of Ambedkar, Gandhi and the Poona Pact. It consisted of a
mixture of caste identity, class, and, from 1944, anti-colonial politics.
The early caste movements in Bengal were marked by contesting
claims of Hinduness, which also meant sanitising the more egalitar-
ian popular religious and social practices and transforming them into
one followed by the elites (Sardar 2012). Roy (2010) laments that the
Matua religion, materialist and pro-producer, was also transformed
over time into a Brahminical one that robbed it of its potential to
be emancipatory. This attempted sanitisation also kept open, in the
context of colonial Bengal, the possibilities for their future (but prob-
lematic) appropriation by caste Hindu interests and ideology. This
appropriation through valorising and imposing the great tradition
was accompanied by a separation and communalisation of social
46
spaces (Menon 2006), as the institutionalisation of the Milan Man-

dirs shows.
The articulate sections of the untouchable castes were involved in
institutional politics from the 1920s and were less involved with the
temple-entry satyagrahas (Sarkar 1987). This space left vacant by the
marginalised caste elites was put to use by the Sangha and others under
the leadership of the Mahasabha (Bandyopadhyay 2011). But in spite
of such efforts, and in spite of marginalised castes participating in riots
against Muslims, one cannot accept the ‘integration’ thesis that Bandyo-
padhyay (2011) puts forward. This alleged ‘integration’ into the ‘Hindu
fold’ was never complete and was contingent on external and internal
factors. The nationalist harping on ‘unity’ and ‘fraternity’ above ‘petty’
distinctions and divisions never could assume hegemonic proportions.
Among the bhadralok, there existed considerable anxiety about mar-
ginalised caste loyalty and integration into the Hindu fold. Integration
could not have been natural, given the quotidian violence that consti-
tuted structures of caste; it had to be created and was always more arti-
ficial than spontaneous and genuine. Correspondence between different
Hindu organisations showed that the untouchable castes were mostly
concerned with the welfare of their own communities and were not nec-
essarily ideologically drawn into Sangathan work (File 279-J/25).
The lure of power and material benefits or hegemony would alone
not be adequate tools to explain the complexities of Dalit politics.
Even though elected on Congress and Mahasabha tickets, Scheduled
Caste legislators continued to seek recognition as a distinct category
and demanded governmental intervention (Sen 2014). The alleged
integration never took place and the formation of the Bengal Parti-
tion League is one testimony to this. Partition was achieved sans an
exchange of population which left the vast majority of Dalit castes in
East Pakistan.
From the archival files on Hindu organisations and their own litera-
ture, it becomes plain that they had firm networks with caste Hindu
bhadralok and their resources. Their patrons were upper-caste zamin-
dars, bhadralok professionals and politicians and their organisations
(File No. 531/26). This networking was a function of what I earlier
termed ‘caste capital’ and caste homophily: personnel, identities, inter-
ests and ideologies of these organisations were caste Hindu in nature;
it was a kind of closed network, which was in no sense even remotely
available to the Dalits. In a sense, these caste Hindu networks were
inherited by these organisations; they had to spend little time, effort
and resources to build them. The involvement of the patrons acted
47
as capital that drew more members of ‘respectable’ society into these

organisations in different capacities. The same could be said of the
Sangha except that its tentacles were tougher and longer. Over time,
the Sangha expanded its networks and rose to an authoritative posi-
tion capable of calling the shots (Chatterjee 2002). These organisations
functioned on the principle of division of labour, which significantly
also took care of leadership struggles. For instance, although Saty-
ananda and Pranabananda parted ways over issues of leadership, one
never came in the way of the organisational work of the other. Dalit
politics had no such resources at hand, it had to build them; and even
then, they were – in all senses – inferior and fragile in nature. Caste
Hindu networks worked in various ways and at different levels, but
their politics was premised on foiling attempts by Scheduled Castes
at gaining economic, social and political autonomy from caste Hindu
society and polity.12
Scheduled Caste politics, particularly of the Federation, clearly
underscored the position that political, economic and social equality
were inseparable from one another for the true emancipation of the
Dalit population. From the Assembly Proceedings of this period, Sen
(2014) shows that the Scheduled Castes were drastically underrepre-
sented in public employment and illustrates the ways in which this sta-
tus quo was being sought to be maintained. For instance, one was the
lack of government patronage with only INR 30,000 set aside for spe-
cial scholarships and stipends for the Scheduled Castes. Schools insti-
tuted for Scheduled Caste students or by Scheduled Caste people were
often not taken up by caste Hindu-run District School Boards and
land leases not renewed on time; legislation was enacted to prevent
election of Scheduled Caste persons on district boards (which was suc-
cessfully challenged by some Scheduled Caste MLAs) (Mandal 2003);
according to Scheduled Caste legislators, funds were unduly diverted
towards towns and city hospitals, universities and colleges at the cost
of advancing cheap credit to poor and Dalit cultivators or instituting
separate funds for Scheduled Caste education. Muslim League min-
ister Tamizuddin Khan spoke about an ‘under-government’, which,
he alleged, managed to bypass and circumvent government rules and
circulars in terms of maintaining the communal ratio in recruitment
and also in proper implementation of grants and so on to non-caste
Hindu interests (Sen 2014).
12 Sen (2014) calls this the non-recognition of Scheduled Caste politics.
48
The notion of ‘under-government’ aptly summed up the deep-

rooted and powerful networks of the caste Hindus. Even when Sched-
uled Caste and Muslim ministers had political will, they could not do
much; at best, the implementation of policies for Scheduled Castes and
Muslims were limited to the ministries they held (Sen 2014). This was
because the implementation rested with the bureaucrats, who were
necessarily caste Hindus. Sarkar (1998) wrote about the increasing
sense of anxiety among the bhadralok from the early 20th century over
the sharing of scarce resources with Muslims and marginalised castes.
Sen refers to Partha Chatterjee’s argument that urban resources were
increasingly becoming the only source of livelihood for the bhadralok.
Hence, for their survival, access to these sources had to be restricted to
the bhadralok. In this context, increasingly organised non-bhadralok
aspirations led to a sense of siege among the bhadralok; they, to a large
extent, successfully prevented this catastrophe by developing network
closures. The year 1941 provided them with a golden opportunity to
thwart the rule of communal ratio in recruitment to public services.
With the government busy with World War II, the task of collecting
census data was left almost without supervision to the field person-
nel, who were predominantly caste Hindus. Prior to the census, caste
Hindu organisations and their print media carried out incessant cam-
paigns so that the Depressed Classes ‘returned themselves’ as Hindus
only, and not as Scheduled Castes or by their respective caste designa-
tions. As a result, from 1931, the Depressed Classes population came
down by more than 1.8 million in Bengal (Sen 2014).
Class, Communist Party and Dalit politics

The Communist Party of India (CPI) was formed in 1920 to give a new
direction to the anti-colonial struggle. I here concentrate on the agrar-
ian front of the CPI, the Bengal Provincial Kisan Sabha, its politics,
agenda and mobilisation of the agrarian population, and compare it
with Dalit organisations, namely, the Bengal Provincial Scheduled
Castes Federation (henceforth ‘Federation’) and their politics. I also
refer to Ranadives’s tract (2011) on Scheduled Caste politics.
The CPI shot to fame in Bengal by waging, through the Kisan Sabha,
what is known as the Tebhaga movement. Yet, as Ghatak (1987)
points out, the Kisan Sabha was divided over the issue of tebhaga and
over the pressing question of whether the bargadars (sharecroppers)
would be given ownership rights over the land they cultivated. At the
Kisan Conference in Panjia, Jessore, in June 1940, for the first time,
49
the Kisan Sabha upheld the demand for tebhaga and gave a call for
launching the movement. It passed these resolutions:
• Right of bargadars to two-thirds of the produce (tebhaga)

• Abolition of zamindari without compensation to zamindars
• Land to the tillers
• Fixing the minimum price of jute at current market prices but not
the maximum price at which the peasants can sell
• Reducing taxes on tenants
• Abolition of chowkidari tax
• Fixing minimum wages for daily wage labourers
All these demands were also made by the Federation, headed by

Jogendranath Mandal. Earlier, Mandal, as an MLA from 1937, had
consistently placed demands for the improvements of the conditions
of peasants and tillers, distribution of khas land to the landless (mostly
Dalit) cultivators, supply of cheap rural credit to the agriculturalists
and the setting up of new cooperative societies that would help the
marginal peasants and tillers (Mandal 2003).
Yet, there was no official alliance between the Federation, the Kisan
Sabha and the CPI. The Federation called upon the small peasants,
sharecroppers and landless labourers to unite and fight for their rights.
Jaagaran, the weekly organ of the Federation, carried full-length arti-
cles on the struggles of bargadars and adhiars (sharecroppers), who
were mostly Scheduled Castes. Hence, these two organisations should
have been able to work together especially since the CPI, before late
1945, demanded legislative activism from the elected representatives –
something which Mandal was already doing. Jaagaran (4 May 1946;
11 May 1946; 15 June 1946) also systematically carried out propa-
ganda against the zamindari system, especially after the famine of
1943. Jaagaran reported consistently on the plight of the cultivators
and called upon the government to take measures to combat the situa-
tion. The Federation also demanded the distribution of khas cultivable
lands to the landless tillers (Jaagaran 25 May 1946; 1 June 1946; 1 Febru-
ary 1947; Mandal 2003).
In 1945, when the CPI was refused a Congress platform for sup-
porting the government during World War II, it gave a call for
renewing the movement in favour of the bargadars’ demands; but
according to Ghatak (1987), the call for movement was given without
any preparations being made – it all remained as mere resolutions.
Ghatak (1987) claims that, in January 1946, the CPI adopted the
50
tactic of not antagonising landed interests in order to win the upcom-

ing elections. Yet, out of the eight workers’ seats and the four general
seats it contested, it won only two workers’ and one general seat,
showing that the CPI was not in a position to have a major impact
on national politics through parliamentary means. Communal riots
prevented the Kisan Sabha from holding meetings, but on 29 Sep-
tember 1945, the Kisan Sabha decided to wage a direct movement
in support of the tebhaga demands. It officially took off towards the
end of 1945 (Ghatak 1987).
Dalit politics and questions of class

The tebhaga movement, however, had a longer history. Sarkar (1987)
shows that this demand for tebhaga had begun on a sustained basis in
the eastern districts of Bengal from 1928, and in several cases, involved
united movements against zamindars and jotedars. One of the features
of the movements was to boycott caste Hindus and deny them ser-
vices. Since caste Hindus would not perform menial tasks, it was seen
by bargadars (mostly Muslim and Namasudra) as a potential weapon
in their struggle for both economic and social gains. The Congress
tried to intervene and deflect the movement into the domain of ritual
gratification and started a temple satyagraha, but it evoked little inter-
est from the untouchable population. Being actual tillers with a mate-
rialist, anti-other-worldly (Matua) philosophy, they were more into
seeking economic betterment. This could be seen also in Hindu Sanga-
than work: religious satyagrahas and shuddhi movements had to be
necessarily accompanied by material reforms, such as building schools
for untouchable castes and ensuring their functioning (File 279-J/25).
This was more prevalent among the Namasudra caste.
The 1922 Pirojpur Namasudra Conference demanded, among sev-
eral things, (i) abolition of the zamindari system; (ii) making actual
cultivators owners of homesteads and lands they worked on; (iii) abo-
lition of intermediate interests and redirecting that revenue towards
expanding education and health services for the poor (Das 1922).
From 1943, the Federation, as an organisation, demanded the abo-
lition of zamindari, distribution of land to the landless tillers and
the distribution of cultivable land left fallow to those who lost their
marginal holdings in the famine of 1943. This Conference reflected a
nuanced understanding of the complexities of the agrarian structure.
At its first Provincial Conference at Gopalgunj in 1945, the Federa-
tion (in conformity with the resolutions adopted at the 1942 Nagpur
51
AISCF Conference) resolved to fight for the reorganisation of the vil-

lage system by setting up Scheduled Caste villages through a distribu-
tion of government-owned cultivable land to the landless Scheduled
Caste communities; it further demanded: abolition of zamindari and
all intermediate interests and the enactment of legitimate agricultural
taxes to be collected from the cultivators directly by the government;
cooperative farming and establishment of cooperative business centres;
fixing legitimate and minimum rates for agricultural products; fixing
minimum wages for the workers in municipalities and other industries;
ensuring insurance for the unemployed, the aged, and the sick; in cases
of married women employees, the mill/factory owners must bear a part
of their maternity expenses, and the government should take up exten-
sive relief work to make the conditions of famine-affected people less
miserable (the relief measures of the private relief committees directed
by caste Hindus had not, it was claimed, reached the poor and land-
less Scheduled Castes) (Mandal 2003). Unlike the CPI the Federation
seemed to be operating within a broad understanding of class, which
was informed and tempered by their own experiences of being lower
class, but vitally also by their ‘untouchable’ caste backgrounds. There-
fore for the Federation, caste was not merely a superstructural entity, it
was material and symbolic at the same time.
In its initial years of formation, the Federation could bring in radical
agendas for building economic and political unity. But not even after
20 years of existence could the CPI take a stand on the most pressing
problem of that time: the question of land in rural Bengal. Bhattacha-
rya (1978) points out that the CPI finally gave the go-ahead for the
tebhaga movement in 1945 mainly because the Comintern had almost
written it off as a potential political force. The CPI could not develop
a truly radical politics because it lacked the presence of an organic
leadership. Had the CPI been formed and led by the downtrodden, its
course could have been vastly different.13
Ranadive (2011) seems to be the only full-length discussion on
AISCF politics by the CPI. Highlighting some of the major features of
this discussion, I shall argue that the class-caste background of the CPI
leadership prevented a more nuanced reading of Scheduled Caste poli-
tics, and thereby precluded any alliance between the CPI and AISCF.
13 A point developed in conversation with D. K. Biswas, October 2012, S. K. Roy,

December 2012, and January 2013.
52
Starting off with the deplorable conditions under which the Sched-
uled Castes lived for centuries, and the injustice that the Congress and
Gandhi’s Poona Pact had done to them, Ranadive pointed out that
despite all this, it was Gandhi who, in 1920, made the removal of
untouchability an integral part of Congress’ campaign for swaraj. He
indicated that the Congress made the cardinal mistake of not giving
adequate attention to their political and economic aspirations on the
grounds that these demands were inimical to national unity. Placing
his argument around the primary concern of securing Indian inde-
pendence, Ranadive emphasised the necessity of a united nationalist
movement against British imperialism. Studies (e.g. Henningham 1982;
Chakravarti 1986) have shown how damaging this unity was for the
interests of the marginalised sections of the population, in particular,
that of Dalit/Adivasi sharecroppers, landless labourers and industrial
workers. Their interests were never really accommodated by the domi-
nant interests in organisations such as Kisan Sabhas and trade unions
(Omvedt 2008). After critiquing the Congress for its shortcomings, his
exposition moves on to AISCF politics.
It begins with doubting the claims of AISCF that it represented and
had the full support of 50 million Scheduled Caste people, but then
quickly added that it nonetheless commanded the support of a sub-
stantial section of that population and that all the intellectuals were
with this organisation. It admitted that the emancipatory goals of the
AISCF and its aim of attaining an equal share in the country’s politi-
cal resources were democratic and indeed legitimate. Highlighting
the resolutions adopted at the Nagpur Session, 1942, Ranadive made
the point that the Scheduled Castes wanted equality in the life of the
country. Its demands for adequate representation in political bodies,
in public employment and in education were all just demands that
should be met. In support of this argument, Ranadive explained the
root causes of untouchability and exploitation as resting in the village
system and claimed that the AISCF demands of a complete overhaul-
ing of the existing system along lines adopted at Nagpur were also
logical. In this section, he took a positive, but somewhat condescend-
ing, stand on Scheduled Caste demands only to demolish their very
basis in the following sections.
In the section on the problems of the AISCF, Ranadive pointed out
that from the very beginning, it was riddled with major faults – the
fundamental one being that it did not demand complete indepen-
dence. In a Mahasabha-Congress-like nationalist tone, it argued that
their emancipation could only be assured and implemented when the
53
country gained full independence. Surely, as a Communist theoreti-

cian, Ranadive must have understood that there was a lack of political
will on the part of the Congress and that lack had a political economy
of its own? If his understanding was not naive or misdirected, then on
what grounds could he claim that only in an independent India could
the problems of the Scheduled Castes be meaningfully addressed?
In 1945–46, the AISCF explicitly engaged with the question of what
independence meant to the Scheduled Castes and how it could be
attained (Sreechanakya 1946). For them, the colonial rulers and
the Congress-Mahasabha shared a symbiotic relation with capital. The
Congress-Mahasabha represented the exploiting segment of the Indian
population and the colonial rulers were eager to transfer power to the
Congress because only in the hands of the Congress would the interests
of foreign capital be safe; the Congress intended to establish the rule of
Indian capital, but it also required the support of foreign capital. The
Congress-Mahasabha, more than any other party, depended on colo-
nial support to establish a capitalist state in India. The demands of the
AISCF were not in conjunction with the interests of capital, foreign or
indigenous. Hence, the Federation made clear that its political imagi-
nation differed from that of the Congress-Mahasabha14 insofar as the
former wanted to secure the rights of the workers, peasants, landless
labourers, while the latter wanted to strengthen the grip of exploitative
interests. It doubted the sincerity of Congress and Mahasabha activi-
ties and believed that they were purely instrumental, aimed at break-
ing the Dalit movement. It had, in contrast, a sympathetic approach
to the Communists (Jaagaran 11 May 1946).True independence, it
argued, could only be attained when the downtrodden and exploited
would be freed from all forms of exploitation and oppression. Was
that also not the goal of the communists in India? And yet, this CPI
document betrays its closeness to the Congress. The AISCF position
that they were not against India’s independence so long as independent
India undertook to safeguard Dalit interests was seen as carrying with
it a negative attitude about India’s national struggle (Ranadive 2011).
It is striking to note how the CPI sought not to take into account
the Federation’s critique of the class basis of the Congress and the
14 Mandal was drawn to the Congress by Sarat and Subhas Bose. Initially, he opposed
the idea of separate electorates and believed that the Congress as the all-India nation-
alist organisation would work for the benefits of all people. His disillusionment began
with the Congress’ open opposition to the interests of the Scheduled Castes, and by
1941, he was convinced that Congress was indeed a caste Hindu organisation.
54
nationalist movement (Patni 1946; Sreechanakya 1946).The CPI, in

this document appeared less critical of Congress and Gandhi than of
the victims of Congress politics – the Scheduled Castes. In February
1946, the CPI, with reference to AISCF politics, urged everybody to
unite under the banner of the Congress, which was, according to the
CPI, a broad platform for all classes and which commanded the sup-
port of the great majority of workers, peasants, middle classes and
intellectuals. This subsuming of differences for a larger cause had been
an inherent problematic of the CPI, which prevented it from arriving
at a nuanced understanding of systems of stratification. This incapac-
ity again stemmed from their class–caste position. While Communist
leaders and intellectuals desired and attempted to declass and de-caste
themselves (Dasgupta 2005), their position in the social hierarchy and
the bhadralok orientation of the Bengal communists acted as barriers
in this. In contrast, this broad platform was what the AISCF and the
Federation in Bengal found problematic as it feared that the interests
of the worst-off would not be adequately handled.
In a strange echo of the Mahasabha, the CPI necessitated the AISCF
to declare that their twin tasks would be to fight for India’s indepen-
dence as well as the rights of the Dalits, together and in harmony
with all other political parties. According to CPI this change of atti-
tude towards non-Dalit organisations became a requirement for the
AISCF since their movements had gone beyond the confines of ritual-
istic demands and entered the arena of politics and state power. Like
the Sangha earlier, the CPI called upon the AISCF to leave behind
its politics of separatism and hatred against the Congress and caste
Hindus and join hands with the nationalist cause for their own good.
Yet, when it came to the Muslim League, the CPI’s position was much
more tempered and it called upon the League and the Congress to
search for compromise solutions. In discussing the anti-democratic
nature of the AISCF and its hostility towards the Constituent Assem-
bly and the Cabinet Mission, the document appeared to be summarily
dismissive of the AISCF’s arguments. When the CPI accused AISCF for
being anti-democratic it did not highlight the most important objec-
tions AISCF raised against the Cabinet Mission – that in every respect,
the proposals were harmful to the interests of the Scheduled Castes. In
spite of this, AISCF was asked to suspend its anti-Congress stand and
unite with the Congress for the sake of Indian independence.
The Federation in Bengal stated that the only way ahead of the sar-
bahara/proletarian Scheduled Caste population was to wage an armed
struggle against the colonial government and zamindar-bourgeoisie
55
combine. From 1946 onwards, Jaagaran featured the long mobilisa-

tion meetings the Federation and its sub-organisations carried out to
prevent a takeover of politics by exploitative interests. Its immediate
struggle was against the Partition of Bengal. The CPI should have
found in the Federation a natural ally, but the CPI made plain its ani-
mosity to Dalit politics.
Ranadive claimed that Scheduled Caste political leaders came from
the same class background as the Congress leaders (except that the
former had to overcome severe odds), and hence, implicitly argued
that they both had common class interests. In this way, he tried to
delegitimise the politics of the AISCF/Federation. This politics was
accused of splitting the class-based workers and peasants organisations
and struggles. However, CPI never felt the need to introspect its own
class politics in search of an answer to this problem. It failed to realise
that among different structures that organised class relations in India,
caste was an important one. As far as the leadership of the Congress
and the CPI went, both came from bhadralok backgrounds. Quite a few
leaders of the CPI came from zamindar-jotedar families who had expe-
rienced a comfortable urban life and could afford a foreign education;
and they were mostly caste Hindus. They shared common class–caste
features with the Congress as opposed to the Scheduled Castes. That
they were not declassed and ‘de-casted’ adequately became clear when,
unlike the tillers of the 1920s, they were unable to take an unequivocal
position on issues of zamindari abolition and on the rights of peas-
ants and sharecroppers over the lands they cultivated. Neither could
they leave behind their Brahminical values that prevented them from
meaningfully engaging with Scheduled Caste politics. While launching
its agrarian struggles, such as the tebhaga movement, the CPI depended
on its urban-bred middle-class leaders to direct the struggles. But mid-
dle-class radicalism could not match the spirit and history of resistance
of the marginalised sections to landed interests that Sarkar (1987) has
written about. Middle-class control of the tebhaga movement failed to
lead it towards bringing about a total transformation – a transforma-
tion the CPI also possibly never envisaged.15 Later histories of the dif-
ferent communist parties similarly show how they remained entrapped
in middle-class understandings and practices of politics.
15 According to some senior Scheduled Caste participants in my research the CPI lead-
ership did not want a total transformation for that would have adversely affected its
class interests.
56
Conclusion
So long as the demands of the marginalised castes were restricted to
the ritual spheres, the response of caste Hindu society was rather mild.
But from the early 20th century, when these castes refused to partici-
pate in the Swadeshi movement, and with their later attempts at
organised political assertion, Hindu bhadralok organisations actively
sought to appropriate their political assertion to serve nationalist
goals. Besides initiating reforms and developing welfare measures, the
bhadralok sought to redirect Dalit hatred for the upper castes towards
the Muslims as the ‘other’ of the united ‘Hindu self’. In addition, they
were being enlisted as foot soldiers of the nationalist project. The aim
was threefold: to wean them away from alliance with Muslims; pre-
vent their development through any kind of dependence on the colo-
nial state and its resources; and to weaken their politics by incorporating
them into the nationalist movement. Simultaneously, caste Hindu net-
works were utilised to prevent the upward mobility of these castes so
that the efforts of Dalit leaders to seek recognition for these castes as
a separate category were continuously foiled.
On the other hand, class mobilisations by the CPI led to political
work among the marginalised castes because caste-structured class
relations of production made it inevitable that the lowest rungs of
the Indian class structure were occupied mostly by these castes. Class,
which seemed to transcend pre-modern identities such as caste, was
more appealing to a vast section of the bhadralok. But the CPI engaged
in class politics without adequately engaging with the question of
caste. This problem was compounded by their high caste background
and their origins in landed interests. Because of this, they could not
develop a meaningful and more grounded critique of Brahmanism
and caste (and thereby, of class relations); hence, they were unable
to become organic leaders of the class-based movements, such as teb-
haga. They remained distant to those they mobilised. It was damaging
for the organisation and the movement because hierarchies, critically
that of caste, were not broken and requisite trust could not be estab-
lished. As a consequence, one finds a familiar division of labour: men-
tal labour performed by the bhadralok leadership and manual tasks
(of implementing the policies, decisions, strategies) performed by the
lower rungs. Class politics thus shielded caste politics. Even when
the antagonism between Scheduled Caste and nationalist politics was
increasingly becoming irresolvable, the CPI kept denouncing the for-
mer as being against the interests of the nation. Since the demands
57
of Scheduled Caste politics took cognisance of class inequalities and

enunciated a programme of removing them in a way the CPI’s poli-
tics never did, this condemnation sacrificed the interests of the most
downtrodden sections of Indian society. When one compares the ways
in which Hindu bhadralok and bhadralok communists responded to
Dalit political assertion in the colonial period, one finds uncanny par-
allels between these two. What united these seemingly different and
even contradictory political responses was their bhadralok connection:
both were politics which not only failed to imagine non-bhadralok
beings as agents of their own transformation but by tempering the
structures of inequality they also allowed for their reproduction.
Bibliography
Atmananda. 2004. Hindu Samaj Samanvyay. Kolkata: Sangha.
Bandyopadhyay, Sekhar. 1990. Caste, Politics and the Raj: Bengal 1872–1937.
Calcutta: K. P. Bagchi.
Bandyopadhyay, Sekhar. 2011. Caste Protest and Identity in Colonial India:
The Namasudras of Bengal 1872–1947. Kolkata: Oxford University Press.
Bedananda. 1996. Sangha Sadhana. Calcutta: Sangha.
Bourdieu, Pierre. 2002. Outline of a Theory of Practice. Cambridge: Cam-
bridge University Press.
Bhattacharya, Jnanabrata. 1978. ‘An Examination of Leadership Entry in Ben-
gal Peasant Revolts, 1937–1947’, Journal of Asian Studies, 37(4): 611–36.
BSS. 2006. Sangha Gita. Kolkata: Sangha.
Chakravarti, Anand. 1986. ‘The Unfinished Struggle of the Santhal Bataidars
of Purnea District, 1938–42’, Economic and Political Weekly, 21(23 and 24):
1847–65.
Chatterjee, Joya. 2002. Bengal Divided: Hindu Communalism and Partition,
1932–47. New Delhi: Cambridge University Press.
Das, Rajanikanta. 1922. Sabhapatir Abhibhasan, Pirojpur Namasudra Con-
ference. Dhaka: East Bengal Press.
Dasgupta, Rajarshi. 2005. ‘Rhyming Revolution: Marxism and Culture in
Colonial Bengal’, Studies in History, 21: 79–98.
Ghatak, Maitryeya. 1987. ‘1946–47 er tebhaga andolon er koyekti dik o
kichhu proshno’, Bortika, July–December.
Ghosh, Anjan. 2001. ‘Cast(e) Out in West Bengal’, Seminar, 508.
Ghoshbarma, Sudhangshu. 1926. ‘Hindu Sangathan O Shuddhi’, Hindu
Patrika, 33(6).
Henningham, Stephen. 1982. Peasant Movements in Colonial Bihar. Can-
berra: Australian National University Press.
Juktananda. 2007. Hindu Milan Mandir. Kolkata: Sangha.
Juktananda. 2008. Hindu Rakshi Dal. Kolkata: Sangha.
58
Kotewal, Jehangir. 1944. The India Charter. Karachi: Karachi Press.

Mandal, Manindranath. 1922. Bangiya Jana Sangha.Medinipur.
Mandal, Jagadis Chandra. 1999. Mahapran Jogendranath O Babasheb Ambedkar.
Calcutta: Biswas Publisher.
Mandal, Jagadis Chandra. 2003. Mahapran Jogendranath (Vols 1 and 2). Kol-
kata: Chaturtha Duniya.
Menon, Dilip. 2006. The Blindness of Insight. Chennai: Navayana.
Mukherjee, S. N. 1975. ‘Daladali in Calcutta in the Nineteenth Century’,
Modern Asian Studies, 9(1): 59–80.
Omvedt, Gail. 2008. Seeking Begumpura: The Social Vision of Anti-caste
Intellectuals. Chennai: Navayana.
Patni, N. C. 1946. ‘Congress O Taphashil Samaj’, Jaagaran, 7 December.
Ranadive, B. T. 2011. ‘The Nature and Trend of the Scheduled Caste Move-
ment’, in A. Ghosh (ed), Janajuddha, Deshbhag O Communist Party Vol. 4.
Kolkata: Seribaan.
Roy, Sunil Kumar. 2010. Bangajana Sabhyata O Namajatir Itihaas. Kolkata:
Janamon.
Sardar, Raicharan. 2012. Deener Atmakahini ba Satya-Pariksha. Sonarpur:
Paundra Mahasangha.
Sarkar, Sumit. 1998. Writing Social History. New Delhi: Oxford University
Press.
Sarkar, Tanika. 1987. Bengal 1928–1934: The Politics of Protest. New Delhi:
Oxford University Press.
Sen, Dwaipayan. 2012. ‘ “No Matter How, Jogendranath Had to Be Defeated”:
The Scheduled Castes Federation and the Making of Partition in Bengal,
1945–47’, Indian Economic and Social History Review, 49(3): 321–63.
Sen, Dwaipayan. 2014. ‘Representation, Education and Agrarian Reform:
Jogendranath Mandal and the Nature of Scheduled Caste Politics, 1937–
1943’, Modern Asian Studies 48(1): 77–119.
Smart, Niniyan. 2010. Prophet of a New Hindu Age. Kolkata: Sangha.
Sreechanakya. 1946. ‘Rajnoitik Andoloner Dhara’, Jaagaran, 21 December.
Srinivas, M. N. and Andre Béteille. 1964. ‘Networks in Indian Social Struc-
ture’, Man, 66: 165–68.
West Bengal State Archives, File Nos. 83/47; 191/46; 279-J/25; 388A/40;
531/26; 75/47.
59
3
PARTITION, DISPLACEMENT,
AND THE DECLINE
OF THE SCHEDULED
CASTE MOVEMENT IN
WEST BENGAL
Sekhar Bandyopadhyay and

Anasua Basu Ray Chaudhury
When the government-sponsored midday meal programme for pri-

mary schools was introduced in West Bengal in 2004, it was widely
reported in Kolkata newspapers that in a number of districts, caste
Hindu parents objected to their children eating cooked meals prepared
by volunteers from the Scheduled Castes.1 The situation embarrassed
the leaders of the ruling Left Front government, as it clearly showed
that caste had not disappeared from West Bengal. If we take this as a
direct evidence to argue that caste still matters, then why has there
been no strong Scheduled Caste movement in this province, when such
movements were so powerful before 1947? There can be many expla-
nations, such as the absence of caste-based aggressive landlordism; the
ambivalence of Bengali modernity; or the advent of leftist ideology,
land reforms and the panchayati raj, which eradicated extreme forms
of untouchability, contained violence against Dalit and resulted in
their (limited) empowerment.2 While there is some truth in all these
explanations, this chapter seeks to point out that the Scheduled Caste
movement before 1947 was most powerful in east and north Bengal.
Therefore, the loss of that spatial anchorage as a result of Partition
and the consequent physical displacement and dispersal of a large sec-
tion of the Dalit peasant population of Bengal certainly had an adverse
1 Anandabazar Patrika, 24 September and 28, 29 and 30 December 2004.

2 A broader discussion of this question is provided in Bandyopadhyay (2014).
60
THE DECLINE OF THE SCHEDULED CASTE MOVEMENT
impact on their social and political movements, which were now over-
shadowed by their struggle for resettlement.
Partition had a long history in Bengal and it needs to be shown how
directly the Scheduled Caste peasants were involved in it and how deeply
they were affected. Their large-scale migration from East to West Bengal
did not begin until 1950, but then, it continued incessantly. The pres-
ence of these Dalit peasant refugees changed the texture of politics in
postcolonial West Bengal, as they resisted an official rehabilitation plan
to disperse them over a wide geographical region outside Bengal. This
plan would deprive them of that spatial capacity for social mobilisation,
which they enjoyed before Partition and which was so crucial for the
strength of their movement. Within this political context, these displaced
Dalit peasants acquired the new identification of being ‘refugees’ –
the only publicly identifiable oppressed group in a new post-Partition
discourse of victimhood. While the refugees were never a homogenous
category, in the interest of a united struggle, their left-liberal and pre-
dominantly high-caste leadership deliberately purged the vocabulary of
caste from their language of protest, which could then be more eas-
ily appropriated into the modern tropes of social justice deployed by
the mainstream political parties and the state. This did not imply that
the caste question was resolved; this only meant that caste became less
conspicuous, though not non-existent, in the public discourse of social
justice and political protest. The Dalit refugees also adjusted to the new
realities of post-Partition West Bengal in myriad ways, and so, there are
multiple narratives of Dalit victimhood, protest, and agency. This chap-
ter also seeks to highlight that plurality of Dalit experience.
Caste movements before 1947

Social movements that began to assert Dalit identities in Bengal started in
the 1870s and were located in two very clearly identifiable geographical
spaces and two communities were at the forefront of these movements.
One was the Rajbansi community, which lived mainly in the north Bengal
districts of Rangpur, Dinajpur, Jalpaiguri, and the princely state of Cooch
Behar (Basu 2003). The other community, the Namasudras of east Ben-
gal, lived mainly in the districts of Bakarganj, Faridpur, Jessore, and
Khulna but were also scattered in other eastern and central Bengal dis-
tricts (Bandyopadhyay 2011). The other Scheduled Caste group that was
also organised was the Paundra-Kshatriyas of south Bengal. But they
were numerically small and organisationally not as powerful as the other
two groups. Apart from these three major communities, the other
61
S E K H A R B A N D Y O P A D H Y AY A N D A N A S U A C H A U D H U R Y
Scheduled Castes of Bengal were demographically so dispersed and eco-

nomically so backward and dependent that we do not see much of their
presence in organised Scheduled Caste politics in the colonial period.
So, when the Scheduled Caste political movement started in Ben-
gal in the early 20th century, it remained heavily focused on east and
north Bengal, and the Namasudras and Rajbansis provided the major-
ity of its leaders and supplied its main support base. For both these
communities, their close geographical location in contiguous regions
was a major factor behind successful social mobilisation. However, in
the last days of the Raj, their movements lost homogeneity and were
heading in a variety of directions as a result of the pressures gener-
ated by the politics of Partition. The Namasudra leader Jogendranath
Mandal, who had started the Bengal provincial branch of the Sched-
uled Caste Federation in 1943, believed that the Dalit and Muslim
peasants in east Bengal had similar interests, and so, a Dalit–Muslim
political alliance was in the best interest of the Dalit. But his pro-
Muslim League stance was not acceptable to many of his fellow Dalit
leaders, who were wary of their future in a Muslim-majority Bengal.
This was because the history of Dalit–Muslim peasant relationship
was not an uninterrupted story of harmony and cooperation; it was
regularly interrupted by violent riots. So, around this time, there were
two other rival Scheduled Caste organisations. The Depressed Classes
League, headed by another Namasudra leader, Pramatha Ranjan
Thakur (popularly known as P. R. Thakur), supported the Congress,
while the Depressed Classes Association, led by yet another Namasu-
dra, Birat Chandra Mandal, was more directly aligned with the Hindu
Mahasabha.3 In other words, the Scheduled Castes, and the Nama-
sudra community which led them, remained intensely divided on the
Partition issue in 1946–47; they lost their autonomy and were strate-
gically aligned with various mainstream political parties, such as the
Muslim League, Congress, and the Hindu Mahasabha.
The 1946 riots in Calcutta and Noakhali, in which the Dalits were
both victims and perpetrators, brought further damage to their move-
ment. Mandal remained a minister in the Muslim League ministry of
H. S. Suhrawardy, widely believed to be the mastermind behind the
Great Calcutta Killing, and this made him intensely unpopular among
3 ‘Extract from File 1164–44 Genl.’ and ‘S.S.1’, Government of Bengal, Intelligence
Branch Records, West Bengal State Archives, Kolkata [hereafter IB Records], F. No.
191/46.
62
the caste Hindus. Mandal and the Bengal branch of the Scheduled
Caste Federation advocated a united sovereign Bengal and opposed
the Partition campaign launched by the Hindu Mahasabha. However,
many Namasudra leaders such as P. R. Thakur actively supported the
Mahasabha campaign to create a Hindu majority province of West
Bengal within the Indian Union, which the Congress endorsed after
the Tarakeswar Convention in April 1947. Their major concern at this
stage was to keep their habitat – the districts of Bakarganj, Faridpur,
Jessore, and Khulna – within the Hindu homeland of West Bengal
(Bandyopadhyay 2004). In other words, the Bengal Dalits were clearly
divided on the Partition issue and were deeply concerned about the
political future of their physical and cultural spaces.
However, it will be wrong to suggest that the Bengal Dalits were only
preoccupied with the Partition question at this stage. In 1946, when
the communist-led sharecroppers’ movement, known as the Tebhaga
movement, started on the demand for two-thirds share of the produce
for the sharecroppers or adhiars, the Rajbansi peasants in north Ben-
gal and the Namasudras in east and central Bengal became its main
protagonists. Its impact on the Rajbansi movement was more divisive,
as it involved direct confrontation between the Rajbansi jotdars (large
land-holding peasants who were the main targets of this movement)
and the Rajbansi adhiars – thus rupturing their caste solidarity along
class lines. Most of the prominent Scheduled Caste leaders remained
remarkably silent on this poor peasants’ movement, clearly suggesting
that class was another issue that was weakening their social cohesion
at this crucial historical juncture (Bose 1986: 286). In the last days of
the Raj, the Dalit movements in Bengal were thus heading in myriad
directions.
Partition and Dalit migration

Partition brought new complications and compulsions for the Bengal
Dalits. It did not solve their problems, as despite their vehement pro-
testations, all the districts where the Namasudras lived went to East
Pakistan. The position of the Rajbansis was even more complicated, as
their habitat was divided by the new international political boundary.4
But most of them did not – or could not – migrate immediately. In Bengal,
4 See ‘Report of the Bengal Boundary Commission’, Government of India, Reforms

Office, F. No. 68/47-R, National Archives of India, New Delhi.
63
those who migrated in 1947–48 were the more wealthy classes, mostly
the upper-caste Hindu gentry and the educated middle classes with
jobs, including many of the Namasudra middle classes as well, who
could sell or arrange exchanges of properties (Chatterjee 1992: 72; Ban-
dyopadhyay 1997: 66–69; Chatterji 2011: 211–19). Very few Dalit
peasants could afford to move because migration required resources,
which they lacked. Also, leaders such as Mandal and the Federation
advised them to stay put, because they were assured by Jinnah that
they would get a fair deal in Pakistan, where ‘every man would be
equal’ (Mandal 1999: 112). Many Scheduled Caste peasants who
stayed back in Pakistan believed him.
But a second wave of refugee influx was triggered by a particular
incident in the Bagerhat subdivision of Khulna district in East Bengal
in December 1949. A police party came to a Namasudra village called
Kalshira in search of a few communists and were resisted by the vil-
lagers, resulting in the death of one police constable. Two days later,
a large police force, assisted by Ansars and other elements, attacked
not just this one, but 22 other neighbouring villages inhabited by
Hindu Namasudras (Amrita Bazar Patrika, 24 February 1950). The
Calcutta press immediately picked it up – the fact that the victims of
this incident were communists and Namasudras was lost – they all
became Hindus. And this media frenzy resulted in the outbreak of a
fierce riot in Calcutta and Howrah – for the first time after indepen-
dence – and the Muslims from West Bengal began to flee. This led to
retaliatory violence in East Pakistan, where the rioting spread from
Khulna to Rajshahi and Dacca, and then, to Mymensingh and Barisal
districts. The main victims of these riots were not the high-caste Hindu
bhadralok as many of them had already left, but the Dalit and tribal
peasants such as the Namasudras and the Santhals, who were now
forced to migrate to India (Biswas and Sato 1993: 34–44).
This was the final breakdown of the Dalit–Muslim alliance in East
Bengal as the Dalit peasants were deliberately targeted in this post-
Partition upsurge of violence. At the Bongaon railway station, the first
batch of Namasudra refugees of about 500 families arrived in the first
week of January 1950 (The Statesman, 21 January 1950) – and since
then, thousands of them began to arrive every day. They either came
through Bongaon and then moved on to the Sealdah station in Cal-
cutta, from where they were despatched to various refugee camps, or
they arrived at the border districts of Nadia or 24-Parganas, where
they began to settle down as the local Muslims began to flee across the
border. As one report suggests, in the early months of 1950, about
64
10,000 refugees were arriving every day through Bongaon and settling
down in Gaighata, Baduria, Habra, and other places.5 By the begin-
ning of 1951, following the disturbances in Khulna, about 1.5 million
refugees had arrived in West Bengal (Chatterji 2007: 111) – the major-
ity of them were Namasudra peasants.
The migration of Namasudra peasants continued incessantly
throughout the following years. There was a steady trickle – about 25
to 30 a day – until the beginning of 1952. Most of them were com-
ing by rail through either Bongaon or Banpur, and a few arrived by
road.6 This number began to rise dramatically from July 1952. In July,
August, and September, on an average, more than 6,500 refugees were
arriving every month, rising to more than 10,000 in October and con-
tinuing thereafter at that rate.7 According to official statistics, nearly
2.1 million refugees arrived in West Bengal between 1950 and 1956.
There was a lull for a few years after this, and then, following the Haz-
ratbal riot in 1964, 419,000 people migrated from East Pakistan to
West Bengal (Chatterji 2007: 112). These official figures are not often
reliable, as they account for only those who registered themselves and
were eventually despatched to various refugee camps. Anecdotal evi-
dence suggests that there were probably many more, who just crossed
the border and settled down in various places in the border districts
of Murshidabad, Nadia, and 24-Parganas. No one knew their exact
numbers.8
As these refugees started pouring in, the dominant popular dis-
course represented all these refugees as ‘minority Hindus’ fleeing from
Islamic Pakistan and their predicament tended to displace all other
public discourses of victimhood (Amrita Bazar Patrika, 30 Decem-
ber 1949; 6 February 1950). But the Congress government refused
to accept this migration as permanent or legitimate and wanted them
to go back at an appropriate time. Nehru, therefore, signed the Delhi
Pact with the Pakistani Premier Liaquat Ali Khan on 8 April 1950
5 Report on refugee situation for week ending 5/2/50, IB Records, F. No. 1838/48,
Part III.
6 Report regarding influx of East Bengal refugees at Sealdah R/S dated 3/6/52, IB
Records, F. No. 982/48-Sealdah.
7 Report on the exodus of Hindus from East Pakistan, dated 8.10.52, IB Records, F.
No. 982–48.
8 See, for example, the letter from the refugees of Murshidabad to S. P. Mukherji, dated
3/5/50, S. P. Mukherji Papers, Sub files, No.34, Nehru Memorial Museum and
Library, New Delhi [hereafter NMML].
65
to ensure their safe return. The Hindu Mahasabha, however, vowed

not to allow the refugees to be sent back and launched a campaign to
demand an exchange of population as the only solution to this ‘Hindu
minority’ problem.9 In West Bengal’s public space, the rhetoric of the
victimhood of the Hindu refugees thus tended to silence all other dis-
courses of identity at this juncture.
But were they all ‘Hindu’ refugees fleeing from the atrocities in
‘Islamic’ Pakistan? And how important was the caste factor in this
story of continuous exodus? ‘About 95 per cent of the refugees are
Namasudras’, reported a police intelligence report in June 1952.
They were mostly cultivators or day-labourers or belonged to vari-
ous professions, such as washermen, fishermen, weavers, petty busi-
nessmen, small jotdars, and talukdars.10 They came from all parts of
East Bengal, but mostly from Barisal, Faridpur, Jessore, and Khulna,
where the Namasudra peasants had been living side by side with
their Muslim neighbours for a long historical period. So why did this
relationship break down? Overwhelmed by the number of incoming
refugees, the police intelligence officers started questioning those who
arrived at Sealdah or at Bongaon and Banpur railway stations. The
stories they narrated were interesting and varied and do not allow
us to reconstruct a simple narrative of communal or caste conflict in
rural Pakistan. But they also make it clear that this was no ordinary
economic migration, as the government of India thought it to be (Roy
2012: 185–95).
Without going into the details, we can summarise here the main
reasons behind this mass Dalit peasant migration in Bengal, as stated
by the migrants themselves.11 It becomes clear from these testimonies
that the Namasudra peasants did not migrate because of only eco-
nomic reasons. They decided to leave their home and land because,
at this crucial point, the serious resource crunch in East Bengal
9 ‘Brief History of Sri Ashutosh Lahiry’; also, ‘History sheet of Shri Ashutosh Lahiry’,
IB Records, S.No. 45/1920, F. No. 210/20; see also the intelligence report on
Ashutosh Lahiry in IB Records, S. No. 158/20, F. No. 210/20; FR No. 9 for period
ending 4 May 1950, IOR: L/P&J/5/320, IOR; Amrita Bazar Patrika, 27 April 1950;
The Statesman, 2, 8, and 10 August 1950.
10 ‘Report regarding influx of East Bengal refugees at Sealdah R/S’ dated 3/6/52, IB
Records, F. No. 982/48-Sealdah.
11 However, one should keep in mind that we get these statements through the media-
tion of the police intelligence officers. The following description is based on numer-
ous statements in two IB Records Files, Nos. 982/48 and 982/48-Sealdah.
66
destroyed whatever goodwill there was between the two communi-

ties of peasants. The low-level routine violence, which Haimanti Roy
has observed in East Bengal since 1948–49 (Roy 2012: 147–76), was
escalated manifold after 1950. The Namasudra small peasants, who
owned some land, increasingly felt the aggressive assertion of their
Muslim employees and more powerful Muslim landed neighbours,
who all wanted their land. The provocations to leave were numer-
ous, ranging from unlawful occupation of land to public humiliation
of women and direct instruction to leave the country if they wanted
to save their lives and honour. Complaints about atrocities against
women were almost universal. However, if we go into the details,
it seems to have been more verbal abuse than actual physical rape,
although rape cases were reported as well. Almost everyone men-
tioned the rising numbers of armed robberies – or ‘dacoities’, as they
called them – in their houses. In some cases, the robbers parted with
as little as INR 60 in cash and earthen utensils, but almost in all
cases, they harassed the women and asked them to leave the country.
And then, when the local Namasudras did not anymore feel safe in
their own homes, came the scare of passport. The Pakistan govern-
ment proposed to introduce passports from October 1952 and many
Hindus thought that this would close all opportunities of moving to
India in future. This panic was also systematically fanned by inter-
ested groups with an eye on the properties, and led to an avalanche
of migration. Thus, competition for scarce resources was possibly a
potent factor behind this breakdown of Dalit–Muslim relationship
in East Bengal. To complement it, there were rumours circulating
that the government of India was waiting to offer them a lucrative
rehabilitation package, with offers of land, if they once managed to
cross the border. But their decision to leave their land and home was
not because of this economic lure; it was due to a pervasive sense of
insecurity.
For many of them, the migration turned out to be a traumatic experi-
ence that brought a permanent rupture with their past. Many of them
came with horrific memories of violence in the riots of 1950 (Basu Ray
Chaudhury 2004). Those who came after 1951 had to undergo the
upsetting experience of train journeys, where the trains were stopped
at the border checkpoints and the refugees were stripped of all their
possessions. They were only allowed to take INR 50 per head and
all their belongings were taken away. So, people tried to hide things,
particularly valuable possessions and ornaments, and consequently,
were subjected to further indignities by the custom officials and border
67
guards.12 Those who tried to cross the border on foot or by boat were
also stopped at various points, searched and stripped of their belong-
ings. Where they tried to escape, they were fired upon.13 The very
experience of migration was, thus, qualitatively so very different from
anything they faced before that it created for them a permanent rupture
with their past, in the sense that most of them never wanted to go back
to their homeland again, and ended up with a new identity of ‘refugee’.
Struggle for rehabilitation

However, the refugee was hardly a homogenous category as the expe-
riences of the low-caste post-1950 refugees were very different from
those of the early arrivals. When the first wave of mainly high-caste
Hindu bhadralok refugees arrived in West Bengal, they resettled them-
selves in squatter colonies in and around Calcutta, and the govern-
ment – after initial hesitation – endorsed that mode of rehabilitation.
But when the Dalit peasant refugees arrived in their thousands after
1950, they were first despatched to various refugee camps in different
districts.14 In allocating space in the camps, caste identity did play a
part, despite persistent official denial. At Sealdah Station, they were
asked about their identity, given a registration card, and sent by train
to a refugee camp.15 They would prefer to go to the camp where they
knew they would find their relatives, village acquaintances, or com-
munity members who had migrated earlier.16 As a result, the refugee
camps developed their own specific community demographies: in cer-
tain camps, such as Cooper’s Camp, Dhubulia Camp in Nadia, or
12 Note dated 15/10/52; ‘Statement of Sri Sarat Kumar Haldar’, IB Records, File
No. 982/48.
13 From Deputy Director, S.I.B., Calcutta, to Deputy Director (A), Intelligence Bureau,
GOI, dated 29 October 1952, IB Records F. No. 982/48.
14 At the peak of the inflow of refugees from across the border with East Pakistan, the
government mainly set up three types of camps, namely, women’s camps, worksite
camps, and Permanent Liability Camps.
15 It is reported that the registration desks issued three coloured cards – white-coloured
cards for those, who wanted to take shelter in the camps; red-coloured cards for those,
who were able to take care of themselves and were not willing to go to the camps; and
blue-coloured cards for those, who only needed initial assistance for their travel before
their own rehabilitation on the other side of the border. Jugantar, 26 and 27 March 1950.
16 Interview with Mahesh Mahato [name changed] on 18/3/2013. Also see, Extract
from W.C.R. of the Superintendent of Police, Murshidabad for the week ending
30 July 1949, IB Records, F. No. 1809/48 (Midnapore).
68
Bagjola Camp in the 24-Parganas, the Namasudras constituted more

than 70 per cent of the residents.
In these camps, they lived a shared commune life – sleeping in large
living spaces with no privacy, sharing common toilets (80 for 70,000
refugees in Cooper’s Camp), and standing in long queues for dole
and ration. Caste officially and apparently had no place in this camp
life, marked by shared poverty and suffering; yet, it was there all the
time. Caste was certainly taken into consideration in allocating space
within the camps, in the sense that the few caste Hindu refugees who
came to these camps preferred to stay in separate rooms – as far as
practicable – and their wishes were respected.17 This is not to suggest
that some form of caste segregation was maintained in the camps; this
is to indicate, however, that to low-level state functionaries, caste still
mattered in matters of governance. And for the refugees themselves, in
everyday social relations, even in extreme situations of privation, and
despite the currency of a levelling discourse of victimhood, caste had
not become irrelevant.
But within the camps, these refugees had little opportunity to orga-
nise along caste lines, as the structural compulsions of the struggle
for rehabilitation imposed on them a language of unity that could
bring all the residents of the camps to the same barricade lines. Their
indignation and sense of frustration were co-ordinated into agitations
by the emerging Bastuhara Samitis (refugee associations), their initial
protests being against camp maladministration, unacceptable quality
of ration, or high handedness of the camp administrators.18 When,
after 1950, the Namasudra peasant refugees began to arrive in large
numbers, there were targeted attempts to mobilise them,19 but most
often they agitated under non-Dalit local refugee leaders.20
Their main grievance was against the official rehabilitation policy.
The government argued that there was not enough vacant land in West
Bengal to rehabilitate these agriculturists. So, it decided to rehabilitate
them in the neighbouring provinces of Assam, Bihar, and Orissa, and
further offshore in the Andaman Islands. From 1953, there was an
explicit policy of selecting only the Namasudra peasant refugees for
17 Interview with ladies camp inmates, Cooper’s Camp, on 8/2/2013.

18 Interview with Amit Mukherjee [name changed] at Cooper’s Camp on 18/3/2013.
19 Report on the refugee situation for the week ending 19/11/50, IB Records, F. No.
1838/48, Part III.
20 Confidential report of 4/4/52 regarding the arrival of the refugees of Ranaghat Cooper’s
Camp, at Sealdah Railway Station on 3/4/52, IB Records, F. No. 982/48-Sealdah.
69
resettlement in Andaman (Thakur 2007: 33). Then started the forc-

ible transfer to other provincial camps, to which the refugees pro-
tested with force.21 In early 1956, the government announced the
Dandakaranya Scheme of rehabilitating them in a region consisting of
78,000 square miles of inhospitable unirrigated land in the tribal areas
of Orissa and Madhya Pradesh. They were to be forcibly transported
to these regions, their camps were to be closed and if they refused to
go, their doles were to be stopped (The Statesman, 23 August and
26 September 1950; Kudaisya 1995: 73–94; Chatterji 2007: 127–41).
Gradually, localised protests against this rehabilitation policy came
under large, umbrella-type organisations such as the United Central
Refugee Council (UCRC) or the Sara Bangla Bastuhara Samiti (SBBS – All
Bengal Refugee Association), and gradually, they came under the shadow
of the mainstream political parties as all of them had eyes on these
unending streams of refugees. On the right, the Hindu Mahasabha, Jan
Sangh, and the Rashtriya Swayamsevak Sangh criticised the govern-
ment’s ‘appeasement policy’ towards Pakistan and advocated economic
sanctions against Pakistan as a solution to the problems of East Bengal’s
Hindu minorities. At the centre, the Praja Socialist Party (PSP) started
actively organising the refugees and came to dominate the SBBS,22 while
on the left, the Communist Party of India (CPI) came to a controlling
position within the UCRC by 1951. Both groups organised mass pro-
tests against the Dandakaranya scheme, as also against the threatened
stoppage of cash doles and closure of refugee camps, but failed to col-
laborate. In March–April 1958, the UCRC and the SBBS organised
parallel satyagraha campaigns, lasting for about a month without any
tangible gains and resulting in the arrests of 30,000 refugees. Most of
them were camp refugees and nearly 70 per cent of them were Nama-
sudras. Caste once again played a role here, as the caste Hindu refugees
from the squatter colonies in Calcutta refused to join them. And what
is equally important, although the movements started initially under a
high-caste leadership of the UCRC and SBBS, it eventually witnessed
the emergence of a Dalit refugee leadership (for details, see Chakrabarti
1999: 177–81; Tan and Kudaisya 2000: 152–53). However, given the
political dominance of the mainstream political parties over the refugee
movement, they could hardly hope to have any autonomy.
21 Report on the refugee situation for the week ending 1/10/50, IB Records, F. No. 1838/
48, Part III.
22 Annual Administration Report 1953, IB Records, F. No. 32/28, Serial No. 220/1928.
70
From around 1955, we hear the names of Dalit refugee leaders such
as Ratish Mullick, Jatin Saha of the Cooper’s Camp, or Hemanta
Biswas of Bagjola Camp, taking prominent roles in refugee agitations.
Initially, they were working under the umbrella of the UCRC,23 but
gradually, they began to lose faith in it (Mandal 2003: 50). The rea-
sons are probably not difficult to surmise, as for many leaders of the
UCRC, the refugee movement was becoming an electoral constituency-
building exercise. In the first election of 1952, the UCRC officially
started enlisting support of the refugees for votes for the CPI against
the Congress (Pal 2010: 77–78). But the CPI was not alone; the Hindu
Mahasabha and its leaders such as N. C. Chatterjee were also trying
to tap into their refugee support base during elections.24 But the CPI’s
additional concern was about maintaining amity between the refugees
and the local peasants – their Kisan front – over the issue of forcible
land occupation.25 It was not surprising, therefore, that at a meeting in
Wellington Square on 12 April 1957, Hemanta Biswas had to appeal
to all political parties not to exploit the refugees to achieve their own
political goals.26 Jogendranath Mandal at this stage emerged as the
major spokesperson for the Dalit refugees in these camps and his lead-
ership, in some ways, undercut the level of support for the leftists. But
he too could not bring back the caste question to the centre stage of
the refugees’ struggle for rehabilitation.
Mandal and Dalit refugees

Jogendranath Mandal, who had migrated to West Bengal on 8 Octo-
ber 1950 after resigning from his ministerial position in the Pakistani
central cabinet as a mark of protest against the continuing repression
of Hindu minorities in East Pakistan (The Statesman, 9 October 1950),
soon found among the Dalit camp refugees a new support base to
reinvent his political leadership as a refugee leader. He initially became
a member of the SBBS and worked in collaboration with the UCRC,
but his relationship with the latter began to deteriorate from late 1957.
23 WBPA, dated 19/1/55, 19/2/55, 19/3/55, 2/4/55, IB Records, F. No. 1483/32.

24 ‘Addendum to the dossier of N. C. Chatterjee (Ex-MP) (written from 1.1.62 to
31.12.63’, IB Records, F. No. 238/42(1), Part III.
25 This was quite evident during the Bagjola Camp sqatting incident in July 1959.
See ‘Copy of an IB Officer’s Report dated 5/7/59’, IB Records, S.No. 288/46,
F. No. 820/46
26 Extract from S.B.D.N, dated 15/4/57, IB Records, F. No. 1483/32.
71
In the early months of 1958, as an SBBS leader, he began to mobilise

the camp refugees in the districts of 24-Parganas, Nadia, Murshid-
abad, Howrah, Burdwan, and Birbhum, preparing them for a civil
disobedience campaign to be launched on 17 March against the Dan-
dakaranya scheme.27 When the 1958 campaign failed, he established a
new organisation, the East India Refugee Council (EIRC), which he
described as a ‘non-political organisation’, truly representing the inter-
ests of the refugees.28 He also temporarily severed his connections
with the SBBS.
At this time, he sought to introduce the caste question into the dis-
course of refugee rehabilitation in West Bengal. As one police report
notes:
In Bolpur and Uttartilpara camp meetings on 23rd and 24th

February [1958], he [Mandal] openly accused caste Hindu
employees and caste Hindu people for sending refugee fami-
lies to Madhya Pradesh outside West Bengal. He accused Govt
to make West Bengal a caste Hindu state.29
The counter-attack that came from the other refugee leaders was
also sharp and virulent, as they were too concerned about maintaining
the unity of the refugee movement. The SBBS leaders, including some
Dalits, condemned him for creating a rift among the refugees by rais-
ing the caste question.30 The CPI leaders, privileging class over caste,
condemned him for establishing a separate organisation only with the
Scheduled Caste refugees.31 But despite these attacks, Mandal continued
to ask difficult questions. In December 1959, a few bus conductors were
recruited from among the refugees in Cooper’s Camp. ‘How many of
them are from the Scheduled Castes’, Mandal asked the camp adminis-
trator. He did not get an answer, but for asking that question, he again
got the flak of the leftist leaders and was branded as ‘communal’.32
27 Extract from the report in connection with the disturbances created by the refugees
in Vishnupur Court on 18.3.58, IB Records, F. No. 1483/32 (P.F.).
28 Report of a D.I.B. officer dated 25.2.58; Copy of Report of a D.I.O dated 26.7.58,
IB Records, F. No. 1483/32(P.F.).
29 Copy of I.B. Officer’s report dated 15/3/58, IB Records, F. No. 1483/32.
30 Copy of a report of a D.I.O. dated 1/7/58, IB Records, F. No. 820/46.
31 Copy of Memo No.7008/57–58 from Additional Superintendent of Police, 24-Par-
ganas to S.S.,I.B.,C.I.D., West Bengal, Calcutta, IB Records, F. No. 88–39(1) P.F.
32 Copy forward under No.5/19–59(1), dated 2/1/50 from Supdt of Police, D.I.B.,
Nadia to S.S., I.B., C.I.D., West Bengal, Calcutta, IB Records, F. No. 998/44.
72
Thus, in the name of unity, the left-liberal – and, of course, caste Hindu –
leadership deliberately purged the caste question from the discourse of
refugee movement, although it remained relevant as ever.
The refugees too were often in serious dilemma over the caste
question, as in the interest of their struggle, unity among all refugees
seemed essential. A police intelligence report in March 1958 showed
that while a rift was clearly visible among the camp refugees in Burd-
wan district, a delegation went to Calcutta to meet the leaders of both
the UCRC and the Jogen Mandal group ‘with a view to bring amity
between the two to strengthen the refugee movement’. But the leaders
refused to listen, and their followers remained divided.33 A refugee
activist in Cooper’s Camp in his recollection of those days of strug-
gle sought to privilege a generalised refugee identity over caste: ‘The
Namasudra or the other lower caste people participated in this move-
ment to fulfil their demands not as lower caste community members
but as refugees’ (Basu Ray Chaudhury 2010: 72). At a group meeting
with the former residents of the Bagjola Camp, the participants vehe-
mently asserted that caste did not matter in their movement – they
were fighting as a united front for all refugees.34 A frequently used
slogan in the refugee demonstrations of this period, ‘Amra kara? Bas-
tuhara’ (Who are we? Refugees) was a powerful statement that privi-
leged their refugee identity over their caste.35
But possibly, the social dynamics of the refugee camps also imposed
compulsions on the Dalit refugees to eschew the caste question. We have
the description of an interesting incident in Sealdah refugee camp on
5 February 1958, when a mixed group of refugees protested against
Mandal raising the caste issue, which they thought was detrimental to
the unity of their struggle, and refused to participate in the procession
he was planning to organise. So, Mandal left the camp angrily without
a single person accompanying him. However, it was later revealed that
five or six Namasudra refugees loyal to Mandal later surreptitiously left
the camp and joined the rally at Subodh Mullick Square; their other
camp mates did not know. In the evening, the UCRC leaders came to
Sealdah camp and severely criticised Mandal for raising the caste issue.36
33 Copy of No.896(7)/120–48 dated 14.3.58 from Superintendent of Police, D.I.B.,

Burdwan to S.S., I.B., C.I.D., West Bengal, Calcutta, IB Records, F. No. 96–49, Part II.
34 Group meeting with Bagjola Camp residents on 22/6/13.
35 Ext. from an IB Officer’s Report, dated 24.3.58, IB Records, F. No. 96–49, Part II.
36 Copy of secret report no. nil dated 6.2.58 from R.I.O Sealdah; Copy of I.B. Officer’s
report dated 6.2.58, IB Records, S. No. 288/46, F. No. 820/46.
73
The incident clearly shows that in their new existence in the refugee
camps, the Namasudras often felt pressured not to articulate their caste
identity for the sake of unity of their movement for rehabilitation.
Contingency too played a part. Given the realities of political
power, the refugees knew very well that they could not fight for their
rehabilitation without the support of the established political parties.
Mandal, therefore, could not ultimately lead an exclusively Scheduled
Caste protest or a completely apolitical movement. He had to strate-
gically align with the other non-Congress opposition groups. While
he shunned the left, as he considered them to be a greater threat to
the interests of the refugees, he was moving more to the centre-right,
like the Hindu Mahasabha, the Jan Sangh, and the PSP, as he walked
shoulder-to-shoulder with their Brahmin and caste Hindu leaders.
While in the changed circumstances of post-Partition Bengal, this caste
Hindu leadership supported Mandal’s cause and even some of them
went to jail with him, they also exerted a moderating influence on the
Dalit refugee leadership and kept them within constitutional boundar-
ies and the Gandhian mode of non-violent satyagraha.37And it was
because of them that the Dalit leaders had to refrain from using the
idioms of caste in a movement that was supposedly for all refugees.
Mandal, at this stage, became the undisputed leader of the camp ref-
ugees. From late 1959 till September 1961, he led a series of satyagra-
has with camp refugees in Calcutta and the districts (Bandyopadhyay
2011: 259–60). But these were not movements to assert Scheduled
Caste identity or project their exclusive interests – these were move-
ments of the refugees, which witnessed cross-caste mobilisation and
multi-caste leaderships at both local camp level as well as provincial
level. These were organised with the support of various mainstream
political parties. But this political support ultimately failed the ref-
ugees and the agitation was withdrawn in October 1961.38 We do
not exactly know why this movement was withdrawn. More than 50
years later, a group of residents of Bagjola Camp felt that their leaders
abandoned them before the goals of their movement were achieved.39
After this, the government forcibly despatched many of them to
37 See, for example, ‘File No. 1808–58(24-Parganas), p. 49. Meeting report’; ‘Report
of a secret source dt. 17.3.50 regarding Sara Bangla Bastuhara Sammelan (PSP) –
forwarded from DC, SB, Calcutta under Memo No. 5765(4)/PM553/58’, IB
Records, F. No. 96–49.
38 Group meeting with Bagjola Camp residents on 26 June 2013.
39 Ibid.
74
Dandakaranya, where, by 1965, 7,500 refugee families were settled

(Tan and Kudaisya 2000: 151). Thus, a sizeable section of the Dalit
population of Bengal was now dispersed across large parts of eastern
and central India. Mandal’s leadership of the refugee movement could
not help him launch a new political career in West Bengal. He failed to
win a single elected office and died on 5 October 1968 while conduct-
ing an election campaign (Sen 2010).
Thakurnagar and reinvention of space

While the Namasudra refugees in post-Partition West Bengal thus lost
their physical space, and with that, lost their spatial capacity to orga-
nise articulate protests, they were also imagining a new spiritual space
where they could reinvent their identity more in a social rather than
political sense. And it was happening through the initiatives of the
other prominent Namasudra leader of Bengal, P. R. Thakur. He had
remained loyal to the Congress during the trying days of Partition and
migrated soon after, and for some time, remained outside organised
politics, as he lost – like other political leaders from east Bengal – his
electoral constituency. At this stage of his political career, he also
devoted time to the cause of the refugees, but his ways were very dif-
ferent from those of Mandal, as he supported the government rehabili-
tation policy and relied more on self-help.
In December 1947, Thakur bought a piece of land in north 24-Par-
ganas between Chandpara and Gobordanga and started the Thakur
Land Industries Ltd, with himself as the chair of the Board of Direc-
tors. This was the beginning of Thakurnagar, the first Dalit refugee
colony in India started by an independent Dalit initiative. It was a
small hamlet near the Indo–Pakistan border, about 63 kilometres
from Calcutta. Within the next 10 years, around this place, in lands
reclaimed from the marshy tracts, more than 50,000 Dalit refugees
settled down. In 1951, Thakur received a government grant of INR
80,000 to develop the infrastructure of the colony, including roads and
supply of drinking water, and each family received INR 200 and two
bundles of corrugated iron for building houses (Biswas 2012: 516).40
Many of the Namasudra peasants who migrated after 1950 – and
continued to migrate thereafter – settled in the two border districts
40 In this context, also see West Bengal Legislative Assembly Proceedings [hereafter
WBLAP], Vol. 20, No. 1, June–August 1958, pp. 64–65.
75
of North 24-Parganas and Nadia (Nakatani 2011: 66087), where

more than half of the Namasudra population in West Bengal now live.
Thakurnagar grew into a major cultural centre for these Dalit refugees.
The main reason for this was not, however, Thakur’s politics, but his
role as the Guru of the Matua Mahasangha (MM), a Vaishnava reli-
gious sect which his great grandfather Harichand Thakur had founded
in the late 19th century and grandfather Guruchand Thakur had organ-
ised in the early 20th century. It was through this religious movement
that Namasudra social protest first started and Scheduled Caste politics
had been organised in East Bengal in the early 20th century.41 By the
early 1930s, Pramatha Ranjan had taken over the leadership of the
MM, which was registered in 1943. After Partition, it functioned from
two centres: the East Bengal chapter operated from its original hub in
village Orakandi in Faridpur, while Thakurnagar became the cultural
centre of the Matua devotees who migrated after 1950 (Biswas 2012:
518). In this period, when large-scale Dalit migration started, Thakur
began to visit the camps, not so much to organise agitations, but in his
role as the spiritual leader of the MM, which many of these refugees
were affiliated with.42 By the early 1960s, he was widely recognised as the
hereditary guru of the MM, the followers of which were concentrated
in the two border districts of 24-Parganas and Nadia.43
As a Congress leader, Thakur was encouraging Namasudra ref-
ugees to settle down in other parts of India where they could get
land to resettle. The Namasudra pioneer cultivators had, in the
past, reclaimed the bil (marshy) tracts of east Bengal and the for-
est lands of the Sunderbans, he argued. So, if they could get vacant
land, they could build a new Bengal in Dandakaranya or Andaman
Island. According to one report, he personally visited refugee camps
and persuaded the Dalit refugees to move to Andaman (Biswas 2011:
517). His political support for the Dandakaranya Scheme did not
endear him to the camp refugees, but they remained ambivalent. In a
group meeting with the Bagjola camp refugees, one of them proudly
declared that 95 per cent of the residents of their camp were Matuas,
yet they did not like Thakur’s support for the Dandakaranya scheme –
thus making a clear distinction between the political leader and the
spiritual guru.44
41 For details, see Bandyopadhyay (2011: 30–63).

42 Interview with Amit Mukherjee [name changed] on 18/3/2013.
43 See petitions from Matua Mahasangha devotees in IB Records, F. No. 2076–50.
44 Group meeting with Bagjola Camp residents on 26/6/13.
76
The point of departure for Thakur came in 1964, when, following

the Hazratbal riots, panic-stricken refugees began to arrive in Sealdah
Station by Down Barisal Express, with harrowing tales of atrocities.45
In all, in 1964, more than 400,000 Dalit peasants crossed the border as
refugees into West Bengal (Chatterji 2007: 112). This fresh influx and
the horror stories of riots in East Pakistan heated up the atmosphere
in West Bengal too – particularly in the border districts of 24-Parga-
nas and Nadia, as well as in Calcutta, where full-scale anti-Muslim
riots started from 10 January. As a precautionary measure, dawn-to-
dusk curfew was clamped on Namasudra majority areas in 24-Parga-
nas and Nadia (Amrita Bazar Patrika, 10, 11, and 16 January 1964).
When further police repressive measures specifically targeted the
Namasudra refugees, Thakur resigned from the Assembly on 6
March 1964 as a mark of protest against the inaction of the Congress
government against attacks on refugees in Bongaon and participated
in a series of meetings with the opposition leaders.46At this stage, he
was accused of supporting the activities of the Save Pakistan Minori-
ties Committee, which was proposing an economic blockade of East
Pakistan.47 So, on charges of inciting public disturbances, the West
Bengal Government arrested him on 19 April under the Defence of
India Rules; a few days later, on 30 April, Jogendranath was arrested
too under the same rules.48 Both were kept incarcerated in Dum Dum
Central Jail until 3 June.49
But even at this stage, Thakur’s political manoeuvres were of less
significance than his career as a spiritual leader. When he was arrested
in 1964, he was described in a police report as a leader with ‘consid-
erable influence upon the refugees of East Pakistan particularly upon
the Namasudra community’.50 But this influence was not of a political
45 Prophet Muhammad’s sacred relics were reported stolen from the Hazratbal mosque
in Srinagar in Kashmir. In retaliation, riots broke out in Khulna from 4 January,
spreading to Jessore the next day. Amrita Bazar Patrika, 5, 6, and 7 January 1964.
46 IB Records, F. No. 29/26, pp. 232–231; Copy of meeting report dated 28.3.64,
F. No. 353/24.
47 ‘A note on Shri Pramatha Ranjan Thakur . . . of Thakur Colony, P.S. Gaighata,
24-Parganas’, IB Records, F. No. 2076–50.
48 Note sheet, dated 22/5/64, IB Records, F. No. 2076–50. Also see Jugantar, 20 April
and 1 May 1964.
49 From Deputy Inspector General of Police, I.B., West Bengal to Deputy Secretary,
Home (Special) Department, 5 June 1964, IB Records, F. No. 2076–50.
50 ‘A note on Shri Pramatha Ranjan Thakur . . . of Thakur Colony, P.S. Gaighata,
24-Parganas, IB Records, F. No. 2076–50.
77
kind, as all those who protested against his arrest, participated in

hunger strikes and sent petitions for his release, described him as a
religious guru, as the Mahasanghadhipati of the MM, whose incar-
ceration had seriously jeopardised the religious activities of the sect.51
In other words, his influence was not because of his politics, but his
hereditary role as the spiritual guru of the sect, which he not only kept
alive, but expanded after the 1950s.
Thakur’s religious role helped to unite a community that had lost
its physical space and were now dispersed across the subcontinent.
He started visiting Dalit refugee camps and colonies, not only in
West Bengal but also in Andaman and Madhya Pradesh. In 1986, he
had the MM formally registered in West Bengal as a socio-religious
organisation to preach the messages of Harichand and Guruchand.
It was devoted to mobilise the dispersed Namasudra community and
to convert Thakurnagar into a new cultural and spiritual hub for a
Namasudra renaissance, reminiscent of the olden glorious days of
Orakandi. After his death in 1990, his son Kapil Krishna established
a centre at village Ashti in the district of Gadchiroli in Maharashtra.
This centre was intended to mobilise the Namasudra refugees who
had settled down in central and south India in the 1960s. In 2008,
the MM had 6,755 branches all over India; and in 2010, it claimed
to have nearly 50 million members, belonging to 100,000–120,000
families.52 On the occasion of baruni mela – the major festival of
the sect – thousands of devotees from all over India visit Thakur-
nagar in their annual pilgrimage. The reinvented MM, thus, seems to
have offered a new imagined space to a geographically dispersed and
socially divided community, trying again to recover their collective
self in post-Partition India. The weight of their numbers compelled
all mainstream political parties in West Bengal to take note of the
MM by 2009.
The leadership of this reinvented MM comes from an upwardly
mobile ambitious Namasudra middle class. The members of the com-
munity who have stayed in West Bengal have done remarkably well
51 See several petitions in IB Records, F. No. 2076–50.

52 See MM publications such Matua Matobad o Sree Sree Hari-Guru-Chand Thakur
(2008); Matua Mahasangher 23tama Barshik Sammelan, Kendriya Karya-nirbahee
committeer pakshe Sadharan Sampadaker Pratibedan (2009); and Matua Maha-
sangher Sangbidhan (n.d.). However, as the general secretary of the MM admits,
these are approximate numbers as they do not have proper membership records.
Interview with Ganapati Biswas, General Sceretary, MM, Thakurnagar, 10/01/10.
78
since the 1980s.53 But despite their educational, economic, and social
progress, they have remained politically marginal in a state where the
higher echelons of political power even today are monopolised by a
high-caste Hindu bhadralok elite (cf. Lama-Rewal 2009). Therefore,
conscious of their strength of numbers in a democratic polity, the lead-
ership of the MM since 2009 have sought their political empowerment
through clever negotiations with the mainstream political parties, such
as the Trinamul Congress and the CPI(M). But while doing so, they
have not projected an exclusive Scheduled Caste agenda and empha-
sised the more universalist anti-caste approach of Guruchand Thakur.
In that sense, while the MM remains predominantly Scheduled Caste
in its membership, it is not a Scheduled Caste movement in a conven-
tional Ambedkarite sense (Bandyopadhyay 2011: 265–73).
Conclusion
Conventional histories of Dalit movements in India have rarely looked
into the importance of space. In Bengal, for both Namasudras and
Rajbansis, their close geographical location gave them that crucial
spatial capacity for social mobilisation. Partition was, therefore,
extremely significant in their history, as the loss of habitat and subse-
quent dispersal were major reasons for the decline of their movements.
The Rajbansis were, comparatively speaking, less dispersed as parts of
their traditional habitat remained in West Bengal, where many of them
settled eventually. This gave them that capacity to mobilise once again
for a more specific demand for an autonomous ethnic space in the
form of Kamtapur. The dynamics of that movement remains outside
the scope of this short chapter. But the point that can perhaps be made is
that because of this historic shift in the life trajectories of these two
large Dalit communities, the organised Scheduled Caste movement,
which was based largely in east and north Bengal in the colonial
period, lost its momentum in post-Partition West Bengal.
Also, Dalits in Bengal have never been a homogenous community
and the Partition exacerbated their internal divisions. The bone of
53 The census of 2001 recorded the literacy rate among the Namasudras in West Bengal
to be 71.93 per cent – ahead of the provincial average of 68.64 per cent. The male
literacy rate among the Namasudras was 80.58 per cent and female literacy rate was
62.76 per cent. In terms of their occupational structure, 21.4 per cent were land-
owning peasants, 16.9 per cent agricultural labourers, and 61.7 per cent were in
various professions (Rana and Rana 2009: 21, 25).
79
contention was the question of their alliance with the Muslim League.
While the Federation and its leader, Jogendranath Mandal, thought
that such an alliance was in the best interests of the Dalit, others such
as Thakur dreaded its consequences and lent their support to the
Congress–Mahasabha campaign for partitioning Bengal. But Partition
in the long run helped neither group, and the peasants were ultimately
uprooted from their land and home. The experience of migration and
camp life made them into refugees, with a different kind of struggle at
hand – the struggle for resettlement. As their leaders got embroiled in
that struggle, the caste issue receded to the background. The specific
political dynamics and the left-liberal ideologies of the refugee move-
ment and its caste Hindu leadership worked to suppress the caste
question from West Bengal’s public life.
This Dalit refugee struggle had a fragmented history, as the two Dalit
leaders, Mandal and Thakur, represented two different approaches to
the issue of resettlement. Mandal wanted to reclaim physical space
in West Bengal and was prepared to lay down his life in battle for
that. Thakur, however, preferred negotiation and self-help in mat-
ters of resettlement. In the end, frustrated by the insensitivities of the
Congress government, he devoted his energies to invent a spiritual
space in the form of MM, where a dispersed Namasudra community
could eventually unite and reinvent their collective self. The struggle
to reclaim physical space almost died with Mandal, in the face of stiff
opposition of a Congress government and the apathy of the Hindu
Bengali society, which wanted to get rid of these unwanted cultivat-
ing refugees, who were eventually dispersed across the country. Their
dream of reclaiming physical space in West Bengal was finally killed
by the Marichjhanpi massacre of 1978, masterminded by a CPI(M)-
led leftist government. But their quest for spiritual space continued
behind the glare of public attention and political contest. However,
partly as a result of this chequered historical trajectory of the Nama-
sudra community – the main powerhouse behind organised Scheduled
Caste movement in the colonial period – that movement was signifi-
cantly weakened in post-Partition West Bengal.
Bibliography
Bandyopadhyay, Saradindu. 1997. ‘The Riddles of Partition: Memories of the
Bengali Hindus’, in Ranabir Samaddar (ed), Reflections on the Partition in
the East, pp. 59–72. New Delhi: Vikas Publishing House.
Bandyopadhyay, Sekhar. 2004. Caste, Culture and Hegemony: Social Domi-
nance in Colonial Bengal. New Delhi: Sage Publications.
80
Bandyopadhyay, Sekhar. 2009. Decolonization in South Asia: Meanings of Free-

dom in Post-independence West Bengal, 1947–1952. London: Routledge.
The Namasudras of Bengal, 1872–1947 (2nd edition). New Delhi: Oxford
University Press.
Bandyopadhyay, Sekhar. 2014. ‘Does Caste Matter in Bengal? Examining the
Myth of Bengali Exceptionalism’, in Mridula Nath Chakraborty (ed), Being
Bengali: At Home and in the World, pp. 32–47. London: Routledge.
Basu, Swaraj. 2003. Dynamics of a Caste Movement: The Rajbansis of North
Bengal, 1910–1947. New Delhi: Manohar.
Basu Ray Chaudhury, Anasua. 2004. ‘Nostalgia of “Desh”, Memories of Par-
tition’, Economic and Political Weekly, 39(52): 5653–60.
Basu Ray Chaudhury, Anasua. 2010. ‘Politics of Rehabilitation in Struggle
of the Lower Caste Refugees in West Bengal’, Voice of Dalit, 3(1): 61–82.
Biswas, Manosanto. 2012. ‘Banglar Namasudra Andolone Pramatha Ranjan
Thakurer bhumika’, in Manju Chattopadhyay (ed), Itihas Anusandhan 26,
pp. 511–20. Kolkata: Paschimbanga Itihas Samsad.
Biswas, Sukumara and Hiroshi Sato. 1993. Religion and Politics in Bangla-
desh and West Bengal: A Study of Communal Relations. Joint Research
Programme Series No. 99. Tokyo: Institute of Developing Economies.
Bose, Sugata. 1986. Agrarian Bengal: Economy, Social Structure and Politics
1919–47. Cambridge: Cambridge University Press.
Chakrabarti, Prafulla K. 1999. The Marginal Men: The Refugees and the Left
Political Syndrome in West Bengal. Calcutta: Naya Udyog.
Chatterjee, Nilanjana. 1990. ‘The East Bengal Refugees: A Lesson in Survival’,
in Sukanta Chaudhuri (ed), Calcutta: The Living City, vol. 2, pp. 70–77.
Calcutta: Oxford University Press.
Chatterji, Joya. 2007. The Spoils of Partition: Bengal and India, 1947–196.
Cambridge: Cambridge University Press.
Kudaisya, Gyanesh. 1995. ‘The Demographic Upheaval of Partition: Refugees
and Agricultural Resettlement in India, 1947–67’, South Asia, 18: 73–94.
Rise of the Plebeians? The Changing Face of the Indian Legislative Assem-
blies, pp. 361–92. New Delhi: Routledge.
Mandal, Jagadishchandra. 1999. Mahapran Jogendranath o Babasaheb Ambed-
kar. Kolkata: Biswas Publisher.
Mandal, Jagadishchandra. 2003. Mahapran Jogendranath, vol. 6. Kolkata:
Chaturtha Dunya.
Nakatani, Tetsuya. 2011. ‘Partition Refugees on Borders: Assimilation in West
Bengal’, in Abhijit Dasgupta, Masahiko Togawa, and Abul Barkat (eds),
Minorities and the State: Changing Social and Political Landscape of Ben-
gal, pp. 66–87. New Delhi: Sage Publications.
Pal, Babul Kumar. 2010. Barisal theke Dandakaranya: Purbabanger Krishijibi
Udbastur Punarbashaner Itihas. Kolkata: Mitram.
81
Rana, Santosh and Kumar Rana. 2009. Paschimbange Dalit o Adibasi. Kol-
kata: CAMP.
Roy, Haimanti. 2012. Partitioned Lives: Migrants, Refugees, Citizens in India
and Pakistan, 1947–1965. New Delhi: Oxford University Press.
Sen, Dwaipayan. 2010. ‘A Politics Subsumed: The Life and Times of Jogen-
dranath Mandal. Himal South Asian, April, http://www.himalmag.com/
A-politics-subsumed_nw4416.html (accessed 3 April 2010).
Tan, Tai Yong and Gyanesh Kudaisya. 2000. The Aftermath of Partition in
South Asia. London: Routledge.
Thakur, Kapil Krishna. 2007. ‘Dalits of East Bengal: Before and after the Parti-
tion’, in Biswajit Chatterjee and Debi Chatterjee (eds), Dalit Lives and Dalit
Visions in Eastern India, pp. 26–38. Kolkata: Centre for Rural Resources.
82
4
PARTITION AND
THE MYSTERIOUS
DISAPPEARANCE OF CASTE
IN BENGAL
Partha Chatterjee
It is often said, with some measure of self-satisfaction, that caste has

disappeared from West Bengal politics. Unlike in colonial Bengal of
the past and other parts of India today, there is no political formation
that is identified with any caste group or bloc. Caste issues do not
feature in election campaigns. There are no political movements that
make caste demands. This chapter will argue that this situation has
been created specifically because of the peculiar consequences of the
partition of Bengal in 1947 with its ramifications on migration, urban-
isation, the re-established social dominance of the urban upper-caste
Hindu elite, and the new ideological formations. Since the politics of
caste operates in terms of the relations between castes, the possibility
of lower-caste mobilisation in West Bengal today must be considered
in relation to the social foundations of upper-caste dominance.
Upper-caste dominance
One of the peculiarities of public life in West Bengal is that while there
is much talk about caste discrimination and caste politics in other
parts of India, any mention of caste practices in West Bengal is virtu-
ally taboo. It is not acceptable in polite urban conversation to bring up
the topic of caste and one who does so is deemed either not civil
enough to know that such things are not talked about among gentle
folk, or deliberately rude and provocative. Those doing field research
in West Bengal feel a sense of embarrassment when they have to ask
the question, ‘What is your caste?’ and are relieved when the answer is
a stock ‘General’, ‘SC’ (Scheduled Caste), or ‘OBC’ (Other Backward
83
PA R T H A C H AT T E R J E E
Classes), without the specific caste name being mentioned. It is as if

caste has been relegated to the interior zones of the private into which
an outsider should have no legitimate access. In the field of electoral
politics, while the demands of the Muslim minority are often an issue
on which parties and leaders have taken positions, especially in recent
years, demands of caste groups, whether of Dalits or OBCs, have
never figured in the political arena. In fact, from the extreme right of
the political spectrum to the extreme left, no party in West Bengal is
willing to discuss the contemporary caste system in the state.
It could be argued that this is a mark of the success of the modernis-
ing pedagogical project of Bengali intellectuals and social reformers
since the 19th century: the absence of conversation about caste could
mean that it has largely disappeared from the consciousness of the
people of West Bengal. But there are, as the several contributions to
this volume demonstrate, many reasons to think that that is not the
case. On the contrary, while practices of caste privilege and discrimi-
nation continue, their designations have been displaced on to other
conceptual categories, such as education and cultural accomplishment.
That practices of caste privilege continue is easily demonstrated by
the near-complete dominance of the upper castes in virtually every polit-
ical institution, including those where the leadership is elected, and in
every modern profession. The contrast between the caste composition
of ministries in West Bengal with those of most other states in India is
striking. From the first Congress governments after independence to the
Left Front (LF) governments between 1977 and 2011 to the present Tri-
namul Congress (TMC) government, the cabinet has been dominated
by leaders from the Hindu upper castes. What is striking is not only
the absence, except for the token one or two, of ministers from Dalit
castes, but the paucity of leaders from the middle castes. One would
have assumed that with the political mobilisation of the rural peasantry –
a process in which West Bengal certainly did not lag behind the other
states of India – there would have been many more leaders from the rel-
atively better-off peasant castes in positions of state-level leadership. But
in none of the major political parties has this happened. Why did not
the political mobilisation of the peasant castes lead to the rise of peasant
leaders from the dominant middle castes in rural areas? This constitutes
one aspect of the mysterious disappearance of caste in West Bengal.
In the absence of systematic research on the subject, it is difficult to
estimate the scale of upper-caste dominance in contemporary West Ben-
gal. If one extrapolates from the proportions of the Hindu upper castes
(Brahman, Baidya, and Kayastha) in the West Bengal districts in the 1931
84
PA RT I T I O N A N D D I S A P P E A R A N C E O F C A S T E
Census, which is the last time caste enumeration was carried out in the
Census, their proportion should be slightly under 10 per cent. It is possi-
ble that with the immigration of upper-caste Hindus from East Pakistan
and Bangladesh, this proportion has increased a little – by how much
is anybody’s guess. But even the most cursory observation of the insti-
tutions of higher education, professional bodies, literary societies, and
cultural associations will show that they are overwhelmingly populated
by the upper castes. Santosh Rana narrates his experiment of counting
the first 100 names listed in the directory of literary personalities pub-
lished by Ganashakti, the newspaper of the Communist Party of India
(Marxist) (CPI(M)). He found that as many as 90 were from the upper
castes (Rana and Rana 2009). I once tried in the early 1980s to estimate
the caste distribution of Bengali Hindu names in the Calcutta telephone
directory (before the age of the mobile phone). I found that 70 per cent
were definitely upper-caste names, only 5 per cent were definitely middle
or lower castes, and the remaining 25 per cent could not be identified by
caste. Of this last category, we know that common names such as Chaud-
huri, Ray, or Majumdar – all associated with land proprietorship – were
mostly used by the upper castes, while names such as Sarkar, Biswas, or
Das, though not exclusively restricted to the upper castes, were also used
by them. Hence, my unscientific study told me that at least 75 per cent of
the Bengali middle class of Calcutta (even if one included the very small
Bengali Muslim middle class) consisted of Hindu upper castes.
If one looks at long-term historical trends, there is something sur-
prising about the prevailing structure of Hindu upper-caste dominance
because it represents a sharp reversal of those historical trends. As in
other regions of India, the initial dominance of upper-caste Hindus in
middle-class occupations during the colonial period came under severe
challenge in Bengal in the last two decades before independence. The
rise of a new educated middle class from among the superior peasantry
and popular political mobilisation led to an assault on the institutions
of upper-caste privilege. But the consequence of independence and the
Partition of the province was that the erstwhile dominance of upper
castes was re-established in West Bengal. The reversal happened dur-
ing the lifetime of a single generation without anyone talking about it.
A social counter-revolution took place behind a veil of silence.
The late colonial challenge

The distribution of caste and religious groups in the western Bengal
districts as given by the 1931 Census is shown in Table 4.1.
85
Table 4.1 Distribution of caste and religious groups as percentage of total

district population, 1931.
District Upper- Middle- Depressed Tribes Muslims Other

caste caste classes
Hindus Hindus
Burdwan 9.81 29.07 34.50 7.69 18.56 0.37

Birbhum 6.44 22.45 36.50 7.82 26.69 0.10
Bankura 11.29 39.70 31.81 12.51 4.59 0.10
Manbhum 7.07 37.50 22.42 16.29 6.01 0.71
Midnapur 6.31 58.17 19.15 8.54 7.59 0.24
Hooghly 10.33 41.84 17.00 4.56 16.17 0.10
Howrah 10.48 46.58 20.86 0.42 21.27 0.39
24-Parganas 6.04 27.61 29.95 1.97 33.65 0.78
Nadia 4.84 21.99 10.19 0.54 61.77 0.67
Murshidabad 3.79 25.31 13.24 2.00 55.56 0.10
Malda 1.41 25.84 10.21 8.25 54.28 0.01
Dinajpur 1.32 32.78 5.76 9.48 50.51 0.15
Jalpaiguri 1.80 43.82 4.07 22.50 23.99 0.88 (+ 2.94)
Darjeeling 3.12 14.50 2.38 14.51 2.63 13.62 (+ 49.24)
Cooch Behar 1.87 59.48 2.74 – 35.33 0.58
Source: Computed from Census of India, 1931, vol. 5 (Bengal and Sikkim) and vol. 7
(Bihar and Orissa). Compared with the present districts of West Bengal, 24-Parganas
in this table excludes the Bongaon subdivision (then in Jessore district); Nadia includes
Kushtia subdivision (now in Bangladesh); Dinajpur includes Dinajpur subdivision (now
in Bangladesh) and excludes part of Purnea district in Bihar, which was transferred to
West Dinajpur district in West Bengal; Cooch Behar was then a princely state outside
British India; and Manbhum was in Bihar, of which the Purulia subdivision was trans-
ferred to West Bengal to form the present Purulia district.
It shows that except for the districts of Bankura, Hooghly, and

Howrah, the proportion of upper-caste Hindus is everywhere less than
10 per cent. Nirmal Kumar Bose had used these same census figures to
show that the three upper castes – Brahman, Baidya, and Kayastha – fully
dominated the field of education and the higher professions in 1931
(Basu 1949). The middle and lower castes, however, were entirely
confined to occupations involving manual labour (see Table 4.2). Not
unexpectedly, therefore, the Hindu upper castes constituted almost the
entire educated middle class of Bengal at that time, while the others
were engaged in agriculture, crafts, or small business.
In what respects should we expect the current situation to be dif-
ferent? It is possible that because of the continuous influx of refugees
from East Pakistan/Bangladesh, the proportion of upper-caste Hindus
86
Table 4.2 Literacy and occupation of selected castes as percentage of the total
population of the various castes in Bengal, 1931.
Caste Literate Agriculture Industry Higher

professions
Brahman 37.28 15.38 4.50 30.76

Baidya 51.74 6.04 1.85 49.40
Kayastha 32.90 20.03 5.16 22.42
Goala 10.17 37.49 7.28 5.42
Kamar 14.51 21.81 56.11 5.32
Bagdi 1.92 81.74 5.03 1.17
Bauri 0.77 65.94 4.07 0.78
Source: Basu (1949), also available in English translation as Bose (1975).
Table 4.3 Proportion of Scheduled Castes, Scheduled Tribes, and Muslims as

percentage of total district population, 2001.
District Scheduled Castes Scheduled Tribes Muslims
Burdwan 26.98 6.41 19.78

Birbhum 29.51 6.74 35.08
Bankura 31.24 10.36 7.51
Purulia 18.29 18.27 7.12
Midnapur 16.40 8.31 11.33
Hooghly 23.58 4.21 15.14
Howrah 15.42 0.45 24.44
North 24-Parganas 20.60 2.23 24.22
South 24-Parganas 32.12 1.23 33.24
Nadia 29.66 2.23 25.41
Murshidabad 12.00 1.29 63.67
Malda 16.84 6.90 49.72
Uttar Dinajpur 27.71 5.11 24.22
Dakshin Dinajpur 28.78 16.12 33.24
Jalpaiguri 36.71 18.87 10.85
Darjeeling 16.09 12.69 5.31
Cooch Behar 50.11 0.57 24.24
Source: Computed from Census of India, 2001 (courtesy: Sohel Firdos).
in the border districts of 24-Parganas, Nadia, Dinajpur, Jalpaiguri, and

Cooch Behar has increased. If we compare Table 4.1 with Table 4.3,
we will see that except for Murshidabad, the proportion of Muslims in
the border districts of West Bengal has decreased since 1931.
However, the proportion of SC (called Depressed Classes in 1931)
has increased in Hooghly, Burdwan, Bankura, and especially in Nadia.
87
The main reason is the immigration of Dalit refugees from East Paki-
stan/Bangladesh. The proportion of SC in the northern districts also
show a significant rise, but this is mainly because the Rajbanshi, who
were not counted within the Depressed Classes in 1931, are now listed
as a SC. Other than this, it is difficult to estimate how the proportions
of different castes in the different districts may have changed since
1931. For this, we will have to await the results of the ongoing caste
census. But let us move a little further in interpreting the 1931 results
in the context of historical developments since.
The social and political dominance of the three upper castes was
based on their proprietorship of zamindari and intermediate tenurial
rights and access to English education. The traditional Muslim nobil-
ity in Bengal did not take to English education in the 19th century. As
a result, the Hindu upper castes were in command of public life in Ben-
gal at the beginning of the 20th century. The nature of this dominance
has been documented by Anil Seal (1968). That this dominance was
intact in the early decades of the century is indicated by the population
distribution within the city of Calcutta. Needless to say, as the admin-
istrative, commercial, educational, and cultural centre of the province,
Calcutta represented the summit of middle-class life in Bengal.
We see from Table 4.4 that of nearly 1.5 million people living in Cal-
cutta in 1931, Hindus constituted 70.34 per cent, Muslims 25.03 per
cent, and Christians 3.43 per cent. Of the Hindu castes, the proportion
of the three upper castes was 26.77 per cent of the city’s total popula-
tion and 38.15 per cent of the Hindu population. Clearly, but not sur-
prisingly, the upper castes were far more concentrated in Calcutta than
in the districts of West Bengal. Most among the Brahman, Baidya, and
Kayastha groups were literate, and bearing in mind that women from
these groups were less likely to be literate than men, the male upper-
caste population in the city was certainly overwhelmingly literate. Not
only that, most were also literate in English. Other than Christians, no
Indian population group in the city was as literate in English. It is easy
to conclude that the educated middle-class professions were dominated
by the three Hindu upper castes.
This dominance came under challenge in the 1930s. With the spread
of the Khilafat movement in the 1920s, Muslim peasants in the Ben-
gal countryside, led by a new crop of Muslim mass leaders, began to
be organised within the Congress. Under C. R. Das’s leadership, the
Swarajist Congress in Bengal forged a ‘Hindu–Muslim pact’ by which
it went into municipal elections with an understanding on sharing seats
88
Table 4.4 Religion/caste and literacy in Calcutta, 1931.
Religion Percentage of Literacy as Literacy in

total Calcutta percentage of English as
pop. religion/caste pop. percentage of
religion/caste pop.
Hindu 70.34 39.57 18.34

Muslim 25.03 27.68 8.46
Christian 3.43 70.63 63.64
Brahman 13.24 59.60 32.39
Kayastha 12.29 65.39 38.06
Baidya 1.24 71.91 50.64
Mahishya 5.28 32.13 12.09
Saha 0.84 38.70 19.55
Kamar 0.71 39.36 13.71
Goala 2.72 27.75 6.91
Napit 1.08 30.55 9.17
Dhopa 0.95 13.89 2.35
Chamar 2.18 8.50 0.84
Source: Computed from Census of India, 1931.
and official appointments between the two communities. After Das’s

death in 1925, the pact fell apart and the Muslim leadership gradually
moved away from the Congress. In the 1930s, this rural leadership,
located mainly in the eastern Bengal districts, began to organise Mus-
lim tenants against landlords and moneylenders under the banner of
the Krishak Proja Party. Further, the establishment of a university in
Dacca significantly increased the number of Muslim graduates in east-
ern Bengal and expanded the Bengali Muslim middle class.
At the same time, the peasant mobilisation produced by the Civil
Disobedience movement in the western districts of Midnapore,
Hooghly, Bankura, Birbhum, and Burdwan brought leaders from
the Mahishya, Sadgop, and other peasant castes to the district level
of the Congress organisation. In some districts, a conflict emerged
between Congress leaders from the ranks of zamindars and the new
village leaders of the peasant movement. There were incidents when
peasant leaders from the districts faced insult and humiliation from
upper-caste Congress leaders of Calcutta. Birendra Nath Sasmal, the
most popular Congress leader of Midnapore, for instance, had to put
up with persistent jibes and rude behaviour from the elite Congress
89
leaders of the city, even though he was himself a prominent barrister

of the Calcutta High Court. The peasant leaders of Arambag were
sidelined and hardly ever allowed to speak at party meetings by the
zamindars who led the Congress in Hooghly. Hitesranjan Sanyal’s
studies carried out in the 1970s – posthumously published in Sanyal
(1994) – provide ample documentation of this aspect of the Congress
movement in south-western Bengal. These oppositions were marked
both by agrarian conflicts and the cultural discriminations – open as
well as subtle – of caste. But unlike in southern India or Maharashtra,
there was no anti-Brahman movement in Bengal. In my view, there
were two reasons for this.
The first is that there were no dominant peasant castes in Bengal
of any size or spread that could have made a significant impact on
provincial politics. In an earlier essay, I showed from the 1931 Census
figures that except for the Mahishya in Midnapore (31.56 per cent),
Howrah (24.92 per cent), Hooghly (15.74 per cent), and 24-Parganas
(12.14 per cent), there was no numerical predominance of any single
middle caste in any other district (Chatterjee 1997). The influence of
the Mahishya caste was established in the Midnapore Congress during
the nationalist movement, but it made no significant mark on the pro-
vincial Congress. There was no dominant caste in Bengal that could
compare with, let us say, the Lingayat and Vokkaliga in Karnataka,
or the Vellala in Tamil Nadu, or the Kamma and Reddy in Andhra,
or the Maratha in Maharashtra. Similarly, the Jat, Rajput, Ahir, or
Goala castes, spread across several states of northern India that have
given rise to the powerful political force of the Jat, Rajput, or Yadav
movements of recent years, have no parallel in Bengal. Except for the
short-lived Bangla Congress in the 1960s, there has been no successful
attempt in West Bengal to give independent political voice to the rich
farmer interest.
The second reason is connected to the first. The role of the domi-
nant peasant caste in the transformation of provincial politics was
played in Bengal by the Muslim peasant mobilisation of the Krishak
Proja movement. The attempt to oust the Hindu upper castes from
their positions of privilege in landed property, political influence, and
cultural dominance was launched in earnest in the 1930s. With the
formation of the provincial ministries led by the Krishak Proja Party
and the Muslim League after 1937, the dominance of the Brahman,
Baidya, and Kayastha elites was under serious threat. In the Muslim-
dominated districts of eastern and northern Bengal, they began to lose
their traditional hold over the municipalities, the union boards, the bar
90
associations, and even their control over schools and colleges. Even in
Calcutta, they had to make way for new Muslim claimants to positions
in government offices, the Calcutta Corporation, and the Calcutta Uni-
versity. This is not dissimilar to the challenge to Brahman dominance
in southern and western India in the same period, except that it was
a challenge mounted not against caste, but religious discrimination.
Why Partition?
Before we look at how the partition of Bengal ended this challenge and
re-established upper-caste dominance on a new basis in West Bengal,
we should briefly recount why the partition of the province took place
at all. There is complete amnesia today in West Bengal on this episode
of history: the prevailing belief, repeated endlessly in textbooks and
public oratory, is that the partition of Bengal was foisted on Bengalis
by outside forces – the British, the All-India Congress, and Jinnah. Yet,
the plain truth is that once it became clear in early 1947 that indepen-
dence would be accompanied by a partition of the country, it was the
Hindu political leadership of Bengal that demanded that their prov-
ince must be partitioned. The upper-caste elite was alarmed by the
prospect of Muslim-majority Bengal joining Pakistan. Even the United
Bengal proposal floated by H. S. Suhrawardy and Sarat Chandra Bose
was quickly aborted when Shyama Prasad Mookerji, putting on his
realpolitik glasses, pointed out that if, at a later date, the Muslim
majority in a sovereign United Bengal voted to join Pakistan, what
option would the Hindu minority have then? The option had to be
exercised right now, namely, by demanding the partition of Bengal
with the Hindu-majority districts, including Calcutta, joining the
Indian Union.
It is remarkable that the entire spectrum of Hindu political opinion
in Bengal from the Hindu Mahasabha on the right to all factions of
the Congress to the Communists on the left were unanimous in 1947
on the necessity to partition Bengal.1 A public opinion survey pub-
lished in the Amrita Bazar Patrika of 23 April 1947 showed that 98
per cent of Hindus were in favour of dividing the province. One need
not quibble too much about the scientific quality of the poll or the
1 Most communists have later recounted that in 1947, the pressure to recognise the
inevitability of a communal partition of Bengal was overwhelming. See the survey of
communist literature and reminiscences in Sengupta (1989).
91
exact percentages reported; that Hindu opinion was massively against

joining Pakistan, and hence, in favour of partition is beyond dispute.
Apart from the blatantly communal arguments, there were sophisti-
cated arguments that insisted that it was the Indian Union that was
more likely to uphold modern democratic traditions, launch social
and economic development, and safeguard the rights of all communi-
ties. Many argued for an administratively supervised transfer of popu-
lations between West and East Bengal to preclude chaos and violence.
These views were reflected in the telegram sent on 7 May 1947 to the
Secretary of State for India in London by some of the leading lights of
Bengal’s intellectual world, including Jadunath Sarkar, Ramesh Chan-
dra Majumdar, Meghnad Saha, Sisir Kumar Mitra, and Suniti Kumar
Chatterjee, supporting the partition of the province (Chakrabarty
1993).2
Those who opposed the partition of Bengal were the Muslims of
the province. They were in favour of an undivided Muslim-majority
Bengal joining Pakistan. But even there, some Muslim intellectuals
were doubtful about the feasibility of that proposal, suspecting that
the Hindus of West Bengal would not be persuaded to join Pakistan.
Consequently, they envisioned as a more homogeneous and realistic
option an East Pakistan excluding the western districts. The influential
East Pakistan Renaissance Society drew a map of Pakistan as early
as 1943, in which the Hindu-majority western districts were left out
(Shamsuddin 1968).
To assess the significance of the re-establishment of Hindu upper-
caste dominance in West Bengal, it is necessary to realise that the
threat to that dominance was not from a rebellion of the lower castes,
but a political and social challenge from a mobilised Muslim forma-
tion. That is what was averted by partition. The two powerful social
movements among the untouchable castes in Bengal in the 19th and
20th centuries were among the Rajbanshi and Namasudra communi-
ties.3 The Namasudra movement acquired a distinct political stamp
in the last years before independence only by its association with
B. R. Ambedkar’s Scheduled Castes Federation and its alliance with
the Muslim League. This political strategy crafted by Jogendra Nath
Mandal made sense only in the context of the three or four eastern
2 The general history of events leading up to the partition of Bengal is surveyed in Chat-
terji (1995).
3 These have been studied by Bandyopadhyay (1990; 1997; 2004).
92
Bengal districts where the Namasudra community was numerous. It

made little sense in the rest of Bengal. Only five of the 30 SC members
of the Bengal Assembly voted against the partition of the province.
With partition and the separation of Hindu-majority West Bengal,
Mandal’s strategy no longer made sense even in Pakistan. Barely three
years after independence, he resigned from his position in the Pakistan
cabinet and moved to West Bengal.4
One could argue, as a historical counterfactual, that had Bengal
remained united, an alliance of Muslim and lower-caste peasants
led by the rising middle classes from those communities might have
ousted the upper castes from political power, much like what has hap-
pened in southern and western India. The question of whether such
an alliance might have successfully negotiated the divide of religion
will, of course, have to remain unanswered. But this counterfactual is
useful precisely in highlighting what was forestalled by the partition
of Bengal. The possibility of an overwhelming alliance against Hindu
upper-caste dominance was suppressed for the next several decades
in 1947.
The consequences of partition

My arguments here will have to be somewhat schematic and often
hypothetical, since there is so little reliable empirical evidence on
macro-level caste structures in contemporary West Bengal.
First, partition meant that the abolition of zamindari was carried
out separately in the two parts of Bengal. In East Bengal, it was the
political fruition of two decades of anti-zamindari movements. In
effect, with the emigration of the vast bulk of propertied Hindus, it
amounted to their effective expropriation from the land. In West Ben-
gal, however, zamindari abolition was carried out through legal and
administrative methods without the backing of a political movement.
As is well-known, this left numerous loopholes through which propri-
etors were able, for a long time, to retain their control over land and
local power.
Second, because of this crucial difference in how zamindari aboli-
tion was carried out in the two parts of Bengal, its consequence was
not merely a restoration but a new formation of the middle class in
4 I have discussed this point at greater length in Chattopadhyay (2012), now reprinted
in Chattopadhyay (2013).
93
West Bengal – one part retaining its ties to rural landed property and
the other becoming purely urban. The migration of upper-caste Hin-
dus from East Pakistan to Calcutta and its suburbs was not a uniform
experience. Those who had prior attachments to the city or access to
finances or those who could arrange the exchange of urban property
between West and East Bengal were able to buy their way into estab-
lished middle-class locations. Others illegally occupied public or pri-
vate land in the outskirts of the city to establish refugee colonies. For
all these migrants, however, their new urban lives were radically sun-
dered from the traditional patterns of rural life in East Bengal, which
were deeply shaped by caste and communal relations. Most urban
refugees no longer owned landed property, nor were they serviced by
attached labour. The fact that refugee colonies tended to be roughly
homogeneous by caste and the district of origin only naturalised the
solidarity of common loss and struggle. The discriminations of caste
practice in rural society receded into a distant memory and were not
transmitted to the next generation. Those growing up in the refugee
colonies of Calcutta and its suburbs had no conception of what it
meant for the upper-caste manib or karta (master) and his family to be
serviced by the Namasudra or Muslim praja (tenant). They encoun-
tered in the city a different service population of poor migrants from
Bihar and Orissa, many of whom were upper caste themselves and
who jealously protected their own cultural claims to ritual purity. For
the younger generation of East Bengal refugees, therefore, the under-
standing of the chhotolok or the lower orders was no longer exclu-
sively, or even prominently, denominated by caste and religion. Caste
difference was something the family elders would only bring up when
negotiating marriages, at which time not only would caste endogamy
be invoked but also, perhaps even more importantly, the overwhelm-
ing stricture not to marry into a West Bengal family.
Third, partition migration afforded the possibility of individual
caste mobility. Once again, there is much anecdotal evidence of this
phenomenon but no reliable study. Yet, it is not difficult to imagine
that given the anonymity of the new circumstances in which many
refugees found themselves after their arrival in West Bengal, the
opportunity was available to assume a new caste identity. There is
one documentary source from which this can be verified. Following a
court decision that allowed any individual to change his or her name
through a sworn affidavit before a magistrate, Bengali newspapers
began to carry from the early 1950s a regular column of notices from
people who declared that they had changed their last names. Usually,
94
the name chosen was caste-neutral or that of a higher caste. We have

no idea of the scale of this phenomenon, but it suggests a possibility
of individuals from the lower castes entering the portals of urban mid-
dle-class life by changing their names. We must also remember that,
historically, families, lineages, and even specific sub-castes in Bengal,
mostly from the middle rungs of the caste ladder, have often sought
to translate a rise in economic or political fortunes into a higher-caste
status legitimised by priests and genealogists. There are many exam-
ples of this from the 18th and 19th centuries. What we see now is the
possibility of individual caste mobility through spatial migration.
Fourth, what we do know, however, is that the refugee movement
of the 1950s and 1960s, voicing mainly the demands of upper-caste
refugees in and around Calcutta, was led by the CPI. The United Cen-
tral Refugee Committee (UCRC) emerged as the principal organisa-
tion mobilising settlers in the refugee colonies to take to the streets
to make demands on the state. Alongside the organisation of white-
collar workers in government and private offices, this opened up a
new mobilisation of the urban middle class and a new crop of mainly
upper-caste leaders of the Left parties. These leaders were very different
from those of the peasant movements in the districts or the industrial
trade unions. But they came to exercise considerable influence over
the parties of the Left, underlining the overwhelming importance of
the city of Calcutta and its urban hinterland in the politics of the state.
They also crafted a new political rhetoric of the class struggle of the
urban middle class, the industrial working class, and the rural masses
against capitalists and landlords. The UCRC leaders recognised that
by remaining a closed movement of the immigrant population, they
would fail to draw the sympathy of the local population. They had
illegally occupied private and state lands and rumbles were already
being heard that the influx of foreigners was threatening the property
of the locals. The UCRC demanded the recognition by the govern-
ment of the refugee colonies as part of a larger demand of abolition of
landlordism and action against land speculators (Chakrabarti 1990).
This language of progressive political modernity had no place for the
backward identities of caste.
Fifth, until the 1970s, upper-caste formations in the western dis-
tricts did not undergo the same radical dissociation from the land
or the traditional structures of rural dominance as did the uprooted
upper castes from the east. That had to wait until the full impact of
land reforms and panchayat government was felt in the 1980s. By the
end of the 1980s, upper castes in the rural areas of West Bengal began
95
to move permanently into urban occupations. Today, it is hard to find

upper-caste families in most villages; those who still maintain a rural
home only visit on holidays and festivals.
Sixth, the period of the LF rule saw the supplanting of virtually
every social institution in the villages by the local branch of the politi-
cal party. Whereas earlier there would be the landlord’s drawing room
or the village club or the school committee or the caste council or the
puja committee, where party leaders would congregate to mobilise
votes before elections, now, every such social institution came to
be constituted merely as an extension of the political party. Where
there was competition between a Left party and the Congress at the
level of the village, there would be merely a rival club or a rival puja
committee. Instead of local social dominance being translated into
political power, there was the complete dominance of the political over
the social. The traditional sociological understanding of local political
power being a reflection of economic control over land and productive
resources or superior social status was reversed. Socially dominant
local groups came under the control of the Calcutta-based leadership
of the political party. Only in some of the Muslim-dominated districts
was there an autonomy of social institutions. This meant that the Cal-
cutta-centred upper-caste dominance of the Left parties was extended,
through the party structure, to dominance over local politics every-
where in the state. The defeat of the LF in 2011 has not meant a reas-
sertion of the autonomy of local social institutions. Rather, the TMC,
in the districts of southern Bengal, where it is now dominant, appears
to be keen to adopt the LF model of the dominance of the political
over the social and exclude the CPI(M) from local power. Since the
TMC is effectively an even more centralised party than the CPI(M),
the Calcutta-centred structure of upper-caste dominance of the entire
political space continues unchallenged.
Seventh, the emergence of a progressive modern political idiom
devoid of any association with caste or region facilitated the displace-
ment of bhadra or respectable status from upper-caste identity to the
insignia of education and the standardised genteel culture of Calcutta.
There is an entire cultural history that remains to be written of the
emergence in post-independence Calcutta of the standard language of
polite society – one that has shed all traces of rural roots in the vari-
ous district dialects and even of the old urban accents of traditional
Calcutta. It is a language of respectability and a genteel lifestyle that
goes with it that is, as it were, nobody’s patrimony, but one that each
person has to acquire through learning in order to become bhadra.
96
Such is the power of this cultural construct that even Bangladesh has
adopted it as the standard language of polite communication. It serves
to make urban middle-class respectability something that is earned
rather than inherited. In principle, it makes the status open to all.
Finally, all of these features of the new middle-class formation indi-
cate the creation of a dominant culture that is, in Antonio Gramsci’s
sense, hegemonic. It offers a cultural repertoire that claims to be the
normative standard for Bengalis from all regions and social ranks –
one that is open to all to acquire and use. Indeed, even non-Bengali
residents of the state are invited to learn the skills of the high Bengali
culture in order to be included in the ranks of the bhadra. One should
note here that unlike many other states of India, there have been no
serious attempts in West Bengal to impose a mandatory use of Bengali
in secondary schools or universities; instruction is carried out in Eng-
lish, Hindi, Urdu, Nepali, and other languages in both government
and private schools and colleges. Possibly, this is an indication of the
sturdiness and self-confidence of the hegemonic cultural formation
built by the upper-caste Hindu elite in West Bengal. It is also impor-
tant to stress that the location of this hegemony is mainly in the cul-
tural sphere since there is as yet no Bengali industrial or commercial
bourgeoisie, and, as we have explained above, the earlier foundation
of upper-caste dominance on land ownership has now disappeared.
Relying principally on its cultural capital, the Hindu upper castes con-
tinue to exercise their dominance over the entire spectrum of public
and political life in West Bengal.
Why is there no resistance?

This dominance cannot be explained simply by asserting that upper-
caste Hindus have conspired to block off all avenues of upward mobil-
ity for others. It is entirely true, of course, as analysed elsewhere in this
volume, that government policies of refugee rehabilitation in the 1950s
were strongly tilted in favour of upper-caste refugees. Forced migra-
tion to Dandakaranya and the Andamans were confined only to SC
refugees. When some of them wanted to return to West Bengal after
the formation of the LF government in 1977, they were violently
repressed at Marichjhapi. But these decisions only show that the upper
castes were able to protect and promote their particular interests. Why
were the far more numerous middle and lower castes unable to push
forward theirs? One must remember that of all the states of India,
West Bengal has the third largest proportion of Dalits to the population,
97
after Punjab and Himachal Pradesh; if one adds the proportion of

Muslims, a majority of the population of the state is either Dalit or
Muslim. Therefore, to put the question the other way round, how are
the numerically tiny upper castes able to continue their political and
social dominance in an electoral democracy based on universal fran-
chise, without an economic foundation in landed property or indus-
trial and commercial capital? That is the peculiar problem posed by
upper-caste dominance in contemporary West Bengal.
No dominant social group can produce a justificatory ideology and
simply expect it to be embraced by subordinate groups. Besides, if the
ideology achieves a position of hegemony, it does so without a sub-
stantial use of coercion; that is to say, the subaltern classes are made to
give their consent to it more or less voluntarily. It is not as though the
particular interests of the dominant group are thereby surrendered.
On the contrary, they are promoted, but only by successfully present-
ing them as the general interest of society as a whole. How was this
done by the upper-caste elite in post-Partition West Bengal?
It may be easier to see how the ideology of progressive political
modernity – unmarked by region, caste, or religion – came to achieve
hegemonic status by pointing out a few counterfactual instances of
events that did not happen in West Bengal in the second half of the 20th
century. The first was open conflict between the millions of migrants
from East Pakistan and Bangladesh and the local residents of West
Bengal. Cultural antagonism and prejudice between the regions east
and west of the River Padma were no less deep than those between
upper and lower castes or Hindus and Muslims. These were intensi-
fied by the sudden emergence of dense settlements of East Bengal refu-
gees on forcibly occupied land in the vicinity of Calcutta. Upper-caste
refugees made claims on the limited and intensely competitive sector
of urban white-collar employment. Agriculturist refugees moving into
the border districts sometimes managed, through superior enterprise
and farming skills, to displace local farmers from the land. While all
this caused adverse reactions among the older settled population of
West Bengal, they were rarely voiced in public. There is no doubt that
the fact that the demands of the refugees were phrased by their lead-
ers in the general terms of social and class justice, and encompassed
within a programme of progressive economic and political change,
and not as the claims of a particular ethnic group, went a long way
in preventing the deep-rooted cultural divide from acquiring the form
of a political conflict. One has only to compare the case of Muhajirs
in Karachi or Palestinians in Jordan or Lebanon to realise what could
98
happen when large groups of refugees congregate at one place and

organise politically to make claims on scarce local resources.
The second development that did not take place in West Bengal was
ethnic conflict against propertied businessmen and industrialists from
northern India. Given the loss of eastern Bengal and the democratic
consolidation of a culturally homogeneous state, this might have been
a populist option before the Bengali upper-caste leadership. To pro-
vide a comparison, a violent movement to throw out propertied and
prosperous Bengalis did occur in neighbouring Assam in the early
1960s. Once again, the ideology of universalist modernity standing
above narrow cultural divisions turned out to be of greater value to
West Bengal’s political leadership.
Similarly, there was also no pogrom against what might have been
a much weaker target, namely, the immigrant workers from Bihar,
Orissa and Uttar Pradesh. East Bengal refugees, even those belonging to
the upper castes, were pushed into seeking employment in skilled and
unskilled manual labour in both formal and informal sectors. Here, they
were in competition with migrant non-Bengali workers. But unlike the
Shiv Sena in Bombay, which launched its political career as champions
of the Marathi people through violent conflicts against south Indian
labourers, the political leadership in West Bengal, including that of the
refugee movement, chose to stick to its ideological commitment not to
encourage ethnic divisions within the working class. It had a greater
stake in establishing its credentials as upholders of the general interest.
This ideological orientation was consolidated after the LF came
to power. Now, the political leadership sought to speak for the state
as a whole. Earlier in the 1950s, the communists had supported the
integration, on linguistic grounds, of the Purulia subdivision of Man-
bhum district in Bihar, dominated by a Bengali-speaking majority,
with West Bengal. In the 1980s, however, it went back on its earlier
support for the political autonomy of the Nepali-speaking people of
Darjeeling, and instead, sought to defend the territorial integrity of the
state of West Bengal. Its repeated endorsement of a progressive cul-
ture unmarked by linguistic, religious, or caste particularities allowed
it to reconcile its anti-capitalist rhetoric with allowing the non-Bengali
bourgeoisie to dominate the urban economy as long as it stayed away
from local politics. Indeed, as we have explained before, the hege-
monic cultural formation enabled the Calcutta-based upper-caste elite
to present itself as a universal class, standing above all cultural and
local particularities, and moreover, as one that was open to entry by
those who acquire the necessary cultural accomplishments.
99
In short, the upper-caste elite culture became hegemonic precisely

because it was not exclusively about caste. Its persuasive power came
from its ability to create and defend larger social consolidations. Had
it been only about caste dominance – a hypocritical attempt to cling
on to age-old privileges – it would have been easily unmasked. Con-
sequently, the question of resistance – or lack of resistance – to this
dominance has to be answered in terms of the possibilities of counter-
hegemonic struggle. Ranabir Samaddar (2013) has recently raised this
question by noting that the most numerous Dalit groups in West Ben-
gal today are the Rajbanshi (18.4 per cent), the Namasudra (17.4 per
cent), the Bagdi (14.9 per cent), and the Paundra (or Pod) (12.0 per
cent). There is little common cause shared between these groups. The
Namasudra, an immigrant group from the east, were engrossed in its
demands of relief and rehabilitation, and, until the 1980s, largely sup-
ported the Congress. The Rajbanshi were internally differentiated into
jotdars and adhiars and divided in terms of political loyalty between
the Congress and the Left parties. The Bagdi are still mostly landless
and among the poorest sections of rural society. The Paundra of south-
ern Bengal have entered in a big way the urban informal economy of
Calcutta and its suburbs. Where is the ground for a common mobilisa-
tion against the dominance of the upper castes? Samaddar concludes
that no lower-caste movement in West Bengal today is in a position
to launch a counter-hegemonic campaign. His proposed answer is a
bahujan strategy of alliance with the middle castes. To what extent
that is a feasible option remains an unanswered question.
In the meantime, there are some features of contemporary caste
mobility that are of interest. Santosh and Kumar Rana (2009) have
pointed out from census figures that for some Dalit castes, such as the
Shuri, Jele, Kaibartta, and Dom, their absolute numbers have fallen
between 1991 and 2001. What could be the reason for this? They sug-
gest that many from these communities no longer wish to be regarded
as having Dalit identities. Again, the dependence of the Namasudra on
agriculture has declined from 56.7 per cent in 1991 to 38.3 per cent
in 2001, while that of the Paundra has declined from 69.5 per cent to
35.3 per cent.5 These figures suggest that new avenues of social mobil-
ity may have opened up for these two large Dalit communities living
in the environs of Calcutta.
5 I have discussed the findings of this study at greater length in Chattopadhyay (2010),
now included in Chattopadhyay (2013).
100
To conclude, it is difficult to see the emergence of a mobilised lower-

caste movement aimed against the dominance of the upper castes in
West Bengal today. There may, of course, be local mobilisations of
the kind seen recently in the Lalgarh region of West Midnapore. As
Samaddar has pointed out, the movement there produced a local
consolidation of Santal and other tribal groups led by the Mahatos,
an OBC group, against principally the Utkal Brahman contractors,
moneylenders, and small businessmen who had become the patrons of
the local CPI(M). Significantly, however, with the return of electoral
politics to the area during the recent panchayat elections, that consoli-
dation appears to have disintegrated. Therein lies the real mystery of
the absence of caste in West Bengal politics: the immensely superior
control exercised by the upper castes over the mechanisms of electoral
democracy through their dominance of the party system. That domi-
nance is not in any serious danger for now.
Bibliography
Bandyopadhyay, Sekhar. 1990. Caste, Politics and the Raj: Bengal 1872–1937.
Calcutta: K. P. Bagchi.
The Namasudras of Bengal 1872–1947. Richmond: Curzon.
Basu, Nirmalkumar. 1949. Hindu samajer garan. Calcutta: Viswabharati.
Bose, Nirmal Kumar. 1975. The Structure of Hindu Society, tr. André Béteille.
Delhi: Orient Longman.
Chakrabarti, Prafulla. 1990. The Marginal Men: The Refugees and the Left
Political Syndrome in West Bengal. Calcutta: Lumiére Books.
Chakrabarty, Bidyut. 1993. ‘The 1947 United Bengal Movement’, Indian Eco-
nomic and Social History Review, 30(4).
Chatterjee, Partha. 1997 (1982). ‘Caste and Politics in West Bengal’, in Partha
Chatterjee, The Present History of West Bengal: Essays in Criticism, pp.
69–86. Delhi: Oxford University Press.
Chatterji, Joya. 1995. Bengal Divided: Hindu Communalism and Partition.
Chattopadhyay, Partha. 2010. ‘Paschimbanglay jati o janajati’, Baromas, autumn.
Chattopadhyay, Partha. 2012. ‘Jogen mandaler ekakitva’, Baromas, autumn.
Chattopadhyay, Partha. 2013. Janapratinidhi. Kolkata: Anustup.
Rana, Santosh and Kumar Rana. 2009. Paschimbange dalit o adibasi. Kol-
kata: Camp.
Samaddar, Ranabir. 2013. ‘Whatever Has Happened to Caste in West Bengal’,
101
Sanyal, Hitesranjan. 1994. Svarajer pathe. Calcutta: Papyrus.

Seal, Anil. 1968. The Emergence of Indian Nationalism: Competition and
Collaboration in the Later Nineteenth Century. Cambridge: Cambridge
University Press.
Sengupta, Amalendu. 1989. Uttal challis, asamapta biplab. Calcutta: Pearl
Publishers.
Shamsuddin, Abul Kalam. 1968. Atit diner smriti. Dacca: Naoroz Kitabistan.
102
5
AN ABSENT-MINDED
CASTEISM?
Dwaipayan Sen
What became of the caste question in West Bengal is often asked, but
seldom considered at any great length in scholarly literature.1 Since
that state has not experienced caste-based mobilisations of the kind
witnessed in Maharashtra, Uttar Pradesh or Tamil Nadu, much less
the centrality of caste-related issues in legislative politics, a fairly well-
dispersed common belief has developed that West Bengal, somehow,
was able to relieve itself of such ‘backward’ attachments. Whatever
animosities that existed in the past were dissolved by the exceptional-
ity of the Bengali social – whether the Congress paternalism, which
reigned shortly after 1947 or the supposed compact with the commu-
nist regimes that followed. As it stands, our understanding of this sub-
sumption is depicted as the outcome of consent and mutuality born of
nationalist and communist hegemony. Indeed, we have little documen-
tation, much less discussion, of what was, and what is, bhadralok
casteism in West Bengal.2
In this chapter, I explore the analytically vexed problem of agency
with respect to the following anomaly: the domination of this state’s
political, social and cultural domains by the upper castes, even as it
was surely proclaimed that caste did not matter; indeed, the perpetua-
tion of caste inequality by those who disavowed the salience of caste.
Who, or what, is the agent of this domination? What is the ‘biography
1 This is a revised version of an essay presented at the Inter-Asia Five-college faculty

workshop at Amherst College, the annual Contemporary India Seminar on ‘Caste
Today’ at the University of Oslo, and published in India Seminar. I thank participants
at both venues, the co-editors of this volume, Tanika Sarkar, and Andrew Sartori for
their suggestions and discussions.
2 To date, for instance, we have no monographic historical treatment of caste in
postcolonial West Bengal.
103
D W A I P AY A N S E N
of the general category’ in West Bengal (Deshpande 2013)? These

questions require consideration because the resumption of upper-
caste domination and concomitant decline of the political visibility
of caste have been explained primarily as a consequence of social
structure in the first case, and acquiescence and accommodation in the
latter. After reviewing two key sets of contributions to the extant liter-
ature, I pose these questions to three dimensions of the caste question
in postcolonial West Bengal, which curiously enough, have – for the
most part – elicited limited comment: a brief account of the Congress
and Communist governments’ observation of reservation policies for
the Scheduled Castes; contemporary activist discourse about upper-
caste domination and Dalit inequality; and the prospects for alliances
between Dalit and Muslim communities.
In brief, I argue that upper-caste domination was far more willed
and coercive than we are given to believe, and that the presumed hege-
mony is less secure than we might think. As I hope to demonstrate, the
history of reservations policy as well as contemporary accounts of dif-
ferent aspects of relations between the upper castes and Dalits, index
a prejudicial systematicity, indeed an ideological formation commit-
ted to the preservation of the practical structures of domination, on
the one hand, and a pervasive discontent, on the other. We must con-
sider the possibility that the disproportionate influence commanded
by the upper castes of West Bengal and the related silence about the
caste question was also the outcome of their exercise of social agency.
Additionally, I hope to show that Dalits do not regard the making of
upper-caste domination in West Bengal as a process the upper-caste
bhadralok stumbled on through sheer contingency and structural
constraint alone, reluctant, or absent-minded casteists, as it were. I
wish to reconsider the notion that what we have seen in West Bengal
is upper-caste dominance with hegemony – an order to which Dalits
have seemingly acquiesced, accommodated and consented through
persuasion. A growing body of evidence suggests that political imagi-
nations have been disabused of continued faith in upper-caste Bengali
Marxist ideology.
Whither the caste question?

A key paradox facing us is this: Bengal was a province where the Dalit
movement was among the strongest during late colonial India. Fol-
lowing Partition and independence in 1947, these political energies
104
A N A B S E N T- M I N D E D C A S T E I S M ?
never quite re-emerged in a West Bengal subsequently dominated by

the upper castes.3 How have scholars explained this change? While the
relative neglect of the historical study of caste was perhaps a conse-
quence of the view that it was not a particularly useful category of
analysis for 20th-century Bengali society, especially given the supposed
rapprochement attained by mid-century and the overwhelming atten-
tion accorded to the study of religious communalism, there are signs
that a gradual shift in this consensus is under way.4
Undoubtedly the pioneering historian of caste politics in 20th-
century Bengal, Sekhar Bandyopadhyay recently extended the insights
of his previous works on the colonial period and the stress he placed
on Dalit integration and ideological hegemony within Indian nation-
alism into the postcolonial (Bandyopadhyay 2004: 237). He argued
that the Partition violence and refugee influx:
led to a rephrasing of the idioms of victimhood and resistance,

placing less emphasis on caste and focusing more on the pre-
dicament of displacement and the struggles of the refugees.
These idioms could be more easily absorbed into the modern
tropes of social justice deployed by the left-liberal ideologies
of the state. Hence, while caste discrimination did not disap-
pear, it was subsumed in a different idiom, marked by the
dominant discourse of class and religion.
(Bandyopadhyay 2009: 456)
3 That the decline of caste in West Bengal is held to be coterminous with the return of
the upper castes to political power is perplexing. No doubt, logic similar to that
which associates race with African Americans is at work in the idea that caste means
Dalits. In a complete turning of tables, might one argue that in fact, far from going
into decline after 1947, caste-based politics have enjoyed a robust life in West Bengal?
Note, for instance, that only leftist parties have seen an increase of Brahmins in their
cadres in recent years. See Sen (2012a) for a study of how the transition from colonial
to postcolonial rule affected the very possibility of Dalit politics. This chapter develops
some of the themes explored therein.
4 See the following for recent contributions that challenge the purported irrelevance of
caste to Bengal: Rana and Rana (2009); Roy (2011); Bandyopadhyay (2012); and
Chandra and Nielsen (2012). That said, it cannot be denied that compared, for
instance, to the historical study of Dalit movements elsewhere in India or the over-
whelming centrality of race to humanities and social sciences scholarship in the case
of 19th- and 20th-century United States, interest in this topic has been limited. The
Subaltern Studies, for instance, did not engage at any great length with the theme of
caste in 20th-century Bengal.
105
Bandyopadhyay showed how in East Pakistan, Dalits who had not

migrated prior to Partition, like many among the upper-caste gen-
try, increasingly became the targets of anti-Hindu intimidation and
harassment and were apprehended as part of the Hindu minority (see
also Sekhar Bandyopadhyay, this volume). The Congress and Mahas-
abha increasingly assimilated acts of violence against them as evidence
of the anti-Hindu policies of Pakistan, and, as Joya Chatterji (2007)
has shown in her book dealing with the post-Partition context, led
to retaliatory violence on Muslim minorities in West Bengal by Dalit
refugees who fled East Pakistan. Bandyopadhyay thus argued that,
‘caste mattered less in Bengal at this juncture’ (Bandyopadhyay 2009:
460). This decline in the importance of caste was also apparent in the
formal domain of politics. In his broad overview, ‘the discourse of
class, alongside the discourses of nation and religion, displaced that of
caste at this historical juncture, marking the onset of freedom and Par-
tition’ (Bandyopadhyay 2009: 463).5 From the 1950s onwards, Dalits
were drawn to the rising communist tide over the course of the refugee
movement, which, Bandyopadhyay argues, was able to channel their
class-based grievances and retain their electoral loyalty.
A new postscript in the second edition of Bandyopadhyay’s (2011)
study of the Namasudras in colonial Bengal sheds more light on the
dynamics of caste in postcolonial West Bengal. In particular, he asserts
that there was no one story of how Dalits negotiated the Indian Repub-
lic: Jogendranath Mandal’s trajectory was apparently but one story
among others. To this end, he stresses the recent rise in importance
of the Namasudra Matua sect in West Bengal politics and its novel
engagement with both contemporary power blocs, and, furthermore,
suggests that theirs is a story of ‘Dalit agency and empowerment’, in
contrast to what he sees as the anti-Hinduism and minoritarianism of
Dalit politics elsewhere in India. Even as he stresses the multiplicity of
experiences, his interpretations have consistently underlined integra-
tion, ideological hegemony, appropriation, absorption, and the Dalit
de-emphasising of caste in favour of idioms palatable to ‘modern tropes
of social justice deployed by left-liberal ideologies’. Although there are
key overlaps in factors they emphasise, where Bandyopadhyay sees in
developments with the Matua Mahasangha the ‘return of an organ-
ised Dalit voice in the postcolonial politics of West Bengal’, Partha
5 One is not entirely sure how discourses, rather than the actors deploying them, can
effect displacements.
106
Chatterjee claims it is ‘unclear that this represents a political resur-

gence of caste in Bengal politics’ (Bandyopadhyay 2011: 272; Chat-
terjee 2012: 70). The latter does not believe that the historic change in
political order in 2011 will result in any related transformation.
Chatterjee’s own explanations for the ‘absence of caste articula-
tion of organised political demands’ include the social-structural, with
emphasis on the complete dominance and preponderance of the upper
castes in the capital Calcutta in particular, largely cut off from sub-
stantial ties to the land (Chatterjee 1998: 81). His point is that their
overwhelming dominance in the city, which, in any case, exerted a
disproportionate influence over the rest of the province in a sense,
consigned them to this role. While he claimed that ‘caste was never
a factor’ among the provincial leadership of the Bengal Congress on
the eve of 1947, he nonetheless noted that a persistent observation
about modern Bengali society and polity ‘strangely enough, has con-
cerned the phenomenon of upper-caste domination’ (Chatterjee 1998:
69). Particularly after Partition, however, ‘when battle lines have been
drawn, the upper-caste intelligentsia were to be found in leading roles
in every contending party – the ruling party and the party of the oppo-
sition, parties of status quo and parties of change’ (Chatterjee 1998:
81). Despite such a situation, he identified the continued salience of
caste not with the upper-caste political classes but within the ‘political
consciousness of the people’ – what he called ‘peasant consciousness’
(Chatterjee 1998: 84–85). While Chatterjee did not quite elaborate
in great detail on how upper-caste domination came into being in the
immediate post-Partition context, much less be maintained thereafter,
and appeared to suggest that the upper castes somehow could not
help but be dominant by virtue of his structuralist interpretation, more
recently, he added that to view the phenomenon as a conspiracy would
be to understand a complex process in a ‘very simple manner’ (Chat-
topadhyay 2010: 105).6 Instead, he pointed to a range of other factors
to explain the decline of caste: the disruptive effect of Partition itself
on the Dalit movement, the absence of a single dominant caste, the
waning of caste consciousness through social mobility, and the Dalit
consent to the upper-caste communists’ leadership. As he put it, ‘is
there any imperative that one undertake an autonomous caste-based
movement? Why should one desire that in connection with progres-
sive politics in any case?’ (Chattopadhyay 2010: 104) In his view, the
6 All translations from Chattopadhyay’s (2010) Bengali original are my own.
107
misrecognition of caste stemmed from a sincerely held proposition:

‘That West Bengal’s intellectuals and political leaders deeply believed
that caste inequality was a superstition of the middle-ages and class
equality was the future course of human history, it is not difficult to
conceive’ (Chattopadhyay 2010: 105). One of the implications of this
view could certainly be that upper-caste domination was, therefore,
the unintended consequence of otherwise noble intentions. There was
nothing especially sinister in the disavowal of caste.
Taken together, the broad narrative at hand indeed goes a long way
towards explaining the apparent subsiding of caste consciousness in
postcolonial West Bengal. Yet, the story appears to suggest that the
upper castes themselves did not quite actively seek out their suprem-
acy over West Bengal. Due to the entire series of factors recounted
above, the analysis remains, in a sense, agreed that Dalits effectively
consented, despite unresolved contradictions, to an unintended upper-
caste rule, itself the consequence of multiple circumstances. Upper-caste
domination and the decline of Dalit political assertion came about
through a series of contingencies, resolutions and structural accidents.
By implication, one of the constituent explanatory factors for these
changes becomes the Dalits themselves. Like the Indians who purport-
edly enabled an absent-minded British colonialism, they also became
collaborators in upper-caste domination by virtue of their consent,
complicity and apparent lack of resistance.
Alongside, even in tension with interpretations that stress the
mutual consent between Dalits and Congress or communist politi-
cal formations – of which, lest I am misunderstood, there is indeed
evidence – is the making and reproduction of upper-caste domination
as a deliberate and intentional process, unable to rely predominantly
on the element of persuasion. My consideration of reservations poli-
cies and contemporary public discourse about upper-caste domina-
tion and political possibilities for its overcoming suggests a far greater
degree of forethought and coercion than received wisdom allows.
Upper-caste domination required the exercise of political will. Fur-
thermore, we will see in subsequent sections that contemporary West
Bengal furnishes evidence not of hegemony, but of aspirations, how-
soever constrained and groping, towards its overthrow. Indeed, one is
at a loss to understand how relations that have encompassed the base
exploitation of labour, sexual, corporeal, and psychological violence,
segregation, humiliation, inferiorisation, atrocity, massacre, multiple
forced relocations and dispersals from territorial jurisdiction, electoral
and census fraud, coercion, bribery, and the concerted reproduction
108
of caste inequalities can be favourably compared to the social and his-

torical context of bourgeois aspirations in Risorgimento Italy.
Reservations in West Bengal

The history of reservation policy in West Bengal is a subject that,
surprisingly, has not garnered extensive comment in connection with
discussions seeking to understand the quandaries of the caste ques-
tion. Barring scanty official data, we have little examination of how
and why the upper castes so strenuously refused to give full effect to
constitutional fiat. It could be helpful at this juncture to remind our-
selves of the social composition of West Bengal. True, there is no accu-
rate data for the number of upper castes and OBCs – due to upper-caste
political refusal in the 1930s and 1980s to enumerate these communi-
ties – but one can hazard a fairly well-educated guess with the data at
public disposal. According to the 2001 Census, the population of the
state was 80,176,197, of which Scheduled Castes were 18,452,555
(23.02 per cent) and Scheduled Tribes were 4,406,794 (5.5 per cent).
Muslims were 20,240,543 (25.25 per cent), OBCs were ‘estimated’ to
be 39 per cent, and other religious minorities were 1,775,924, or
about 2.2 per cent.7 If these figures are roughly accurate, no more than
the remaining approximately 5 per cent of the total population is com-
posed of the upper castes.8
Surely one of the reasons the history of reservation policy should
take on a special significance, given the considerations of this chap-
ter, was the exceptionality of the Dalit relationship to the state. B. R.
Ambedkar once explained the singularity of Scheduled Castes’ entry
into the public services, for instance, as one of ‘life and death’, not
only because of the impossibility of securing career employment in the
domains of trade and industry but because it is ‘only in Government
service that they can find a career’ (Ambedkar 1942). In addition, we
might recall Congress’s long-standing claim, which made the robust
fulfilment of reservations policies contingent on the accomplishment
of national freedom. This deferral took on another layer of meaning
in Bengal, where Dalit agreement to Partition rested on the promise
7 Data retrieved from http://www.anagrasarkalyan.gov.in/htm/state_data.html#sdc; and

http://censusindia.gov.in/Census_Data_2001/Census_data_finder/C_Series/Population_
by_religious_communities.htm (accessed 11 December 2013).
8 See also Lama-Rewal (2009).
109
of social equality (Sen 2012b: 356). Finally, in my view, any discus-

sion of caste domination in West Bengal must engage with the history
of reservations in this state because the presumed implementation of
these measures as a matter of constitutional entitlement formed a cru-
cial possibility for further constituting politics. How does one explain,
for instance, why reservations have seen far greater success in other
Indian states than in West Bengal? What is clear is that from the very
first days of the republic, upper-caste political elites have consistently
expressed their refusal of the exceptional provisions of differentiated
citizenship.
During the debate on the draft Constitution among members of the
West Bengal Legislative Assembly in 1948, for instance, the leading
lights of the Congress unanimously and baldly assailed the provisions
being considered for the Scheduled Castes in particular. Prafulla Chan-
dra Ghosh, the first chief minister of West Bengal declared:
The sooner this reservation goes the better. This reservation

stands in the way of a democratic procedure in the country
and it fosters the idea that we are separate. So long as this sys-
tem of reservation will continue, there can be no one common
country for us. (Cries of ‘hear’, ‘hear’) Therefore the sooner
it goes the better. If all our friends agree, we should make a
recommendation that there should be no reservation either
for my Muhammadan friends or even for the Scheduled Caste
friends. My friends will excuse me if I call it a vested interest.
(Government of West Bengal 1948: 21)
Ghosh concluded: ‘We are all free now, and there should be no
reservation’ (Government of West Bengal 1948: 22). Many others
joined in the chorus; the removal of Scheduled Caste privileges was
an ‘absolutely necessity’. At the dawn of postcolonial constitutional-
ity then – the horizon to which the Congress gestured in response to
Dalit leaders’ grievances and demands during the late colonial years –
Indian nationalists once again professed their reluctance to concede
their logic. Nothing they saw in their society justified the arrange-
ments proposed by the draft Constitution.
As anyone who may have investigated will know, the fairly steady
stream of petitions over the decades since independence for the redress
of grievances about the non-implementation of reservations quotas and
related procedures issuing not only from politically unaffiliated activ-
ist organisations, from Dalit MLAs of various political parties, as well
110
as innumerable prospective students and employees, speaks to the ‘for

all intents and purposes’ abrogation of this dimension of the law. The
data collated by the Commissioner for Scheduled Castes and Scheduled
Tribes and other centrally appointed agencies testifies to the ubiquity
of the caste bias in the governance of West Bengal, and the widespread
subversion of the rationalities required for reservations to have mean-
ingful consequence. Despite the abatement of caste consciousness in
communal violence, the refugee movement and the camaraderie of
the communist upsurge, neither Congress nor Communist Party rule
meant that these policies – the very laws for which Dalit movements in
colonial India and Bengal struggled – were observed. This is hardly to
suggest that the substantial developments and transformations which
have taken place among numerically and political dominant Scheduled
Castes, such as the Namasudras, for instance, are of no significance,
even if they must be assessed alongside similar indices for other com-
munities for them to acquire any meaningful social significance.9 Yet,
neither can there be any doubt that the systematic withholding of con-
stitutional entitlement stemmed from subjective certitude.10
For it is possible to trace a sustained pattern of especially poor per-
formance on this score, especially in comparison to other Indian states.
In 1974, several decades into Congress rule, a committee appointed
by the West Bengal government found that most municipalities in the
state had diverted both central and state funds intended for ‘Harijan
sweepers’ towards other expenses. Only 243 of the intended 600 liv-
ing quarters had been constructed over the last decade (The Times of
India 1974). A couple of years later, the same committee noted that the
reservations policy for the admission of Scheduled Caste students to
undergraduate institutions had ‘utterly failed to achieve its objectives
and no attempt has been made to discover the reasons’ (The Times of
India 1976). The government circular stipulating the reservation of
15 per cent of seats for Scheduled Caste students in undergraduate
institutions ‘seems to have been taken as a routine one by the heads of
the institutions concerned without any effort to appreciate the bigger
9 This, again, is not quite possible, due to the serious lack of socio-economic and
quantitative data about the upper castes.
10 Anecdotally at least, the strategies of refusal have included the forging of surnames
and caste certificates, the sustained resistance to repeated appeals for observing the
enumerative rationalities required for even the nominal success of reservations,
invoking the discourses of meritocracy and suitability, indifference, non-compliance,
humiliation, and violence.
111
social considerations which prompted this concession to be extended

to the underprivileged class’. A few years after the onset of communist
government, the report of the Commissioner for the Scheduled Castes
and Scheduled Tribes revealed that West Bengal consistently ranked
among the states recording the lowest percentages of prescribed quo-
tas in all classes of central government cadres. In terms of state-level
performance, West Bengal was placed ‘unsatisfactory’ (The Times of
India 1980). In late 1982, when the University Grants Commission
requested different universities to set up a special cell to strengthen the
implementation machinery for the planning, evaluation and monitor-
ing of reservations for Scheduled Caste and Scheduled Tribe students,
the CPI(M) and Left Front-dominated Calcutta University syndicate
managed to stall the process over the next five years (Lahiri 1988:
1836). Using data from 1994, a study published in the Indian Journal
of Economics examining the relative positions of SCs and STs vis-à-
vis other social groups in nine selected states, employing some stan-
dard measures of poverty and income inequality, showed that West
Bengal had the second-most-severe incidence of poverty after Orissa
(Adhikary and Mazumder 2010). In 2002, the initial findings of a
University Grants Commission-sponsored study on social mobility
patterns of the Scheduled Castes in West Bengal indicated ‘very little’
progress in connection with the state and central governments’ wel-
fare programmes (The Statesman 2002). Towards the end of the com-
munist innings in 2009, the state government failed to provide figures
to the National Commission for Scheduled Caste employees (Indian
Express 2009). Surely, such evidence cannot be chalked up to solely
the benign intentions of nationalist and Marxist progressivism.
Consider how Upendranath Biswas, the current Backward Classes
minister appointed under the new dispensation, summed up what he
inherited from the outgoing Left Front government over six decades
since the passage of laws regarding reservations: a backlog of 136,000
cases for the issuance of caste certificates, non-observance of reserva-
tions policy in almost all government departments, non-functioning
and inefficient programmes in his own department, and a research
wing with 55 vacant posts, a vehicle junked through disuse, its sign-
board on the ground (Backward Classes Welfare Department, Govern-
ment of West Bengal 2011). Responding to a question in an interview
with Outlook magazine, he explained:
The dominance of upper castes is so intense in Bengal that low

castes don’t dare to even launch an agitation against them – their
112
dominance is accepted as divine dispensation. The situation is

worse than in Bihar. Bengal has not produced a Jagjivan Ram,
Ramvilas Paswan, Nitish Kumar, Lalu Prasad or Mayawati.
Upper castes – Brahmins, Kayasths and Baidyas – comprise
hardly 20 percent of the population but are ruling over
80 percent. Bollywood is open to all classes and castes. But
is Tollywood? When it comes to hypocrisy, nobody can beat
the Bengali bhadralok. Have you seen the matrimonial ads
in Ananda Bazar? What is the caste composition of Calcutta
Club? What about the performing arts or journalism?
(Abdi and Biswas 2012)
Implicit in Biswas’s response is the allusion to domination wrung

from coercion; Dalits do not even ‘dare’ launch agitation. Accord-
ing to his reasoning, if West Bengal has not produced a major Dalit
figure, it is because of the upper castes. In reply to the perennial ques-
tion about why all chief ministers of West Bengal have been upper
caste: ‘Social discrimination has prevented others from becoming CM.
Equal opportunity is not available to everyone. Ours is a closed and
non-inclusive society’ (Abdi and Biswas 2012). The critique extends
equally, he adds, to his own party. These are the views of the most
prominent Scheduled Caste minister in the current government. Upper-
caste domination has hinged, among other reasons, on the sustained
violation of constitutional provisions.
Contemporary Dalit discourse

How do Dalits themselves, then, explain and understand this phenom-
enon? There is, of course, any number of ways to go about answering
this question. Here, I will briefly consider the views and writings of
Sukriti Ranjan Biswas and Manoranjan Byapari.11 Biswas is a retired
political activist and Byapari has made the most significant contribution
to Dalit autobiography in West Bengal in recent memory. Undoubtedly,
there are a great many differences between them, but taken together, I
believe they represent a meaningful range of opinions that jointly offer
an insight into the problem with which this chapter wrestles.
11 This is not an entirely arbitrary choice. While both are public figures in their own
right who work in different fields, they are also individuals whom, as with Dr Nazrul
Islam (see below), I met during the course of my research.
113
Before proceeding, I wish to point out that while electoral results

have often been taken as conclusive evidence of Dalit consent to Con-
gress and Communist government, there is, in addition, a fairly expan-
sive domain of various kinds of activism, which is yet to be considered
in any serious detail with respect to questions of hegemony and con-
sent. One might include the principles and activities of organisations
throughout the state, such as the B. R. Ambedkar Mission, All India
Backward and Minority Communities Employee Federation (BAMCEF)
(West Bengal), the Democratic Action Forum of Dalits, Women and
Minorities (DAFODWAM), the Barisal Namasudra Seba Samiti
(West Bengal), All-India Udbastu Unnayanshil Samiti, the Bangla
Dalit Sahitya Sanstha, or the Udbastu Kalyan Sangha; the considerable
number of Scheduled Caste and Scheduled Tribe employee councils in
multiple state institutions; or the work of publishing houses such as
Caturthha Duniya (Fourth World) and the growing number of little
magazines and independent publications that have documented the lit-
erature of Dalit West Bengal, within this ambit.12 These organisations
comprise an entire world of the Dalit public in West Bengal, whose
existence has hardly garnered wide scholarly attention. Biswas and
Byapari offer but two openings into this seemingly subterranean milieu.
Sukriti Ranjan Biswas was born in Jessore district in East Pakistan
in 1955, and like many in the Namasudra community, migrated to
West Bengal in the wake of the dislocations wrought by Partition on
both sides of the border. He initially associated with the CPI(ML),
but broke his connection with that party over ideological differences
about B. R. Ambedkar’s thought and Dalit politics, and gave himself
over to what he calls ‘the Dalit movement of West Bengal’ (pascim-
banger dalit andolan). As is well known, this is disillusionment that
many have experienced in their engagement with various commu-
nist formations elsewhere in India. Since the mid-1990s, he has been
involved with the Republican Party of India, both nationally and at the
state-level, the All India Scheduled Castes Bank Workers’ Federation
and the All India Union Bank Scheduled Caste Workers’ movement.
More recently, he has joined the Bahujan Mukti Party. Biswas has
been active in seeking official intervention in the ongoing difficulties
over citizenship status for many Dalit refugees from East Pakistan and
12 I should note the difficulty in resolving the tenor and substance of informal conversa-
tions with various individuals associated with these organisations with the persuasive-
ness and consent believed to accompany the ideological hegemony of the status quo.
114
Bangladesh, even meeting former prime minister Manmohan Singh as

part of a delegation.13
Many of Biswas’s essays – several of which are speeches – deal
squarely with the problems and prospects for Dalit politics and exis-
tence in West Bengal and India, past and present. In ‘Possibilities for
the Dalit Movement in West Bengal’, Biswas begins with a contrastive
gesture – the relatively large proportion of upper castes in states such
as Uttar Pradesh and Bihar (25 per cent) versus West Bengal (3–5 per
cent).14 For him, the plausibility of domination appears greater in the
former than the latter. What then, explains its strength in West Ben-
gal? In Biswas’s view, ‘in this state the “upper castes” have sustained
their domination through artifice (sukaushale)’ (Biswas 2007: 122).
Where Uttar Pradesh and Bihar have seen the rise of Dalit political
elites, ‘nobody from West Bengal has had the opportunity to do so.
No, from none of the political parties’ (Biswas 2007: 122). The roots
of this inability are economic as well. Whereas the upper castes in
Uttar Pradesh and Bihar are dependent on the land, this is not the case
in West Bengal:
Which is why they can speak of a smattering of land redis-

tribution over here. But they are unwilling to give Dalits any
part in their salaried professions, education, journalism, the
world of letters, brokerages, contracts, businesses, politics,
higher education, etc. Thus the harassment with issuing caste
certificates, the infringement of educational and employment
reservations policies, a backlog of a lakh or so positions, the
unfulfilling of various welfare schemes for the Scheduled are
the ordinary working of governance in this state.
(Biswas 2007: 122)
Biswas goes on to state that Dalits have been ‘deluded by Marxist

propaganda’ and ‘unable to detect this shrewd contrivance’ (Biswas
2007: 122–23). But as difficulties have intensified, ‘the actual form of
13 At a BAMCEF meeting in Nagpur in 2005, Biswas also attributed the apparent lack
of growth in West Bengal’s population according to official census figures (over a
period which saw the mass-migration of predominantly Dalit refugees from East
Pakistan and Bangladesh to India) to the attempt of ‘Brahminical forces’ to snatch
voting rights from these communities. See: https://www.youtube.com/watch?v=
pquXZ_Wr3Ew.
14 The percentages are Biswas’s figures.
115
the conspiracy is becoming evident to ordinary people in many ways’

(Biswas 2007: 123).
It should be clear that according to Biswas, conspiracy does indeed
bear purchase for grasping their present domination in West Bengal.
For him, there is no question about its existence. Does his view that
there has been a conspiracy prove as such? Not at all. But what do we
make of the fact that this is precisely how he chooses to perceive the
matter? I am not sure that one can simply wish the choice away.
Just as relevant, in my view, are Biswas’s thoughts on how to trans-
form the seeming impasse. Like his ideological predecessor, Jogendra-
nath Mandal (who founded the West Bengal branch of the Republican
Party of India), he believes that only a democratically elected broad-
based coalition of Muslim, OBC, SC, and ST political representatives –
the desired yet unfulfilled ‘third-front’ of politics – can genuinely offer
radical change. This is an aspiration shared by many other activists as
well, and Biswas has gone so far as to establish alliances with Siddiqul-
lah Chaudhury’s Peoples’ Democratic Conference of India and the All
India United Democratic Front, West Bengal (Haque 2012). Biswas’s
views and activities do not strike me as reflecting consent to the per-
suasions of Bengali communist ideology. Rather, his writings speak to
the deep-seated discontent and disappointment with the status quo.
Like Biswas, Manoranjan Byapari, the acclaimed author of Itibritte
Chandal Jiban, also believes that only with a ‘genuinely Communist’
political party, which includes representatives of the largest communi-
ties of West Bengal in the key positions of leadership that looks to the
interests of the majority, will any considerable progress occur with
respect to Dalits feeling as though they actually have some say in the
nature of their state, perhaps even agency. Byapari is sceptical, how-
ever, of the mere acquisition of political power and no proponent of a
solely caste-based analysis of society or identitarian fundamentalism.
He is critical of extant Dalit political formations both within West
Bengal and elsewhere in India.
Itibritte Chandal Jiban is nothing short of an event in Bengali liter-
ary circles. Indeed, it is an account that will likely take quite some time
with which to come to terms. Byapari narrates the jagged course of
his existence as a ‘Chandal’ in exceptional and unprecedented detail –
his many lives as a refugee from East Pakistan, cowherd, dishwasher,
tea-stall assistant, cook, prisoner, labourer, munshi for road construc-
tion projects, labourer at a locomotive workshop, godown security,
sweeper-cum-scavenger, bookseller, and rickshaw-puller. His writings,
‘a mirror of the class and caste divide we are still associated with’,
116
come from ‘the anger of my life and the people around me’ (IANS
2013). Byapari presents an overwhelming abundance of experiential
portraiture. I select but two instances below where his narration draws
back from the episode to a more declarative stance.
Like Biswas in yet another respect, Byapari spent his youth growing
up in refugee-camp West Bengal and came to confront the yawning
discrepancies between the treatment upper-caste refugees received from
the Congress government and the distinctly less humane consideration
refugees from Scheduled Caste communities received from the same.
Of particular note was the forced removal of solely Dalit refugees out-
side West Bengal, or most egregiously, the mass killing of the same at
Marichjhanpi. Byapari writes that he cannot solve the equation. He
raises a question, which, in the words of one sympathetic reviewer, ‘will
haunt and ring in the hearts of millions of the underprivileged for long’
(Biswas 2013). He submits: ‘Now the one question is, why did the rul-
ing classes display two different attitudes towards these two groups of
people?’ His reply: ‘Behind this lies that eternal caste-disgust. If those
people were not Namasudras, Pods, Jeles, but Brahmins, Kayasthas
and Baidyas, whatever government it was they could never have com-
mitted such beastly oppression’ (Byapari 2012: 53).
The standard account on refugee rehabilitation grasps the discrep-
ancy in more secular terms. The beleaguered government of West Ben-
gal worked with considerations of suitability and propriety in mind in
the allotment of properties and lands within and without West Bengal
to its various categories of refugee. These mapped squarely onto the
caste divide. For Byapari, and many others besides, this seeming coin-
cidence was not merely accidental, but issued from a socially available
grammar: ‘That eternal caste-disgust’. For him, and as reflected in a
growing body of literature of various genres, there is little identity
between Dalit and upper-caste refugees’ experiences of the crisis.
Byapari addresses the subject of domination in another important
register: the historical efforts to change the caste name from Chandal
to Namasudra in the late 19th and early 20th centuries. He wonders
whether the change in name in 1911 ever amounted to much. As he
asks, ‘after having been freed from the name Chandal, and having
received entrance to the chaturvarna order of Hindu dharma, have
they received their plundered social respect in return? Are they receiv-
ing that humane conduct from the varna-lords?’ (Byapari 2012: 29)
Byapari concedes that perhaps in the course of day-to-day relations in
professional settings, Namasudras may receive some esteem and distinc-
tion due to them by virtue of their qualifications. Yet, he goes on to
117
doubt whether upper castes genuinely extend an equal respect. ‘With

lowered faces, muted laughter, slighting words, don’t they mock? The
experiences of those notable Namasudra figures with whom I’m famil-
iar in this regard are very bitter, of great mental hurt’ (Byapari 2012:
29). Byapari thus feels that the movement to change their name in the
past was of no use:
Not a change in name, what was needed was a movement for

a transformation in the capacity for humanity. Because that
didn’t happen, the Namasudras have remained at that con-
temptible, neglected and undignified level. There is no reason
to believe that their respect has increased by even an inch.
(Byapari 2012: 29–30)
There is an abruptness to this statement from one of the most

renowned Bengali Dalit literary figures alive, which might inform our
assessments about how the nature of upper-caste domination is con-
strued. We often teach that self-consciousness is constituted through
being recognised as self-conscious by the other. Byapari’s conviction
that upper-castes’ regard for Namasudras has not budged ‘even an inch’
speaks to the obstruction in the dynamic between Hegel’s figurations of
lord and bondsman in West Bengal – the transformation in the capac-
ity for humanity, which did not occur due to upper-caste refusal. His
account of the past century or so of their relations is characterised by
the sustained withholding of respect, the seeming upper-caste incapacity
to see self-consciousness in the Dalit. In a recent interview, Byapari feels
that before all else, what is required is a battle – violent if necessary – for
securing what he calls atma-samman (literally, ‘soul-respect’).15
Possibilities for Dalit and Muslim alliance

My closing set of considerations concern the reappraisal of the idea of
cohering political unity among Muslims, SCs, STs, and OBCs in recent
years. I have already indicated how both Biswas and Byapari share
this aspiration. This is not a vision, however, exclusive to Dalits alone.
According to senior IPS officer and Additional Director General of the
West Bengal Police Dr Nazrul Islam, for instance, ‘it is time for
15 Harichand Thakur interview with Manoranjan Byapari: https://www.youtube.com/

watch?v=KVYMo2cW-N8&list=UUBNY7bkgoRIhWYFNPdkymvQ&index=2.
118
Bengal’s SCs, STs, OBCs and Muslims to get together and launch a
political battle. We will then be the king-makers and the chief minister
of the state will be a Muslim, SC or OBC’ (Mazumdar 2013). What is
implied in the desired political vision is the emergence of a political
party, which, in turn, is brought to power by the majority of the state’s
electorate. In theory, the argument goes, this is possible.
Islam is author of a controversial book titled Musolmaner Koroniyo
(literally, ‘What Muslims Should Do’), which embroiled him in diffi-
culties with the present government. Therein, he criticised the attitude
of the present government towards the Muslim community of West
Bengal. Revealingly, in a section of his book titled ‘What Needs to
Be Done’, comes his view that ‘We need to form one society that
will include not just Muslims, but also the low-caste Hindus – those
belonging to the Scheduled Castes, Scheduled Tribes, Other Backward
Classes and others’ (Sharma 2012).
At a symposium organised by the West Bengal Muslim Association
last year, Islam elaborated his rationale as follows: ‘Not only Muslims,
not one Schedule Caste and Schedule Tribe sits on any of the important
post in West Bengal. Only 2 per cent Brahmins in West Bengal have
occupied all the important ministries since Independence till now.’ He
offered the example of his native village, where ‘political leadership of
the area whether CPI(M), Congress, Trinamul Congress, RSP, SUCI all
are non-Muslim. But the fact is that whenever there is a clash between
groups it is always Muslim or SC/ST who get killed’. Islam underscores:
‘If we can sit together, we can launch a new political party for Muslim
along with Schedule Caste and Schedule Tribes [sic]’ (Haque 2012).
There is a deep irony about this return of the idea of Dalit and
Muslim alliance to the present conjuncture in West Bengal, given how
often the notion has been scrutinised for its implausibility, indeed, fail-
ure. Even if a major development is yet to emerge in this regard, there
is evidence of a growing number of meetings discussing this very pos-
sibility. I think it is meaningful that some version of this political aspi-
ration is shared by individuals whose own arenas of work are quite
distinct: Biswas, Byapari, and Islam. For that convergence indicates
the awareness that upper-caste domination can ultimately only be met
with a democratic and political response. With them, there seems little
doubt that it is precisely because of the upper castes that their com-
munities are insufficient and unequal partners in the various domains
of public life. There is no ambivalence on this score.
It should be of some significance then, that the upper-caste leader-
ship of both political parties recently and unanimously delegitimised
119
any premeditation on their part. In response to Rezzak Mollah’s and

Kanti Biswas’s public criticisms of systematic exclusions in August
2013, they have also ‘denied any conspiracy of exclusion’ (Bhattacha-
ryya and Rana 2013: 13). Mollah and Biswas were two of the most
prominent Muslim and Scheduled Caste leaders within the CPI(M).
The former caused a public relations storm in early August 2013
because he stated, quite frankly: ‘The Dalits and minorities are giving
their lives for the party while the Brahmins and Kayasthas are boss-
ing over.’ Kanti Biswas’s view: ‘It is unfortunate that since the days of
the undivided Communist Party to today’s CPI(M), nobody from the
scheduled communities could find a place in the state secretariat.’
In a telephone conversation with the authors of an article analys-
ing recent panchayat elections, Mollah added: ‘Have I said anything
wrong? Have you seen the reaction (from various political parties)?
Mouchake dhil poreche (a stone has struck the beehive)! All the upper-
caste leaders are furious!’ And then comes his final assessment:
You see, West Bengal has been a secular state because of its
working class, and Muslims form a majority of that. But the
upper-caste leaders who pretend to be progressive, secular, etc,
have been serving the interests of their own people – not of the
working class. That is the real problem.
(Bhattacharyya and Rana 2013: 13)
At a symposium titled ‘Implementation of Article 30 of Constitu-

tion of India for Educational uplift of Muslim Minority’ in mid-2012
at Sishubikash College of Education in Chakberia-Makrampur, South
24-Parganas, Mollah expressed his intentions of forming a party
like the Krishak Praja Party of Abul Kasem Fazlul Huq, comprised
of Muslims, SCs, STs, and OBCs. He will celebrate his birthday, he
added, along with Jogendranath Mandal (Haque 2012).
In West Bengal, we have a situation where some of the senior-most
Dalit and Muslim political representatives of both parties are expos-
ing the persistent misalignment between community and capital,
indeed, the hollowness of upper-caste justifications for their continued
supremacy. Whether these sharpening contradictions lead to genuine
transformations in the political dynamics of the state remain to be
seen. Whatever its past configurations, however, I am not sure of how
meaningful it is to speak in terms of hegemony, when the emperor so
plainly wears no clothes.16
120
Conclusion
An image I recently came across on Facebook features all the (upper-
caste) chief ministers in West Bengal’s history; the title reads: ‘Non-
Bengali rulers in Bengal’. The associated comments offer an instructive
window into socially mediated understandings of the present. They
underscore the unthinkability of a Dalit chief minister in a state that is
home, as per the 2001 Census, to the second largest number of Sched-
uled Castes in any Indian state and third highest as per a percentage of
its population, yet has never seen a significant number of Dalit minis-
ters assume control of any of the major departments or positions in
government.
This is not to suggest that there is no fragmentation among Dalits
themselves or to posit a monolith. Many who have benefitted from
the political process are perceived as opportunistic and self-serv-
ing, unconcerned with Dalit interests per se. While anxieties over
the articulation of politics, factionalism, internal critique, and the
search for alternative sources and networks of leadership have
remained, there is little sense in Dalit public opinion that upper-
caste domination or the present political, social, and economic
conditions of their communities are merely the consequence of a
cascade of circumstance.
What has happened in West Bengal is a process roughly analogous to
what the sociologist Eduardo Bonilla-Silva (2009) ironically termed ‘rac-
ism without racists’ – the reproduction of racial inequalities in the United
States despite the denial of racist attitudes; in West Bengal, the persis-
tence of caste-based inequalities despite the disavowal of casteism:
casteism without casteists. And like the ‘sincere fictions’ and ‘colour-
blind racism’ Bonilla-Silva analyses, in our case, the powerful expla-
nations Indian nationalists and communists marshalled for explaining
inequalities effectively exculpated the upper castes from any responsi-
bility for the historical present. It would appear these arguments have
begun to lose their appeal.
16 Since the time this essay was initially written, the emergence of the Samajik Nyayabi-
char Mancha, an organisation that Rezzak Mollah hopes will give West Bengal its
first Dalit chief minister, would seem to confirm the growing lack of dissidence to
upper-caste-led politics in the state.
121
Bibliography
Abdi, S. N. M. and U. N. Biswas. 2012. ‘Bengal Hasn’t Produced a Jagjivan
Ram or Even a Mayawati’. Outlook India, 10 August, http://www.outloo-
kindia.com/article.aspx?281957 (accessed 11 December 2013).
Adhikary, Maniklal and Ritwik Mazumder. 2010. ‘Income Inequality and
Poverty among Scheduled Castes and Scheduled Tribes in Selected States’,
Indian Journal of Economics, 90(3): 997–1010.
Ambedkar, B. R. 1942. Grievances of the Scheduled Castes. http://www.ambedkar.
org/ambcd/44B.%20Grievances%20of%20the%20Scheduled%20Castes.
htm (accessed 11 December 2013).
Backward Classes Welfare Department. ‘Population Distribution Data’, http://
www.anagrasarkalyan.gov.in/htm/state_data.html#sdc (accessed 11 Decem-
ber 2013).
Backward Classes Welfare Department, Government of West Bengal. 2011.
Annual Administrative Report, 2010–2011, http://www.anagrasarkalyan.
gov.in/pdf/bcwd-annual-report-10–11.pdf (accessed 11 December 2013).
Bandyopadhyay, Sekhar. 2009. ‘Partition and the Ruptures in Dalit Identity
Politics in Bengal’, Asian Studies Review, 33: 455–67.
Bandyopadhyay, Sekhar. 2011. Caste, Protest and Identity: The Namasudras
of Colonial Bengal, 1872–1947 (2nd edition). New Delhi: Oxford Univer-
sity Press.
Bhattacharyya, Dwaipayan and Kumar Rana. 2013. ‘West Bengal Panchayat
Elections: What Does It Mean for the Left?’ Economic and Political Weekly,
47(37): 10–13.
Biswas, A. K. 2013. ‘Memoirs of a Chandal Jeevan: An Underdog’s Story’,
Mainstream, 51(17), http://www.mainstreamweekly.net/article4116.html
(accessed 11 December 2013).
Biswas, Sukritiranjan. 2007. Dalit Mukti Andolan: Ripablikan Barta Samka-
lan. Kolkata: Caturthha Duniya.
Bonilla-Silva, Eduardo. 2009. Racism without Rracists: Color-Blind Racism
and the Persistence of Racial Inequality in the United States. Lanham: Row-
man and Littlefield.
Byapari, Manoranjan. 2012. Itibritte Chandal Jiban. Kolkata: Priyashilpa
Prakashan.
Chatterjee, Partha. 1998. The Present History of West Bengal: Essays in Politi-
cal Criticism. Delhi: Oxford University Press.
122
Chatterji, Joya. 2007. The Spoils of Partition: Bengal and India, 1947–1967.
Chattopadhyay, Partha. 2010. ‘Pascimbanglay jati o janajati’, Baromas (sara-
diya), 101–07.
Deshpande, Satish. 2013. ‘Caste and Castelessness: Towards a Biography
of the “General Category”’, Economic and Political Weekly, 47(15):
32–39.
Government of West Bengal. 1948. Extract from the Assembly Proceedings,
Official Report, West Bengal Legislative Assembly, Debate on the Draft
Constitution. Alipore: West Bengal Government Press.
Haque, Zaidul. 2012. ‘ADG Criticizes Brahmins’ Hold on West Bengal
Politics’. TwoCircles.net, 27 June, http://twocircles.net/2012jun27/adg_
criticizes_brahmins%E2%80%99_hold_west_bengal_politics.html
Haque, Zaidul. 2012. ‘AIUDF Gains Strength in West Bengal with Merger
of State RPI and INL’. TwoCircles.net, 22 June, http://twocircles.net/node/
291098 (accessed 11 December 2013).
Haque, Zaidul. 2012. ‘CPI-M Leader to Launch Party for Muslims, SC/STs
and OBCs in West Bengal’. TwoCircles.net, 19 June, http://twocircles.net/
node/290729 (accessed 11 December 2013).
Haque, Zaidul. 2012. ‘Siddiqullah’s Political Party Merged with the AIUDF
in West Bengal’. TwoCircles.net, 29 May, http://twocircles.net/node/288545
IANS. 2013. ‘A Rickshaw-Puller’s Journey from Prison to Books’. The Times
of India, 7 November, http://articles.timesofindia.indiatimes.com/2013–
11–07/people/43773251_1_rickshaw-mahasweta-devi-prison (accessed 16
January 2014).
Indian Express. 2009. ‘Plan Panel Pulls up Bengal on SC, ST Welfare’, 26
November.
Lahiri, Souparna. 1988. ‘Discrimination by Default’, Economic and Political
Weekly, 23(36): 1836–37.
Mazumdar, Jaideep. 2013. ‘The Caste Club’. The Times of India, 3 February,
http://articles.timesofindia.indiatimes.com/2013–02–03/special-report/
36721094_1_upper-castes-brahmins-obcs (accessed 11 December 2013).
Rana, Santosh and Kumar Rana. 2009. Paschimbange Dalit o Adibasi. Kolkata:
Camp.
Sen, Dwaipayan. 2012a. ‘The Emergence and Decline of Dalit Politics in Ben-
gal: Jogendranath Mandal, the Scheduled Castes Federation, and Partition,
1932–1968’, unpublished Ph. D. dissertation, University of Chicago.
123
Sen, Dwaipayan. 2012b. ‘ “No Matter How, Jogendranath Had to Be Defeated”:

The Scheduled Castes Federation, and the Making of Partition in Bengal,
1945–1947’, Indian Economic Social History Review, 49(3): 321–64.
Sharma, Anuradha. 2012. ‘Sharp Criticism for Mamata Banerjee in Muslim
Officer’s Book’. New York Times, 5 September, http://india.blogs.nytimes.
com/2012/09/05/sharp-criticism-for-mamata-banerjee-in-muslim-officers-
book/ (accessed 11 December 2013).
The Statesman. 2002. ‘Still Children of a Lesser God’, 26 August.
The Times of India. 1974. ‘Funds for Harijans in Bengal “Diverted” ’, 4 November.
The Times of India. 1976. ‘Bengal College Seat Quota for Harijans Flops’,
19 October.
The Times of India. 1980. ‘Harijan Representation in Government Cadres
Poor’, 17 March.
124
6
THE POLITICS OF CASTE
AND CLASS IN SINGUR’S
ANTI-LAND ACQUISITION
STRUGGLE
Kenneth Bo Nielsen
Caste has long since acquired the status of a ‘foundational category’ in

anthropological thought (Mathur 2000: 97). As one of the discipline’s
most important gate-keeping concepts (Appadurai 1986), it has his-
torically constituted perhaps the most widely used lens through which
scholars have viewed social life in India. As several of the contribu-
tions to this volume point out, however, West Bengal is often por-
trayed as an exception when one speaks about caste in India – and
particularly, if the question of caste is tied to a discussion of politics.
As Partha Chatterjee wrote in his important 1982 essay on ‘Caste and
Politics in West Bengal’:
On several aspects of contemporary politics, West Bengal appears

to fall outside the general all-India pattern. Ask an informed
observer whether caste has any great influence on present-day
West Bengal politics and he is likely to reply: ‘Very little’.
(Chatterjee 1997: 69)
Chatterjee’s imaginary informed observer is likely to find evi-

dence for his claim in the success of the Left parties in West Bengal
in mobilising support for their programmes across the lines of caste
and community, thus precluding the formation of linkages between
caste and the structure of organised politics (Chatterjee 1997: 84).
Unlike in many other Indian states, the political parties that are
dominant in West Bengal today do not have clearly identifiable caste
bases: most castes and communities are represented in most of the
political parties (Lama-Rewal 2009: 63). Nor are there parties of
125
KENNETH BO NIELSEN
any significance that explicitly, and exclusively, champion the cause

of particular castes or caste conglomerates. Lastly, the prolonged
hegemony of the Left, whose leaders often proclaim ‘the irrelevance
of caste in the struggles of the downtrodden’ may well, as Sarbani
Bandyopadhay (2012: 73) has argued, have made the language of
caste illegitimate in political discourse. This is reflected in, as Lama-
Rewal has recently noted, the continuing widespread acceptance
of the proposition that ‘caste is irrelevant in West Bengal politics’
(Lama-Rewal 2009: 363).
While some have taken Chatterjee’s 1982 essay as evidence for the
‘truly unique’ (Sinharay 2012: 26) nature of the caste–politics nexus
in West Bengal, the keyword to reading Chatterjee’s quote above is
the little word, ‘appears’. While Chatterjee does indeed analyse the
absence of caste articulation in organised politics, the concluding
pages of the essay reflect on the continued importance of the catego-
ries of caste in ‘the apparently uninstitutionalized world of what may
be called politics among the people’ (Chatterjee 1997: 83). Over the
past decades, a growing body of ethnographically informed literature
on politics in rural West Bengal has examined precisely this political
world and has amply demonstrated the continued salience of caste,
thus effectively exploding what Dayabati Roy (2012: 948) calls the
‘long-held political myth’ about its irrelevance. Indeed, a critical eth-
nographic approach to the issue of the politics of caste is, as Uday
Chandra and I have argued elsewhere (Chandra and Nielsen 2012),
likely to expose caste as a social fact that continues to shape local
relations of power and influence. As discussed in the introduction to
this volume, Arild Ruud’s (2003) work on the rural responses to com-
munist mobilisation in the 1950s and 1960s, for instance, brings to
light the significance of caste and caste stereotypes in producing con-
siderable variations in the way people perceived and responded to the
message of the Marxists. Dayabati Roy’s (2012; 2014) contemporary
village ethnography documents the prevalence of caste hierarchies in
village society at large, including within political parties. And so does
Mukulika Banerjee’s (2010) study of the political work of a CPI(M)
comrade in Birbhum.
This chapter seeks to contribute to this growing body of ethnographic
literature on the significance of caste in rural West Bengal by zooming in
on the role of caste in one of the most talked-about rural movements in
the state in recent years – namely, that in Singur. The Singur movement
erupted in the summer of 2006 in response to the Left Front (LF) gov-
ernment’s plan to acquire 997 acres of agricultural land in the area to
126
POLITICS OF CASTE AND CLASS IN SINGUR
make way for a new car factory, operated by Tata Motors. In late 2008,
the movement – supported by the Trinamul Congress (TMC), the then
leading opposition party in the state – succeeded in driving Tata Motors
out of Singur. While much has been written on the Singur movement,
the role of caste in it has, as I have argued elsewhere (Nielsen, forth-
coming), remained a neglected issue. While the scant attention generally
accorded to the politics of caste in West Bengal may form part of the
explanation for why this should be the case, another reason is that sup-
port for the Singur movement (or lack thereof) on the ground did not
neatly follow caste lines (Roy 2014). Yet, as I demonstrate below, the
everyday politics of caste was indeed operative in shaping both the local
organisation of the movement and the articulation of its agenda. I begin
below by briefly introducing the Singur movement. I then analyse how
the politics of caste that could be seen to operate within it was continu-
ous with how caste and class operated as organising principles in every-
day life. I do so through a detailed ethnography of local caste and class
relations in two of the project-affected villages.
The Singur movement

When the LF government announced its intent to acquire 997 acres of
agricultural land from Singur as part of its overall programme of rap-
idly reindustrialising the state, a section of the land losers – described
by the media as ‘unwilling farmers’ because of their unwillingness to
part with their land in lieu of cash compensation – organised a move-
ment to prevent the land acquisition from going ahead. Under the
aegis of the Singur Krishi Jomi Raksha Committee (SKJRC – The
Committee to Save the Farmland of Singur), the unwilling farmers
took out rallies, organised meetings and public hearings, petitioned
the local administration, and liaised with civil society groups and
political parties as part of their campaign. The SKJRC adopted slo-
gans such as ‘we will give our blood but never our land’ and empha-
sised the unbreakable bond that exists between a farmer and his land,
the emotional and economic value of which could never be compen-
sated in cash term. As Ananya Roy has rightly pointed out, the Singur
movement drew heavily on images of rural West Bengal as Sonar Ban-
gla (Golden Bengal), a land of fields of gold and plenty, invoking the
potent pastoral motif of the owner-cultivator (Roy 2011: 271) living a
harmonious life intimately connected to the land. The SKJRC’s cam-
paigns were, thus, predominantly carried out under the sign of the
farmer, that is, the owner-cultivator.
127
KENNETH BO NIELSEN
Through its campaign, the SKJRC succeeded in drawing the atten-

tion and support of social and human rights activists, intellectuals,
and politicians of many persuasions from across the state and other
parts of India and abroad (Roy 2007: 3326), thereby propelling the
land acquisition in Singur to a matter of national importance. In
December 2006, however, under heavy police cover and amid much
violence, the administration went ahead and acquired and fenced the
land earmarked for the Tata factory; and in early 2007, possession
was formally handed over to Tata Motors. Employees of Tata Motors
performed bhumi puja – a Hindu ceremony performed before com-
mencing the construction of a house or building – on the site, and
the company announced that the factory would be ready and up and
running in just over a year. Yet, as is well known, this never happened.
Throughout the construction period, the unwilling farmers kept up
their protests, which culminated in August 2008, when thousands
of protesters from across the state, led by opposition leader Mamata
Banerjee, descended on Singur, demanding that the acquired land be
immediately returned to its erstwhile owners. In October the same
year, Tata Motors had had enough. Decrying the obtrusive tactics of
the unwilling farmers – and of Mamata Banerjee, in particular – the
company decided to call it quits and headed for greener Gujarati pas-
tures. As this is written, however, little has changed on the ground as
Tata Motors retains the right of use of the land in Singur.
In late 2007, I moved into the village of Shantipara in the heart of
the project-affected area. Having arrived to conduct fieldwork for
my doctoral dissertation, I had managed to rent a room in a spacious
two-storey, nicely painted concrete house owned by one of the most
important local SKJRC organisers, Prasanta Das. Welcoming me in
an air-conditioned upstairs bedroom, Prasanta informed me that my
choice of basing myself in Shantipara was a good one: ‘In our vil-
lage you will find that almost everybody is against the land acquisi-
tion’, Prasanta said. We talked a while about the adverse impact of
the land acquisition on local livelihoods in Shantipara, and Prasanta
explained that many households were now finding it increasingly
difficult to make ends meet – especially those who had no sources
of off-farm income. Later the same day, Prasanta invited me to come
with him to the nearby village of Nadipara: ‘It may be interesting
for you’, he argued, ‘the inhabitants of Nadipara are all support-
ers of our movement too!’ A few hours later, I was seated on the
back of Prasanta’s motorbike. The drive to Nadipara took no more
than a couple of minutes, but the contrast between it and Shantipara
128
was stark. The first house we passed driving into Nadipara was
low, made of mud and clay, a mix of straw, tin and tiles combining
into a sagging roof. The crumbling foundation of the house, which
appeared to have been damaged during the monsoon, was reinforced
with sand-filled plastic bags. The next couple of houses we passed
were no different, and although Prasanta’s house in Shantipara was
among the larger ones, the norm in Shantipara was bricks, concrete,
and tiles – not mud, clay, and tin, as in Nadipara. As a local organ-
iser of the SKJRC, Prasanta had gone to Nadipara to lead a gram
baithak (an informal village meeting) to inform the residents about
recent developments of importance to their common struggle against
the land acquisition. Prasanta and I were offered plastic chairs to sit
on during the gram baithak, while the residents squatted or sat on a
large tarpaulin spread out on the ground. If Shantipara radiated rela-
tive affluence, Nadipara definitely radiated relative deprivation, the
crowd which had assembled to listen to Prasanta composed of short
and frail bodies marked by a life of hard labour.
During the course of my fieldwork, I would visit Nadipara often.
The village was inhabited almost exclusively by people who made
their living as khet majur, that is, as agricultural labourers without
land to their name, who gain their livelihood primarily from work-
ing on other people’s land (see Thorner 1991: 265). Even though
it is often the case that the real wages and material well-being of
agricultural labourers increase significantly when new employment
opportunities open up in the non-agricultural sector (Lindberg 2012:
67) – for instance, in the guise of a new car factory – the large major-
ity of Nadipara’s khet majur had, from the outset, rallied behind the
SKJRC to oppose the land acquisition in Singur. With few exceptions,
the khet majur belonged to the SC Bauri caste. During my interviews
with the Bauri, my interviewees would often stress how important
agriculture was for them – work as an agricultural labourer was
widely available, they said – at least on a sufficient scale for them
to avoid living in abject poverty. That was why they rallied behind
the SKJRC’s demands for the return of the acquired land to its for-
mer owners. At the same time, however, the Bauri would repeat-
edly grumble about how the local SKJRC leadership, none of whom
resided in Nadipara, paid only scant attention to the increasing dif-
ficulties the Bauri faced on a day-to-day basis. During one particular
gram baithak held in early 2008, this latent dissatisfaction suddenly,
and somewhat unexpectedly, erupted. Ajay, an industrious and vocal
Bauri in his twenties, had complained that nobody in the SKJRC’s
129
KENNETH BO NIELSEN
leadership cared about the plight of Nadipara’s khet majur. The

whole anti-land acquisition mobilisation, he had alleged, was domi-
nated and led by local chasjomi malik (land owners) and its agenda
was shaped by their desire to retain their agricultural land in the face
of expropriation. Although they owned hardly any land, Nadipara’s
khet majur had supported the landowners in this struggle because
they wanted to fight for their right to make a living through tilling the
land. But so far, they had gotten nothing in return, and their concerns
were hardly ever raised by the SKJRC in public fora. What Ajay, in
effect, appeared to be saying was that the concerns of the khet majur
Bauri and the chasjomi malik, or simply chasi, who predominantly
belonged to the Mahishya caste of owner-cultivators, differed signifi-
cantly: should the SKJRC ultimately fail in securing the return of the
acquired land to its erstwhile owners, the landowners would still be
legally entitled to financial compensation from the government. But
the khet majur, who were nowhere officially recognised as part of the
project-affected population eligible for compensation, would be left
jobless and empty-handed.
By the time the meeting ended, Ajay had repeatedly been assured by
those SKJRC leaders who had attended the meeting, which included
the SKJRC President as well as Prasanta, that this was all a miscon-
ception on his part. The movement, the president explained, was
genuinely concerned about the plight of the khet majur, and would
continue to fight for their rights (adhikar), no matter what the out-
come of the movement would be. But as most of those present could
hear from the quarrel he was presently having with Prasanta, Ajay
remained unconvinced. Prasanta explained that in a movement such
as this, everybody should reconcile their differences and unite like
brothers; but Ajay retorted that this was easier said than done as long
as the khet majur and their concerns were not given adequate atten-
tion. Prasanta replied that surely the khet majur had a stake in this
struggle too since, if the land was returned to its original chasi own-
ers, the khet majur would once again have their old jobs back. Ajay
said that Prasanta was, of course, right – but should this mean that all
other concerns, ‘our concerns’, as he put it, should never be raised at
all? The SKJRC president eventually intervened to briefly explain that
the matter would be resolved soon, and that this disagreement should
not escalate any further. Prasanta thus abandoned his quarrel with
Ajay and drove off into the night on his motorbike to attend another
gram baithak while he angrily mumbled ‘khet majur lok!!!’ (Labour-
ing folks) just loud enough for Ajay to hear.
130
This latent conflict between the labouring Bauri and the land-owning
Mahishya surfaced regularly as the two groups struggled to construct
and manage a common political platform from which to challenge
the land acquisition that had deprived the latter of their property and
the former of their livelihood. When I later spoke to Ajay about the
reasons for his discontent with the chasi leadership, he said that all
they offered was ‘talk’. The leaders only came to Nadipara when they
wanted the Bauri to participate in an event or a meeting – apart from
that, they showed little concern about how poverty and distress was
on the rise in Nadipara. Reflecting on his own role in the Singur move-
ment so far, Ajay testified that he had mostly found himself marching
in rallies or charging the police with a lathi in his hand until somebody
in the leadership had told him to fall back. He found it particularly
insulting that he was made to march in rallies and shout slogans like
‘we will never give up our land!’ – as a landless labourer, Ajay had had
no land to give in the first place. The most pressing concern for Ajay
and the other Bauri khet majur was rather the acute lack of alternative
forms of employment that he badly needed to make ends meet while
waiting for the long-term outcome of the Singur movement. He also
feared that the chasi leaders would eventually cave in, stop their move-
ment, and claim the compensation, leaving the khet majur – who were
not entitled to any compensation from the government – high and dry
and left to fend for themselves. In sum, the stakes for the khet majur
were both different, and in an important sense, also higher than they
were for the chasi. And this was not properly reflected in the cam-
paigns, demands, and slogans coined by the SKJRC leadership, which,
as I indicated above, were first and foremost carried out under the sign
of ‘the farmer’.
As I have detailed at length elsewhere (Nielsen, forthcoming), Ajay’s
sporadic attempts at mobilising the Bauri behind a more explicit khet
majur agenda within the SKJRC were most often met with a mix of
disapproval, scorn, and intimidation by the chasi leaders, who accused
him of undermining the ‘brotherly unity’ of the chasi and khet majur.
To his already extensive list of woes, Ajay could have added even more.
For instance, the Bauri hardly ever attended the central committee
meetings of the SKJRC, during which future plans and strategies were
arrived at. Such meetings were routinely attended by representatives
from urban civil society, and thus, offered important opportunities for
networking and establishing new supra-local alliances. To a consider-
able extent, the Bauri thus found themselves effectively excluded from
important decision-making fora.
131
KENNETH BO NIELSEN
The partial exclusion of the Bauri, the perceived lack of atten-

tion on the part of the Mahishya to the everyday challenge of mak-
ing ends meet in Nadipara and the unwillingness of the Mahishya
SKJRC leaders to actively champion the cause of the landless as part
of their struggle to reclaim the land are indicative of how the poli-
tics of caste and class imbued the Singur movement with an internal
core–periphery structure (see Nilsen 2010: 160) that saw the chasi-
Mahishya lead, organise, and represent, and the khet majur Bauri
follow. In this regard, it is important to note that although Ajay’s
argument with the SKJRC leadership was articulated primarily in
class rather than caste idioms, using vernacular terms such as khet
majur, chasi, and chasjomi malik, which denote a particular relation-
ship to property and a means of production, caste and class relations,
in fact, formed ‘a reciprocal nexus’ (Madan 2002: 11) at the village
level.1 This caste–class overlap was such that the politics of caste can
only be analysed as one co-constituent of multiple and overlapping
inequalities operative both within and beyond the anti-land acqui-
sition movement. The following sections map out how these over-
lapping hierarchies, and the core–periphery structure they produced,
were continuous with how the politics of caste operated in everyday
life. To this end, I offer a description of the two villages of Shantipara
and Nadipara, based on fieldwork data and two village surveys that I
conducted in late 2008 and early 2009.
Caste and class in everyday life:

the chasi and the khet majur
Shantipara had, at the time I conducted my survey, a total of 158
households with a population of 838. In total, 127 households who
had lost land to the Tata project had joined the anti-land acquisition
movement, while 21 households had collected the compensation. The
remaining 10 households had not been directly affected by the land
acquisition, either because they were already landless or because their
land fell outside the area acquired. 148 households belonged to the
Mahishya caste. Seven households were made up of members of the
Chakrabarty family of Brahmans, while three belonged to the SC Bagdi
caste.
1 This is often, although not exclusively the case in rural India (Thorner 1991; Shah et
al. 2006; Herring and Agarwala 2008).
132
Literacy among the chasi inhabitants of Shantipara stood well above

the state average at nearly 90 per cent, slightly lower among women
than among the men. Newspapers were widely read, and the village
club had a small library of books in Bengali. Most families reported
that they considered their life to have been good and improving, in
material terms, in the years preceding the land acquisition. Although
the distinctive and highly diversified agricultural region in which Sin-
gur is located has a very long history of commercialised agriculture
(Kelly 1981), most Mahishya families told me that it was only with
the transformations brought about by the modernisation and diver-
sification of agriculture over the past three or four decades that their
quality of life had improved dramatically. Agricultural transformation
was, in turn, routinely narrated to me as the key driver of a range
of other desirable local social transformations that had come about
within the span of just a few generations. As agricultural techniques
and input had developed from traditional to modern, and cultivation
gone from monocrop to multicrop, the Mahishya of Shantipara had
gone from ‘walking barefoot’ to ‘wearing shoes’; from living in mud
huts to building concrete houses; from dreaming of one day owning
a bicycle to actually driving a two-wheeler; from crowding around a
battery-driven transistor radio to watching Bollywood films on colour
TV; and from being illiterate to pursuing higher education in the town
or city.
Nearly all the families who owned land had been self-sufficient
in basic foodstuffs and had generated a marketable surplus. While
monthly incomes are notoriously difficult to assess in a semi-
agricultural and semi-informal economy, and underreporting in
surveys is common, a surveyed household (comprised by an aver-
age of 5.3 persons) had an average monthly income of just over
INR 4,100. Just over 85 per cent of the houses were of bricks,
with roofs of concrete or tiles. On average, a house in Shantipara
would have approximately three rooms. Nearly two-thirds of the
households had Below Poverty Line (BPL) cards, but nearly half
of the households also reported having some savings. Given that
the survey was conducted more than two years after the de facto
land acquisition, and that many families had had to live off their
savings during this interval, it is likely that both the prevalence and
size of savings would have been even greater prior to 2006. The
presence of so many BPL cards among the residents of Purbopara
may therefore equally be a reflection of the village’s relatively good
political connections with the local gram panchayat, rather than of
133
KENNETH BO NIELSEN
widespread poverty.2 And a range of consumer goods were common:

nearly all households possessed one or more bicycles; 25 per cent
owned a two-wheeler; and four families owned a car. Barring a few
exceptions, all homes were equipped with electric fans, and 68 per
cent with TV sets. Fridges, gas stoves, and computers, however, were
rare, but not entirely absent.
Apart from the 10 families who had always been landless, which,
tellingly, included all three Bagdi households, all other families in Shan-
tipara had owned land prior to the land acquisition. Several Mahishya
families could tell of a past where their forefathers had owned many
acres. But average land holdings at the time of the land acquisition in
2006 had declined to only 2.39 bigha (less than one acre) – a reflection
of the steadily escalating fragmentation of land holdings in rural Ben-
gal. In addition to owning land, just over one-fourth of the households
had, before the land acquisition, entered into sharecropping arrange-
ments with other landowners.
The small average size of landholdings, however, was not indicative
of the Mahishya being tied to a life in marginal cultivation. Many
families had, in fact, been able to successfully diversify out of agri-
culture, so that by 2009, as many as 72 per cent of all households
were ‘pluri-active and straddled the agricultural and non-agricultural
sector’ (Lindberg 2012). The range of non-farm sources of income
found in Shantipara included inter alia gem-polishing and making or
selling jewellery, carpentry, and factory work, but also various kinds of
salaried office work and small-scale business and trading. In addition,
17 of the surveyed households had a family member in government
service. Hence, even though the proportion of households having non-
agricultural sources of income had increased somewhat after the land
acquisition as several families had responded to the loss of agricultural
land by seeking to raise incomes elsewhere (Ghatak et al. 2012: 28),
many families had already – prior to 2006 – sought a deliberate diver-
sification of livelihood strategies.
Compared to Shantipara, Nadipara was both a more recent and con-
siderably smaller settlement. One of my Mahishya informants said that
the Bauri of Nadipara were the descendants of a group of agricultural
labourers who had been brought to Singur from the district of Bankura
2 Pattenden (2005: 1982) similarly notes from south India how caste, influence, party
political loyalty, and bribery were important factors that determined who was issued
ration cards by the local gram panchayat.
134
many years ago to work on the field of a local large Kayastha land-
owner, but I was unable to confirm his story. At the time of the survey,
Nadipara consisted of 74 households with a population of 292. Three
families were of the Bagdi caste, while the remaining were Bauri.
Literacy among the khet majur of Nadipara stood abysmally low
at 35 per cent, and only two (male) adults out of a population of
292 had passed class 10. Education was, in fact, available locally as
Nadipara did have its own school, set up by the Belur Ramkrishna
Mission; but most children were withdrawn after a few years, as soon
as they were old enough to work on the fields (Sinha 2007: 1). Prior
to the land acquisition, only eight households (out of 74) had owned
any land, and four of them had been rendered landless by the land
acquisition. The remaining four landowning families owned a mere
1.1 bigha (one-third of an acre) between them, while all other families
were and remained landless. The main livelihood in Nadipara was
instead derived either from sharecropping or from agricultural labour.
Prior to 2006, 34 households (46 per cent) had sharecropped plots
of land of a size ranging from one bigha to three acres, while sev-
eral other households would lease land, either for a full year or for
one agricultural season.3 Income from cultivation would be supple-
mented with fishing and collecting snails, both of which could be used
for personal consumption or sold at the local bazaar. But apart from
that, the diversification of local income sources had been very limited.
Some industrious young men, including Ajay, had purchased cycle
vans, which they used for transporting crops and fertiliser from field
to house or vice versa, or for transporting bricks and cement to per-
sons who were building or expanding their houses. Using the cycle van
for transporting building material to Mahishya families in, for exam-
ple, Shantipara had, in fact, been good business in the recent past.
Many prosperous Mahishya families in these villages had, over time,
enlarged their dwellings to such an extent that they had encroached
considerably on the already narrow lane that connected Shantipara to
the larger road running through the local bazaar. This encroachment
3 Over the past few decades, leasing land has become increasingly common. This
arrangement of farming is different from the sharecropping system because the entire
exchange process is monetised; the duration of the agreement is limited; and the land-
owner is entitled to a fixed amount as rent rather than a proportional share of the
harvest (Sarkar 2012). Lessee cultivation may thus be more risky (but also more
profitable) and is divested of any moral content that may characterise ‘traditional’
sharecropping arrangements.
135
KENNETH BO NIELSEN
had meant that lorries of a certain size were no longer able to drive
into Shantipara. They, therefore, had to park near the bazaar, from
where Ajay and other cycle van owners would transport the bricks
and cement to its rightful owner. Another Bauri who had successfully
diversified his livelihood strategies was Dulai Patra, a man in his late
thirties, who still, even after the land acquisition, retained a small plot
of land that he cultivated as a bargadar. He had invested some of his
modest savings in setting up a tea shop near one of the factory gates,
where he sold tea, biscuits, cigarettes, biri, khoini (chewing tobacco),
sweets, and other snacks, mostly to the guards and the police stationed
within the factory site. A few khet majur had also on and off been
hired to do construction work, but as is often the case in rural India,
the ability to successfully diversify out of agriculture had proven very
class-dependent (see Bardhan, cited in Sharma 2009: 360), and the
many social transformations that had played out in Shantipara over
the past several decades had progressed only haltingly in Nadipara, if
at all. The result was that life and labour in Nadipara remained closely
tied to the agricultural economy.
Nearly all of the khet majur I surveyed in Nadipara appeared to
have been content with their situation in life prior to 2006. Agricul-
tural work, they said, was readily available, and their labour was often
in demand. But even then, there were several lean seasons in the agri-
cultural calendar. During such times, the khet majur would resort to
borrowing from local moneylenders who charged very high rates of
interest, or they would pawn their valuables (Sinha 2007: 2–3). The
relative poverty of Nadipara compared to Shantipara was similarly evi-
dent in the physical appearance of the village space itself. In Nadipara,
85 per cent of the uniformly single-storey houses (with one exception)
were of mud, and while nearly three-quarters had tile roof, asbestos,
and straw were also common. Sixty out of seventy-four houses con-
sisted of just one room. The village was only electrified around 2002
as one of the last villages in the area. And overall, 75 per cent of all
households reported a monthly income of less than INR 1,000. In this
regard, it is important to note that while the average reported income
in Shantipara was thus upwards of four times as high, one-third of
the households in Shantipara did, in fact, report a monthly income of
INR 1,000 or less. This supports Lindberg’s (2012: 65–66) argument,
based on research in rural Tamil Nadu, that the difference between
SC and non-SC populations is not primarily identifiable at the level of
average household income, but rather, in terms of landownership and
household assets (see Table 6.1).
136
Table 6.1 Frequency of household assets per 100 households, calculated as

the total number of an asset relative to the total number of house-
holds. Figures over 100 indicate that some households possessed
more than one of the items in question.
ASSET Nadipara Shantipara
Bicycle 44.6 164.01

2-wheeler 4.0 28.17
Car 0 2.82
TV 27.0 76.76
Radio 14.86 36.62
Fan 48.65 212.68
AC 0 1.41
Fridge 2.70 6.34
Gas stove 1.35 21.13
Computer 1.35 3.52
Mobile 20.27 92.25
Music system 6.76 9.86
Source: Survey by the author.
While the distinction between the chasi and the majur is thus evi-
dently economic and material, these two terms also capture many of
the symbolic forms of rural distinction that would go uncovered by
an exclusive focus on patterns of landownership and income (Ray
1983: 309). Arild Ruud (1999: 270) describes the desired way of
life among the chasi as modelled on the Bengali bhadralok, the edu-
cated, respectable, and genteel upper classes – and often, Hindu upper
castes – for whom manual labour is anathema (Mukherjee 1991). This
desired chasi way of life – the ‘chasi model’, as Ruud calls it – therefore
revolves inter alia around notions of frugality, moderation, hard work,
cleanliness in dress and manners, dedication to the land, a concern
with harnessing female sexuality, and abstaining from excesses such
as drinking, womanising, brawling, and slandering (Ruud 2003: 126–
27; see also, Bandyopadhyay 2004). The attainment of chasi status
for the Mahishya was achieved through considerable upward caste
mobility, which had entailed the appropriation of both bhadralok
and upper-caste norms and values, which often overlapped (Bandyo-
padhyay 2004). The Mahishya caste was originally formed by the
Chasi Kaibartta, the cultivating section of the lowly ranked Kaibartta
caste, which also included fishermen, known as Jele Kaibartta. The
Chasi Kaibartta adopted the name Mahishya to escape the stigma
attached to the parent Kaibartta caste and to distinguish themselves
137
KENNETH BO NIELSEN
from their former caste compatriots, the fishermen. From the late 18th
century onwards, entrepreneurial and prosperous Mahishya families
had spearheaded an assertive caste movement that aspired for higher
social status. Mahishyas actively sought to promote and gain recogni-
tion for a shared sense of Mahishya identity (Bandyopadhyay 2004),
and successfully established themselves as a ritually clean Sudra or
Jalacharaniya caste, with a reputation for being thrifty, frugal and
industrious (Sanyal 1988: 355). In the process, the Mahishya – mainly
in the south-western part of undivided Bengal – had formed their own
society, the Bangiya Mahishya Samiti, which brought out its own jour-
nal, the Mahishya Samaj, through which a shared sense of Mahishya
identity was promoted. The spread of education among Mahishyas
was similarly encouraged through the Mahishya Siksha Bistar Bhan-
dar (fund to promote education among the Mahishya) (Sanyal 1988:
365).4 Because of the emphasis on education, respectability, sexual
morality, and restraint, a chasi may often look down upon the majur
as inferior chhotolok, or ‘small people’ – as mere uncultured labourers
found lacking in most of the elements that combine to constitute the
desired chasi way of life.
Thorner’s (1991) observation from the 1970s that, in Bengal, the
chasi were drawn primarily from the cultivating or artisan castes,
while the majur belonged primarily to the scheduled castes, fits
remarkably well with contemporary Shantipara and Nadipara. As a
formerly untouchable caste, the caste status of the Bauri was natu-
rally very different from that of the Mahishya. In his study of the
Bauri in the neighbouring state of Orissa, Freeman describes how
a Bauri untouchable would not speak to a high-caste person until
spoken to (cited in Guha 1999: 47). A Bauri would also crouch
‘so that one hand touched the ground’ with his face ‘toward the
ground’ whenever he passed by higher caste people (Guha 1999: 56).
Similarly, in his literary portrayal of the Kahar Bauri, a sub-caste
of the Bauri, penned by the Bengali novelist Tarashankar Banerjee
in 1948, the Kahar Bauri appear as the very embodiment of every-
thing that is considered immoral and excessive by the chasi castes.
The Kahar Bauri had few eating inhibitions, were overwhelmingly
4 In certain parts of rural Bengal, the Mahishya used their ‘remarkable experience of
caste mobilisation’ and the social and political authority this had generated to move
into positions of influence in the Congress and the Indian movement for indepen-
dence from the 1920s (Chatterjee 1997: 73–77).
138
illiterate (99.23 per cent of them, according to the 1931 Census),

practised extensive commensality, drank heavily, had defiling occu-
pations, lived in a world of superstition and ignorance, and wor-
shipped gods who ranked low in the divine hierarchy (Ray 1983:
303–4). Since Banerjee wrote thus of the Kahar Bauri, some formerly
untouchable castes in West Bengal have, particularly over the past
several decades of communist-led LF rule, reformed their lifestyles
towards the respectable chasi-cum-bhadralok model by, for instance,
abstaining from drinking, brawling, and womanising (Ruud 1999).
But particular caste and class histories and stereotypes of the Bauri
as ignorant, excessively emotional, and undependable still circulated
among the Mahishya.
Because of the overlapping class and caste hierarchies that charac-
terised the relationship between chasi and majur, the former, in many
ways, considered the latter as inferior, although they would rarely say
so explicitly in public. When I spoke to the Mahishya about caste, they
would usually say that it did not matter much today. Some of my elder
Mahishya informants recalled how, in the past, untouchable labour-
ers would have their pay dropped into their palm by the landlord so
as to avoid any polluting physical contact. But that had ceased long
ago, they said. Nonetheless, they sometimes referred to Nadipara as
Kulipara, a somewhat derogatory term meaning ‘the coolies’ neigh-
bourhood’. And although they would hire people from Nadipara to
perform agricultural labour for them during the peak planting or har-
vesting seasons when extra labour input was required, they described
them as phakibad, that is, lazy shirkers, who – unlike the industrious
Mahishya – did not care much for hard and dedicated work.
That the Bauri were lazy and ignorant drunkards also explained
why, from the point of view of the Mahishya, the Bauri would remain
stuck at the bottom of the social hierarchy and would never success-
fully aspire to move up. Gopal Koley (67) from Shantipara said:
It is simple. If you are landless, whatever you earn in a day

you will have spent by midnight. You work for a day and have
70 rupees in your hand; but before you have reached your
house, the money will be gone because you have to purchase
everything you need: Food, clothes, everything. If you also
have to spend money on drink and biri there will be nothing
left. It will all be gone by nightfall. How will you make any
progress in your life? To move up [prosper] you must be able
to save and invest in land and in education for your children.
139
KENNETH BO NIELSEN
With land in your possession things are different. You work

hard; you produce your own food, sell some on the market
and get cash to spend. You can save, improve your house,
educate your children in a good way. If you don’t have that
base, don’t aspire to have it, you can never move up.
Apart from emphasising the importance of landownership and hard

work, Gopal Koley evidently also alluded to the fact that Nadipara had
seen a good deal of heavy drinking in the past. When I spoke about this
to Ajay’s wife, she rolled her eyes and exclaimed ‘bap-re-bap!’ (Oh my
God! – literally ‘father of father’) – she had been shocked at the amount
of heavy drinking going on all around her when she had moved to
Nadipara several years ago. This fact was readily acknowledged
among the Bauri, and although alcohol consumption had declined con-
siderably in recent decades, it remained a part of life in Nadipara. The
Mahishya of Shantipara, in contrast, would rarely consume alcohol,
and certainly not in public. Differences in educational status and an
unequal distribution of the knowledge of how to make wise financial
decisions, to save and invest in the future, were similarly highlighted as
important differences between the Bauri and the Mahishya.
Differences in caste status between the two villages manifested
themselves in other ways too. As stated, several Brahman families
lived in Shantipara and provided ritual services, both in temples and
the individual homes. No Brahmans lived in Nadipara, which was
instead provided with ritual services from a family of patita (fallen/
degraded) Brahmans. The Mahishya, by and large, kept only cleaner
animals such as cows and ducks, but not goats and chicken, which
defecated to a larger degree in both private and public spaces. Goats
and chicken were, in contrast, found in abundance in Nadipara. Most
houses in Shantipara had proper toilets, whereas the Bauri would def-
ecate in the adjacent fields. And whereas nearly three out of four Bauri
women worked as agricultural labourers, the nexus of women and
menial physical work among the Mahishya was, in accordance with
their superior caste status, generally allowed to exist only under very
difficult financial circumstances.
Patronage, hierarchy, and organisation

Although the Bauri and the Mahishya thus lived in very different
worlds, they did not live in separate worlds. Rather, the overlapping
hierarchies described above extended well into the material and
140
political world to tie the Bauri and the Mahishya together in complex
relations of patronage and dependency. Not only did Prasanta and
other land-owning Mahishya from Shantipara sometimes hire the
Bauri from Nadipara to work on their fields during the peak planting
or harvesting season, they also acted as small-scale patrons and lead-
ers in the local party political arena. Many of the local Mahishya
SKJRC leaders were leading figures in the local TMC – some held
seats on the gram panchayat or the zilla parishad; others led the
(largely inactive) TMC youth wing or peasant front, while others
again exercised their leadership in a more informal way. But regardless
of the formal or informal nature of their leadership, the Mahishya
were seen as both figures of authority and sources of patronage from
the point of view of the Bauri. Because of its poor and marginalised
position, Nadipara had historically had only weak and fragile rela-
tions to the local gram panchayat through which considerable devel-
opment funds and state resources flow. For the past decade or so,
however, Nadipara had constituted itself as the vote bank of the TMC,
voting almost – although not entirely – en bloc for the TMC at the
time of elections. When I interviewed the Bauri, they were all in agree-
ment that to get by in life as a poor, landless agricultural labourer, one
‘needed a party’. It could be any party, they added – the TMC, the
CPI(M), the Congress – as long as it was an influential and generous
party, attentive to the needs of the poor. Because political parties dom-
inate the rural socio-political scene and almost the entire field of trans-
actions and mediations in rural Bengal (Bhattacharya 2011), poor,
landless agricultural labourers such as the Bauri needed to cultivate
and maintain whatever links they were capable of establishing in order
to tap into the local flow of resources and patronage.
The chasi leadership of the SKJRC also knew well that they were
expected to act as patrons of the khet majur in return for their support
for the anti-land acquisition movement and the TMC. During the gram
baithak held by the SKJRC in Nadipara, one would often hear local
chasi SKJRC-cum-TMC leaders warn the khet majur that they should
be aware that the CPI(M) would come to them with ‘bribes and offers’.
‘Do not be tempted by them!’ the Bauri were instructed. They were also
reminded of how neither the TMC nor the SKJRC had ever collected
chanda (subscription/donation) from Nadipara – a practice for which
the CPI(M) is well-known. In other words, the chasi leaders had done
their duty as good patrons by giving to, rather than taking from, their
Bauri clients. When I interviewed chasi leaders about the situation in
Nadipara, they would often say that there was a constant pressure on
141
KENNETH BO NIELSEN
the chasi SKJRC organisers from Shantipara to keep the khet majur
content. They knew that, otherwise, the CPI(M) would move in and
promote their twin agenda of dividing the common anti-land acquisi-
tion platform and luring the khet majur away from the TMC:
The people of Nadipara are poor people. They are not edu-
cated. If you give them something and tell them something
they will believe whatever you say and they will think, ‘OK,
we will go with you’. That is why we always have to be con-
cerned with Nadipara and see to it that the people there do
not grow unhappy. If that happens, the CPI(M) can move in
and give them something, and they can join with the CPI(M)
like that.
From the point of view of the chasi leaders, the purpose of a gram
baithak was to bring news from the SKJRC to the khet majur, and to
ensure that the khet majur knew about future rallies and were able to
attend in good numbers. Although such meetings were held in Nadip-
ara, and sometimes in the village club, non-local chasi leaders such as
Prasanta would regularly appear as the host or moderator, welcoming
both the audience and the speakers to the meeting. As I discovered
during my first visit to Nadipara, Prasanta (and I) would be offered
plastic chairs to sit on while the Bauri would squat on the ground.
Although indicative of the due deference and respect a client shows
towards his superior or patron, Prasanta often complained about this
in private. According to Prasanta, he and other chasi SKJRC leaders
had to shoulder the responsibility of ‘leading Nadipara’ because the
Bauri had produced ‘no local leader or people to organise them’ – had
they had a proper local leader, he implied, he would not have had to
spend his time travelling to and from Nadipara as often as he did.
Prasanta also complained that it was he who had to pay from his own
pocket for the tea and biscuits that would customarily be served to vis-
iting VIPs whenever such came to Nadipara, partly because it was seen
as his obligation as a patron and SKJRC leader, and partly because
nobody from Nadipara could really afford to.
The flow of resources from SKJRC organisers to the khet majur also
included a share in the emergency relief – food, blankets, clothes, medi-
cine, and so on – that poured into Singur from non-local sympathisers.
Most of it arrived through TMC’s party network and was channelled
through the SKJRC organisers in Shantipara, who therefore controlled
the local distribution of it. In Shantipara, the local SKJRC organisers
142
claimed that they always made a point out of first distributing the
emergency relief among the impoverished Bauri of Nadipara – only
then would they search for eligible and needy households in other vil-
lages. However, the khet majur saw things in a different light, claiming
that they mostly received what was left after the Mahishya had taken
what they wanted. A particular sources of discontent was, moreover,
the dishonourable manner in which Bauri women were made to line
up under the scorching sun ‘like beggars’, as Ajay put it, as they waited
for the relief to be distributed.
While Mahishya leaders such as Prasanta, thus, saw the Bauri as
incapable of organising themselves, and as easily tempted and lured
by more or less empty promises or immediate rewards, without the
intervention of outside responsible leaders like himself to take charge,
the Mahishya also considered the Bauri difficult to control and easy
to anger. During an interview with a chasi leader from Shantipara, my
interviewee claimed that the Bauri – as well as the Bagdi who were
numerically dominant in other project affected areas, and who also
supported the SKJRC – were ‘very militant’: unless the SKJRC leader-
ship had instructed them to exercise restraint, the Bauri would have
charged the police head-on without hesitation or regard for their own
safety on any number of occasions:
We have to explain to them that attacking the police cannot

be the strategy here. It could in Nandigram, but not here. Any
movement must have a method, and random attacks on the
boundary wall cannot be your method. Confronting the police
or the guards head-on would only result in much violence on
us. That is why we instruct them to hold back.
Such forms of representation draw on caste stereotypes of the Bauri

as lacking in restraint, insight, and ‘method’, and as people whose
passions are easily roused and who lose their cool. By implication,
this would not only justify using the Bauri mostly as movement foot-
soldiers; it would also serve to legitimise the almost paternal leader-
ship of the more restrained, calm, controlled, and rational Mahishya.
Conclusion
The case presented here has sought to bring out how the everyday
politics of caste and class in Singur successfully segued into the anti-
land acquisition politics of Singur’s unwilling farmers. As Ajay’s
143
KENNETH BO NIELSEN
discontent described earlier indicates, overlapping hierarchies rooted

in caste, class, and patronage between the Mahishya and the Bauri
combined to create a situation in which the Mahishya provided local
leadership, planned rallies and meetings, and liaised with important
non-local supporters, whereas the Bauri mostly received rather than
produced information, heeded rather than made decisions, marched as
foot-soldiers and not as leaders, and only rarely liaised with non-local
constituents. This ability of local caste hierarchies to permeate and
shape the organisational structure of grassroots movements in India is
well known from other settings too. Thus, Nilsen (2010: 160) describes
from the Narmada Bachao Andolan how caste relations in the locality
produced a local core–periphery structure that effectively excluded
Dalits from important spheres of influence. Similarly, Pattenden
(2005) writes of how the Karnataka Rajya Raitha Sangha’s (KRRS)
rural mobilisation in Karnataka against the local state and merchants –
which managed to provide ‘temporary cover’ for local differences of
caste and class – eventually ended up reproducing the very relations of
inequality it promised to supersede.
Much like the many Indian new farmers’ movements that mush-
roomed in the 1980s (see Brass 1995), the Singur movement can be
seen to have publicly projected an image of brotherly rural unity –
unfractured by caste and class hierarchies – threatened with displace-
ment and destruction by an exploitative nexus of industrial and state
interests working in tandem. Even if caste and class hierarchies were
thus successfully airbrushed from the Singur movement’s public politics
of representation, the present ethnographically grounded examination
of the micro-politics of its organisation and claim-making reveals that
the fact that the movement took place ‘under the sign of the farmer’
was not incidental. Rather, it reflected the dominance within the move-
ment of the chasi, and more specifically, of the Mahishya.
Bibliography
Appadurai, Arjun. 1986. ‘Theory in Anthropology: Center and Periphery’,
Comparative Studies in Society and History, 28(2): 356–61.
Bandyopadhyay, Sekhar. 2004. Culture and Hegemony: Social Dominance in
Colonial Bengal. New Delhi: Sage Publications.
Banerjee, Mukulika. 2010. ‘Leadership and Political Work’, in Pamela Price
and Arild Engelsen Ruud (eds), Power and Influence in India: Bosses, Lords
and Captains, pp. 20–43. New Delhi: Routledge.
144
Bhattacharyya, Dwaipayan. 2011. ‘Party Society, Its Consolidation and Crisis:

Understanding Political Change in West Bengal’, in Anjan Ghosh, Tapati
Guha-Thakurta, and Janaki Nair (eds), Theorizing the Present: Essays for
Partha Chatterjee, pp. 226–50. New Delhi: Oxford University Press.
Brass, Tom (ed). 1995. New Farmers’ Movements in India. Essex: Frank Cass.
cal Criticism. New Delhi: Oxford University Press.
Ghatak, Maitreesh, Sandip Mitra, Dilip Mukherjee, and Anusha Nath. 2012.
‘Land Acquisition and Compensation in Singur: What Really Happened?’
Working paper. Boston: Institute for Economic Development, Boston Uni-
versity.
Guha, Ranajit. 1999. Elementary Aspects of Peasant Insurgency in Colonial
India. Durham: Duke University Press.
Herring, Ronald J. and Rina Agarwala. 2008. ‘Introduction: Restoring Agency
to Class: Puzzles from South Asia’, in Ronald J. Herring and Rina Agarwala
(eds), Whatever Happened to Class? Reflections from South Asia, pp. 1–23.
Delhi: Daanish Books.
Kelly, Kenneth. 1981. ‘Agricultural Change in Hooghly, 1850–1910’, Annals
of the Association of American Geographers, 71(2): 237–52.
Lindberg, Staffan. 2012. ‘Rural India 1970–2005: An Arduous Transition to
What?’ The Indian Journal of Labour Economics, 55(1): 61–75.
Madan, Vandana. 2002. ‘Introduction’, in Vandana Madan (ed), The Village
in India, pp. 1–30. New Delhi: Oxford University Press.
Mathur, Saloni. 2000. ‘History and Anthropology in South Asia: Rethinking
the Archive’, Annual Review of Anthropology, 29: 89–106.
Mukherjee, S. N. 1991. ‘Bhadraloks of Bengal’, in Dipankar Gupta (ed), Social
Stratification, pp. 176–82. New Delhi: Oxford University Press.
Nielsen, Kenneth Bo. Forthcoming. ‘Managing “Communities” of Resistance:
Negotiating Caste and Class in an Anti-Land Acquisition Movement in
West Bengal’, in Uday Chandra and (eds), Staking Claims: The Politics of
Social Movements in Contemporary Rural India. Oxford: Oxford Univer-
sity Press.
Nilsen, Alf Gunvald. 2010. Dispossession and Resistance in India: The River
and the Rage. London: Routledge.
Pattenden, Jonathan. 2005. ‘Trickle-Down Solidarity, Globalisation and
Dynamics of Social Transformation in a South Indian Village’, Economic
Ray, Rajat K. 1983. ‘The Rural World of Tarashankar Banerjee: Social Divi-
sions and Psychological Cross-Currents’, in Peter Robb (ed), Rural India:
145
KENNETH BO NIELSEN
Land, Power and Society under British Rule, pp. 274–312. London: Curzon
Press.
Roy, Ananya. 2011. ‘The Blockade of a World-Class City: Dialectical Images
of Indian Urbanism’, in Ananya Roy and Aihwa Ong (eds), Worlding Cities:
Asian Experiments and the Art of Being Global, pp. 259–78. Basingstoke:
Blackwell.
Roy, Dayabati. 2014. Rural Politics in India: Political Stratification and Gov-
ernance in West Bengal. New Delhi: Cambridge University Press.
Ruud, Arild Engelsen. 1999. ‘From Untouchable to Communist: Wealth,
Power and Status among Supporters of the Communist Party (Marxist) in
Rural West Bengal’, in Ben Rogaly, Barbara Harriss-White, and Sugata Bose
(eds), Sonar Bangla: Agricultural Growth and Agrarian Change in West
Bengal and Bangladesh, pp. 253–78. New Delhi: Sage Publications.
Ruud, Arild Engelsen. 2003. Poetry of Village Politics: The Making of West
Sanyal, Hitesranjan. 1988. ‘Congress in Southwestern Bengal: The Anti-Union
Board Movement in Eastern Medinipur, 1921’, in Richard Sisson and Stan-
ley Wolpert (eds), Congress and Indian Nationalism: The Pre-ndependence
Phase, pp. 352–76. Berkeley: University of California Press.
Sarkar, Swagato. 2012. ‘Between Egalitarianism and Domination: Governing
Difference in a Transitional Society’, Third World Quarterly, 33(4): 673–88.
Shah, Ghanshyam, Harsh Mander, Sukhdeo Thorat, Satish Deshpande, and
Amita Baviskar. 2006. Untouchability in Rural India. New Delhi: Sage Pub-
lications.
Sharma, Shalendra D. 2009. ‘A Political Economy of the Indian State in the
Era of Globalisation: A Review’, South Asia, 32(3): 354–73.
Sinha, Kuver. 2007. Singur: Losses Beyond Compensation. Kolkata: Manthan
Samayiki Publication.
Weekly, 47(34): 26–27.
Thorner, Daniel. 1991. ‘Agrarian Structure’, in Dipankar Gupta (ed), Social
Stratification, pp. 261–70. New Delhi: Oxford University Press.
146
7
BUILDING UP THE
HARICHAND-GURUCHAND
MOVEMENT
The politics of the Matua Mahasangha
Praskanva Sinharay
On 28 December 2010, thousands of Matuas, the followers of a devi-

ant sect – the Matua Dharma – hit the streets of Kolkata with their red
flags held high, dancing to the beats of their traditional instruments:
donkas and kashis. They gathered to attend a huge public meeting
called by their frontal organisation, the Matua Mahasangha, at Espla-
nade in the heart of Kolkata. It was before the 2011 state assembly
elections in West Bengal. The meeting was attended by top-notch rep-
resentatives from all the major political parties, who shared the dais
with Baroma Binapani Devi, the nominal head of the organisation,
and the other prominent leaders. The objective of the meeting was to
demand the repeal of the Citizenship Amendment Act 2003, which
denied citizenship to those refugees who migrated to the Indian side of
the border after 1971, considering them as ‘illegal migrants’. The law
posed a threat to the lives of a large number of Dalit refugees that
included the Matuas, who had migrated from Bangladesh. The mes-
sage of the Mahasangha in the meeting was clear. Since the Matuas
under the leadership of the Mahasangha had already, by then, made
their presence felt as a solid vote-conglomerate in the panchayat polls
in 2008 and the Lok Sabha elections in 2009, the community, which
was now capable of affecting the outcome in about 35 state assembly
constituencies, would extend support only to those candidates in the
forthcoming 2011 elections who would pledge to meet their demands
in the future (The Telegraph, 29 December 2010). As all the politicians
present on the stage sympathised with the demands of the Matuas and
147
P R A S K A N VA S I N H A R AY
extended all-out assurances to meet them, it became evident that the

Matuas had now emerged as a force that no political party in West
Bengal could ignore (The Times of India, 29 December 2010).
The emerging trends in the political landscape of contemporary West
Bengal, thus, compel us to rethink the caste question in the state’s pol-
itics. It had been comforting so far for the urban, upper-caste Bengali
bhadralok to describe the postcolonial politics of the state as truly unique
with regard to the discourses of exclusion and discrimination in terms
of caste, particularly in comparison with the other Indian states. Such
a political standpoint of the bhadralok leadership in West Bengal was
clear from one of the statements by Jyoti Basu made in 1980. Basu said
that ‘viewing the social scene from the casteist angle is no longer relevant
for West Bengal’ (cited in Jaffrelot 2003: 255). However, the situation
has changed in the recent past, and we can now see the emergence of an
organised movement of the Dalits1 in the politics of West Bengal. For
instance in 2009, Mamata Banerjee – then, the leader of the opposition –
commented: ‘I shall work for the Matuas as long as I am alive. I was
moved when baro ma told me how her people were being looked down
upon as most of them belonged to lower castes’ (The Times of India,
6 December 2009). Even the CPI(M)-led Left Front, which once refused
to consider caste as a significant category could no longer maintain
its erstwhile position. In 2010, the Left Front government instituted an
award in the name of Harichand-Guruchand Thakur – the local icons of the
Dalits – and honoured the former head of the Matua Mahasangha for his
contribution to Dalit upliftment. Such recent developments clearly show
us that the Kolkata-based upper-caste politicians of West Bengal now felt
the need to change their way of looking at society and politics, thereby
acknowledging the particular identity of a community. The change in the
attitude of the bhadralok leadership was no one-fine-morning phenom-
enon; rather, the Matuas have asserted their presence both organisation-
ally and politically as a significant factor in the state’s politics in the last
few years. Evidently, it seemed that caste could no longer be shelved as a
hush-hush social category; rather, it has evolved into a determinant factor
in the formal domain of political activities in the state.
The politics of Matua Mahasangha, therefore, I shall argue in this
chapter, has introduced ‘a new politics of caste’ (Sinharay 2012) in
1 The term ‘Dalit’ in this chapter has been used to define the Matuas, constituted almost
entirely by the Namasudras who were once untouchables. The Matuas use the term
Dalit/patit/pichhiye pora manush (Dalit/downtrodden/backward section of the popu-
lation) alternatively to define themselves.
148
BUILDING UP THE HARICHAND–GURUCHAND
contemporary West Bengal – the state where the institution of caste

had otherwise been treated either as an ‘irrelevant’ or a ‘pre-modern’
category since the post-Partition days, both by the Congress and
then by the decades-long ruling Left. Although primarily a religious
organisation that challenges the basic tenets of Hinduism, the Matua
Mahasangha, on the one hand, I intend to show, gained prominence
in the state’s politics strictly on political grounds by being the media-
tor of the bulk of Dalit refugees. On the other hand, the organisation
has undertaken the task of building up the Harichand-Guruchand
movement in order to recreate the alternative spiritual and cultural
space once initiated by its preceptors. The Mahasangha today aims
to mobilise the Namasudras under its aegis on the grounds of caste
loyalty by rejuvenating their collective self as a distinct non-Hindu
political subject. The following study shall therefore try to conceptu-
ally understand the politics of the Matua Mahasangha and its role as
a community organisation in the popular politics of contemporary
West Bengal.
Re-emergence of Matua Mahasangha in politics

The Matua movement ‘needs to be recognized as the first organized
dalit activity in Bengal’, argued Manoranjan Byapari, one of the
prominent Dalit writers in contemporary West Bengal (Byapari and
Mukherjee 2007). The religious movement that was initiated by Har-
ichand Thakur did not only, as we know from Sekhar Bandyopadhy-
ay’s (1994; 2008; 2011) work, remain as an ideological opposition to
the caste structure of Brahminism; rather, it metamorphosed into the
social and political movements of the Namasudras under the leader-
ship of Guruchand Thakur in colonial Bengal. However, the Dalit
movements in Bengal were eventually fragmented and almost came to
an end on the eve of Partition. Thereafter, there had been an eclipse of
the caste question in the politics of the state.
The disappearance of caste from the formal domain of politics has
been analysed by different scholars, and I do not intend to revisit
those analyses here (see Chatterjee 1997; Ghosh 2001; Bandyopad-
hyay 2011; Sen 2013). Instead, our focus shall be on the present
political activism of the Matuas that has effectively disrupted the
continued eclipse of the caste question, and perhaps now signals the
re-emergence of Dalit political activism in West Bengal. The foremost
question therefore is: how can we conceptually explain the recent
emergence of the Matuas?
149
Since 2006–07, it became inevitable that the long-standing Left Front

regime was on the verge of collapse after the much-criticised incidents
of forcible land acquisitions in Singur, Nandigram and elsewhere. The
policies of rapid development and crass industrialisation severely dam-
aged the ‘pro-poor’ and ‘pro-people’ image of the mainstream Left,
which had been responsible for its unprecedented success in the state’s
politics for over three decades (Chatterjee 2009; Chakrabarty 2011).
When the Left Front came to power after the class-based violent resis-
tance of the Naxalbari days against the old Congress regime, it tried to
establish its authority from ‘participation in political movements and
by the fact that they represented the “party” ’ (Chatterjee 2009: 44).
It broke the erstwhile political authority of the Congress at the rural
level that derived its privileged status from landed property, caste loy-
alties or religious associations. Thus, the mainstream Left established
its authority in the state ‘autonomously in the political domain’ (ibid.).
Second, the Left parties received the overwhelming support of the bulk
of refugees, including the Dalit refugees, through their various politi-
cal initiatives, such as the United Central Refugee Council (Chakrab-
arti 1990). Third, since the communists and the mainstream Left had
always linked the struggle against untouchability and the caste system
with the need for an agrarian revolution, redistributive economic poli-
cies, land reforms, and so on (Ranadive 1991: 6–10), the Left Front
government immediately after coming to power in 1977 launched
two path-breaking policies. On the economic front, it dismantled the
concentration of land in a few hands through land reforms and the
redistribution of small plots of land to the rural proletariat; and on
the administrative front, it introduced the three-tier decentralised local
government institutions (Bhattacharyya 2009: 59). The result was the
gradual formation of ‘party society’2 in the West Bengal countryside,
where ‘the older forms of patron-client relationship based on social
2 Party society, as Dwaipayan Bhattacharyya (2011: 230–32) has written, is ‘the modu-
lar form of political society in West Bengal’s countryside’. Political society is a concep-
tual tool formulated by Partha Chatterjee in order to understand the vast domain of
political activities that includes ‘large sections of the rural population and the urban
poor’ who ‘make their claims on government, and in turn are governed, not within
the framework of stable constitutionally defined rights and laws, rather through tem-
porary, contextual and unstable arrangements arrived at through direct political
negotiations’ (Chatterjee 2008: 57). Bhattacharyya’s ethnographic study makes a dis-
tinction between the politics of negotiation practiced in urban and rural settings, and
shows that during the Left Front regime, the sole agent/negotiator/mediator of the
rural poor in West Bengal was ‘the party’.
150
and economic hierarchies’ was replaced by the organisational grid of

the political party that had now emerged as the ‘chief mediator . . . in
settling of every village matter: private or public, individual or collec-
tive, familial or associational’ (Bhattacharyya 2009: 68–69). There-
fore, the new arrangement under the Left Front regime swallowed up
every local institution, which meant that the ‘Kolkata-centred upper-
caste dominance of the left parties was extended, through the party
structure, to dominance over local politics everywhere in the state’
(Chatterjee 2012: 70).
The stable party society faced a crisis from 2006–07 as the Left par-
ties failed to reproduce the initial conditions of being and became ‘an
apologist for corporate capital’ (Bhattacharyya 2011: 232–40). The
crisis of party society, as anticipated by Bhattacharyya, could lead to a
politics of community, based on locally constituted networks of caste,
ethnicity and religious associations (245). The anticipation proved to
be quite convincing as the politics of West Bengal began to witness a
host of autonomous community-based political assertions, each with
their specific demands.3 Such a phenomenon was more overtly vis-
ible when the Left Front faced electoral defeats in the 2009 general
elections and in the 2010 civic polls (Rana 2010). The Left changed
its political stance vis-à-vis the question of reservations, amended the
Panchayat Act in order to ensure reservation for the OBCs and also
instituted an award in 2010 for contributions to Dalit upliftment (The
Times of India, 16 September 2010). Consequently, the different com-
munity organisations, now capable of securing votes, in turn, achieved
immense political salience and engaged in a politics of negotiation
with the world of party politics. Moreover, since the present ruling
party, the Trinamul Congress (TMC), came to power capitalising on
the spontaneous popular movements of Singur, Nandigram, Lalgarh,
and elsewhere, and given that it does not adhere to any revolutionary
ideology nor has any preexisting well-orchestrated party machinery,
it could not immediately reproduce a similar structure in rural West
Bengal. Therefore, the TMC had no other option but to tap into vari-
ous community organisations for electoral and other political benefits,
not least because these local institutions largely ensured its ascent to
power in 2011.
3 For example, the emergence of the Kamtapur People’s Party, Gorkha Janmukti Morcha,
Jamiyat-e-Ulema-e-Hind, and the ‘ethnicity and religion-based mobilizations’ of the
Rajbanshis, Gorkhas, and Muslims (Bhattacharyya 2011: 250). The mobilisation of the
Matuas on the grounds of caste loyalty adds to the list of such political formations.
151
The Matua Mahasangha emerged as the political mouthpiece of

the Matuas and the Dalit refugees during this crisis of party society.
By proving its organisational strength as the guarantor of a massive
vote bank since the 2007 by-elections in Bongaon constituency, the
Mahasangha gained a solid political reputation. Thus, on the one
hand, as a political mouthpiece, it emerged as a negotiator of the bulk
of Dalit refugees, representing them on the issues of migration and
refugee rehabilitation, demanding the repeal of the 2003 Citizenship
Act. On the other hand, as a religious organisation, the project of
the Mahasangha today is to remember, recollect and revive the glori-
ous counter-history of a religious protest movement through a series
of organisational activities and by re-organising the Namasudras as
a cohesive community that can counter the dominant socio-religious
and cultural markers of bhadralok politics.
Building up the Harichand-Guruchand movement

Drawing on my own ethnographic work, I shall try to develop a cri-
tique of the politics of Matua Mahasangha, focusing on three aspects:
one, its organisational history and contemporary structure; two, its
role in the Dalit literary movement; and three, its political activism as
the community organisation of the Matuas.
Although one can trace the organisational history of the Maha-
sangha in various local literatures and oral narratives, an official doc-
umentation of the history was published by the Thakurbari4 in 2008,
titled An Approach to Matuaism, written by Debdas Pande, a Matua
literary activist. Pande’s work introduces us to Matua cosmology, but
more importantly, we get to see how the Mahasangha represents its
own organisational history.
The Matua Dharma was introduced as a deviant sect of the Nama-
sudras, who were once called the Chandals and were untouchables, by
Harichand Thakur and was given an organisational structure by his
able son, Guruchand Thakur in 1931–32.5 At that point of time, it was
4 Thakurbari in Thakurnagar (North 24-Parganas, West Bengal) is the headquarters of

the Matua Mahasangha and also the house of P. R. Thakur and the Thakur family.
5 The official declaration of Matua Mahasangha today is that the organisation was
founded by Guruchand Thakur in 1931/32 (See Matua Mahasangher 25 tama Barshik
Sammelan: Sadharon Sampadoker Protibedan, published 10 November 2011; and Sree
Dham Thakurbari; Matua Mahasangher Sanbhidanba Gathantantro, Matua Maha-
sangha, Sree Dham Thakurbari). But Sekhar Bandyopadhay (2008: 192) mentions that
the organisation was formed around 1915 by Tarak Goshai, an ardent Matua follower.
152
known as the Sree Sree Harichand Mission. The partition struggles

not only disoriented the Namasudra movement, but it also disrupted
the political project of the Matuas. Although P. R. Thakur, the first
sanghadhipati (organisational head) of the organisation appointed
by Guruchand, took the initiative to revive the organisation repeat-
edly since 1949, the project met with numerous hurdles and did not
evolve into anything substantial. One major reason for this failure was
political differences within the leadership of the organisation imme-
diately after Partition. Mahananda Halder, the former secretary and
editor of the magazine, Thakur, resigned from his post and formed his
own independent organisation, called the Harichand Seba Sangha in
1965. The two organisations functioned separately for over a decade.
Finally, in 1980, under the initiative of Susil Kumar Biswas, the two
separate outfits merged to form the Harichand Matua Seba Sangha.
In 1983, a change in name of the central organisation was proposed
by Baroma Binapani Devi, the wife of P. R. Thakur, which was finally
accepted by the working committee meeting of the organisation. It
was thus renamed as the Matua Mahasangha in 1986. Since then, the
organisation has regained its full life and devoted itself to the task of
preaching the doctrines of Harichand-Guruchand amid the scattered
Namasudra population across India. The Mahasangha was registered
in 1988 with its headquarters at Thakurnagar, North 24-Parganas,
about 50 km away from Kokata. In 1990, the organisation launched
its women’s wing, called Santi Satyavama Nirman Committee.
On its cover page, the Matua Mahasangha Patrika, the official
mouthpiece of the organisation, defines the sect as:
A socio-economic religion with spiritual outlook based on

the uplift of the down-trodden class including ideal family
life and for their mass salvation in the way of eradicating the
barriers of casteism and untouchability, social and economic
disparities.
Organisationally, the Mahasangha has taken the structure of a

‘modern voluntary association’ with ‘offices, membership, incipi-
ent bureaucratisation, publications and a quasi-legislative process
expressed through conferences, delegates and resolutions’ (Nigam
2000). We shall see shortly how the Mahasangha today, like any
other caste association, plays ‘a mediatory role between the “illiterate
mass” and the political system’ (ibid.).The primary site of mobilis-
ing the Matuas is the religious space, particularly the Harimandir,
153
where the dals (small groups of devotees) assemble for regular eve-
ning kirtans (devotional songs), hold periodic uplift meetings and
look into the various organisational activities. The Matuas, wherever
they might reside, are instructed to form dals, each of which is to be
led by a dalapati/dalanetri (leader of the dal). All of these groups
are affiliated to the central organisation – the Matua Mahasangha.
These dals are assigned a uniform set of functions by the central
organisation such as constructing a Harimandir in every neighbour-
hood, organising regular meetings and mobilising the devotees for
political agitations and movements under the banner of the Matua
Mahasangha.
An integral part of the present political movement led by the Maha-
sangha is its relentless efforts to develop an autonomous consciousness
among its followers that efficiently challenges the cultural and religious
markers of the caste Hindu order. The Matuas, noticeably, alienate
themselves from the upper-caste Hindu ritual order. The organisation
has introduced a booklet on a separate set of rituals and practices,
opposed to the Brahminical doctrine. The Matuas follow these ritual
practices in their quotidian lives in order to mark their independent
religious identity. In other words, the agenda of the Mahasangha’s
contemporary activism is to revive the sect as an autonomous non-
Hindu religious order that aims to eradicate the caste hierarchy. For
instance, another crucial development that needs to be marked in the
organisational history of the Mahasangha is a change in name of their
annual festival from ‘Baruni Mela’ to ‘Matua Dharma Mahamela’.
It is assumed by the followers that because the origin of Baruni6 (the
sea deity) coincides with the birth of Harichand Thakur, the annual
assembly of the Matuas came to be known as Baruni Mela. But since
the Matuas assemble on that day to worship their guru Harichand
Thakur, the Mahasangha decided to change the name. The change in
name of one of the biggest religious festivals in West Bengal is defi-
nitely a step towards establishing the religion autonomously outside
the shackles of Hinduism.
The act of renaming the festival faced serious criticisms from
the hindutva forces. Two of my respondents, who were volunteers
of the All India Refugee Front (established in 2004) at the time of
6 According to Matua cosmology, Baruni is the deity of sea and the daughter of Devata
and Asura.
154
the festival, showed disagreements regarding the change in name.

They complained that the present leadership of the Mahasangha
is self-seeking, and hence, trying to project the Matuas as follow-
ers of a separate religion, thus ostensibly deviating from the stand-
point of P. R. Thakur. Interestingly, they were reluctant to talk about
Guruchand Thakur and his politics, and repeatedly asserted the inte-
grationist political position of P. R. Thakur, particularly at the time
of Partition. In their view, the Matuas are a community (sampraday
was the word they used) within the Hindu religious order. Later on,
I came to know them as supporters of the Bharatiya Janata Party
(BJP), who have also been trying to establish their political influence
among the Matuas and the Dalit refugees for quite some time now.
It seemed that the political history of P. R. Thakur is, at present, act-
ing as a negative catalyst in the process of building the autonomous
movement.
How, then, is the Mahasangha at present looking at P. R. Thakur
and his political career? Swapan Biswas,7 the editor of the Matua
Mahasangha Patrika, shared his views on this issue during an inter-
view. For him, their revered Thakurmoshai (as P. R. Thakur is affec-
tionately called by many of his followers) was a pragmatic man who
took the correct decision at the time of Partition by supported the
ruling Congress party to ensure the proper resettlement of the Nama-
sudra refugees. However, being the leader of the Dalits, Thakur had
fought for the inclusion of a couple of Dalit-populated districts within
the Indian border at the time of Partition, but unfortunately failed due
to the upper-caste dominance in political decisions. This history is the
official claim of the Mahasangha at the moment. In the 66th issue of
the official magazine of the Mahasangha, the speech delivered by Kapil
Krishna Thakur at the meeting on 28 December 2010 – a description
of which opened this chapter – was published. In his speech, Kapil
Thakur quoted his father thus:
It is extremely important to acknowledge that we the Dalits

are original settlers in this country . . . It is a fact that India
is ours and we shall not allow our own Motherland to get
divided among the caste Hindus and the Muslims.8
7 Interview with Swapan Biswas, 13 November 2012.

8 Translation from Bengali is mine.
155
What we can derive from such official claims is that the Maha-
sangha is rewriting its organisational history in which they portray
the Matuas as Dalits of Bengal, and as the original inhabitants of the
land, who had always been marginalised amid the communal tensions
between Hindus and Muslims.
Although a religious organisation, as Debdas Pande’s book officially
suggests, the Mahasangha has ‘alternative plans’ for establishing a
political party if and when it feels it is necessary. The movement unde-
niably aims to capture political power for which, as Pande argues,
they need to form a political party that will consist of members who
will be of ‘religious mind and good moral character’. Swapan-babu,
an official spokesperson of the organisation, also reasserted similar
political ambitions:
The names of Harichand-Guruchand Thakur whose ideals are

followed by so many people had always been suppressed in
the history of Bengal. The caste Hindu leaders of West Ben-
gal politics had never allowed the Dalits to escalate the upper
echelons of the political parties. Now we shall flourish. Our
religion is a protibadi dharma (religion of protest).
Another Matua activist from Ashokenagar (North 24-Parganas),

Sushanta Bala,9 also talked about the organisation’s future plans
to introduce booth committees within their organisational struc-
ture (see Table 7.1). In other words, the Matua Mahasangha has
organisationally been quite successful so far in uniting the dispersed
Namasudra refugees, which enabled them to politically represent
the community. It has efficiently launched a movement to reunite
the Matuas on the grounds of caste affinity and shared histories
of marginalisation and victimhood within the caste Hindu order.
Dr Sukumar Halder, a local Matua leader, believes that ‘the Maha-
sangha and the Thakurbari has provided the rootless Namasudras
an identity’.10
9 Interview with Sushanta Bala (name changed), 10 November 2012.

10 Interview with Dr Sukumar Halder, President, Ashokenagar Block Matua Maha-
sangha, 12 November 2012.
156
Table 7.1 Organisational structure of the Matua Mahasangha.
CHIEF ADVISOR Religious Head SANGHADHIPATI Organisational Head

Central Executive Committee
1 2 3 4 5 6 7 8 9
Religious Education Social Socio- Youth Women’s All India Regional Family
Branch and Cultural Welfare Economic Welfare Branch Branch Mandir Sangha
Branch Branch and Politico- Branch Committee Committee
Economic
Branch
Source: Matua Mahasangher Sanbhidan ba Gathntantro, 2nd Edition, Sree Dham Thakurbari, Thakurnagar.
The Matua literary movement

If one is to talk about Bengali Dalit literature today, a reference to what
is known as Matua-sahitya (Matua literature) and which began in the
19th century is a must (Byapari and Mukherjee 2007). Literary activ-
ism has historically been an indispensable part of the politics of the
Matuas. The Mahasangha incessantly promotes literary activities
among its followers. Apart from the primary texts, such as Sree Sree
Harililamrito and Guruchandcharit, there are a host of periodicals
published statewide by the various branch organisations of the Maha-
sangha. These include the Matua Mahasangha Patrika, Matua Bandha
and Matua Darpan, among many others. Although one can find a con-
stant critique of the caste-Hindu-dominated Bengal politics in this
Matua literature – which aims to produce authentic knowledge about
the experiences of caste oppression and perpetual victimhood – it has
other salient features as well. Unlike the Dalit literature in other parts
of India, the core of the contemporary Matua literature is its attempt to
posit Harichand-Guruchand as counter-elites in bhadralok-dominated
West Bengal and local icons of Dalit identity by moving beyond Ambed-
karite political assertions. For example, it is a common grudge among
the Matuas that the name of Guruchand Thakur, unlike other upper-
caste reformers of 19th- and 20th-century Bengal, has never been
included in the cherished history of social reforms and reformers in
Bengal. Although his contribution towards the education of the Nama-
sudra community is unmatchable, his name was never allowed to flour-
ish because of his lower-caste status. Moreover, Guruchand led a
delegation of Namasudras in 1907 to ensure the recruitment of lower-
caste people in government jobs. But none seem to remember the
contributions of this great Matua leader. Therefore, in order to com-
memorate their guru, the centenary year of this event was celebrated
with enthusiasm by thousands of Matuas in 2007 at Ashokenagar,
where the periodical Rajogrihe Satobarsho (Hundred Years in the Gov-
ernment Sector) was launched. The idea behind such initiatives is to
uphold an alternative history that was never allowed to germinate in
the past because of the preponderance of the upper-caste bhadralok in
all avenues of public life in Bengal.
The rise of the Matua Mahasangha to political fame also influenced
other Dalit literary mouthpieces to acknowledge the former’s current
activism as a crucial step towards a future of organised Dalit poli-
tics. For instance, in 2009, when the Matuas were receiving enormous
attention, one of the issues of the Chaturtha Duniya, published by
158
the Bangla Dalit Sahitya Sanstha, had as its theme Matua Dharma
O Darshan (Matua Religion and Philosophy). This is no mere coin-
cidence, but rather an acknowledgement of the contemporary Matua
movement by its comrades. We also find different analogies drawn in
many contemporary Dalit writings between the political outlooks of
Guruchand and Ambedkar. The former resisted Brahminical Hindu-
ism by being a Matua, while the latter embraced Buddhism. Again,
the Matua sect has been interpreted by many as a religion that drew
its sap from Buddhism. Drawing such correlations between the two
religions in the recent Dalit literature can surely be read as an attempt
by Bengali Dalit activists to unite the Matua movement with Dalit
activism in other parts of India. Chaturtha Duniya also commemo-
rated Harichand Thakur in one of their issues in 2012 on the eve of
his 200th birth anniversary.
Other small Matua organisations such as the Harichand Mission
actively contribute to Matua literature in spite of political differences
with the Mahasangha. Established in 1977 by Debendralal Biswas
Thakur, after he had a rift with the Sree Sree Harichand Seva Sangha,
the Mission had launched a magazine called Harisevak. Deben Thakur’s
major works include Matua-ra Hindu Noy: Dalit Oikyer Sandhane
(Matuas are Non-Hindus: In Search of Dalit Unity), Swadhinata 50:
Iye Azaadi Jhutha Hai (Fifty Years of Independence: The Freedom is
a Farce), Dharmadando Mormokatha (An Essence of the Dialectics
of Religion), Kolir Kahini (The Story of Kaliyuga), his autobiography
titled My Life History in Brief, and many others. In Matua-ra Hindu
Noy, he sought to unite Dalits and put forward Guruchand Thakur as
the local icon of Dalit politics in Bengal, elevating him above Ambed-
kar. For him, Harichand and Guruchand Thakur had been the true
liberators of the Dalits in Bengal as they provided them an alternative
spiritual space outside Brahminism. He also writes that although Hin-
duism had historically played a politics of inclusion by incorporating
various deviant faiths into its fold, the ‘clever Hindus’ never recog-
nised Harichand Thakur as an incarnation. Such a move, he argued,
would have dismantled their ideology. In Swadhinata 50, he elabo-
rated on the discriminatory treatment that the Dalits had faced in West
Bengal, even under the so-called progressive communist regime. The
title of the book, published in 1997 – the golden jubilee of India’s inde-
pendence – suggests that independence had been a farce for the Dalits
of the country. Drawing his case studies from West Bengal, he shows
how the Dalits here have received little material benefits compared to
that enjoyed by caste Hindus. He pointed to the sufferings of the Dalit
159
refugees, the harsh conditions in which they were resettled outside

Bengal and how they were systematically made the prey of state vio-
lence in Marichjhapi in 1979. It is the failure of the Indian state that it
could not eradicate poverty among Dalits, and could not provide them
with the opportunities of education and employment.
These Dalit literary and political formations which still dwell on the
margins do share organisational differences with the Mahasangha, yet
they have common concerns. Manohar Mouli Biswas,11 an eminent
Dalit writer and member of Chaturtha Duniya, admitted that even
though sections of the Namasudra community do not share a comfort-
able rapport with the Thakurbari, they are with the Mahasangha in
their movement to repeal the 2003 Act and share their concerns for
the Dalit refugees. Debendralal Biswas Thakur12 too shared this view:
Yes, I support them in their present movement against the

2003 law. If somebody does something right, then how can
we refrain from supporting them? Then we are not human
beings. I support them because they have the strength which
I do not have.
Matua Mahasangha: a community organisation in

popular politics
Let us now finally move to the role of the Matua Mahasangha as a
community organisation in the popular politics of contemporary
rural West Bengal. The Mahasangha, as already mentioned, could
hardly attract the public gaze in the erstwhile party society, and only
gained political salience recently. Since the days of P. R. Thakur, the
Mahasangha made repeated attempts to posit itself as the frontal
organisation of the Namasudra community. But success evaded it
for a long time. Thakur died in 1990. In 1994, the organisation held
a large public meeting at the Shahid Minar in Kolkata to bring the
Namasudras under the doctrine of Matuaism. Since then, the organ-
isation engaged in different forms of community service at the local
level. It extended relief aid to the flood victims in Nadia district
during this period, and passed a resolution to establish a primary
school, named as Pramatha Ranjan Prathamik Bidyapith. In 2001,
11 Interview with Manohar Mouli Biswas, 6 June 2013.

12 Interview with Debendralal Biswas Thakur, 29 April 2013.
160
the Mahasangha held a mass agitation before the Bangladesh High

Commission in Kolkata to protest against the atrocities committed
on the Hindu minority in Bangladesh. However, even if such activi-
ties on the part of the Mahasangha indicate its new zeal to evolve
into an important political actor, the organisation could not estab-
lish a strong foothold in the formal domain of politics. The Matua
Mahasangha received recognition as the political mouthpiece of the
Dalit refugees only when 21 Matua devotees pledged to a fast unto
death in 2004 to protest against the Citizenship Amendment Act of
2003. Ganapati Biswas, the present general secretary of the organ-
isation recollects:
In 1994, we held a meeting in Kolkata near Shahid Minar.

Nobody knew us much at that point of time. After the law
was passed in 2003, I led a fast here at Thakurnagar in 2004
demanding the amendment of the law. We were noticed for
the first time and delegates came from Delhi to listen to
our demands. Then took place the historic gathering on 28
December 2010. Leaders of all the parties came to us. They
were surprised to see so many people united under Baroma’s
leadership. The media also highlighted us. Moreover, it was
before the elections.13
In the 2009 general elections, the Matua Mahasangha showed its

organisational strength as a guarantor of votes in the Matua-populated
areas of Bagda, Bongaon and Gaighata. The strength of the Maha-
sangha in the Dalit-populated areas brought the bhadralok leadership
to their knees. All political parties adopted a ‘politics of compensation’
vis-à-vis the Mahasangha by providing material gifts and promising
administrative support to flatter the community. The Left Front sanc-
tioned the plan of a government college at Chandpara and also pro-
vided land for the Sri Sri Harichand-Guruchand Research Foundation
in Rajarhat. The TMC, in turn, upgraded the Thakurnagar railway
station, promised government jobs, a railway stadium and a hospi-
tal. Above all, the party offered the youngest son of Baroma, Man-
jul Krishna Thakur, a candidature in the 2011 elections, and, finally,
appointed him the minister for Refugee Rehabilitation and Relief. All
13 Interview with Ganapati Biswas, General Secretary, All India Matua Mahasangha,
12–13 November 2012.
161
the prominent political parties quite openly offered cash for the devel-
opment of the organisation; the Mahasangha, quite ingeniously, pub-
lished the amounts received in their official annual report. In 2011, the
Harichand-Guruchand Research Foundation came up with a two-fold
demand: ‘Chai nagorikotyo, chai jatipatro’ (we demand citizenship,
we demand caste certificate). The last grand initiative taken by the
Matua Mahasangha was an all-India rathyatra (chariot procession) to
mark the 200th birth anniversary of Harichand Thakur. The leaders of
the organisation led the procession in the Namasudra-populated areas
across India. The idea behind the initiative was to reunite the commu-
nity under the influence of Matua Mahasangha, which now aspires to
emerge as the pan-Indian religious organisation of the Namasudras.
The re-emergence of the Matua Mahasangha as the vanguard of the
Dalits, in turn, allowed it to play a new role within the community.
Below, I offer five brief case studies, selected from my ethnographic
material and newspaper reports. A reading of these cases shall help us
both understand and acknowledge the new role of the Mahasangha as
a community organisation in rural West Bengal.
I
One can get a comprehensive account of the patterns of migration of
the Namasudra refugees, and their consequent Indianisation, in
Ranabir Samaddar’s study The Marginal Nation (1999). According to
Samaddar, the refugees migrate, settle, choose some local occupation,
receive absolute support from the local people due to caste affinities,
eventually obtain their ration cards and caste certificates, get enrolled
in schools and colleges as Scheduled Caste candidates, and then, they
vote (Samaddar 1999: 96–106). Such a pattern of migration has
largely remained unchanged till date. During the Left Front regime,
the refugees were recognised in a clandestine manner, and the clout of
political parties was unquestionable (Bandyopadhyay 2011). How-
ever, a major role at present is played by the Matua Mahasangha vis-
à-vis the migration of Dalit refugees. Saroj Bepari,14 a cycle van-puller
at Thakurnagar, revealed that if today, some Namasudra migrant who
took refuge in and around Thakurnagar has to get her or his voter
identity card issued, a broker close to the government officials at the
nearby village of Chikanpara does it at a price of only INR 6,000. The
14 Interview with Saroj Bepari (name changed), 12 November 2012.
162
Mahasangha extends initial support to these refugees by granting

membership. And once you are a Matua, it is easy to get other things
done. Moreover, since the 2003 Act was passed, there had been mul-
tiple cases of arrest of the Dalit refugees. The Matua Mahasangha had
continuously resisted such arbitrary arrests and torture of the refu-
gees. Swapan Biswas, the editor of the Matua Mahasangha Patrika,
revealed that the Mahasangha once ensured the release of 87 Namasu-
dra refugees who were arrested in Gushkara in Burdwan as illegal
migrants.15 Thus, we see how the Mahasangha today efficiently deals
with the formal institutionalised world of politics to safeguard the
interests of Dalit refugees in West Bengal.
II
On 29 December 2010, a report titled ‘The Importance of Being a
Matua’ was published in a leading English-language daily after the
huge gathering organised by the Mahasangha the day before. The
report mentioned that there was an ‘assurance by CM that no Matua
who crossed over from Bangladesh would be pushed back’ (The Tele-
graph, 29 December 2010). It was evident that because of the sheer
organisational strength of the Mahasangha in terms of votes, the iden-
tity of being a Matua could now ‘trump’ the identity of an ‘illegal
migrant’. After the poribortan (change) at Writers’ Building, the TMC
government appointed their victorious candidate from Gaighata,
Manjul Krishna Thakur, as the minister of state for Refugee Rehabili-
tation and Relief. To know about the role of the Matua Mahasangha
and the state government vis-à-vis the refugee problem, I interviewed
the minister. He replied: ‘There is no problem on the issue these days.
Earlier there were arrests and all, which has now been absolutely
stopped. We cannot change the law; it’s a central government’s issue.
As of now there is no problem.’16 The response implied that although
the national law upholds the 2003 Act, the refugee problem at present
is dealt with largely on the grounds of illegality. If we consider the
minister’s version of the situation to be true, the refugee problem is, in
effect, addressed by the state administration through ‘temporary, con-
textual and unstable arrangements’ by moving beyond the scope of the
legal-bureaucratic rationality of the modern liberal state (Chatterjee
15 Interview with Swapan Biswas, All India Matua Mahasangha, 13 November 2012.
16 Interview with Manjul Krishna Thakur, 13 November 2012.
163
2008: 57). And such adjustments are arrived at through direct politi-
cal negotiations between the Mahasangha, on behalf of the Dalit refu-
gees, and the formal world of institutionalised politics.
III
The next case reconfirms the everyday existence of casteism and caste-
based discrimination in West Bengal. One can seldom find news on
such matters in the reports of the leading media while sitting in Kolk-
ata. A local weekly called Simanta Bangla (dated 8 November 2012),
published from Bongaon in North 24-Parganas, reported the following
incident. The headlines of the weekly newspaper read: ‘College student
verbally abused as Namasudra, Matuas hit the streets of Bongaon,
F.I.R.’17 The report claimed that the general secretary of the Bongaon
College students’ union, run by the Trinamul Chhatra Parishad (stu-
dents’ wing of TMC), mentally and physically assaulted two students
of the Namasudra community because of their lower-caste status. Fol-
lowing the incident, the Matuas, under the banner of the Mahasangha,
took to the streets of Bongaon holding demonstrations and protest ral-
lies. The organisation called a public meeting at Bongaon Lalit Mohan
Bani Bhavan to condemn the incident, and demanded the arrest of the
general secretary. Moreover, the Bhagawan Sri Sri Guruchand Thakur
Janmo-Jyanti Uthjaapan Samiti of Bongaon had informed the chief
minister about the incident and asked her to ensure a public apology.
Incidents of caste-based discrimination are nothing new in the state.
But earlier, such local disputes were either covertly resolved by the
party, or were efficiently suppressed at the local levels. Now, the role
played by the Mahasangha in the wake of this incident, and the fash-
ion in which it negotiated with the formal domain of politics on behalf
of the community, clearly indicates its role as the community organisa-
tion of the Namasudras. It seems that such discriminatory practices at
the rural level can no more be silently mitigated.
IV
The following case is that of a local dispute over agricultural land.
During my interview with Ganapati Biswas, the general secretary of
Matua Mahasangha, a group of people who were Matua followers
17 Translation from Bengali is mine.
164
came to meet him from a nearby village under Bagda police station
limits. I was a silent observer. They came to complain about some
recent arrests following a dispute over 11 bighas of agricultural land.
They were bullied and forced to leave the land by another group close
to the present ruling party, the TMC, having been warned that ‘puro
Nandigram hoye jabe’ (‘a new Nandigram will happen’) if they did
not follow orders. The local police intervened and eight people – four
from each group – were arrested, only to be released the next day.
Now, they had come to Ganapati-babu to fix the situation, and ensure
‘security’ in case of any further incidents. Ganapati-babu, after listen-
ing to them, replied that he already ‘knew everything’ from his own
sources. What he then suggested to them left me awestruck: ‘First you
erect a Harimandir there; then see what happens’ was his reply. Evi-
dently, the existence of a Harimandir expands the organisation’s
strength at the grassroots level. But most importantly, we can infer
from the incident that the Mahasangha today is a local power centre
that the Matuas turn to in order to settle their disputes and negotiate
on their behalf with the police and the party.
V
The Harimandir is perhaps the most important site in the everyday lives
of the Matuas, not only because of its symbolic significance but also
for being the site for political socialisation. This became clear to me
when I visited the Sri Sri Harichand-Guruchand Seva Sangha at Ashra-
fabad Government Colony at Ashokenagar, North 24-Parganas. The
Harimandir being a popular site, it had been cunningly used by local
political leaders from all parties for their evening adda (gossip sessions).
As the political parties had also extended funds to upgrade the mandir
premises, the local leaders had taken different initiatives, such as setting
up a library or arranging a blood donation camp that centred on the
mandir. However, such political interventions were a source of dissatis-
faction among the local people. A dispute finally started regarding the
use of the mandir as the local leaders appointed a priest on their own,
and denied the local people access to the premises.
The local dalnetri of the Matuas, Asima Biswas,18 told me that
the mandir had been there since the 1980s. In 1988, the government
divided the plots in their neighbourhood and a plot was kept by the
18 Interview with Asima Biswas (name changed), 11 November 2012.
165
local devotees. The mandir was registered only in 2009 in the name
of Sri Sri Harichand-Guruchand Seva Sangha, something which the
local people now feel had been a mistake: ‘We should have registered
it mentioning the word Matua Mahasangha. They do not allow us to
go in. Now politics has entered the temple’, said Asima Devi, point-
ing to how the mandir today had been hijacked by local politicians.
When asked about the solution to their problem, Asima Devi said:
‘If Baroma tells Mamata about this incident, everything will fall into
place.’ What is interesting is not the strategic control of the mandir
by the local politicians, but the popular belief among the people that,
at the moment, the Thakurbari has the powers to negotiate with the
formal world of politics and solve their everyday problems.
Conclusion
A careful analytical reading of the aforementioned cases necessitates a
rethinking of the nature of popular politics in rural and semi-urban
West Bengal. With the crisis of ‘the party’ as the chief speaker in the
domain of popular politics in the West Bengal countryside since 2006–
07, a void was created. This crisis of party society has, in turn, created
new spaces for ‘a politics of community’, formed on the basis of caste,
religion, language, or tribe. The present politics of the Matua Maha-
sangha confirms this emerging trend.
With the gradual evolution of the Matua Mahasangha as the repre-
sentative of the Dalit refugees and the guarantor of a large number of
Dalit votes, the organisation too was strengthened. The sudden atten-
tion that it received from the world of party politics unquestionably
boosted its own confidence as a political actor, and eventually led it to
opt for an autonomous political position outside the ambit of bhadral-
okism. The Mahasangha, being the representative institution of a pro-
tibadi dharma, looks at itself as the vanguard of the Dalit refugees
and the underprivileged. Simultaneously, a popular trust among the
common people vis-à-vis the organisation has also developed. Such
popular confidence acted as a positive catalyst and complemented the
Mahasangha’s plan to project itself as the community organisation of
the Namasudras. As the aforementioned cases show, the Mahasangha,
on the one hand, has emerged as a political mouthpiece for the Dalit
refugees; and on the other, it has evolved into a powerful institution
in rural areas. The organisation has started to play a new role within
the community by replacing the role hitherto played by the party. The
present role of the Matua Mahasangha in rural West Bengal, therefore,
166
is that of a mediator or negotiator that positions itself between the

formal world of modern liberal politics and the world of collective
social thinking. In other words, it carries out a ‘new’ politics of media-
tion in the West Bengal countryside. Along with that, the most impor-
tant aspect of the politics of the Matua Mahasangha is its systematic
programme of building up the Harichand-Guruchand movement.
By projecting Harichand and Guruchand as the local icons of Dalit
politics, the Mahasangha is, on the one hand, striving to expand its
organisational strength, and on the other, trying to create a counter-
discourse that can stand outside, and at the same time, efficiently chal-
lenge the bhadralok dominance over the Dalits.
While it is beyond doubt that the Matua Mahasangha has emerged
as one of the key political actors in West Bengal over the last couple
of years, the question whether they can maintain their current politi-
cal status is an open one. The leadership of the Mahasangha today
has very close ties to the TMC. The two brothers – the elder brother
and sanghadhipati, Kapil Krishna Thakur and the younger, Manjul
Krishna Thakur – are, in spite of their personal and political differ-
ences, aligned with the TMC; the former was elected as a Member
of Parliament in the 2014 Lok Sabha polls, while the latter is still a
minister in the TMC-led state government. Some may see this as a
co-optation of the Mahasangha leadership by the TMC and as an indi-
cation of how the new political spaces opened up by the crisis of party
society are now rapidly closing again. Yet, it is also possible to think of
the Mahasangha’s close links with the TMC as a temporary and stra-
tegic alliance, which can be broken anytime in the future. Regardless
of how one views these alignments, however, one cannot deny that the
politics of the Matua Mahasangha has effectively created a situation
where the voices of those who dwell on the margins can no longer be
ignored, and where the silence on the caste question has finally been
broken in West Bengal. Perhaps there lies the real contribution of the
present politics of the Matuas.
Bibliography
Bandyopadhyay, Sekhar. 1994. ‘Development, Differentiation and Caste: The
Namasudra Movement in Bengal 1872–1947’, in Sekhar Bandyopadhyay,
Abhijit Dasgupta and Willem van Schendel (eds), Bengal: Communities,
Development and States, pp. 90–119. New Delhi: Manohar Publishers.
Bandyopadhyay, Sekhar.2008. ‘Social Mobility in Colonial India: The Nama-
sudras’, in Ishita Banerjee-Dube (ed), Caste in History, pp. 181–96. New
Delhi: Oxford University Press.
167

The Namasudras of Bengal 1872–1947. New Delhi: Oxford University
Press.
Bandyopadhyay, Sandip. 2011. ‘Who Are the Matuas?’, Frontier, 43(37). http://
frontierweekly.com/archive/vol-number/vol/vol-43-2010-11/vol-43-37/
matuas-43-37.pdf (accessed on 18 August 2014).
Bhattacharyya, Dwaipayan. 2009. ‘Of Control and Factions: The Changing
“Party-Society” in Rural West Bengal’, Economic and Political Weekly,
44(9): 59–69.
Bhattacharyya, Dwaipayan.2011. ‘Party Society, Its Consolidation and Crisis:
Understanding Political Change in Rural West Bengal’, in Anjan Ghosh,
Tapati Guha-Thakurta and Janaki Nair (eds), Theorizing the Present:
Essays for Partha Chatterjee, pp. 226–50. New Delhi: Oxford University
Press.
Byapari, Manoranjan and Meenakshi Mukherjee. 2007. ‘Is There Dalit Writ-
ing in Bangla?’, Economic and Political Weekly, 42(41): 4116–20.
Chatterjee, Partha. 1997. The Present History of West Bengal. New Delhi:
Chatterjee, Partha. 2008. ‘Democracy and Economic Transformation in India’,
Chatterjee, Partha. 2009. ‘The Coming Crisis in West Bengal’, Economic and
Chakrabarti, Prafulla K. 1990. The Marginal Men: The Refugees and the Left
Political Syndrome in West Bengal. Kalyani: Lumiere Books.
Chakrabarty, Bidyut. 2011. ‘The Left Front’s 2009 Lok Sabha Debacle in West
Bengal, India’, Asian Survey, 51(2): 290–310.
Ghosh, Anjan. 2001. ‘Cast(e) out in West Bengal’, Seminar, 508.
Jaffrelot, Christophe. 2003. India’s Silent Revolution: The Rise of the Low
Castes in North Indian Politics. New Delhi: Permanent Black.
Nigam, Aditya. 2000. ‘Secularism, Modernity, Nation: Epistemology of the
Dalit Critique’, Economic and Political Weekly, 35(48): 4256–68.
Ranadive, B.T. 1991. Caste, Class and Property Relations. Calcutta: National
Book Agency.
Rana, Santosh. 2010. ‘The Crisis of Identity’, Frontier, 43(12–15).
Samaddar, Ranabir. 1999. The Marginal Nation: Transborder Migration from
Bangladesh to West Bengal. New Delhi: Sage Publications.
Sen, Dwaipayan. 2012. ‘The Emergence and Decline of Dalit Politics in Ben-
gal: Jogendranath Mandal, The Scheduled Castes Federation and Partition,
1932–1968’, unpublished Ph.D. dissertation, University of Chicago.
Sen, Dwaipayan. 2013. ‘An Absent-Minded Casteism?’, Seminar, 645: 48–54.
Weekly, 47(34): 26–27.
168
8
TRANSFORMATIVE POLITICS
The imaginary of the Mulnibasi in
West Bengal
Indrajit Roy
Some people call this the Sharadiyo Utshob. Others know it as

Durga Puja . . . These names carry within them thousands of years
of chokranto [conspiracy].
(Charan Besra, President, Majhi Pargana Gaonta,
Convocation speech, Kanturka (Maldah),
Mahishasur Smaran Sabha, October 2012)
At our very first meeting, Lerka Hembrom asked me what my caste

was.1 When I told him I was Kayasth, he smirked, ‘Ah, you tried to be
Brahmon but failed, eh? Welcome to the Mulnibasi fold.’2 He intro-
duced himself and his eight other friends as Mulnibasi. ‘We were born
into different castes. But we call ourselves Mulnibasi, the autochtho-
nous [in English] people of this country.’ And he clarified, for good
measure, ‘Anyone who is not Brahmon, Khotriyo and Baishyo is Mul-
nibasi.’ Introducing us to each other was Shyamsundar Sarkar, my
host in a rural locality of West Bengal’s Maldah district. Shyamsundar
had been keen that I meet Lerka Hembrom because he was convinced
that meeting him would help me understand the changes in people’s
lives in the locality, the vaguely stated purpose for my being there.
1 I would like to acknowledge the help provided by Masum Reza and Selina Shelley in
the translation of the Bangla pamphlet on which some of the discussion in this chap-
ter is based. Thanks also to Nandini Gooptu for general suggestions and to Sumeet
Mhaskar for timely provocations. The usual disclaimers apply.
2 All conversations, unless otherwise stated, occurred between 28 and 30 December
2009.
169
INDRAJIT ROY
Evidently admiring every word that Lerka spoke, Shyamsundar told

me, ‘You are living through important times. If you stay through till
autumn, you will see us reclaim our history.’ Lerka was of the Saotal
community, a self-consciously Adibasi community that is classified as
a Scheduled Tribe by the Indian government. Shyamsundar, however,
was of the Desiya community, most members of which considered
themselves ritually ‘cleaner’ than the Adibasi and Dalit communities
with whom they shared the locality.
Shyamsundar Sarkar was not exaggerating. It appears that during
autumn 2010, in conjunction with local organisations such as the
Majhi Pargana Gaonta, the Mulnibasi Samiti organised the first-ever
Asur Memorial Day to commemorate the martyrdom of Mahishasur,
portrayed in several mythologies as an evil demon. However, some
members of the Asur and Yadav communities regard him as an egali-
tarian and a just king. The following year, a poster was circulated in
Delhi’s premier Jawaharlal Nehru University, extolling Mahishasur’s
virtues and condemning his assassin, worshipped by many Indians as
the goddess Durga. The poster was accompanied by an article titled,
‘Who are the bahujans worshipping?’, and called on the ‘backward
classes’ to cease worshipping Durga, the annihilator of their hero (The
Telegraph, 1 November 2011). The piece was authored by Prem Kumar
Mani, Hindi litterateur, an ex-parliamentarian and one-time activist
of the Communist Party of India (Marxist-Leninist/Liberation). I am
unsure if the commemorations were observed in 2011, but YouTube
clips available on the internet suggest that they certainly were held in
three locations during the autumn of 2012.3 And a recent report on
social media (Roy and Biswas 2013) suggests that, at least in Maldah,
the observances continued this year, organised under the auspices of
the Mulnibasi Samiti and the Majhi Pargana Gaonta. If anything, the
observances appear to have proliferated, with reports of observances
from at least 15 districts across West Bengal, Bihar, Jharkhand, Uttar
Pradesh, and Odisha (The Indian Express, 14 October 2013).
In this chapter, I want to explore what it might mean for Shyam-
sundar Sarkar, and other activists affiliated with the Mulnibasi Samiti,
to say that they were reclaiming history. What meanings are made of
these narratives by young men such as those I had the opportunity
3 See https://www.youtube.com/watch?v=NyXE2KRaySY. Charan Besra suggests in

the clip that concurrent memorials were organised in Maldah’s Habibpur block,
Puruliya and Shodhpur, Kolkata.
170
T R A N S F O R M AT I V E P O L I T I C S
to meet? What subjectivities are inhered in these meanings? Of parti-

cular interest is the transformative idiom engendered by these narra-
tives and subjectivities, and the manner in which these are deployed by
young activists such as Sarkar, Hembrom and their friends. Much of
the literature on transformative politics in West Bengal has tended to
focus on class, with a near-complete silence on the salience of caste as
a trope of discrimination as well as of mobilisation. However, the rec-
lamations of history about which my interlocutors in Ditya enthused
were underpinned by the assertion of caste equality. That the language
of caste emerges as salient despite the language of modernity perme-
ating the state’s cultural and political spaces for nearly two centuries
appears somewhat paradoxical. Indeed, the near-total silence of the
caste question in West Bengal’s contemporary polity has been cited as
evidence of progressive politics and political modernisation by authors
such as Atul Kohli (1996; 2012), although others, such as Baviskar
and Attwood (1995), have been less sanguine about it.4
Modernist pretensions notwithstanding, caste remained a central
idiom of the state’s social life. Indeed, even as Bengal’s political and
social elites claimed to have imbued modernist principles, my inter-
locutors suggested they perpetuated a regime of discrimination and
marginalisation vis-à-vis the lower classes (nimno borno/chhotolok),
even after the establishment of communist rule. Indeed, the all-
pervasiveness of the ‘modernity’ idiom effectively prevented the state’s
Dalits from recognising, labelling and interrogating what are often
disparaged as primordial identities. They invoked comparisons with
Dalit movements in Uttar Pradesh and Bihar to direct attention to
the relative voice asserted there by marginalised groups against caste
discrimination – a phenomenon often thought to be absent from this
state. While the relationship between modernity (real or spurious) and
the persistence, entrenchment and reproduction of caste in West Ben-
gal is discussed elsewhere in this volume, my analysis focuses on the
transformative politics being advanced by a new political generation
in rural West Bengal.
Ernesto Laclau (1994) reminds us that the struggle over meanings
lies at the heart of politics: political interventions perform a consti-
tutive role in the making of meanings, the forging of subjectivities
4 Their contrasting views become clear from their opposing interpretation of the fact
that the privileged castes in the State increased their representation in the state cabinet
from 70 per cent during 1952–62, to 90 per cent during 1977–82.
171
INDRAJIT ROY
as well as the crafting of identifications. Speaking of identifications

helps me to emphasise the continuous and fragmented nature of self-
making, without folding the process into a teleological narrative. Fol-
lowing from this insight, I want to direct attention to the manner in
which members of a self-consciously ‘clean’ community such as the
Desiya engage with the discourse of being Mulnibasi and seek affilia-
tion with it. Is their engagement with and appreciation of Mulnibasi
politics a calculated use of identity politics by a few well-educated
individuals? Or, is it a discovery of who they are and an affirmation of
their cultural difference? Or, as I am compelled to argue, based on the
material presented in this chapter, does it represent a political genera-
tion’s shared attempts to found a universal political community? I will
explore these themes drawing on fieldwork I conducted during the
winter of 2009–10 in one rural ward in Maldah district’s Old Mal-
dah block. My fieldwork included elite interviews, group discussions,
observations, as well as ethnographic ‘hanging out’.
The ‘agony’ of being: the contradictory affirmations

of Desiya self-hood
Members of the Desiya community with whom I met during my
research disagreed vigorously with one another on the origins of their
community. Some of my interlocutors claimed to be a southern branch
of the more numerically prominent Rajbanshi caste. Others averred
that they were a branch of the Polia community. The Polia community
had apparently fled north-east to protect themselves from the invasion
of Parshuram, the mythic hero of the Brahman community. Those who
returned to their ‘original homeland’ (desh) called themselves Desiya.
These accounts were laughed at by my interlocutors from other com-
munities. Patras-babu, a clerk in a local office and who identified his
community as Chere (Dalits among Saotals, as he said) elaborated:
Don’t believe a word they say! They were thrown out of their
‘homelands’. Maybe in China or maybe in Mizoram . . . I
don’t know. Because they ran away, we call them Polia. Of
course, once they came here, they started calling themselves
Desiya, to claim that this is where they had always lived.
Under such widely contested interpretations, the finality with which

colonial officials claimed that the Desiyas ‘originated’ further east
(Beverley 1872: 183) is difficult to sustain. Writing in Bangla, Pradyot
172
Ghosh (2007) reminds us that the colonial account was only one of
several, and needs to be considered as such. My Desiya interlocutors
scorned Beverley’s (and Patras-babu’s) suggestions that they originated
elsewhere. However, people such as Lerka Hembrom shrugged their
shoulders and countered my investigations by asking, ‘How does it
matter where they “came” from? Is it not enough that they live here?’
What all my interlocutors as well as the colonial and postcolonial
observers agreed upon was that the Desiyas considered themselves
‘cleaner’ than the Rajbanshis, Polias and Saotals. Their rules of com-
mensality forbid them to dine with members of any of these other
communities – a rule that continued to be followed even after the
communists took over power in West Bengal, and the local panchay-
ats passed into Desiya hands. An important, if somewhat intriguing,
marker of cleanliness, was centred on the consumption of pork. The
Desiyas considered it unclean. However, pork was freely consumed by
households of the other aforementioned communities. That justified,
according to the elderly Desiya men with whom I conversed, the bar
on intercommunity commensality.5 Other rules forbid intercommunity
connubiality. That these rules have been in place since at least the last
one century is testified by accounts such as that presented by noted
scholar Nagendranath Basu in his magnum opus, Bishwakosh.6
A recurring theme during my conversations with middle-aged and
older (individuals aged over 40 years) men of the Desiya community
pointed me to an ongoing controversy over a temple being constructed
within the ward. Some members of my host family and a number of
other Desiyas in the ward were in favour of constructing a temple:
‘Every village should have a temple’, Shyamsundar Sarkar’s father told
me. They said there had never been a temple in the ward, and all
their observances were focused on a huge banyan tree that marked the
heart of the village. A few years ago, members of the Sanatan Dharma
Sabha had come visiting and had been appalled by the absence of a
5 The abhorrence of pork as an item of food is at once surprising, and yet, not unex-
pected. It is surprising because pork is not conventionally a marker of being ‘unclean’,
except among Muslims; and the Desiyas emphatically denied any Muslim influence
over their dietary habits. It is not unexpected, given the ‘cultural intimacy’ (following
Herzfeld 1995) that spatial proximity facilitates. Such cultural intimacy as might exist
between the Desiyas and the Muslims does not, of course, translate into intercommu-
nity commensality. Furthermore, the cultural intimacy between the Desiyas and the
Saotals has not encouraged the former to take up pork or the latter to give it up.
6 Bishwakosh is an encyclopaedia covering 22 volumes and published over a period of
22 years during the late 19th and early 20th centuries.
173
INDRAJIT ROY
temple in the village. They had then offered to build a temple, but the
ward’s elders were ambivalent. As committed communists, they had
not been too enthused about giving too much publicity to a religious
matter. As elders, they worried that the customary observances, such
as Nabanna (the harvest festival), would be ignored, and their own
authority undermined. Rebuffed, the Sabha left, but the idea that a
temple should be constructed gained a sympathetic ear among many.
The issue was particularly popularised by leaders of the Congress, who
launched a vicious campaign against the communist leadership, accus-
ing them of being atheist. They made this an issue during the ensuing
campaign over the 2008 panchayat elections. Although the commu-
nists won these elections, their margin of victory was far slimmer than
what it had been in the past. Many leaders attributed this narrow-
ing margin of victory to their opposition to the temple, although it
needs to be borne in mind that a broader wave against the CPI(M)
swept across the state during that time. When Congress leaders began
to collect public contributions for the temple, the communist leaders
actively supported it, and indeed, tried to outperform their opponents
in displaying their religiosity. When I interviewed the veteran of the
CPI(M) in the ward, he sighed and told me, ‘Look, after all we are all
Hindus. And I am growing old. I will meet my maker soon. What do
I tell him?’ As I will show subsequently, this attitude placed them at
odds vis-à-vis younger members of the Desiya community.
On the one hand, Desiya elders attempt to sustain the putative social
‘cleanliness’ of the Desiyas in terms that resonate with the language of
the caste hierarchy characteristic of the social landscape of West Ben-
gal. The increasing articulation of their religious beliefs with temple-
building and other practices associated with Hinduism is also clearly
discernible. On the other hand, however, often, the same individuals
affirm their separateness from the wider Bengali society by highlight-
ing their linguistic and cultural practices. They proudly narrated their
marriage practices, which they asserted were different from what was
practiced elsewhere in West Bengal. ‘We are our own priests’, they
said. Desiya elders told me that no priest from the Brahman commu-
nity officiates Desiya wedding ceremonies. A priest from within the
community is invited to do so. No rituals involving fire are conducted.
Consequently, the marriage is confirmed with the exchange of grass
garlands. Connubial rules permit village endogamy.
Linguistically, the Desiyabhasha, the identity conferred by the
Desiya elders on their language was claimed to be different from the
Bangla language, which underpins the state’s identity. Offended when
174
I asked if Desiyabhasa was a dialect (anchalik bhasha), my host recited

the following verse and asked me if I, as a Bangla-speaker, could make
any sense at all of it: ‘Bishno re, kalo kalo kopila, Kalo ghas khaye – Na
jaane kopila kon ghaas, Kon kher khaye – Akuaar kanya, pathroch-
hare’. This verse is sung during weddings and I admitted that I did not
understand what it meant, but pointed out that many of the words,
such as ghas (grass), kanya (girl) and khaye (eat) were actually used in
Bangla. My host reminded me that similarity of words did not make
one language a dialect of the other. Would I say that, he challenged
me, Bangla was a dialect of Hindi or Urdu because so many words in
the three languages were similar? When I argued that these languages
used different scripts, my interlocutor was visibly irritated: ‘Hindi and
Nepali share the same script, as do Ahomiya and Bangla: does that
make them dialects of one another?’ he retorted.
The inherently unsettled nature of these identifications reflects the
agonism that underpins these processes of collective self-making. The
attempt to fit into a hierarchy of cleanliness sits tenuously with efforts
to maintain separation from this hierarchy. Are the Desiyas a caste
group or are they a linguistic group? By agonism, I mean the mutual
contradictions, yet imbrications that prevent the constitution of politi-
cal identification from being a matter to be settled through either stra-
tegic calculation (on the one extreme) or antagonistic practices (on the
other). Here, I am thinking about agonism in the sense of a ‘permanent
provocation’ (Foucault 1982), a reciprocal incitation between affirm-
ing a place within caste society, and asserting autonomy from it. This
incitation provokes a conflict between the contradictory pulls of such
identifications. It foments tension between a vision of the community
that is absorbed within a putative hierarchy and an imagination that
posits itself as equals to other similarly categorised communities. The
questions they impel members of the community to ask are very dif-
ferent. In the former instance, the key questions before members are:
where are we on the caste hierarchy; and, are we ‘cleaner’ than X or Y
community? In the latter instance, these questions include: how do we
affirm our uniqueness as a linguistic, or even religious, community?
The agony of becoming: the discourse of

the Mulnibasi
The engagements of the Desiya youth with the Mulnibasi Samiti could
only aggravate this agonism. They were not merely additive to the
ongoing debates within the community, but introduced, as we shall see,
175
INDRAJIT ROY
a substantively different element to them. The Mulnibasi Samiti in

West Bengal is the state body of the Rashtriya Mulnivasi Sangh
(National Mulnivasi Organisation), which functions as the cultural
wing of the All India Backward and Minority Communities Employee
Federation (BAMCEF). The BAMCEF was founded as a trade union in
1983 by the charismatic statesman Kanshi Ram in order to provide a
national-level platform for the public sector employees of the margin-
alised communities. BAMCEF and the National Mulnivasi Organisa-
tion share the same political slogans: Jai Mulnivasi, or victory to
Mulnivasi.
In its imagination, the figure of the Mulnivasi refers to the autoch-
thonous inhabitant of India. Who is the autochthonous inhabitant of
India? The answer is stated crisply on the BAMCEF website: the mem-
bers of the Scheduled Caste, Scheduled Tribe, Other Backward Class,
and religious minorities who have been converted from these clusters
of communities are included in the discursive figure of the Mulnivasi.
The West Bengal-specific website of the Mulnivasi Sangh addresses its
‘SC, ST, BC’ audience to clarify the excess of the population that is
not Mulnibasi: the Brahmins, the Kshatriyas and the Vaishyas, with
supposedly Eurasian antecedents. However, this certainty dissolves in
the face of engagements and negotiations that activists face on the
ground in their negotiations with people. It is not my intention here
to engage with questions of accuracy or objectivity in the Mulnibasi
narrative. For me, a more interesting question is the manner in which
collective meanings are forged and what they tell us about the way in
which political identifications are being crafted. For, the figure of the
Mulnivasi/Mulnibasi encapsulates shared meanings of social discrimi-
nation and oppression.
The Samiti’s work in Maldah was spearheaded by a retired police
officer who had worked assiduously over the previous decade to
build support among the youth for a vision of social transformation.
The Samiti publishes a booklet, which its members used as peda-
gogical aids to help their members educate the general population.
Samiti members use the booklet to foment discussion among their
audiences, which tends to be young men and women with basic lit-
eracy and numeracy skills. Provocatively titled Swadeen Bharater
Paradeen Bahujan (The Dependent Masses of Independent India,
hereafter SBPB), the booklet is priced at an affordable INR 10. Its
subtitle Alpajan Banam Bahujan (The Few against the Many) sig-
nals the incitements that the authors try to provoke. The young men
176
of the Desiya community, who purchased, read, and discussed the

booklet all agreed that they found the booklet not only disturbing
but also eye-opening.
What does it mean to pit the few against the many? It is conve-
nient for analysts to interpret this to refer to the division between
the minority and the majority, potentially building off existing reli-
gious cleavages that are often framed in these terms. However, in
the booklet, the use of the term bahujan approximates the Ambed-
karite appropriation of the Pali term to refer to the people in the
majority, particularly the vast numbers of the different ‘low’ and
‘untouchable’ caste communities who lead lives of precariousness
and face social discrimination on a quotidian basis. Given the over-
laps between the descriptions of the bahujan and the Mulnivasi in
the booklet as well as the associated literature, it is reasonable to
assume that the two terms are used interchangeably in the text. In
this sense, it is a way of thinking about ‘the people’, but with an
important qualification. The term ‘bahujan’ assumes the existence
of an antagonist, the other, the few, the alpajan. It is, thus, distinct
from the word janata, which refers to the people as indivisible. Such
an assumption of indivisibility is absent in the understanding con-
veyed by the term ‘bahujan’, with the result that it is more cogni-
sant and realistic in thinking about conflicts that, in reality, divide
the people. The authors of the booklet explicate the antagonistic
frontier by emphasising the opposition between the 15 per cent and
the 85 per cent. The 85 per cent refers to the combined propor-
tion of the Mulnibasi communities (the aggregate of the three caste
clusters described above), while the 15 per cent includes the others.
For good measure, the authors of the booklet specify that they call
for the freedom of the Mulnibasis from the Brahmon Raj – the rule
by Brahmans, who comprise 2 per cent of the population of West
Bengal.
Although on the surface, the authors of the booklet appear to
provoke antagonism, the first section is a call for an ideological bat-
tle (bichardharar lodai). The contents of that section, however, are
far from antagonistic. Rather, they appear to be a call to initiate a
conversation. Stanley Cavell (1979) reminds us of the importance of
the ‘conversation’ in the ‘founding’ of a community. A conversation
is about expressing an imagination of membership in the political
community and an attempt to work the terms of such member-
ship. A conversation in this sense is not a solipsistic enterprise, but
177
INDRAJIT ROY
refers to a collective negotiation between members of the political

community.
This section does not assume as given a Mulnibasi identity. In fact,
the term is not used anywhere in that first section. Here, the attempt
is to try and negotiate the terms of the founding of the political com-
munity. The Mulnibasi is not a figure whose constitution is settled. His
or her identity is not a priori. The figure of the Mulnibasi is not a fully
fixed one – one which can be mobilised to leverage strategic bargains.
The contents of the first section refer to wide-ranging claims and con-
cerns that are not limited to the Mulnibasis alone. They explicitly
grapple with the multiple dimensions of the problem and make no
effort to flatten the narrative. The apparent certainties that inform
the concerns of the Mulnibasi organisation seem to flounder in the
face of the myriad themes taken up for discussion in the pamphlet.
For instance, the authors discuss themes related to religion, education,
wealth, and equality. This is quite unlike the singular and fully formed
dimension of identities invoked by proponents of identity politics.
These themes are then taken up for detailed discussion in the histori-
cal and contemporary contexts.
An examination of this booklet compels me to distinguish my
approach on identities from the two predominant intellectual streams
of thought on identities. Against the liberal-individual view of identi-
ties, popularised in the writings of John Rawls (1971) and Amartya
Sen (1999), I argue that identifications are not always a matter of
reasoned preferences, to be strategically invoked by aggregates of
individuals in a ‘voluntaristic’ and ‘solipsistic’ way. Against the com-
munitarian view of identities, propagated most famously in the writ-
ings of Michael Sandel (1982) and Charles Taylor (1987), I argue that
identities are not ‘given’ and immutable, waiting to be ‘discovered’.
Rather, the booklet and the engagements of the Desiya youth with
it resonate with the perspectives of authors who view the construc-
tion of identities as a consequence of political interventions (Cavell
1979; Tully 1995; Laclau 2005). It is through the historical narrative
presented in the third section of SBPB that the figure of the Mulnibasi
emerges. In concluding this section, the authors urge their readers to
overcome these gradations, stratifications and differences. It asks of
them to eschew violence against one another, caste distinctions, and to
identify (ektai porichoy) themselves as ‘Mulnibasi’ (SBPB n.d.: 6). The
figure of the Mulnibasi is ‘disclosed’ (Cavell 1979: 27) at the very end
178
of the historical narrative through the making and staking of histori-

cal claims.7
Predicaments
The engagement of young men of the Desiya community with the
Mulnibasi Samiti becomes intriguing in light of all the provocations
apparently directed against the upper castes and considering that the
elders of the Desiya community were explicit about their ‘clean-caste’
status. What motivated people such as Shyamsundar Sarkar to associ-
ate with the Samiti and its activities? I had the privilege of meeting
nearly 15 young men of the Desiya community, aged 25–35 years,
who were involved in the Samiti’s activities. Raju and Lerka teased
Shyamsundar about being with the Samiti because the Desiyas were
trying to get themselves registered as Scheduled Caste: this would enti-
tle them to the government’s affirmative action policies. Lerka
explained, ‘The Desiyas are well educated. Certainly more than we
(Saotals) are. So, they have a fair chance of bagging all the jobs if they
register themselves as SC.’ Raju Pramanik, Shyamsundar’s best friend,
added, ‘The Desiyas have all the privilege in the private sector. Now
they want all the privilege of the public sector.’
Lerka’s and Raju’s friendly banter with their Desiya friends might
appear to carry more than a grain of truth, especially for social scientists
7 I want to make clear that verifying the factual accuracy of the claims advanced
through the text is not of interest to my endeavour here. Among the several excel-
lent accounts of the manner in which caste hierarchies came to be ‘settled’ as late
as the 19th century, reference must be made to Susan Bayly’s (1999) seminal
work. Bayly’s approach draws on Dumont (1970), but differs significantly in that
it is a more historicised and concrete account. However, Dirks (2001) has argued
that caste identities were a product of colonial governmental regimes. Beteille
(1997) rejects Dumont’s notion completely and locates the hardening of caste
hierarchies in the feudal economy of Medieval India. What is interesting is the
way in which the figure of the Mulnibasi is invoked through a retelling of Indian
history, particularly a rendering of history that is sensitive to caste-centred social
conflict. While the SBPB borrows freely from the categories deployed by the
Indian state, it seeks to create a political community by forging a unity among
members identified by the state through these categories. These categories are
defined in relation to their interaction with the state as the product of colonial
and postcolonial governmentality.
179
INDRAJIT ROY
who believe that individual and collective preferences are determined

on the basis of calculations of returns on investments. Indeed, many
of my elderly interlocutors from the Desiya community rued the fact
that they had not registered themselves as Scheduled Castes earlier:
‘We were too stuck up about our shomman (honour), but we lost out
on jobs.’ Some of the more enterprising Desiyas had actually managed
to purchase certificates that registered them as Scheduled Castes, and
on the production of which they would be entitled to the benefits to
which members of this community are entitled. But many others were
contemplating a community-wide application to request the govern-
ment to enumerate them as Scheduled Caste. They sought advice from
their neighbours from the Bind and Napit communities, both of which
were classified as Scheduled Castes. I was present at a Gram Unnayan
Samiti where the convenor, a 58-year-old Desiya gentleman, told his
colleagues from these communities that they were going to make a
representation sometime soon: ‘What can we do? We can’t hold on
to our honour for too long.’ To which his interlocutors – all of whom
were at least 15 to 20 years his junior – without a trace of irritation or
antipathy replied:
Kaka (uncle), you have to understand, your honour will not

feed you. You have to let go of these things. Think of the
future. Think of your grandchildren. This is a new age. Your
children need jobs. Honour will follow.
This exchange was very interesting for the very frank way in which
members of the Desiya asked their neighbours for advice, which was
freely offered. The Dalit members were trying to convince their inter-
locutor with respect, but not deference, that they had to give up their
time-honoured traditions for their own good.
However, the quest for Scheduled Caste certificates and jobs cannot
explain the depth of engagement that the young men of the Desiya
community seem to have with the Mulnibasi Samiti. The Binds and
the Napits, who have been Scheduled Castes for many decades now,
have no affiliation with the Mulnibasi Samiti and are self-consciously
Hindu. The 15-odd youth from the Desiya community, however, have
absorbed themselves in the booklet produced by the Samiti and meet
regularly to discuss its contents. They seek Lerka for clarifications and
debate with him and with others. One of my earliest conversations
with Shyamsundar Sarkar was on religion: once he had verified I was
180
not Christian,8 he asked me what I thought of belief, faith and god. He

said he was atheist and was interested in knowing about Buddhism:
That was the only ‘rational religion’ (juktipurno dhormo), he said.
His room had at least one picture of the Buddha pasted on the inside
of the door.
Over the next few days, as I met more young people from the Desiya
community, it became clear that those who were affiliated with the
Mulnibasi Samiti were deeply interested in Buddhist philosophy. Given
the contents on religion in SBPB, this was not entirely surprising: the
sixth proposition in the first section is a quote from the Buddha, the
only religious quote in the booklet: ‘Know, and only then believe.
Know the truth. Know falsehood’, it says. In the historical section, sev-
eral passages refer to the emergence of Buddhism, its flourishing and
its eventual destruction during the reign of Pushyamitra Sunga (SBPB
n.d.: 4). Members of the Desiya community had an intimate knowledge
of the rituals and observances of their Saotal neighbours, to which they
were regularly invited. But not too many people of the older generation
knew (or cared) about Buddhist philosophy and observances. I am not
sure whether Shyamsundar or any of his friends were practicing Bud-
dhists, but their interest in Buddhism was evident. If all they wanted
was a Scheduled Caste certificate, they need not have engaged so deeply
with Buddhism or with the activities of the Mulnibasi Samiti in general.
The struggle over meaning

By invoking the figure of the Mulnibasi, the authors of SBPB compel
readers to confront deeply shared intersubjective frameworks of subor-
dination and particularism that bind Mulnibasi communities to a
homogenously conceived Hindu culture, with Brahmans at its apex.
They provoke readers to question this framework. An adherence to this
framework has generally led proponents, reformers and antagonists of
caste-based privileges to focus their claims on a putative Hindu culture.
8 In this part of Old Maldah, I came across a ubiquitous suspicion of Christian mis-
sionaries. The knowledge that I was affiliated with a university in England fuelled the
suspicion that I might be one. I heard of various stories about missionaries ‘brain-
washing’ local people. In one strange encounter, a local physician asked me if I was
Christian. I told him I was atheist. He was extremely apologetic, and said, ‘Of course,
that’s entirely your personal belief. I don’t need to know. I just wondered if you were
Christian’. I am still wondering what to make of that comment.
181
INDRAJIT ROY
The orthodoxy, which supports the perpetuation of caste hierarchies,

claims that these are central to the Hindu culture. Reformers point to
the myriad cultural traditions within Hinduism and point to social
activists who preached egalitarian values among putative Hindus. Even
antagonists call for a theological break with Hinduism, favouring con-
versions to other religions. Hinduism – and the role therein of the
Brahman – remains central to the political imagination of all three.
SBPB, however, dislodges the privileged position accorded to Hin-
duism in all these discursive formulations. In two succinctly worded
paragraphs, it dismisses the grand narrative of Hinduism, thereby
destroying its unifying edifice:
There is no such thing as Hindu religion. There never was.

In the Vedas, Upanishads, Brahminical texts, autochthonous
narratives, Ramayan, Mahabharat – nowhere do you find any
reference to the word ‘Hindu’ . . .
The Muslims [kings] referred to the defeated people of this
country as Hindus. ‘Heen’ means defeated. The Brahmans
never considered themselves Hindu, because they thought of
themselves as Aryans and foreigners. When the jaziya was
imposed on the Hindus, the Brahmans were exempt from it.
The term ‘Hindu’ is an abuse. The Brahmans colluded with
the Arabs.
(SBPB n.d.: 5)
Among the discursive moves that dislodge the grand narrative of

Hinduism, the most innovative is the attempt to disaggregate the
very term ‘Hindu’ and to emphasise the negativity and the humili-
ation associated with it. Emphasising the perspective that the term
‘Hindu’ is inextricably linked with defeat allowed readers to engage
critically with its unproblematised usage. Shyamsundar Sarkar nar-
rated the sense of helplessness when he read this passage. His best
friend, Raju Pramanik, at whose shop we would often hang out to
chat, said that he did not believe a word in that passage when he
read it:
But then, we thought about it. Shyam and I, and [three other
young men]. And we thought that these people (the authors
of SBPB) have a point. We don’t have the word ‘Hindu’ in any
of the mantras. Where did it come from? And why do we call
ourselves that?
182
I find the above-cited passage instructive because of the way it dis-

sociates the Brahmans from the Hindus and associates them with the
Arabs. The very notion of the Brahman has been central to grand
narratives in Hinduism. Not only do the authors of the SBPB dislodge
the Brahman from his privileged position, they accuse them of siding
with the invaders (as the Brahmans themselves were). The antagonistic
frontier is redrawn in such a way that Brahmans and Arabs are on the
same side of the frontier pitted against Mulnibasis of different faiths
and communities. In a border district with a slight Muslim majority
(52 per cent), where the Hindu–Muslim cleavage has been the most
salient one since at least 1946, this way of thinking about social divi-
sions is radical indeed. The unsettling of the hitherto antagonistic
frontiers makes the struggle over meanings even more complex, for
it dislodges long-held interpretive frameworks and dislodges ways of
making meaning (‘We were told that Hindus were different from Mus-
lims. But how come the Brahmans always bagged plum jobs under the
Arabs and other Muslim invaders?’). If the ensemble of beliefs that
constitute ‘being’ Hindu are extricated from that overarching frame-
work, and the people supposedly at the apex of it were really never
there, it becomes meaningless to speak about Hinduism at all.
Another theme whose meanings were sought to be reclaimed was
that of representation. The passages on affirmative action for Mulni-
basis asked whether it was fair that ‘reservations for 85% of India’s
population were being determined by 15% of the population’ (SBPB
n.d.: 6). A particular focus of discussion was the situation of the OBCs:
at 52 per cent of the population, the authors of the booklet inform
their audience using statistics of the Report of the Mandal Commis-
sion, they were entitled to only 27 per cent of all public jobs. ‘It is not
clear under which constitutional provision has 52% of the population
been restricted to 27% representation?’ (SBPB n.d.: 7).
In an important discursive move, the authors of the booklet illus-
trate the way in which the state extended this limited concession with
one hand and took it away with another through the liberalisation,
privatisation and globalisation (‘LPG’, written in English in the book-
let, which is then explained – once – as Udarikaran, Besharkarikaran,
Bishwayan) of the economy. The loss of jobs in the public sector as a
result of the reforms meant that the policy of affirmative action was
meaningless. The downsizing of jobs in the public sector, readers are
informed, accompanies the withdrawal of state support to secondary
education. Therefore, ‘the children of one group partake of “mid-day
meal” education [referring to the subsidised school feeding programmes
183
INDRAJIT ROY
to which the poor are entitled] and the children of the other receive
English-medium education. Distance is inculcated from birth’ (SBPB
n.d.: 7). The complete quotation is instructive:
Now there are two Indias. One is the India of the LPG. The
other is the India of the NREGA.9 Those of the higher castes
(ucchoborno) have huge incomes thanks to the wealth created
by the LPG. Whereas for the others (even if they are graduates,
they have the right to work after all), they have NREGA . . .
Now there are no reservations for members of the SC/ST/OBC
community. What we have is reservation for Extraordinary
Rich People (ERP).
(SBPB n.d.: 7)
The struggle over meaning was particularly reflected in the way the
readers of the booklet grappled with questions of representation and
hierarchical difference. For instance, they wanted to understand from
each other how it could be possible that those who represented over half
the population of the country could expect to be content with just about
a quarter of all public sector jobs. As far as they could see, this implied
that 15 per cent of the population had ‘reserved’ (cornered) 49 per cent
of the jobs in the public sector.10 There was, of course, no question of
them ceding space in the private sector. Lerka Hembrom told the group:
At least we (STs) have our quota in proportion to our popula-

tion. Raju’s people here (OBC) have much less. And to think
they are the largest cluster . . . But I tell you, once Raju awak-
ens, no power on earth can prevent India from becoming a
true democracy.
9 The Indian government’s flagship National Rural Employment Guarantee Act

(NREGA), which provides employment for 100 days on public works to any household
that demands it. The booklet’s telling slogan, kaaj noye raaj chayee (We want power,
not work. If we get power, we will get work), has to be contextualised against the sus-
picion harboured by its authors vis-à-vis social protection programmes such as the
NREGA. My interlocutors interpreted such programmes as ‘Band-Aid’ measures insti-
tuted by the state to respond to what was nothing less than economic catastrophe.
10 About 24 per cent of public sector jobs were reserved for members of SCs and STs,
who make up 24 per cent of the population. About 27 per cent of jobs are reserved
for OBCs, who according to the Report of the Mandal Commission represent 52 per
cent of the population. That left 49 per cent of all jobs, to be cornered by the 15 per
cent of the population estimated to be the ‘Upper Castes’.
184
Clearly, the skewed quotas for OBCs were a point of concern for
my interlocutors, only one of whom, Raju Pramanik, was OBC. The
solidarity among members of groups classified by the government as
SC and ST with the travails of the OBCs is remarkable because it dem-
onstrates the incipient cohesion of a Mulnibasi identity.
In explaining the uniqueness of West Bengal and the near-complete
absence of any discourse of social equality, my interlocutors pointed
to the relatively low share of OBCs to the total population. This rep-
resented to them the single largest difference between the situation in
West Bengal and those of the northern states, where such tropes of
political mobilisation were common. The OBCs represent an amor-
phous collection of caste groups to refer to the castes derided in the
Brahmanical literature as ‘Shudras’. For India as a whole, it has been
argued that although individual members of some of the castes clus-
tered together as OBCs have amassed political and economic resources
in recent decades, the bulk of the OBCs continue to face several eco-
nomic and social deprivations, certainly compared with members of
the upper castes (National Commission for Enterprises in the Unor-
ganised Sector 2008). In the northern states of Uttar Pradesh and Bihar,
they have been at the forefront of the efforts of the discriminated and
marginalised communities in instituting what Gopal Guru has called
‘egalitarian protocols’ (Guru and Chakravarty 2005). By politicising
questions of social justice and dignity, politicians with support in these
communities have contributed in no small way to breaking the strangle-
hold of the privileged castes in politics (though, as readers of the SBPB
are told, OBCs continue to be systematically excluded from economic
benefits).11 In West Bengal, however, the relative numerical insignifi-
cance of the OBC population and its further classification into 100-odd
communities reduce the possibilities of them forming coherent electoral
blocs in order to contest the systematic discrimination practiced by the
privileged caste leadership of the different political parties in the state.
11 At the same time, some socially mobile caste groups have tended to emulate the
discriminatory practices of their erstwhile tormentors vis-à-vis the other Mulnibasi
groups (Ilaiah 1996), and actual experiences of coalitions between OBCs and SCs
have been bitter, as the experience of Uttar Pradesh has shown. A few writers have
expressed scepticism in the possibilities of recruiting members of the OBCs to the
cause of the complete dissolution of caste as a system of social hierarchy (Prasad
2003). Nonetheless, organisations such as the BAMCEF, Rashtriya Mulnivasi Sangh
and the Mulnibasi Samiti pin their hopes on the eventual coalition of the three great
caste clusters that will eventually dismantle the prevalent social hierarchies.
185
INDRAJIT ROY
The published articles and the observances noted at the beginning of

this article aimed to do precisely that: to craft an identity that would
encompass all Mulnibasis, and yet, retain the antagonism vis-à-vis the
privileged castes. Given that Mahishasura, the ruler in question, was
associated with the buffalo, the protagonists of the observances argue
that he was rooted in one of the Mulnibasi communities and that his
life deserves to be celebrated too. Unlike the privileged castes, pro-
tagonists of these observances are careful to note that these festivals
are observances and not worship, for that would not distinguish the
Mulnibasis from their tormentors. Calling for an observance presum-
ably makes it easier for Mulnibasis of the Muslim and Christian faiths
to join.
Affirming the universal

The question of representation (protinidhitva) came back again and
again. Who was represented in the population that benefitted from the
LPG economy? Who was represented in the ERP? In the NREGA? In
MDMs? This way of interpreting the lopsided distribution of privilege
and opportunity enabled my interlocutors to think about what it
might take to become what Lerka called a ‘true democracy’. The fact
that there existed a differential access to the benefits of liberalisation,
privatisation and globalisation – a differential access which resulted in
the poor being overrepresented in the MDM and NREGA – perpetu-
ated difference and inhibited the creation of a universal political com-
munity. The greatest obstacles to the creation of such a community
came from the ERPs, as they put it – the Extraordinary Rich People –
who were drawn overwhelmingly from the privileged caste communi-
ties. The Mulnibasis were the torchbearers – in the eyes of the readers
of the book – of the movement that would eliminate the hierarchical
difference that animates the Two Indias and achieve One India: so my
interlocutors suggested.
By raising the question of representation, the authors of SBPB
appear to encourage their readers to consider the ways in which their
concerns and interests were being (mis)represented. It encouraged
them to evaluate prevailing opinion. Publications like the one chal-
lenging bahujans to disregard Durga and observe the day of Mahu-
shasur’s martyrdom aimed to provoke them to think through these
themes even more closely: the searing critique to which the worship
of Durga was subjected called upon them to reflect on their past prac-
tices and give them up. These political interventions called upon a
186
new political generation to transform their social practices. They were

encouraged to eschew the present practice of allowing the alpajans to
represent them in the cultural domain. Indeed, they were being nudged
against conforming to the patterns and practices that were instituted
by the upper castes and were then projected in universalist terms.
‘Politically’, Aletta Norval (2007: 175) tells us, ‘conformism entails
a forgetting of the need to define oneself’. Against such conformism,
interventions such as those by BAMCEF, Prem Kumar Mani and the
authors of SBPB were instigating aversion, or the practice of represent-
ing and identifying oneself (Cavell 1990).
During one of the discussions around the booklet, one of Shyam-
sundar’s Desiya friends asked why it was that the term ‘general’ cat-
egory was applied to the 15 per cent alpajans, and the 85 per cent
bahujan were all scattered among the three different caste clusters.
He pointed out that the term ‘general’ usually refers to the norm, the
universal (sarbojonin), so there was no reason that the few should be
allowed to appropriate that epithet. Although he was also ‘of the few’,
he wanted to know how the others felt about being relegated to the
different categories. The conversation led to further consideration of
the chicanery of the Brahmon-Baidya-Kayasth troika. The exemplar
was the 1947 Partition of Bengal.
Thrity-five-year-old Sushobhan Sarkar, a Desiya school teacher,
pointed to the thesis that Bengal’s Partition in 1947 was, contrary to
popular opinion, enforced by the privileged caste Hindu minority. He
highlighted the manner in which the Scheduled Caste Federation and
the Muslim League sought to retain a United Bengal, and it was the
privileged caste leaders of the Congress, Communist Party of India
and the Hindu Mahasabha that had pressed for division. This episode
of the Partition is recounted in the booklet, and our group members
re-read the relevant passages to emphasise that the caste Hindus con-
spired to partition the province. The ability of the tiny privileged caste
Hindu population to partition Bengal in order to salvage its dominance
in the western half of the province was pointed out several times. My
interlocutors agreed with each other that Partition destroyed the unity
of the Dalits in the province – a unity that had, in coalition with the
Muslims, threatened the privileges of the privileged castes represented
by the Congress, Hindu Mahasabha and the Communists.
The figure of Jogendra Nath Mandal finds a place of prominence in
this context. Described as ‘mahapran’ (Great Soul), he is credited with
having invited Bhim Rao Ambedkar to contest the Constituent Assem-
bly elections from eastern Bengal. My interlocutors discussed among
187
INDRAJIT ROY
themselves that with Ambedkar’s victory, the privileged castes had no

option but to engage with him. They were enraged with the Dalits and
Muslims of eastern Bengal for having elected him and sought to teach
them a lesson. Bengal’s Partition was forced by them to penalise the
Dalits for having elected Ambedkar. Parallels in the political careers
of the two leaders were drawn. While Ambedkar became India’s first
Law Minister, Mandal became Pakistan’s. Both fell out of favour with
their respective prime ministers, and both resigned within a few years.
To my interlocutors, this did not appear surprising at all, as the lead-
ers of both the new nations were essentially derived from the same
sociocultural background.
The narrative of the privileged castes’ role in forcing the Partition
as a means of sabotaging the incipient alliance between Dalits and
Muslims in Bengal provides an important interpretive trope for under-
standing the continued lack of a united effort by the two communities
against the continued dominance of individuals from privileged com-
munities in the leadership positions of extant political parties. A prime
example of the continued policy of ‘divide-and-rule’ followed by the
privileged castes about which my interlocutors discussed passionately
was the massacre enacted by the CPI(M)-led Left Front Government
during the first five months of 1979 at Morichjhanpi, in the Sundar-
bans (Mallick 1999). Here, the government deployed armed Muslim
thugs against unarmed Namasudra settlers who, they alleged, were
violating a new law on tiger conservation and refugee resettlement.
Prospects for alliances between Dalits and Muslims receded consider-
ably thereafter, enabling the privileged castes to project their leader-
ship as one to which all sections of the state consented.
My interlocutors admitted that it represented a failure of the
leadership among both Dalits and Muslims that, despite each repre-
senting nearly a quarter of the state’s total population,12 they were
12 According to the Census of India/West Bengal (2001: Table A-10), the Rajbanshis
contribute 18 per cent to the Dalit population, while the Namasudras represent 17 per
cent. Other numerically significant Dalit communities include: Bagdi (15 per cent),
Pod/Pundra (12 per cent) and Bauri (5 per cent). This represents a considerably
fragmented demographic, as a result of which none of the Dalit communities are
able to appropriate the electoral system to their collective advantage. By way of
comparison, in Uttar Pradesh, the Chamars/Jatavs comprise 56 per cent of the total
Dalit population (Census of India/Uttar Pradesh 2001: Table A-10). Likewise, in
Bihar, the Dusadhs and Chamars each make up a third of the Dalit population (Cen-
sus of India/Bihar 2001: Table A-10).
188
unable to mount a serious challenge to the upper-caste leadership of

the state’s polity. They averred that the upper castes had always suc-
ceeded in projecting their particularistic interests as universal knowl-
edge. They narrated stories about the great reformer Guruchand,
who had established 108 schools for girls before Ishwar Chandra
Vidyasagar did. But the latter’s reforms are far more celebrated. As
they reminded me:
The stories they (‘upper caste’) Hindus tell us. Have you heard
the one about Vidyasagar trying his tiki to the lamp post and
studying (I nodded that I had heard). Don’t tell me you believe
it. He was in school long before the electric pole was intro-
duced. How could he have done anything like that?
In the same way, the celebration of the Durga Puja too was pro-
jected as a universal affair when it was really an observance of a cer-
tain group of people. As far as my interlocutors could see, the naming
of the upper castes in general terms was a logical extension of their
chicanery. It was time that a new political generation sought to change
that. Informing that change is the imagination of a new political com-
munity that young men such as Shyamsundar Sarkar, Raju Pramanik
and Lerka Hembrom are trying to forge. The important times that
Shyamsundar Sarkar mentioned reflected precisely this attempt to
advance the bahujan vision of history, politics and culture as the uni-
versal, and limit the presumptions of universality inherent in the dis-
courses of the alpajan. From their point of view, the advancement of
their cause was being resisted by the haughty arrogance of the elites of
the privileged castes.
Conclusion
Shyamsundar Sarkar and his friends’ interpretive engagement with the
Mulnibasi discourse appears to reflect the forging of a collective self-
hood that leads them to think about the possibility of ‘becoming’ Mul-
nibasi, yet ‘being’ Desiya. The permanent provocations attending to
ongoing contests of identification among the Desiyas are exacerbated
when these enmesh with the interventions of the Mulnibasi Samiti.
While discussing the pedagogical approach of the SBPB, I had pointed
to the disclosive nature of the Mulnibasi category: a figure that comes
into being through conversations with friends and opponents. The
sense of being different from the mainstream Bengali culture is, as we
189
INDRAJIT ROY
have seen, key to the Desiya communal identity. This identity, articu-
lated with the Mulnibasi promise to found a political community, is
perhaps what excites Shyamsundar and his friends about engaging
with the Samiti.
The authors of the booklet leave no doubt of the utopian dreams
they harbour. They are conscious of the imaginations that political
interventions must foment. Additional to promising jobs and benefits,
they call attention to those dreams, and why they should be pursued
at all. As they write:
We are peddlers of dreams. We want to show people a beauti-

ful dream. This is such a dream that if people do want, they
could make it a reality on earth. We want . . . people to love
each other (manusher proti manusher bhalobasha), hard
work (kothor udyom), rational (jukti’r madhyome) thinking,
and not belief . . . Only then will we be able to end the slavery
of the past thousand years. These dreams will then be fulfilled.
(SBPB n.d.: 14)
In an era when it is academic fashion to scorn emancipatory

thought and practice, it may indeed be difficult to sustain an appre-
ciative investigation of political interventions that so honestly seek
to advance such discourses. However, such investigation is of utmost
analytical and empirical value for social scientists who are interested
in developing a concrete understanding of the ongoing processes of
social change and political identification. Emancipatory thoughts and
the prospects of transformative politics they inhere continue to ani-
mate collective self-making. This is all the more so in a region when
universalist ideals of modernity and rationality have camouflaged the
particularistic interests of a privileged minority. I have drawn atten-
tion to the interpretive frameworks through which a new political gen-
eration is making meaning of their history, present circumstances and
transformative possibilities. The history of the Partition, the question
of representation and the contest over cultural tropes together provide
the constitutive materials for the forging of these frameworks.
I also drew cursory attention to the booklet with which my inter-
locutors in rural West Bengal engaged and emphasised the disclosive
nature of the Mulnibasi identity that was being forged through these
engagements. Together, these conversations hold out the possibilities
of founding an alternative political community, of agonistic identifica-
tions against social hierarchies, of affirming universal values, and of
190
imagining transformative politics. Recent pronouncements of Leftist

dissidents such as Abdur Rezzak Mollah point to the growing impor-
tance of these conversations in the electoral arena. Social analysts
ignoring these conversations do so at their own peril.
Bibliography
Baviskar, B. S. and Donald Attwood. 1995. Finding the Middle Path. Boulder:
Westview Press.
Bayly, Susan. 1999. Caste, Society and Politics in India from the Eighteenth
Century to the Modern Age. Cambridge: Cambridge University Press.
Beteille, Andre. 1997. ‘Caste in Contemporary India’, in Chris J. Fuller (ed),
Caste Today, pp. 150–79. Delhi: Oxford University Press.
Beverley, Henry. 1872. A Report on the Census of Bengal. Calcutta: Bengal
Secretariat Press.
Cavell, Stanley. 1979. The Claim of Reason: Wittgenstein, Skepticism, Moral-
ity and Tragedy. New York: Oxford University Press.
Cavell, Stanley. 1990. Conditions Handsome and Unhandsome. Chicago: Uni-
versity of Chicago Press.
Dirks, Nicholas. 2001. Castes of Mind: Colonialism and the Making of Mod-
ern India. Princeton: Princeton University Press.
Dumont, Louis. 1970. Homo Hierarchicus: The Caste System and Its Implica-
tions. Chicago: University of Chicago Press.
Foucault, Michel. 1982. ‘The Subject and Power’, Critical Inquiry, 8(4):
777–95.
Ghosh, Pradyot. 2007. Banglar Jonojati. Kolkata: Pustak Bipni.
Government of India/Bihar. 2001. Primary Census Abstract. New Delhi: Min-
istry of Home Affairs.
Government of India/Uttar Pradesh. 2001. Primary Census Abstract. New
Delhi: Ministry of Home Affairs.
Government of India/West Bengal. 2001. Primary Census Abstract. New
Delhi: Ministry of Home Affairs.
Guru, Gopal and Anuradha Chakravarty. 2005. ‘Who Are the Country’s Poor?
Social Movement Politics and Dalit Poverty’, in Raka Ray and Mary F. Kat-
zenstein (eds), Social Movements in India: Poverty, Power and Politics, pp.
135–60. Lanham: Rowman & Littlefield.
Herzfeld, Michael. 1995. ‘It Takes One to Know One: Collective Resentment
and Mutual Recognition among Greeks in Local and Global Contexts’, in
Richard Fardon (ed), Counterworks: Managing the Diversity of Knowledge,
pp. 118–36. London: Routledge.
Ilaiah, Kancha. 1996. Why I Am Not a Hindu? A Sudra Critique of Hindutva
Philosophy, Culture and Political Economy. Calcutta: Samya.
Kohli, Atul. 1996. ‘Can the Periphery Control the Center? Indian Politics at
the Crossroads’, The Washington Quarterly, 19(4): 115–27.
191
INDRAJIT ROY
Kohli, Atul. 2012. Poverty amid Plenty in the New India. New York: Cambridge
University Press.
Laclau, Ernesto. 1994. The Making of Political Identities. London: Verso.
Laclau, Ernesto. 2005. On Populist Reason. London: Verso.
Mallick, Ross. 1999. ‘Refugee Resettlement in Forest Reserves: West Bengal
Policy Reversal and the Marichjhapi Massacre’, Journal of Asian Studies,
58(1): 104–25.
Mouffe, Chantal. 2007. On the Political. London: Routledge.
Mulnibasi Samiti (SBPB). n.d. Swadeen Bharater Paradeen Bahujan. Place and
publisher unknown.
National Commission for Enterprises in the Unorganised Sector (NCEUS).
2008. Report of Conditions of Work and Promotion of Livelihoods in the
Unorganised Sector. Delhi: Academic Foundation.
Norval, Aletta J. 2007. Aversive Democracy: Inheritance and Originality in
the Democratic Tradition. Cambridge: Cambridge University Press.
Prasad, Chandrabhan. 2003. Dalit Diary: 1999–2003 – Reflections on Apart-
heid in India. Chennai: Navayana Publishing.
Rawls, John. 1971. A Theory of Justice. Cambridge: Harvard University Press.
Roy, Georgy Kuruvilla and Samata Biswas. 2013. ‘Asur Utsav Not Durga
Puja’. Round Table India, 29 October, http://roundtableindia.co.in/index.
php?option=com_content&view=article&id=7012:asur-utsav-not-durga-
puja&catid=119&Itemid=132 (accessed 13 January 2014).
Sandel, Michael J. 1982. Liberalism and the Limits of Justice. Cambridge:
Cambridge University Press.
Sen, Amartya. 1999. Reason before Identity. Oxford: Oxford University Press.
Smith, Anne Marie. 1996. Laclau and Mouffe: The Radical Democratic Imag-
inary. London: Routledge.
Taylor, Charles. 1987. ‘Interpretation and the Sciences of Man’, in Charles
Taylor (ed), Interpretive Social Science: A Second Look, pp. 33–81. Berke-
ley: University of California Press.
Tully, James. 1995. Strange Multiplicity: Constitutionalism in an Age of Diver-
sity. Cambridge: Cambridge University Press.
192
9
FROM CLIENT TO
SUPPORTER
Economic change and the slow change of
social identity in rural West Bengal
Arild Engelsen Ruud
What is caste, really? This is not an easy question. We need to distin-

guish between varna and jati, of course, but need also to look at the
totality of social practices that make up what people experience as
caste. For beyond ritual status, I propose, caste may be seen as a set of
social practices embedded in local history and culture: local discrimi-
natory practices, wealth and landownership patterns, ideologies of
equality or hierarchy and the many in between, and local or state-level
patterns of political mobilisation. These practices vary over space, giv-
ing rise to many different kinds of caste systems that are in constant
evolution and processes of negotiation with other social forces. This
elasticity will also work over time, even relatively short time, changing
as other social forces are changing.
This chapter will investigate this proposition. The material is from a
village in West Bengal where the patron–client relationships that pre-
vailed until a generation or two ago were central in sustaining local
social inequalities of which caste was an element. With time, caste as
an ideology lost legitimacy. This happened with the increasing promi-
nence of the cultural-ideological development that we may term ‘the
Bengali modernist tradition’ – a development that helped give birth to
34 years of Left Front rule in the state and that was further enhanced
by that rule (Ruud 2003). In the more recent age of the welfarism
of the development state and neoliberalist economic changes, new
benefits and new opportunities have further improved peoples’ life
193
ARILD ENGELSEN RUUD
situations. These developments have undermined the clientilistic rela-

tionships of agricultural society and weakened caste as a social indica-
tor and as a practice.
Caste in the order of village society

The village politics literature of the 1950s and 1960s found a close
connection between caste and patronage relations. The dominant
caste held both economic and political power – a position that was
made easier if the caste was not only numerically large but also of
relative high ritual status (Srinivas 1955). In the kingly role that a
landowner of a dominant caste had in village society, as the jajman in
some interpretations, he would, in his person, represent and protect
the whole of society and the others were service castes to him in
exchange for services and loyalty (Mayer 1958; Cohn 1990). The
value system that legitimised this unequal yet reciprocal relationship
constituted a fundamental ideological construct that underpinned and
regulated relationships between the haves and the have-nots in village
society, between the clean or high-caste patrons and the lower or low-
caste dependants. This construct was negotiable, of course (Davis
1983; Bhadra 1989), but could be evoked and employed by both sides if
potentially useful. The ideal king’s place in society may have been under-
mined by the introduction of democracy, although these same sources of
local pre-eminence often allowed dominant castes to convert their clout
into political power in the new democratic era (Frankel and Rao 1993).
The relationship of caste to patronage was never entirely clear-cut,
though. There were many variations of the role of the patron, the jaj-
man, mandal, periyar, or other such denominations. Moreover, the
rivalry between them gave rise to the ideologically awkward pres-
ence of factions. These formations were cleavages that ran through
the dominant caste and very often through dependent castes as well.
‘Lesser’ members of the dominant caste would also be clients to their
more dominant brethren.
However much caste identity was negotiable and however much
the construct of patrons and clients could be manipulated and con-
tain elements of reciprocity, there is ample evidence to suggest that
caste identity was central to one’s status and that low ritual status
was something to be avoided altogether, if possible. Powerlessness and
the ideology that identified you as belonging to a caste that was not
naturally or locally dominant together upheld the morality of your
subordination and the other’s superiority. What is interesting for us
194
FROM CLIENT TO SUPPORTER
here is that caste mobility in the form of sanskritisation was a mat-

ter not only of cultural change, but of economic opportunity. Whole
castes or sub-castes would endeavour an improvement in their lifestyle
once the economic means were within reach. In other words, caste was
a compound status that contained both economic as well as cultural
elements.
In this chapter, I will investigate what happened to caste when the life
situation of the poor and ritually low was improved and their depen-
dency ended. The material I present here, I would like to suggest, is
reasonably representative for large swathes of the West Bengali coun-
tryside. However, I make no claims for it to be a general proposition for
the entire state or for all its communities. I claim simply that sociopolit-
ical changes over the last half-century have brought about tremendous
multi-layered changes. One of the most eye-catching changes is that
dependency has come to an end, and with it, the importance of caste.
The case is from Bardhaman district, and we start in the late 1950s.
To do so, we need first to outline briefly two momentous processes
of change that were already underway as our story begins. Then we
shall address the local implications of these changes, both cultural and
political, but beyond all, economic.
The first of the momentous changes was in the delegitimisation of
caste as a pillar of society in Bengal. Caste ideology’s loss of legitimacy
in Bengal stems from several historical strands, and is perhaps most
lyrically grasped in Bibhutibhushan Bandyopadhyay’s novel, Pather
Panchali, translated as Song of the Road (Banerjee 1968). Here is a
clear rejection of ancient customs, particularly evident in the early
chapters and in the fate of the old aunt. Her life as married in the way
that was traditional in her caste, Kulin Brahmin, to a man she hardly
knew or ever saw, and the slow crushing of her hopes and dreams as
a consequence, served to expose society as backward and tradition
as cruel. The same generation of authors, Saratchandra Chatterjee,
Tarashankar Banerjee, Manik Bandyopadhyay, and others, all wrote
novels that explored superstition, oppression of women, moneylend-
ing, and casteism. These and similar ills all formed part of a way of life
that was abhorrent to the increasingly larger and more vocal society of
the progressive and educated in Bengal (Zbavitel 1976; Ruud 1997).
These novels were widely read and appreciated both before and after
independence, and slowly percolated into the countryside. The abhor-
rence of casteism, gender discrimination and activities such as mon-
eylending and the mamla business (bringing illiterate peasants into
debt and then use the courts to relieve them of land held in collateral)
195
ARILD ENGELSEN RUUD
was something the increasingly educated and often progressive rural

middle class took to heart. They were not many, perhaps, but they had
much influence because they formed part of the rural elite of landlords
and village leaders.
A second momentous development was the introduction of adult
franchise. The democratisation of India after independence consti-
tuted a teutonic shift in society. This impact is still being felt, and
still debated (Manor 2010; Piliavsky 2014). Politics has become less
a matter of vertical mobilisation of vote banks and more an issue of
horizontal appeal and the mobilisation of an increasingly indepen-
dent-minded and untrustworthy electorate. The ideals of equality have
become central to the political relation-building that politicians today
are engaged in and to their rhetoric.
With these two momentous changes in mind, how has caste fared?
What can we learn about caste and its role in society from a dia-
chronic comparison of huge changes? First, we look at the economic
differences between then and now, before turning to the political prac-
tice of caste.
From clientilism to disengagement

‘We used to sit around a lot.’
[amra boshe thaktum].
Elderly man from the Bagdi caste, about the off-season peri-
ods in the old days.
The perhaps most significant change in rural West Bengal over the last
five or six decades is the increased wealth, improved livelihood and
new economic opportunities for individuals and families. Statistics will
show that West Bengal has not done as well in this respect as many
other states, or indeed as well as even the average for the nation. How-
ever, such comparisons across regions hide the fact that the diachronic
changes, over time, have been tremendous. The changes that have hap-
pened in rural West Bengal are fully comparable to those of Punjab or
Tamil Nadu.1 The changes have significantly reduced the dependence
of the poorer sections of society on the richer sections. The changes
have also reduced the capacity of the richer sections of society to
1 For Tamil Nadu, see Djurfeldt et al. (2008).
196
Table 9.1 Landownership in percentage of total village land in 1957.
Land in Clean Sekh Namasudra Bagdi Muchi Santal Total

bighas castes
0–4.9 12.5 16.6 34.8 54.2 81.8 91.7 35.2

5–9.9 25.0 30.8 4.3 25.0 18.2 8.3 23.1
10–19.9 12.5 32.0 43.5 12.5 – – 25.0
20+ 50.0 20.5 17.4 8.3 – – 16.7
Total 100.0 99.9 100.0 100.0 100.0 100.0 100.0
N 8 78 23 24 11 12 156
Percentage 5.1 46.2 14.7 15.4 7.0 7.7 99.9
Source: Field data.
provide patronage to the poorer sections, and these are changes that
have taken place within the last few decades – within living memory, so
to speak.
The village we look to is a moderately sized one, located in the
middle of the vast paddy fields that is most of Burdwan district. The
village dominant community is the Sekhs, who are Muslims and who
account for about half the population. Otherwise, it is populated by
people belonging to the common castes of Burdwan district, the Sched-
uled Castes known by the names Namasudra, Muchi and Bagdi, and
the Scheduled Tribe called Santal, and some clean caste Hindu house-
holds. (I shall refer to all these various communities as jati or caste, as
the Sekhs in all interesting respects in this context behave like a caste.)
To appreciate the extent and significance of the changes this village,
along with innumerable others, has gone through over the last few
decades, we shall first look in some detail at the situation with regard
to landownership in 1957, and then, compare it to the changes that
were to come.2
In 1957, the difference in land ownership between the jatis was
marked (see Table 9.1). The higher end of society consisted of two
2 The field material for this chapter, including the statistics, have been collected during
several sojourns to this village. The older material, referring to 1957 in particular, is
from the village portrait written by one of the villagers as part of his training to
become a teacher. The carefully written and coloured notebook has been kept by his
son and was lent me. Other historical material had been carefully recorded by another
villager in hundreds of notebooks. These contained information on political events,
marriage statistics, landownership, village gossip, etc. Material referring to 1993 (see
Ruud 2003) and 2011 was collected by me with the aid of forthcoming informants.
197
ARILD ENGELSEN RUUD
groups. The clean castes (Brahmins, Bene [Baniya], Kalu, and Kayast-
has) comprised only eight households, but of these, four owned lands
in excess of 20 bighas. Only one family held less than five bighas.
The other jati was the Sekhs. Their 78 households in 1957 constituted
about half the village population. In terms of ownership, this jati was
spread over the whole spectre. They lived in three distinct neighbour-
hoods or paras: the Middle Para (next to the clean caste para), in the
South Para and in the North Para. Some 15 households owned in
excess of 20 bighas each. These mostly lived in the Middle Para. Five
of these households owned land in excess of 50 bighas and two in
excess of 100 bighas – one of whom was an absentee landlord. The
Sekhs in the North Para were mostly landless or land-poor, and the
Sekhs of the South Para somewhere in-between.3
At the other end of the village socio-economic scale, the two com-
munities of Muchi and Santal had very little to their name. Most of
them were, for all practical purposes landless.
In the middle were the two SC communities of Namasudra and
Bagdi. Both had some families in the well-off category. Two large and
joint money-lending Bagdi families owned between 20 and 30 bighas
each, as did three such Namasudra families. The vast majority of Bag-
dis owned less than 10 bighas and more than half owned less than five
bighas. The Namasudras fared somewhat better, with only every third
household owing less than five bighas and a fairly large proportion
holding between five and 10 bighas.
This pattern of stark inequality in land distribution leads us to ask
what consequences this had for the individual household’s capacity for
survival, and what that meant for its economic independence. First,
it is important to note that land was mostly used for paddy and that
there was mostly only one crop, the summer aman. In winter, a little
land might be used for vegetables or fruit for private consumption.4
At this point in history, in the late 1950s, five bighas or less would
be sufficient for a household of one or two. Any larger family would be
3 There was yet another jati, the Muslim Malliks. They comprised six households in
1957. They lived alongside the poorer Sekhs in the North Para. Although they, in
some ways, may be considered a separate jati (no intermarriage for instance with the
Sekhs, and very little interaction), I have, for simplicity’s sake, included them here in
the general Muslim category and see them as part of the North Para.
4 There were also the ponds, in which many had shares and that were used for fish
cultivation. But fishing was a specialised occupation and did not open for local
employment in the village.
198
dependent on additional income to meet its ordinary needs and even

more so for extraordinary needs – such as to tide one over after a
bad season or a period of illness, or to cover expenses for a mar-
riage. Families owning less than 10 bighas would mostly scrape by, but
would need additional funds for certain periods or events, depending
on the number of family members and workers in the family. It also
depended, naturally, on the yearly success of the crop and the health
of family members. However, for a household of four to eight adults
(joint families being the common pattern), the figure of 10 bighas was
held as a rough estimate for what it would take to carry a family over
from one year to the next.
The next category up consisted of people who could rely on their
own land from one year to the next without requiring access to addi-
tional income. They might, however, require access to additional
sources of money every now and then. It could be illness, for instance,
in particular if the working father fell ill, it could be the opportunity to
buy an additional cow or invest in a house, a marriage, or misfortune
such as a fire. Such extraordinary events could require fresh money,
and people in this 10–20 bigha category were still vulnerable in this
sense. Weddings were more of a burden among some jatis than among
others. The Bagdis still mostly practiced a more tribal-like form of
marriage in which the bridegroom’s family paid for the bride rather
than the other way around, whereas jatis such as the Sekhs, the Nama-
sudras and the clean castes regularly incurred debt in connection with
a daughter’s marriage. Families in the 10–20 bighas category were able
to secure loans on more reasonable terms to meet these needs.
The kinds of additional income that were available for the land-
poor involved forms of dependency. The majority secured additional
income working as day labourers, on day-to-day contracts or long-
term contracts of varying length. Most long-term contracts were for
a month or for a season, and only a few were hired for a year or lon-
ger. Contracts were normally confined to the work involved and the
breadth of the patron–client relationship was limited. The contracts
did not oblige the landowner to extend assistance in case of illness or
sudden misfortune. For the most part, those contracted were left to
fend for themselves in case of illness or misfortune. However, credit
could occasionally be advanced in part-payment to labourers in antici-
pation of the upcoming planting or harvest. Labourers thus contracted
were known as ‘tied people’, but these contracts were mostly for just
one season. Only labourers hired for a year or more could expect more
tangible patronage in exchange for being available as an all-purpose
199
ARILD ENGELSEN RUUD
labourer. In this monocrop land, employment on someone else’s land

was mostly available during planting or harvesting, apart from some
occasional weeding and threshing. Up to six months or more per year,
there was little work to be had. ‘We used to sit around a lot.’
Sharecropping was another important source of additional income,
in particular, for families with excess labour. Sharecropping was avail-
able for people of all classes, both landowners and land-poor. The
landlord letting out land would not necessarily be very rich (widows,
families with small children) and patronage would, in such cases, be
close to non-existent. Interestingly, in these parts, sharecropping had
largely ceased to be long-term arrangements over several decades and
generations, and was, instead, limited to two to three years. Legis-
lation that gave certain rights to the sharecroppers had been passed by
the state parliament (the 1956 Tenancy Act), and although implemen-
tation was very slack or non-existent, many landowners still saw rea-
son to be cautious. By the late 1950s, most large landowners actively
avoided letting sharecroppers develop any form of rights to the land.
Another bond with a potential for patronage and exploitation was
moneylending. Moneylending was not uncommon and it was fairly
profitable. The normal interest rate was 3 per cent per month with
land as the common collateral. In consequence, land was often lost
within these three years. There were some moneylenders in the village
who had made good profit in this way, but their number was limited.
The reason for this was simple. Landowners wanted labourers who
could work on their lands during the planting and harvesting season.
For this, they would want labourers they could trust. The interest-
ing social mechanism is that moneylending as a business made you
unpopular and you thus ran a risk of not being able to find reliable
workers. The most valuable asset a landowner had was his land, but
the land was worth little if the rice was not planted properly or har-
vested in time.
Highly unequal land ownership patterns, then, kept the poor in forms
of dependency. Their economic position was reflected in their appear-
ances: very simple and torn clothing, commonly a simple sari for the
women and a mere loincloth for the men, shirtless even in winter. They
lived in squalid and cramped houses, and, to better-off villagers, the
poorer paras were congested and unclean. All the poorer groups, but the
Bagdis, Muchis and Santals in particular, were known as avid consum-
ers of alcohol, and drinking bouts among the Bagdis especially often
turned violent. Their lives were tough, by comparison with their fellow
villagers. Off-season life was tough when money and food were scarce.
200
Fast-forward to the CPI(M) era

In 1977, the Left Front coalition won the election and took over the
government of the state. The coalition government was dominated by
the Communist Party of India (Marxist) (CPI(M)), which had a strong
presence in this village as it did in most of the countryside of the state.
The state government programmes on the rural front encompassed sev-
eral different initiatives that were also implemented in this village. First,
there was the land redistribution that took place around 1978–80. In
this particular village, 53 bighas or 2.8 per cent of village land was redis-
tributed. The reason more land was not redistributed was that the major
landowners had sold off their lands several years earlier, in anticipation.
Most redistributed land was given to landless Bagdi and Muchi families
and some to Santal, Namasudra and Sekh families.
Two other initiatives were more significant. All sharecroppers had
their contracts registered in the Operation Barga in 1978. They were,
thus, given security of tenancy and a larger share of the crop – legal
rights they had had for a while, but that from now on were being
implemented. Another initiative was the rise in the daily wages for
labourers. The positive impact of this was acknowledged even by the
local opposition. In 1977, the daily wages for a labourer had been
rupees 2, 1 kg of husked paddy, 10 country cigarettes, and some body
oil. The wages were raised on four different occasions over the next
few years, each time after a certain amount of conflict, and each time
the CPI(M) was on the side of the labourers even if the local party also
helped reach a consensus. By 1993, the daily wages had been raised to
INR 12 and 2 kg of husked paddy. Calculated in paddy equivalents at
going rates, the daily wages had doubled in a decade and a half.
The figures in Table 9.2 show that landownership patterns had
changed before 1981, but that changes in the subsequent period were
more substantial. And more so in the first 15 years than in the next
18 years. The ritually lower jatis were still underrepresented in the
high end of village landownership. Santals, Muchis and Namasudras,
all held less than their proportion of the village population. The Bag-
dis represented a different picture, though, and now held a portion of
the village lands roughly equivalent to their proportion of the village
population – about 15 per cent.
The clean castes, as a group, had seen their landholdings reduced
by about 40 per cent. Although they never constituted a very substan-
tial ownership group in this particular village, their landholdings in
2011 had become quite marginal. We may also note that the process of
201
ARILD ENGELSEN RUUD
Table 9.2 Landownership by jati in percentage of total land 1957, 1981,

1993, and 2011.
Year Clean Sekh Namasudra Bagdi Muchi Santal

caste
1957 8.6 64.3 16.4 8.9 0.9 0.8

1981 7.3 65.0 14.5 10.8 1.2 0.8
1993 6.7 67.5 8.3 14.2 2.8 0.7
2011 5.0 66.0 9.7 15.1 3.2 1.0
Percentage of village 4.0 56.2 15.0 15.3 4.6 4.7
population 2011
Source: Field data.
Table 9.3 Landownership groups by jati in number of households 2011.
Bigha Clean caste Sekh Namasudra Bagdi Muchi Santal
0–5 2 16 8 10 5 5
20+ – 2 – 1 – –
Source: Field data.
reduction of landholding has been even and uninterrupted. This sug-

gests forces that have to do with more than sudden political changes.
The Sekhs still constituted the largest caste in the village, both in
terms of number of people and in terms of their landholding. Their
share of the village lands did not alter all that much. The population
doubled during the intervening years and land was subdivided. The
richer segment also sold off land to invest in more remunerative enter-
prises. Of the three 20 plus owners in Table 9.3, one held 35 bighas,
one held about 28–30, and one held 25 bighas.
At the other end of the scale, for the two land-poorest jatis of the
village, the Muchis and the Santals, we note that there has been a
marked improvement for the Muchis and hardly any for the San-
tals. The main improvement for the Muchis came between 1981 and
1993.
The two castes in the middle show a very interesting pattern. Where
the Namasudras have lost land, the Bagdis have gained. Whereas the
Namasudras in the 1950s owned about one-eighth of the village land
and the Bagdis less than one-tenth, the situation is now the reverse.
Again, for the Namasudras, the significant period of change is the
202
middle period, between 1981 and 1993. Some lost territory was
regained later on, though. As for the Bagdis, the increase in land
ownership seems to have been a continuous process throughout the
48 years covered. But for the Bagdis, the 1980s were particularly
bountiful.
The contrast
In terms of visible material change in everyday life, the contrast
between 1993 and 2011 is striking, in particular as regards communi-
cation with the wider world. In 1993, there were eight motorcycles
and no cars. There were two or three television sets, but no telephones.
Also, the village lay along a dirt road on which cars could drive only
with difficulty even under the best of circumstances, and not at all dur-
ing the rainy season and for some months thereafter.
Eighteen years later, there were at least 40 motorcycles, four cars,
four tractors, and five or six hand-tractors, an all-weather tarmac
road with a bus connection twice a day, and a colour television set
with cable TV in practically every household. Everybody had their
own mobile phone, and by the road there was a stall selling SIM cards
and recharge. It was a radical change. The village sported four per-
sonal computers, two of which had internet access via modems bought
in a nearby bazaar village. Even the landless labourers had their own
mobile phones. Even illiterates had mobile phones.
In addition, there were changes in healthcare opportunities and per-
sonal appearances in the village. There were four doctors in the vil-
lage in 1993, two of whom were ayurvedic; 18 years later, there were
six plus one ayurvedic. In 1993, there were two kinds of soaps avail-
able in the two village shops; 18 years later, new products had come
to the village with a vengeance. In the three village shops, one could
buy perfume, shampoo, body spray, shaving cream, and facial cream.
If your needs were more advanced, the shops in the nearby bazaar
village – now within a 15 minutes’ drive whereas it used to be an hour
on bicycle – offered body spray, night cream, deodorant, moisturiser,
fairness cream, hair dye, and multivitamins for healthy skin and hair.
The boys and young men of village society constituted a new group
of customers.
Even more impressive was the overhaul of the economy. Most people
as late as 1993 had lived directly off the land as farmers or as labour-
ers, although some also had non-farm occupations. There used to be a
bus driver, five doctors (one with his shop in a neighbouring village),
203
ARILD ENGELSEN RUUD
two shopkeepers, two tailors, nine school teachers, and eight to 10 men
who worked off-season in steel polishing in Calcutta. There was also
a beggar, a carpenter and one police constable. In addition, there were
a number of people who made an extra income trading in cows; there
were a few moneylenders; a woman worked for the local preschool; and
there were two or three itinerant traders in fish or vegetables. Lastly,
two had political positions elsewhere, one of the school teachers had a
son in the United States, and one family owned a steel polish business
in Calcutta – out of which they made substantial money. Altogether,
about 30 households in the village had non-agricultural income, mostly
seasonal or as an addition to income from land. Many of those who had
non-agricultural income were landless and still poor and dependent.
Eighteen years later, in 2011, the number of mini tubewells had
increased from 11 to 35 and about 40 per cent of village lands were
double cropped, up from 20 per cent The green winter fields were
particularly large in the northern end of the village, towards the Bagdi
hamlet, and towards the east. The lands here were owned mostly by
the Sekhs of the North Para, the Bagdis and the Sekhs of the Middle
Para, respectively.
The diversification of employment was striking. Two out of three
households had non-agricultural income or (in many cases, and) one
or several children in higher education. In the village, there were
now a taxi owner, two bus drivers, two bakery workers, two tai-
lors with a shop in the bazaar village, several owners of various
kinds of shops, five or six life insurance (LIC) agents, a veterinarian,
a government-employed surveyor, a lorry driver, a poultry owner,
three van drivers, a Vodafone stall owner, an electrician, a bicycle
repairman, a dealer in used plastics, an ice cream vendor, and many
engaged in workshops out of the village (many in Kolkata, some
in Delhi, one in Ahmedabad). Altogether, at least 120–130 of the
approximately 420 households in the village had a full wage earn-
er’s non-agricultural income.
In the previously rather bleak North Para, the changes were par-
ticularly visible. Many of the earners of non-agricultural income
lived here. There used to be eight or nine men engaged off-season in
workshops in Delhi and Calcutta; 18 years later, there were 12 thus
employed, sending money home. More impressively, one more fam-
ily had set up its own steel polish business in Kolkata and another
family had set up a small fertiliser business. The tailor who lived in
that para had expanded his shop in the nearby bazaar village and had
engaged his brother as well. Three of the LIC agents, the electrician,
204
the bicycle repair man, two drivers and many others lived here in the
North Para.
The changes were also visible in terms of buildings. The temple had
a new structure; the number of local shops had increased from one to
two; and there was a small shop for rental television sets and DVD
players. There were four new brick houses, one of which was quite
small, and five or six existing mud houses with new brick verandas.
One of the brick houses was owned by the tailor brothers, who had
fitted it with a bathroom with a flush toilet.
There are similar developments in the other paras, although with
some variations. The Santal Para and the Muchi Para had both fared
quite well, with more people with non-agricultural income. These
paras looked cleaner in 2011 than previously; there were several new
two-storeyed houses (still mud), of which some were fitted with brick
verandahs, and the Muchi Para had a new brick road running through
it. There were many more large paddy storages, and most youth were
reading in secondary school and a handful even in college.
More significant and substantial changes were visible in the Bagdi
hamlet. The local shop (new) stocked body spray and night cream.
A brick road connected the quite large para to the metalled road so
that the numerous bicycles and motorcycles could traverse the nar-
row spaces at great speed. Interestingly, the Namasudras had seen
changes, too, but somewhat less marked than the Bagdis or the North
Para. There was one new brick house and one new shop. A number of
people had income from outside of agriculture. One worked in steel
polish and one in construction, and two more who worked far away
in Kolkata and Ahmedabad. In addition, there were two high school
teachers, from different families, a vegetable vendor who walked on
foot from village to village, a carpenter with a workshop in the nearby
bazaar village, and one with a position at the area’s panchayat office.
One better-off family had rented a steel polish workshop and was
making some money.
The picture that emerges from this is one of vastly improved liv-
ing standard. Every house had a toilet (there had been a separate
concerted effort some years before, a ‘drive’), children were mostly
in school, and there were no beggars and no starvation. The non-
agricultural income was so widespread that landownership no longer
had any direct correlation to wealth. The few land-rich were still well-
off, but the land poor were not necessarily poor. Shop owners, the van
drivers, the poultry owner, the man who owned a car that could be
hired, and many others were not poor in village terms. The difference
205
ARILD ENGELSEN RUUD
between jatis in terms of wealth had been greatly reduced. There were
still differences between the Muchi or Santal paras, on the one hand,
and the Middle Para, on the other. But the differences were very much
reduced compared to five decades earlier or even 18 years earlier.
To return to our original interest, then, we ask how these economic
changes have impacted on patron–client relationships, and then again,
on the practice of caste. In order to see clearly, we again take a dia-
chronic view and focus first on the negotiated and fluid nature of caste
even back in the 1950s.
The role of caste in village politics

‘We used to be quite naughty.’
[amra bodmaishi chilum].
An elderly man of the Bagdi caste when talking about their

former lifestyle, a little shamefaced, but also a little bemused.
Caste is neither a singular identity nor unchanging, and in the late

1950s, there were significant sociocultural changes going on, both in
this village and more generally. Certain reform movements had been
under way for decades. Some Bagdis rejected that caste name and pre-
ferred to be known as Barga-Kshatriyas. Most Namasudras had long
rejected the original name Chandal. A few Bagdi and other SC families
in this and neighbouring villages belonged to the layman religious
movement called the Satsangha, popular among the sanskritising and
moderately well-off.
The point is that we should not overdo the significance of caste as
an identity for individuals. But we should also not underplay it. As a
group, the Bagdis lived in their separate para, some 300–400 metres
across the fields away from the main village. The Muchis lived in a
separate para into which few others ventured. The Namasudras lived
in their own para. And so did the Santals. And so did the clean castes.
No Bagdi lived in any other para. No person from any other jati lived
among the Bagdis. This was the case for all jatis.
All jatis had particular rituals associated with them, particular
deities, temples (or mosque) and festivals to which few outside the
jati were invited in. There were also all-village festivals, naturally,
and a fair amount of intermingling in general. But there were only
two inter-caste marriages, an eloped couple and a couple that had
moved there after marriage. Children were told not to accept food
206
in certain paras, and men shared their country cigarettes with some,
but not all, jatis. In other words, a great many social and cultural
practices kept the individual jatis apart and underlined their dif-
ferences in daily life. And this difference was a constituent part of
village social fabric.
After independence, the village was politically dominated by the
owner of large lands, the richest man in the village. He was a Sekh
with a zamindar family tradition. He used the village Bagdis as his
lathiyal, his fighters, and he settled the first Santals on his own land
in order to ensure a ready supply of labourers. When he died, village
leadership lapsed to the second largest landowner in the village. This
man also ensured the compliance and collaboration of the village Bag-
dis as his fighters. The Bagdis had a reputation for being violent, and
heavy drinkers, and some of them found additional income as dacoits,
or so it was rumoured. They were regularly involved in local fights,
and other villagers feared them. As a jati, then, the Bagdis were poor
and dependent, but had a small additional leverage in their rowdy
reputation and village leaders’ need for physical support.
The Muchis were very different in this respect. As individuals, they
were strongly identified by their jati because the Muchi jati is consid-
ered particularly defiling to others. Even Muslims tended not to touch
Muchis. Their comparatively small number and the aversion others
held towards them meant they were somewhat impotent in village
politics. When approaching village leaders for succour, they tended
to insist on their loyalty and emphasise their poverty and dependence.
They were normally considered, almost as by default, to belong to
the village leader’s faction. They were too poor and too dependent to
contemplate other options, but they too can be seen to have used their
particular identity for whatever leverage it might have given them.
Again, their caste status framed their role in village politics, although
it was also constituent of their place, and as such, not entirely open to
manipulation and strategy.
Another interesting jati here is the Namasudras. Ritually, they
considered themselves better than the Muchis, certainly, and also the
Bagdis, whom they considered rowdy and uncouth. The Namasudras
had, for quite some time, engaged in ideas of sociocultural reform.
The fact that the term ‘Namasudra’ had gained currency over the
more derogatory ‘Chandal’ suggests that the endeavour had had
some success. Moreover, alcohol consumption was far less common
among the Namasudras than among the Bagdis – again a sign of
success because abstention from drinking is considered concomitant
207
ARILD ENGELSEN RUUD
to an elevated status. But the success was somewhat limited, at least

in this village. Even in the 1990s, the literacy rate was still quite
low, albeit on the rise. As opposed to the village Bagdis, this self-
consciously restrained lifestyle did not easily enable them for a role
in village politics.
In the late 1950s, the situation was changing. A group of younger
men was growing up – the first generation in the village to have college
education. They all hailed from the upper section of village society and
mostly held substantial land. In college, they had become acquainted
with ideas of progress, equality, the ideals of the Nehruvian state and
of the radical Bengali intelligentsia. They had also become acquainted
with Bengali literature, not the least the post-Tagore generation of
authors who wrote about village society.
After college, they set out to change their home village society. They
did two things that are significant in our history. First, they decided,
after discussions, to bring the Congress into the village, and later,
after a rift, the weaker faction started associating with the CPI(M).
This was a classic establishment versus opposition scenario (Nicholas
1963). The larger faction remained allied with the Congress and was
particularly strong during the 1970s, when the executive was given
extraordinary powers over the administration. During these years,
the Congress-affiliated village leader drew on the support of the vil-
lage Bagdis as lathiyals to maintain his position. The village faction
associated with the CPI(M) suffered some brutality from the Bagdi-
supported regime in the village.
When the situation changed in 1977 and the CPI(M) came to power
in the state, the village Congress faction disintegrated and the leader,
long reviled, fled. The village Bagdis who had been the mainstay of
his local power and influence withdrew from village affairs. As the
CPI(M)-allied faction took over the reins and re-established its posi-
tion as the embodiment of the interests of the poor in the village, the
Bagdis remained at arm’s length.
A telling incident from the first year of CPI(M) rule in the state and
the village shows the role Bagdis had in village politics. The village
CPI(M) leadership sought to stop moonshining in the Bagdi Para. This
source of income was an eyesore to the village leadership with its ideas
of progress and of educated refinement. Approaching the Bagdi Para,
a delegation was met with angry shouts and barrage of clay lumps,
and chased away. And moonshining continued, with little or no inter-
action between the new communist leadership of the village and this
large group of poor. An adjustment was needed. The village leaders
208
then decided to make the Bagdis part of the village CPI(M)-affiliated

setup. In the 1978 panchayat election, the village had two representa-
tives – one from each of the two wards the village was divided into. It
was decided that one of the village’s two panchayat members would
be a Bagdi. The other member was the village’s CPI(M)-affiliated (and
later, member) village leader. This situation remained for the next two
decades. In terms of personnel, three different Bagdi men represented
the first ward over the years and two Sekhs – one man, and later, one
woman – the second ward. The other jatis, such as the Namasudras,
the Muchis, the clean castes, and the Santals, went unrepresented. This
was also the case of the North Para Sekhs in a sense they came to
resent.
The disintegration of the CPI(M) leadership

Look. This nala [drain] is all we have [in the North Para]. It is
collapsing there and there [pointing]. There is no one who
gets [NREGA] work in the North Para.
But you are entitled to?
We don’t sign up for it. They will only give work to South
Para. People like that. They will not give to us.
After 34 years of CPI(M) leadership, a large proportion of villagers

had benefitted from various government programmes, including sub-
sidies for buying homesteads, subsidised loans, support for improved
agriculture – including such as irrigation, seeds and fertilisers – and
vastly improved security of tenure for sharecroppers and wages for
labourers. Some were general, such as the wages, and were distributed
to selected families or individuals. The distribution of those that went
to select families was handled by the local leaders as representatives of
the panchayat. They generally sought to distribute such benefits to the
poor, although it was always clear that party loyalty would also be
factored in.
The privileged position the Bagdis, as a group, enjoyed due to their
close association with the CPI(M) raj in the state and in the local-
ity, is a significant circumstance for understanding their eye-catching
improvement in socio-economic status during those years. The situa-
tion was different for some of the other jatis, for instance, the North
Para Sekhs. In 1993, this group was still quite poor and many among
them were landless. Their houses were small and their living condi-
tions cramped. Their share of state-subsidised programmes had been
209
ARILD ENGELSEN RUUD
less than they might have expected and there was, thus, a sense of
dissatisfaction among them. They also resented the fact that they were
somehow looked down upon by members of the more well-off Mus-
lims families of the village. There was little social interaction between
the more well-off Sekhs of the Middle Para and the poorer Sekhs of
the North Para. There was, for instance, only a few cases of inter-
marriage. The dissatisfaction with the Middle Para Sekh leadership
was often formulated in terms of the party leadership having become
aloof, not sufficiently close to the people.
There had been simmering dissatisfaction with the CPI(M) leader-
ship for some years and a factional split emerged within the party in
the village. The man who, for long, had been at the helm of village
affairs retained his position largely due to support from a section of
the village Sekhs and from the Bagdis. His opponent gathered support
from the somewhat dissatisfied North Para and from among the long
side-lined Namasudras. As the two factions each coalesced within the
CPI(M), the North Para was increasingly less able to benefit from the
activities of the party in the village and the para came increasingly to
be associated with a third group in the village, the emerging Trinamul
Congress (TMC).
It was about 2005 that the North Para started veering towards the
state opposition party and became a TMC stronghold in the village.
Eventually, the strength of the local TMC was such that in the pan-
chayat election of 2008, the TMC won in the ward that the North
Para formed part of. This was quite surprising and unusual. Out of the
22 wards in this gram panchayat, only two were captured by the TMC
and the rest retained by the CPI(M). By the time of the state election
in 2011, there were several small TMC flags and banners in the North
Para while the rest of the village was mainly pro-CPI(M) if the wall
painting, banners and flags are anything to go by.
Facing difficult state elections in the winter and spring of 2011,
the CPI(M) leadership in the village used available government pro-
grammes to reach as many villagers as possible. In particular, the
spectacular NREGA, locally known as the ‘hundred days work’ pro-
gramme, was used to give employment to as many as possible on as
easy terms as doable. Men were hired to excavate ponds or build cul-
verts or small feeder roads at a rate well beyond what they would earn
as agricultural labourers. And they were given a full day’s salary for
what, in practice, amounted to half a day’s work.
But it was still clear that the benefit accrued to some groups and
not to others. The Bagdi and Muchi paras plus the South Para were
210
given work at pond excavations, tree plantings and road construc-

tion works in their paras. The Namasudra Para saw its share, but
less. All these paras were in the ward won by the CPI(M) in the last
panchayat election. The ward won by the TMC saw only a few small
projects. Compared to the other ward, this was very little. As a group,
the North Para Sekhs felt side-lined. ‘There is no one who gets work in
the North Para.’ They seemed to have taken a risk together, almost to
a man siding with the opposition, and took a beating together, getting
very little out of the bountiful provisions of the NREGA programme.
There were others also with the TMC, in particular the clean castes
of the village, a fair proportion of the Namasudras and many of the
Sekhs were known sympathisers. In the 2011 election, a little more
than half the village voted for the CPI(M) and the rest for the TMC.
The Bagdis were considered to be mostly behind the CPI(M), and the
North Para Sekhs and the clean castes mostly behind the TMC. For the
rest, it was a mixed picture, even for the Muchis. After the TMC vic-
tory in the state election in 2011, the local panchayat was still CPI(M)
dominated but impotent for all practical purposes. Power locally had
come to rest with the new leadership and the centre of political gravity
in the village had moved to the North Para.
The North Para Sekhs were no longer poor and dependent. The
changes from 1993 were surprisingly substantial. Partly, it had to do
with the advantages of the CPI(M) raj, including sharecropper rights
and higher wages, and massively increased irrigation on lands they
owned or sharecropped. It also had to do with improved levels of
education, which enabled many to get outside work, some of which
required at least some formal education, and it had to do with a gen-
eral improvement in living standards in the state, better communi-
cation and the increased opportunities for work that followed. With
improved assets, the North Para Sekhs were no longer in need of
patronage. Their political loyalty could not be bought or even threat-
ened by the once-mighty CPI(M) leadership. Their independence was
startling compared to the situation years earlier.
Moreover, their leader was incapable of securing them patronage
from his own means. Like the naya netas of elsewhere (Krishna 2007;
Alm 2010), he was young and college-educated. Sekh by jati, he was
related to the former CPI(M) leader, but poorer with some 12 bighas
to his name. Similarly his main opponent was a dissenting CPI(M)
leader and school teacher with little land, although rumoured to have
amassed some illegally. The two were very far from the owners of
massive lands who dominated village life five decades earlier. Now,
211
ARILD ENGELSEN RUUD
supporters had independent income, and very often, income that was
accrued outside of the reach of the village leaders. There were still
fighters, lathiyals, but they were individuals from several caste back-
grounds. And more land would not secure clients because all labourers
were now paid the standard rate, and advances were very rare. The
presence of a larger than before number of migrant labourers dur-
ing harvesting alleviated the landowners of having to rely on local
labour, thus further undermining whatever clientilistic relation-build-
ing there might have been. Sharecropping had vanished, except for
those arrangements that existed before 1978.
Village society has changed into something that may be character-
ised as ‘patronage democracy’, following the suggestion by Kanchan
Chandra (2004). Here, group identity or, in Chandra’s terminology,
ethnicity, are cultural elements that are referred to and used to create
a ‘we’ that can demand patronage or be expected to vote in a particu-
lar fashion. Following this suggestion, we could understand jati iden-
tity in this village history as something that was harnessed politically.
‘Regardless of the good they seek’, writes Chandra (2004), Indian vot-
ers are ‘instrumental actors who invest in an identity because it offers
them the best available means by which to obtain desired benefits’.
The case is overstated. Caste identity, jati, cannot easily be bent to
suit new needs as opportunities change, and certain caste identities
cannot be shed. And yet, jati locally seems to have been invested in
to seek political or economic leverage, and once the circumstances
changed and the demeaning status of patronage-seeking was less
imperative, the jati identity changed and withered.
Conclusion: wither caste?

As all forms of caste discrimination, except the prejudice against inter-
caste marriages, have gone, it may seem that caste as such has gone.
All castes nowadays take part in the all-village festivals, including the
Hindu festivals, on an equal footing. Even the Brahmin priests avoid
expressions of caste aversion (if they have any) after 34 years of Com-
munist rule.
The 54 years period (1957–2011) covered by the material in this
chapter is close to India’s entire postcolonial era. The figures reveal
that in this time, landownership patterns in the countryside of West
Bengal have changed radically. Ten years after independence and
after legislation such as the abolition of zamindari in the state and
the tenancy reform act, there were still landowners in this modest
212
and inconspicuous village who held a hundred bighas of land and

more while a majority in the same village held no land or too little to
sustain themselves. Today, the rich have much less land in terms of
bighas, although perhaps not in terms of output. In terms of paddy
produced in a year, one bigha today yields as much as three bighas
50 years ago if the right implements are present (water, improved
seeds, fertiliser, and pesticides). Even so, the ‘size’ of the rich is
reduced. Land is much more evenly distributed; labourers do not
come as cheap as they did; and a large proportion of the potential
labourers have lands on their own that they need to work on in the
peak season. Thus, the rich may be rich in terms of paddy, but the
resources they control are not anymore sufficient to create factions
of dependent clients. The gap between the rich and the poor has been
reduced significantly.
Another difference to the past is the living conditions of the major-
ity of the village. In addition to the reduced incidence of landlessness,
rights of tenure are strengthened and enforced, the wages have been
doubled, there is a new agricultural season, and the state welfare pro-
grammes, which included subsidised loans and guaranteed work some
of the year, have been implemented. These, combined with the non-
agricultural incomes from the many new opportunities, have contrib-
uted to radically reduced socio-economic differences in village society.
These developments have had two parallel effects: first, the flight
from the countryside of the very rich and a reduced capacity for or
interest in dispensing patronage and keeping personal factions among
the remaining landowners. In short, there has been a change in inter-
est in village politics, which has left the village scene open to a new
type of leader. This new leader has fewer personal sources of power
and relies much more on contacts. Second, the economic changes have
reduced the need for patronage among the poor. Many are still poor
and dependent, but a fair proportion of those who previously were
poor and dependent now find themselves better-off and less in need
for patronage.
Caste as a practice was entangled with the patron–client relation-
ships that made up the village polity. In some cases, ritual identity was
used, perhaps manipulated, to gain an added advantage in the scram-
ble for scarce resources. The village Bagdis made much ado of their
rowdy image and traditions, to become a useful, almost indispensable
asset for the village leader. The Muchis did something similar. And
with their sanskritising name change and altered traditions, the Nama-
sudras in this village and elsewhere were making a statement against
213
ARILD ENGELSEN RUUD
untouchability and their ritually low position. Caste identity and the
nature of that precise identity were matters of conscious moulding and
adaptation within existing socio-economic and political conditions.
Caste, in other words, was conceived in dynamic interaction with
everyday lived society, influenced by circumstances. Caste was part of
how relations between high and low were formed. Ritual identity is
still a reality in village society, but its role and place is undermined by
the fact that one of the legs that sustained it, patron–client relation-
ships, have changed.
Bibliography
Alm, Björn. 2010. ‘Creating Followers, Gaining Patrons: Leadership Strate-
gies in a Tamil Nadu Village’, in Pamela Price and Arild Engelsen Ruud
(eds), Power and Influence in India: Bosses, Lords and Captains, pp. 1–19.
London: Routledge.
Banerjee, Bibhuti Bhusan. 1968 (1929). Song of the Road. Pather Panchali:
A Bengali Novel. Transl. by T. W. Clark and Tarapada Mukherji. London:
Allen and Unwin.
Bhadra, Gautam. 1989. ‘The Mentality of Subalternity: Kantanama or Rajd-
harma’, in Ranajit Guha (ed), Subaltern Studies VI, pp. 54–91. Delhi:
Chandra, Kanchan. 2004. Why Ethnic Parties Succeed: Patronage and Ethnic
Head Counts in India. Cambridge: Cambridge University Press.
Cohn, Berhard S. 1990 (1987). An Anthropologist among the Historians and
Other Essays. Delhi: Oxford University Press.
Davis, Marvin. 1983. Rank and Rivalry: The Politics of Inequality in West
Bengal. Cambridge: Cambridge University Press.
Djurfeldt, Göran, Venkatesh Athreya, N. Jayakumar, Staffan Lindberg, A.
Rajagopal and R. Vidyasagar. 2008. ‘Agrarian Change and Social Mobility
in Tamil Nadu’, Economic and Political Weekly, 43(45): 50–61.
Frankel, Francine R. and M. S. A. Rao. 1993 (1989). ‘Introduction’, in Fran-
cine R. Frankel and M. S. A. Rao (eds), Dominance and State Power in Mod-
ern India: Decline of a Social Order (Volume 1), pp. 1–20. Delhi: Oxford
University Press.
Krishna, Anirudh. 2007. ‘Politics in the Middle: Mediating Relationships
between the Citizens and the State in Rural North India’, in Herbert
Kitschelt and Steven I. Wilkinson (eds), Patrons, Clients and Policies: Pat-
terns of Democratic Accountability and Political Competition, pp. 141–58.
Manor, James. 2010. ‘Prologue: Caste and Politics in Recent Times’, in Rajni
Kothari (ed), Caste in Indian Politics (2nd edition revised by James Manor),
pp. xi–lxi. London: Orient BlackSwan.
214
Mayer, Adrian C. 1958. Caste and Kinship in Central India: A Village and Its
Region. London: Routledge and Kegan Paul.
Nicholas, Ralph W. 1963. ‘Village Factions and Political Parties in Rural West
Bengal’, Journal of Commonwealth Studies, 2(1): 17–32.
Piliavsky, Anastasia (ed). 2014. Patronage as Politics in South Asia. Cam-
bridge: Cambridge University Press.
Ruud, Arild Engelsen. 1997. ‘Of Novels and Dramas: Engaging with Litera-
ture in Bengal and the Making of a Modern Village leader’, South Asia
Research, 17(1): 70–92.
Ruud, Arild Engelsen. 1999. ‘From Untouchable to Communist: Wealth,
Power and Status among Supporters of the Communist Party Marxist in
Rural West Bengal’, in Ben Rogaly, Barbara Harriss-White and Sugata Bose
(eds), Sonar Bangla? Agricultural Growth and Agrarian Change in West
Bengal and Bangladesh, pp. 253–78. New Delhi: Sage.
Ruud, Arild Engelsen. 2003. Poetics of Village Politics: The Making of West
Srinivas, M. N. 1955. ‘The Social System of a Mysore Village’, in McKim Mar-
riott (ed), Village India, pp. 19–32. Chicago: University of Chicago Press.
Zbavitel, Dusan. 1976. Bengali Literature. Wiesbaden: Otto Harrasowitz.
215
10
CRAFT, IDENTITY,
HIERARCHY
The Kumbhakars of Bengal
Moumita Sen
Kumbhakar (pot maker), Kumar, or colloquially Kumor, is the potter

caste of West Bengal, Bihar and Orissa. Around the mid-18th century,
the Kumors of Nadia began to specialise in moulding life-like minia-
ture clay models showing the everyday lives of different groups, castes
and tribes of rural Bengal (Chakravarti 1985). This craft received con-
siderable colonial patronage and distinction through an elaborate net-
work of colonial exhibitions in different parts of Europe, the
proliferation of Western art education in the colony and the role of
individual connoisseurs and collectors (Mukharji 1888). The craft and
its practitioners achieved a rare celebrity during and following this
period, based on the making of realistic models. In this chapter, I look
at how the changing nature of the craft led to significant changes in the
status and the identity of the Kumbhakar mritshilpis (clay-modeller).
The members of the Kumbhakar caste in West Bengal who are
engaged with different professions involving clay work are spread
over different districts. While a portion of this community is no lon-
ger related to clay work of any kind, clusters or individual families of
Kumbhakars in various parts of Bengal are involved in making bricks,
utensils and images,1 among other clay objects. The Kumbhakars as a
jati (caste) are attributed with the making of clay utensils for everyday
use. In fact, the potter’s wheel stands as a marker of their identity.2
Certain clusters of the Kumbhakar community, however, left the wheel
1 Dolls, toys (more appropriately ‘putul’), ritual and decorative images.

2 As seen in a photographic survey commissioned by Lord Canning, ‘The People of
India’ (1868–1875), and the dictionary definition of ‘Kumbhakar’ the Samsad Ben-
gali-English dictionary.
216
C R A F T, I D E N T I T Y, H I E R A R C H Y
and began to specialise in making god-images by moulding unfired

clay around a bamboo and straw frame. In several, if not all, towns
and villages of Bengal, one finds a few families of Kumbhakars who
meet the local demand for ritual murtis (images)3 made in unfired
clay.4 But in some areas of Bengal, clay-modelling has emerged as an
important trade, particularly in Bankura, Nadia and Kolkata.
My study focusses on one such community, which branches out in
two main regions of Bengal: from Krishnanagar in Nadia to Kumor-
tuli in Kolkata. The objects produced by the mritshilpis of contempo-
rary Krishnanagar and Kumortuli can be broadly broken down into
three kinds of images: God-images of varying scale (for ritual use – see
Figure 10.1), portraiture (for exhibitional5 use) and an eclectic selection
of commissioned clay-models ranging from ‘Tom and Jerry’ dustbins to
copies of Michelangelo’s David (for decorative use).
The first settlement of Kumbhakars is located in Krishnanagar,
Nadia – notably Ghurni, among others. The second cluster is situated
in Kumortuli in North Kolkata. The major part of the settlement in
Kumortuli consists of clay-modellers who migrated from Krishnanagar
to the city of Kolkata from the 19th to the early 20th centuries. The
mritshilpis of Krishnanagar, especially those in the Ghurni region, are
responsible for the beginning of a particular kind of craft in Bengal –
that of naturalistic image-making in unfired clay.
Here, I look at how a caste-based community of artisans negotiates
with questions of identity, economic mobility and power through the
practice of its craft. The dominant castes of Bengal and other parts
of India shared certain interests with the British art educationists in
creating a hegemonic picture form, namely Realism or Academic
Realism. A part of the Kumbhakar community appropriated this new
picture form and sought a kind of celebrity status among the elites
(both indigenous and foreign) in the 19th and 20th centuries. But
there was another part of this community that could not master this
picture form. This chapter will look at the construction of hierarchies
3 Murti seems to be the most appropriate general term for the images made by the clay-
modellers as opposed to statue, sculpture or even model. For a detailed etymology of
the term ‘murti’, see Eck (1998).
4 A larger market and network of production of ready-made murtis called chancher murti
used mostly for domestic worship is necessarily limited to the Kumbhakar community.
5 A term used by Christopher Pinney (2002). By exhibitional value, as opposed to cultic
value, Pinney refers to the detached, intellectualised discipline of Western aesthetics
and art appreciation, especially the modernist kind.
217
M O U M I TA S E N
Figure 10.1 Pratima of Goddess Kali being finished at Kumortuli, 2012.

Source: Photo by the author.
within the caste-based artisanal community centred on their capacity

to appropriate this powerful way of moulding reality. The question of
caste becomes a matter of framing and reframing identities within the
relations of power among the elite and the non-elite in the colonial
and postcolonial state. The britti (instinct, propensity or nature) and
pesha (profession)6 of the Kumbhakar allowed him to mould images
of power – this-worldly and other-worldly. The clay-modellers who
mastered the art of portraiture could form alliances with the colonial
state and they became the makers of public statuary for the govern-
ment and political parties in postcolonial Bengal. In contemporary
Bengal, the 19th or early 20th centuries, we see a marked change in
patronage – ranging from the central and state government to the local
clubs. Most, if not all of the public statues and busts – of the great men
6 This distinction between the nature or propensity of the individual belonging to a

caste and his profession is made by Prodyot Ghose (2011). This distinction reflects on
a fundamental one between ‘who one is’ and ‘what one does’.
218
Figure 10.2 Tarit Pal’s Studio, Krishnanagar, 2013.

of the nation such as Gandhi, Netaji, Vivekanada, Tagore, and so on –

populating the public spaces of Bengal are made by the clay-modellers
belonging to the Kumbhakar caste (see Figure 10.2).
In comparison, the others, who continued to make traditional7
ritual images could not secure this kind of socio-economic mobility.
The ability or inability to hone this new skill to make realistic images
affected the socio-economic mobility of the craftsmen. Conversely,
one can also look at this shift in terms of two different regimes of
visual language and the investiture of value, power and efficacy in this
new form. In the struggle for status and identity, the makers of cultic
images could not reach as far as the makers of portraits of powerful
men: the otherworldly lost to the this-worldly.
7 I use the term ‘traditional’ with some reservation because different kinds of images
are referred to as ‘traditional’ in different discourses around clay models.
219
M O U M I TA S E N
Through the story of the Kumbhakars of Bengal, I revisit the

notion of hierarchy – hierarchy not as embedded within the Dumon-
tian binary of ritual purity and pollution.8 Given the varied caste and
caste-like groups in contemporary India and the changing, complex
network of caste alliances and identities, the idea of hierarchy based
either solely on ritual purity or both status and sacredness, does not
suffice. If we were to retain the idea of hierarchy at all, it has to take
into account the opening of new markets, the forging of new profes-
sional identities and political alliances across caste lines. In order to
accommodate these constant reconfigurations, the idea of hierarchy
has to be multiple, idiosyncratic and incommensurable, just like the
‘non-encompassing’ nature of caste identities itself (Gupta 2004). The
story of the Kumbhkars of Bengal urges me to understand hierarchy
in terms of exigencies of power, status and wealth, through a recurrent
trope of success stories I encountered in the field. I employ the concep-
tual category of power to signify two variants pertaining to different
aspects of political life. I understand power in relation to the state
and party politics, the forging of alliances between the image-maker
and the political organisations as also the formation of associations
(variant of a trade union) within the community. But more signifi-
cantly, I understand power in relation to the capacity of individuals
and groups to inculcate a kind of symbolic value in the public sphere
and consequently in public consciousness. A large part of this analysis
will try to negotiate the domain of symbolic politics with other kinds
of politics by looking at the symbolic power of different picture forms
as embedded in larger relations of power among individuals, groups
and institutions.
A note on locations: jati and the craftsmen

Ask an informed observer whether caste has any great influ-
ence on present-day West Bengal politics and he is likely to
reply: ‘very little’. Perhaps he will go on to describe the depth
8 In fact, contrary to Dumont, this chapter will argue against the separateness of the
spheres of religion and politics. This, however, is not to supplant the idea of an over-
arching importance of religion – like a ‘sacred canopy’ – in all spheres of life in South
Asia. The way the clay murtis traverse the religious and the political will point
towards a continuum of the social affects created by these practices in the way they
are performed in the public sphere.
220
of class feelings in West Bengal, both in political organisations

and in the ideas which people have about political issues.
(Chatterjee 1997: 69)
I started my ethnographic study in Kumortuli in 2010, when the mrit-

shilpi community was fighting with the Kolkata Metropolitan Devel-
opment Authority for their rights over the thika land they inhabit. The
conversations I had with the families was about the problems of habi-
tation, the conditions of trade, the dissatisfaction over the samitis,9 the
monsoon and malaria in the new precincts. And never quite about
caste. In my conversations, the members of the Kumbhakar commu-
nity did not refer to themselves as Kumors (the colloquial version of
Kumbhakar). Instead, the prominent members called themselves mrit-
shilpis, which some of them translated as ‘sculptor’ and/or ‘artist’. The
vernacular term ‘mritshilpi’ is officially used in the naming of the two
associations or samitis, which represent this group of artisanal labour
in Kumortuli. Of course, my encounters in the field were contingent
upon my anthropological location as what can be called a ‘native
anthropologist’. My position as a native, Bengali-speaking, upper-
caste and middle-class young woman among older craftsmen created
a series of adjustments in terms of representations of identity. The
politics of status, hierarchy and power that was laid out for me, given
my location in the field, is the one I seek to represent here.
As Chatterjee points out in his writings on caste in West Bengal,
caste hides behind questions of class in this part of the country. The
issue of caste in the field of cultural production in contemporary Ben-
gal is found shrouded in other signifiers. For instance, a recurrent
form of addressing the community was ‘amra Palera’ (us, Pals). The
weight of caste identity, or rather, jati, is borne by the use of this sur-
name. It is a way of attaching a larger history of a community to an
individual. Unpacking the signification of this term, address or mode
of self-identification begins to unravel the importance of caste in this
community of artisans. I trace two shifts in terms of identity: first,
from the identity as Kumbhakar to that of mritshilpi and second,
from mritshilpi to sculptor/artist. The journey of the pot-maker from
the Vedic ‘kulal’ to Kumar, and then, mritshilpi is not just a shift of
pesha and britti – it is, in fact, a story of the evolution of a craft and of
9 Associations of mritshilpis. Presently, there are two such representative associations

in Kumortuli. See Heierstad, this volume, for further details.
221
M O U M I TA S E N
a community. Let me begin this story by talking about the idea of the
Kumbhakar in different texts: oral narratives, ethnographic accounts
and colonial census and newspaper or journal articles.
In H. H. Risley’s (1892) The Tribes and Castes of Bengal, among other
census reports, we find several theories concerning the parentage of the
Kumbhakar jati such as: the Kumbhakar is born of a Kshatriya mother
and a Brahmin father (according to Sir Monier Williams) (Risley 1892:
517–18). Risley, however, identifies these theories as an attempt to rec-
oncile the disparate groups and sub-castes into the fourfold Brahminical
caste system. He shows us the possible logic of arrangement of this caste
group in different parts of undivided Bengal based on: region or place
of origin, specificity of material they work with, nature of produce and
endogamy, among other factors. According to Risley, the Kumbhakars
were recognised as members of the Navasakha group in Bengal and their
social standing was respectable. Here, we could launch into a debate
about whether caste was a colonial construction aiming at more effi-
cient governmentality, or whether it was a matter of inadequate repre-
sentation of the actual practice due to the mechanics and politics of data
collection. However, I would like to focus on how the contemporary
clay-modellers talk about caste. One of the stories that I came across in
other ethnographic accounts, and eventually, in the field was the myth
of origin of the Pals (Heierstad 2009). The most recurrent version of
the story goes like this: at the wedding of Siva and Parvati, there was
want of a pot. And no one knew how to make one. Siva then took two
rudrakhas (beads) from his necklace and made a Kumbhakar with the
first one. With the second bead, Siva made a woman who became the
Kumbhakar’s consort. The name given to them by the lord himself was
Rudrapal, which we now know as Pal. This story of genesis is told and
retold in similar versions among the Kumbhakars all over Bengal.
As noted, there are a few families of Kumbhakars all over Bengal
who do not necessarily practice their hereditary trade. Conversely,
there are other artisanal jatis who took on this craft as there are
individuals from unrelated jatis who learnt this trade because of the
demand for murtis in the market.10 The particular group of Kumb-
hakars I am concerned with is found in different parts of Kolkata
and Nadia. Popular narratives (folklore, mythology and short stories)
suggest that Maharaja Krishnachandra, who ruled Nadia district and
the adjoining areas in the 18th century, and his predecessors were the
10 Interview with Samir Ghosh, puja organiser, Bansberia, 2013.
222
major patrons of the idol-making community of Krishnanagar. Not

only did they disseminate the worship of different Hindu gods and
goddesses, their patronage and aesthetic sensibility were indispensable
to the birth of a form which came to be broadly known as the ‘Krish-
nanagar style’. Sudhir Chakravarti (1985), in his study of this com-
munity, notes that the royal family of Krishnanagar was instrumental
to certain grand festivals in Bengal, such as the Jagatdhatri Pujo in
Krishnanagar and Chandannagar in Hooghly district and Rash-Utsav
of Nabadwip. Accounts of newspapers and anthropological studies of
the late 18th century retell the legend of Maharaja Krishnachandra
bringing in a community of artisans from Natore, in contemporary
Bangladesh (Goldblatt 1979). Some of these clay-modellers migrated
to Kolkata from the end of the 19th century through the first half of
the 20th century (Goldblatt 1979) to create another cluster, which we
now know as Kumortuli.11 Both these clusters, till the early 20th cen-
tury, specialised in moulding god-images of varying scales.
I want to emphasise, at this point, the notable shift in the value of objects
made by the Kumbhakars. Even though pots have ritual significance, the
Kumbhakar community originally made utilitarian objects – utensils for
everyday use. Now, we see them moulding images of cultic value, which,
as Christopher Pinney (2004) puts it, have ‘intimations of power’. It is,
in fact, the artisan who can consecrate or conjure the divine in the clay
body by painting the eyes on it. Through the history of the Kumbhakars
of Nadia, we also witness the change in the identity of this cluster of
Kumbhakars from pot-maker to image-maker. This change leads to the
differentiation in the identity between the Kumor and the mritshilpi.
The coming of the ‘modern’ mritshilpi:

tradition and Westernisation
The craft of the Kumbhakars who made god-images (or ‘efficacious
idols’)12 is inextricably related to Shastra knowledge, especially the
Kashyapasilpa.13 Kashyapa’s treatise lays out precise guidelines about
the technique of making and the form of images. The problem of
11 Contemporary Kumortuli consists of mritshilpis both from Krishnanagar and other

parts of Nadia as well as Bangladesh.
12 According to Pinney (2002), the ritual object or the idol is efficacious as opposed to
the object of aesthetic contemplation.
13 The treatise on clay-modelling in Kashyapasilpa of Kashyapa is echoed later in Sil-
paratna compiled by Srikumara.
223
M O U M I TA S E N
identity of the contemporary Kumbhakar community is also predi-

cated on the continuities of Shastric knowledge in their present prac-
tice, on the one hand, and the use of Western styles and alternative
permanent materials, on the other. The contradictions visible in the
craft as it stands today – one which one may understand in terms of
hybridity or vernacularity – appears to be the contradiction of tradi-
tion and modernity. In this section, I will like to first destabilise the
idea of tradition in caste-based crafts, which romanticises the village
craftsman, on the one hand, and constructs a narrative of regression
and technological stagnation in crafts, on the other. Second, I will
question the idea of progress in the craft as inextricably tied to West-
ernisation, colonisation and modernity. Instead, I will attempt to
locate the ‘modern mritshilpi’ in a continuum through the precolonial
to the postcolonial. What we need to keep in mind with the clay
objects made by erstwhile and contemporary mritshilpis is that they
were bespoke images; therefore, what is made and how it is an effect
of who wants to show what, and to whom. The large-scale clay mur-
tis are always made for patrons or clients, and mostly for public
spaces. The shifts in style and the change in the subject matter, thus,
reflect shifts not only in the taste of the patrons but the ideology of
showing/seeing things in the public space.
The ritual basis of unfired clay-modelling, pertaining to god-images
at least, is clear from the writings of Pratap Chandra Ghosha and
K. M. Verma. P. C. Ghosha (1871) tells us about the elaborate list of
duties of the Kumbhakar during the process of construction of a god-
image, relating the minutiae of the technique with particular hours
and days of the lunar calendar. We find a more detailed account of
material and its ritual significance in K. M. Varma’s (1970) account
of clay-modelling. Each of the seven materials used for image-making
has a metaphorical relationship to the living human body. The murti is
built up, like the human body, in layers from inside to outside.
Ghosha gives us an idea of the relationship between the families (or
the yajmana) and the Kumbhakar or sutradhar, who are responsible
for making the ritual image. In lieu of the chakran lands, which were
granted to their ancestors, they construct the idols for the household
puja. The individuals who accomplish various duties during the puja –
such as the priest, the artisans or image-makers, musicians and so
on – hold rent-free lands granted by the family and their return is in
the form of services rendered during the puja (Ghosha 1871: 2). This
relationship between the consumer and the producer is no longer the
same. First, the artisanal labourer is no longer exclusively tied to the
224
landowner. Second, in the urban areas, the settlement of the mritshil-

pis in clusters forms a market where patrons (yajmanas) can either
purchase ready-made images or commission a mritshilpi of their
choice for what is called a bespoke image. The local clubs, who orga-
nise public festivals, much like the households, have designated mrit-
shilpis in several areas, the Jagatdhatri Puja in Krishnanagar being
one such instance. But the mritshilpi is technically not tied to the club
and can move to any other club or householder of his choice. Most
of these alliances between mritshilpis and local clubs are because of
reasons of prestige, expertise, style and even camaraderie, and not
related to land. With this shift from the attached labour structure to
that of the free market, we also arrive at the idea of maximisation
of profit. The mritshilpis today cannot afford to ritually consecrate
every piece of bamboo slit (the base of the kathamo), given the mass
production of readymade images. With a high demand for images,
an average-sized workshop makes more than 50 murtis during the
puja season. What we need to keep in mind is the seasonal nature of
the trade. The livelihoods of the mritshilpis who make god-images
depend on these three to four months. However, the consecration of
the kathamo on the day of the ratha yatra, according to Ghosha’s
account, is still prevalent, at least in Kumortuli. This consecration
can be understood as a ritual gesture, which symbolically applies to
all the murtis of the workshop. This symbolic gesture is also used
for chakshudan or the ritual consecration of images by the painting
of eyes on the murti. The mritshilpis cannot afford to ritually ‘open’
the eyes of every murti on the day of the mahalaya, at the appointed
hour. Therefore, one or two of the images are consecrated on this
day, often as a public performance for the print and electronic media
of the city. These are some of the changes in ritual gestures, which
demonstrate the way the mritshilpis negotiate the tradition of the
craft and its modern demands. Through these shifts in the significa-
tion of rituals that surround the craft of making god-images, we also
witness the transformation of the ‘bondsman to worker’ as part of
the larger change in the trade from semi-feudal to capitalist condi-
tions (Guha and Spivak 1988: 3). The other important area of nego-
tiation, inextricably related to the status of the artisan, is notable in
the style of imagery.
Varma traces the source of this technique in the Agama texts, pri-
marily Silpasutra and Kashyapasilpa, among others. There is sig-
nificant disagreement over the presence of this Sashtric knowledge
in the eastern part of India.14 But a closer look at the practice of
225
M O U M I TA S E N
contemporary Kumbhakars, rather mritshilpis, gives us a clue about

the relevance of Sashtric knowledge to their techniques. In Kumortuli
and Krishnanagar, the practice of unfired clay-modelling among the
Kumbhakar community has changed in several ways. It will, how-
ever, be incorrect to assert that the scriptural knowledge never existed
or does not exist among the clay-modellers. For instance, two kinds
of clay dough of different consistencies are referred to as harer mati
(clay of the bones) and mangsher mati (clay of the flesh). But while
the core technique of building an image remains the same, there have
been significant stylistic transformations. The image of the goddess
we understand as traditional (claimed to be ‘correct’ both in terms of
metaphorical description and technical prescription) is what can be
called debi murti. However, the debi murti is now one of the styles
in the oeuvre of a contemporary mritshilpis.15 A significant point in
the history of the stylistic evolution of clay-modelling is its encounter
with Western Academic Realism. The body of the goddess became
naturalistic and more dynamic (even theatrical)16 as her face turned
more human-like (putul murti), especially in the hands of Gopeshwar
Pal, a particularly renowned mritshilpi of the early 20th century, who
traversed the two clusters of Ghurni and Kumortuli (Heierstad 2009).
The modern murti, in fact, as a blanket category, would include a wide
array of styles. Under this category a few decades ago, the mritshilpi
would portray the face of a contemporary actress from Bombay cin-
ema whereas now, a modern or ‘theme’ murti might mean anything
from an ‘African-style’ to a ‘Mexican-style’ Durga.17 The oeuvre of
the modern mritshilpi in terms of method, material and style is a rich,
complex hybrid space.
Traditional craft or artisanal labour, even in contemporary dis-
course, has been largely construed as ‘simple’, where the craftsman
uses mostly his hands or rudimentary tools. Technological improve-
ment was in the craft, and according to this understanding, was
14 Jim Robinson (1983) holds a contrary view.

15 In the contemporary taxonomy of styles, the Bangla or debi murti is one of the four
broad categories. While there is no agreement on one such taxonomy, the most com-
monly invoked one lists the following: (1) the debi murti or the Bangla murti; (2) the
putul murti or the naturalistic murti; (3) the oriental murti or the ‘Ajanta’ murti often
inspired by the ‘Bengal School’ Aesthetic; and (4) the modern murti (Pal 2006).
16 The relationship between images of goddesses and popular theatre is noted in Pin-
ney (2004) and Jain (1999).
17 The epoch of the ‘theme pujas’ in Bengal, beginning with Kolkata in 2002 intro-
duces several new stylistic templates.
226
checked by, first, the laws prescribed by the Shastras, which fixed the
method of work; and second, by the caste system, which organised
aspects of the condition of trade. What resulted from these rules and
prohibitions put in place by Shastric knowledge and the caste system
was the rarity of ‘inter-sectoral diffusion of tool techniques’ and tech-
nological stagnation (Sarkar 1998: 131). This idea of stagnation is
reflected by Sheldon Pollock (1985) in his study of the Shastras. Pol-
lock, discussing the place of Shastric knowledge in pre-modern India,
refers to the primacy of theory over practice. He identifies ‘a practical
discourse of power’ in the construction of Sashtric wisdom as tran-
scendent (Pollock 1985: 516). This, he argues, forecloses the possibil-
ity of innovation or progress in the field of intellectual and cultural
production in India.18
However, the fixity of the Sashtric knowledges he claims in pre-
modern India does not apply to most artisanal practices in colonial
and postcolonial India. As Smritikumar Sarkar (1998) shows in his
study of the Kansari’s craft19 in Bengal, there have been significant
changes in both the material and the method used in the craft since the
16th century. It seems plausible that the tools and methods used by the
craftsmen in practice always spilled over the strict theoretical bound-
aries of the Shastras. And if we do not want to build a grand narra-
tive of progress in cultural production, we can at least suture together
significant moments of encounters, even confrontations, which bred
new picture forms. In case of the clay-modellers, I will trace one such
important encounter, which changes not only the fate of a part of this
artisanal community, but also the popular visual culture of Bengal.
And this is also most easily recognised as the moment of modernity in
the mritshilpi community. I would like to argue, however, that neither
was (what is constructed as) ‘traditional craft’ timeless or stagnant,
nor was Academic Realism the first significant shift in terms of style
and content of imagery. The elite of Bengal – the zamindars and the
18 Having recognised the notion of ‘progress’ as an idea born on the Western soil, Pol-
lock posits: ‘From the conception of an a priori Sastra it logically follows and Indian
intellectual history demonstrates that this conclusion was clearly drawn – that there
can be no conception of progress, of the forward “movement from worse to better,”
on the basis of innovations in practice . . . Whatever may be the possibility events
ideological hindrances in its way of the idea’s growth in the absence of these con-
cepts; it is clear that in traditional India there were at all events ideological hin-
drances on its way’.
19 Kansaris are the caste-based practitioners of braziery.
227
M O U M I TA S E N
aristocracy – who formed the core group of patrons for the mritshilpis
wanted elaborate ritual images. This facilitated the movement of a
group of mritshilpis from pot-making to image-making. Again, the
change in the patronage to a new group of elites in the 19th century –
British administrators and Indian civil servants in high ranks, wealthy
individual patrons, both British and native – saw the upsurge of real-
istic imagery of contemporary life, and subsequently, naturalistic
portraiture. What we see now is another shift in patronage – from
the colonial government to the postcolonial state. The contemporary
patrons of large works in public spaces – comprising both ritual imag-
ery and secular portraits – are the central and the state governments
and the local clubs. A fresh market for secular portraits opened up
when fibre glass became popular as a cheaply available permanent
material among the clay-modellers approximately in the late 1980s.
This marks an interaction between a cheap permanent material and
a powerful visual template of reality, which led to a proliferation of
public statuary of great men all over the cities and towns of Bengal.
Academic Realism, from the mid-19th century, proliferated
through an elaborate network of colonial art colleges, salon exhibi-
tions and popular prints, which not only led to new ways of picture-
making, but an active moulding of taste and judgement among the
elite (Guha-Thakurta 2008). Here, we note the second shift in the
value of objects made by mritshilpis, the first one being from pots
to cultic images. Through the application of Academic Realism, the
cultic value or power of images comes to be replaced by what Pin-
ney calls ‘exhibitional value’. In terms of ways of seeing, technically,
this creates a shift from sensorial transformations or corpothetics20
to disinterested visual appreciation or aesthetics (Pinney 2004). In
the following section, I will look at how a training of Academic
Realism inducts the mritshilpis into what Jacques Rancière (2006)
calls the ‘Aesthetic regime in Western art’. It is in the encounter with
this form that one can locate the beginnings of the claim of the mrit-
shilpis towards the identity of the artist. From the vocation of mak-
ing and doing things (Rancière 2006), through the appropriation of
the master’s form, the artisan now sees himself standing alongside
the artist who is not only making and doing things, but is engaged
in creation.
20 Corpothetics is understood in relation to aesthetics as a practice of interaction with

an image which involves all the senses as opposed to vision alone (Pinney 2004).
228
The master-mritshilpi: the struggle towards

bhadralok artisthood
In this section, I will look at the identity, or rather, the struggle for iden-
tity of the master-mritshilpis. This struggle towards the identity of ‘art-
ist’ as opposed to ‘artisan’ is squarely located in the value that the worlds
of art ascribe to the objects they produce. The larger problematic that
surrounds the value of these objects is the hand–head binary that is at
the heart of the aesthetic regime in Western art. The artisan’s work,
within the aesthetic regime, is valued as skill perfected through practice –
understood as mindless repetition. As Richard Sennett (2008) points
out, making and doing things is not considered ‘Art’ because making is
not thinking. The induction of the artisan into the Western aesthetic
regime is tied to colonial patronage and the curriculum of art schools.
An early-20th-century mritshilpi of much repute was Gopeshwar
Pal, who was christened as the ‘Lightening Sculptor’ in the Daily Tele-
graph of London in 1924. The report says:
Taking a handful of clay, he changes it into a horse’s head

within forty five seconds, with a deft touch here and there and
the staid features of the horse are transformed into a snarling,
biting, distorted mask – with ears lying wickedly flat, hair fly-
ing of a wild horse under the first restraint of a Vein. With a
sweep of the hand, Mr. Pal wipes out the image. A poke here
and a twist there, and within thirty seconds the head of a dog
appears placidly contemplating the spectators . . . Someone
wonders if the sculptor is capable of moulding a bust!
(Chakravarti 1985: 38)
The allure of the performance of Gopeshwar Pal is partly the fascina-

tion of a Western audience with this particular native craft. The colour
and malleability of this kind of alluvial clay, the dexterity of the sculp-
tor, the brisk erasure and re-use makes this a ‘performance’ of skill as
opposed to making of a sculpture. The final comment in the report is par-
ticularly striking. After this spectacular performance of turning a lump of
clay into different images in seconds, one still wonders if the ‘lightening
sculptor’ is capable of creating a faithful human form. Gopeshwar Pal
stands at the cusp of the identity between artisan and sculptor located on
a moment of encounter between the Orient and the Occident.
Dulal Pal, a contemporary mritshilpi, points out how, over the years,
there have been several stylistic infiltrations in their community. How-
ever, the one which has proved to be the most economically fruitful is ‘the
229
M O U M I TA S E N
Indian Art College style’. The Indian College of Arts and Craftsmanship
was established in 1893, as one of the colonial pedagogic endeavours ‘to
teach them [the natives] one thing, which through all the preceding ages
they have never learnt, namely drawing objects correctly’ (Mitter 1994:
33). Henry Hover Locke, a former student of the South Kensington
academy, who instituted the syllabus of the Indian College of Art, was
mainly in favour of teaching Academic Realism as a means towards ‘sci-
entific and intellectual progress’ and ‘proficiency in manual skill’ (Mitter
1994: 34). What Dulal Pal indicates by using the shorthand ‘the Indian
Art College’ style is Western Academic Realism, or Realism. This style
of modelling has proven the most commercially lucrative to the mritshil-
pis of Ghurni and Kumortuli. Making portraits of great men – political
figures, thinkers, nationalist leaders, gurus and babas – in permanent
material is the most important aspect of the trade in these clusters. The
mritshilpis become the image-makers for the local clubs and the munici-
pality because they are able to supply cheap (in comparison to the price
asked by ‘sculptors’) ‘realistic’ images (see Figure 10.3).
Figure 10.3 Dilip Pal making a portrait.

230
Beginning from T. N. Mukharji’s (1888) writings to the larger dis-

course surrounding the style of Krishnanagar, the term that is ubiq-
uitously used in order to define the style of the models of Ghurni is
‘realistic’. The Bangla counterpart of the term is ‘bastabdharmi’, liter-
ally, ‘true to reality’. Crucially, it is not the Bangla word but the term
‘realistic’ that is employed colloquially to describe certain aesthetic
qualities of a model. In this case, the distinction needs to be made
between Realism as a movement in Europe and that which travelled to the
colony. Formulated in France, with echoes and parallels in England and
the United States, Realism as a movement was dominant between 1840
and the 1870s–80s. Its aim was to give a truthful, objective and impar-
tial representation of the real world, based on the meticulous observa-
tion of contemporary life (Nochlin 1971). Thinking of Realism along
these lines, the use of this term by, presumably, the British to describe
the figurative modelling of Krishnanagar seems quite likely. The 19th-
century clay-modelling of Krishnanagar was a documentation of the
social life of mostly the lower caste people of Bengal. This documenta-
tion of rural life captured in detail speaks to the most important cause
of Realism: to paint as it is. But if we think beyond the association of
this term with the picture form in the colony, to consider the evolution
of the traditional form through an encounter with Academic Realism,
we need to think about what version of European Realism travelled
to the colony. In her essay on Ravi Varma, Geeta Kapur says, ‘what
Ravi Varma had access to, in actual pictorial terms, is not Renaissance
classicism but a reduced kind of Victorian classicism and it’s pastiche
like quality’ (Kapur 2000: 197). And the curriculum of the art schools
in the colony was structured around this reduced kind of realism. The
subject matter that was laid down as the curriculum of the art school
was characteristic of the conservative European salons of the 19th
century. The radical content of the earlier periods was weaning and
the choice of subject matter was really meant to serve the bourgeoisie
taste and uphold their snobbery.21 Interestingly, the training received
by the artisans who did not attend the art academy either paralleled or
began to echo the syllabus of the academy. The 19th-century masters
21 Geeta Kapur says: ‘The hierarchy worked through the following classification. (I) His-
tory Painting . . . to show their (the students’) skill in painting nudes and draperies alike
after plaster casts of antique statues, and to copy old masters. (II) (a) Historical Land-
scape with a classical motif (b) Portraits (c) Religious Subjects. (III) (a) Still Life (b)
Animal Painting (c) Rural Landscape (d) Genre or Domestic Scenes’ (Kapur 2000: 154).
231
M O U M I TA S E N
had their own gharanas, instituting their own syllabi. The men of their
family would train with the master craftsman, step-by-step, making
toys, then animal figures, then smaller human models, and finally, life-
size portraits.22 In the Ghurni area, there are several myths around
the master-mritshilpis who lived in the late 19th to the early 20th
centuries.
Some of these narratives are strikingly similar to the Ancient Roman
anecdote (77–79 AD), reported by Pliny the Elder in Natural History.
The Greek artist Zeuxis who lived in the 4th century BC challenged
Parhessius, his contemporary, in a contest of painting. When Zeuxis fin-
ished his still-life, a painting of grapes, birds are said to have flown down
to peck at it. But when he tried to unveil Parhessius’s work, he realised
that the curtain itself was his painting. Needless to say, Parhessius
won the contest, having ‘tricked’ the superior human intellect with
its illusionism. The story of the contest between the painters Zeuxis
and Parhessius of Athens is ultimately a championing of the power
of mimesis, and unsurprisingly, it achieved a fabular quality in the
art theory of Neo-Classical Realism (18th–19th centuries), the kind
which filtered into the colony through the training in Academic Real-
ism in the art schools. The stories around Bakreswar Pal, one of the
most legendary mritshilpis of Ghurni, have a similar nature. Bakre-
swar Pal seems to have modelled a cadaver of a cow and left it in the
ground for animal disposal. The contemporary mritshilpis report with
much pride that after a while, vultures started hovering over the ‘dead
cow’. Chakravarti says:
The dead bodies, the bones and the garbage was modelled in
such a realistic manner and the coloring was so accurate, that
vultures started descending on these. Lord Lytton (the erst-
while Viceroy of India), who had come to see this exhibition
was particularly pleased.
(Chakravarti 1985: 48)
Similar stories about birds pecking at clay fruits are told about the
great masters of Ghurni. One of these master craftsmen is Jadunath
Pal. Jadunath Pal went on to study in Italy and teach in the Govern-
ment College of Arts and Crafts, Calcutta in the 1940s.
22 Interview with Dilip Pal, Krishnanagar, December 2012.
232
The tension between the two identities of the gentleman artist and
the subaltern artisan is written into the institutional history of art edu-
cation of colonial India. Partha Mitter shows us the dilemma of the
educationists in deciding whether they want to educate the artisans
towards finer production of industrial arts or inculcate the higher
pursuit of fine arts among the elite. The struggle of the 20th-century
mritshilpi towards the identity of artist or sculptor is mainly two-fold:
first, the distinction between fine art and craft (understood as indus-
trial art); and second, the distinction in the status of the gentleman
artist as opposed to the low-caste artisan. These distinctions are still
present in contemporary clay-modelling. While the successful mritshil-
pis of Kumortuli and Krishnanagar identify themselves as sculptors,23
they are not considered so by the worlds of high art. The claim to
artisthood is established in Kumortuli primarily through the achieve-
ment and display of degrees from art colleges: Government College of
Arts and Crafts and Indian College of Arts and Craftsmanship. In the
cluster of Krishnanagar, however, the master-mritshilpis establish the
finesse of their craft by displaying medals, certificates of merit, and
awards, particularly the Presidential award for excellence in Craft.
Ramesh Chandra Pal is one of the most renowned mritshilpis, his
career spanning the mid-20th to the early 21st centuries. Pal received
recognition because of his public statuary in stone and bronze and
particularly his Durga murtis in South Calcutta, which attracted
thousands of people every year. In an interview, his son, Prasanta Pal
asserts that their workshop, even though it is in the vicinity of the
Kumortuli area, should not be confused with the rest of the cluster.
His father, he says, even though he is a ‘Rudrapal’, was not a ‘home-
trained’ artisan. He came from East Bengal to train in the art college.
He travelled the country looking for inspiration. And finally, he settled
down in Calcutta, near Kumortuli. His Durga pratimas, his son says,
exemplify more than pratimashilpa (loosely, idol-making); they are
indeed ‘sculptures’.
Interestingly, Pradip Rudra Pal, a part of the East Bengal gharana
led by Mohon Bashi Rudra Pal, always adds a signatory note on
his goddess images. While Mohon Bashi Rudra Pal’s signage reads
‘mritshilpi’ as per convention, Pardip Rudra Pal’s note always says
23 The visiting cards of the mritshilpis say ‘Sculptor’ or ‘Artist’, and the sign-boards
read ‘Art Studio’ or ‘Sculpture studio’.
233
M O U M I TA S E N
‘bhaskar’ (sculptor), sometimes followed by ‘Masters in Fine Arts,

M.S. University Baroda’ (see Figure 10.4).
Another case in point is Goutam Pal, the most successful contempo-
rary mritshilpi, who attended the Italian academy of art, and who is
considered a real sculptor by the mritshilpi community. But even Gou-
tam Pal, who calls his workshop ‘Studio de Scultura’ is not merited
in a similar way by the art world. The value ascribed to the work of
mritshilpis is not just reflected in social status, but also the prices they
can demand in the markets they have access to. However, the struggle
for these highly-skilled mritshilpis is not just one for upward economic
mobility; it is a struggle for identity outside of their own cluster.
Inside the cluster, however, they receive not just respect, but also
high positions in the associations or samitis for the community, both
in Kumortuli and Krishnanagar. And their master status creates a dif-
ferent order of hierarchy within the mritshilpi community. While in
the struggle against the worlds of high art, within the aesthetic regime,
the master-mritshilpi is denigrated as only a skilled hand, the posses-
sion of this very skill translates into a ‘master’ status in an altogether
different regime. In this order of things, the master-mritshilpi is not
only the custodian of the rules of making and doing, he is also the
one who is trusted with the power of aesthetic judgment. Simply put,
the master-mritshilpi has the power to either bestow or withhold the
secret knowledge of making and doing, otherwise called ‘technique’.
He also has the power to assert what constitutes a good murti. In the
Figure 10.4 Two signs showing the names of the mritshilpis on pandals. On
the left is Pradip Rudra Pal, whose sign says bhaskar (sculptor), followed
by ‘B.F.A. Govt. College of Art and Craft, Kolkata’, ‘M.F.A. M.S. University
Baroda, Gujrat’. On the right is another sign of shilpi Jatindra Nath Pal of
‘Sambhu Charan Pal and Sons’, Kumortuli.
234
next section, I look at these internal hierarchies that are constituted

because of the distribution of power within the mritshilpi community.
The master-mritshilpi and the poor pratimashilpi:

the discourse of ‘wrong lines’
In a conversation with a pratimashilpi in Krishnanagar, I was told,
‘What do you want to know about us, madam, go to the model-mak-
ers! We are alive because we are not dead yet. . .’24
The story of patronage in Ghurni runs parallel to the other two clus-
ters of Krishnanagar which make god-images. These clusters were not
only met with indifference in terms of patronage, their work was iden-
tified as inferior in comparison to the production of Ghurni. Every
town and village in Bengal has a few Kumbhakar families who make
god-images because of the demand generated by the ubiquitous fes-
tival culture of Bengal (the Barowari Puja in the public spaces) and
the domestic pujas. The other two clusters in Krishnanagar are not
exceptional in that sense. Like the other mritshilpis all over Bengal,
they begin to hone their hereditary trade of pratima-shilpa25 or the art
of making god-images from an early age. Most of these pratimashil-
pis do not experiment, to any remarkable extent, with the form of
the murti. And these limitations reflect majorly on their financial and
social status. The ephemeral nature of clay, lack of storage space and
infrastructure, their own incapacity as businessmen, among other
problems, plague the area of pratima-shilpa.
In Krishnanagar, their work is referred to as thakur banano (god-
making) as opposed to the work of the Ghurni cluster, which is called
model banano (model-making). The use of the English word ‘model’
as opposed to the vernacular term pratima is a comment on the tryst
of the Ghurni cluster with colonial picture forms and patronage. The
production of Ghurni in the present is characterised by different kinds
of portraits of iconic personalities and even god-images. But the god-
images made in Ghurni are not constructed in the conventional way,
using bamboo and straw armature. They make realistic (after the Ravi
Varma school of representation) figures of deities. The pratimashil-
pis, I was told by my respondents in Ghurni, do not understand the
finer nuances of model-making. The model-makers told me that their
24 Interview with Manik Pal, Krishnanagar, 2013.

25 Pratima means an icon, image or idol.
235
M O U M I TA S E N
repertoire being limited to pratimas, they cannot make portraits. And

when they attempt to mould a portrait, their ‘lines are all wrong’.26
What the master-mritshilpis assert through the comment about the
‘wrong lines’ is that the pratimashilpis were never taught the nuances
of Academic Realism.
The dissemination of technique or knowledge within the mritshilpi
community follows the method of training in gharanas. Knowledge,
in this case, skill, is not freely available but interlocuted through the
figure of the master or the guru. Within the mritshilpi community,
especially in Krishnanagar, ‘technique’ is guarded within the family
by its male members. Technique, particular to a gharana, is therefore,
transmitted from the father to the son. Even though there have been
mritshilpis who have been able to learn from a maternal uncle or a
master from another gharana, the predominant story of the great mas-
ters of particularly Krishnanagar is a lineage of fathers and sons.
Within the new organisational orders of the free market, the
pratimashilpis are able to offer a lower price than the model-makers
for a public statue in fibre glass or clay. The patrons who are look-
ing for a cheaper model approach the pratimashilpis. Often, these
portraits are seen as crudely constructed and stylistically closer to
the god-images by the master-mritshilpi. The lack of training and
the difficulty in acquiring this skill does not allow them to break the
monopoly of the model-makers in the market for secular portraiture
easily.
Conclusion
Colonial perspectivalism was also concerned with the inculca-
tion of taste, in the sense of disinterest, i.e., as the suppression
of the concern with the image’s intimacy with power, and its
efficacy as an ‘idol’. In colonial India this break – the precon-
dition for taste – is signified by a severance of the viewer’s eye
contact with the eyes of the deity and the norms of symmetry
and frontality which usually frame the efficacious representa-
tion of a deity.
(Pinney 2002: 115)
Pinney identifies the power of an idol in the fact that it appeals to the
sensorium complex of the viewer. That it trembles, as it were, with a
26 Interview with Dilip Pal, Nadia, 2013.
236
superior uncanny power of the otherworldly. However, the this-

worldly coordinates of power places realism and statue-making in
such a way in the colony that realistic images appear to assert them-
selves through the power of reality – at least, the dominant version of
reality for the colonised or the once-colonised. In the discussion of
craft and craftsmen, I have tried to show how a set of discernible pro-
cesses – the network of patronage, the investment of political regimes
in symbolic capital, the methods of dissemination of techne – create a
new template for seeing and thinking about the real in a community.
I want to conclude this chapter by looking at the reaction of a con-
temporary sculptor, Bimal Kundu, regarding the public statuary in the
city. When asked to comment on the public statues of Vivekananda,
he said: ‘It is a work that inspires respect for the subject.’ As for the
numerous statues encountered on the streets of Kolkata, the less said
the better, he says. ‘They’re not sculptures, they’re clay idols churned
out almost the way Kumortuli churns out Saraswati idols’; he is dis-
missive. ‘The only difference is these saffron-clad images have a pagdi
on their head! Even the works at Vivekananda House or at Gol Park
Mission are not worthy of being described as “art” ’ (cited in Sen-
gupta 2013).
The portraits produced by the pratimashilpis, even by some of the
model-makers, rely more on iconography than likeness in order to
establish recognition (of the subject with the representation). There-
fore, Tagore is identified by his robe and his beard, Netaji by his uni-
form, Vivekananda by the saffron robes, turban and so on. And much
like the gods they make, the identity of the subject is also iconographi-
cally produced and not based on physiognomic likeness. Often, the
works of these clay-modellers are dismissed as idol-like or lacking in
anatomical correctness. They are however characterised with a frontal
gaze, wide ‘god-like’ eyes, re-establishing the viewer’s eye contact, to
return to Pinney, with the new secular deities who grace our public
spaces now. In the larger reception of these images, the local nuances
of the craft are completely lost. In the eyes of the ‘fine artist’ or the
‘sculptor’, the entire gamut of clay-modellers – including the master-
mritshilpi and the ‘inadequately trained’ pratimashilpi – remain idol-
makers. Only now, they appear to have diversified into making other
kinds of idols: ‘ugly’ secular gods in fibre glass.
I have tried to demonstrate, through the story of the Kumbha-
kars of Krishnanagar and Kumortuli, the struggle for identity and
wealth, through their relationships with different patrons. Here, the
idea of hierarchy appears through conflicts between groups who are
237
M O U M I TA S E N
simultaneously empowered and disempowered in relation to other

groups. As I pointed out earlier, these relations are in flux, being
contingent upon political regimes and their investment in symbolic
capital. The pot-makers of the 18th century found patronage in the
kings and the aristocracy of Bengal and they diversified into image-
making. The encounter with colonial patronage and art education
led to another momentous shift in the nature of bespoke images they
began to make. In postcolonial Bengal, the central and state govern-
ments, along with the local clubs, created a new network of demand
and patronage for secular portraits and god-images in the community
of mritshilpis. The value of these objects is, however, a problematic
that is best understood in terms of regimes of art. While the worlds
of art continue to judge the work of the craftsman as not worthy of
the status of sculpture, within the mritshilpi community, the standards
of aesthetic judgement are not the same, even though there is a clear
relationship between the two. By looking at the relationship between
the sculptor of the art world and the master-mritshilpi, I have tried to
put forth the logic of the discrepancy in the ascription of value to the
craftsman’s work within the aesthetic regime. And then, by looking at
the system of training within the community, I have tried to bring out
the internal hierarchies within the mritshilpi community.
Bibliography
Bandyopadhyay, Sekhar. 2004. From Plassey to Partition: A History of Mod-
ern India. New Delhi: Orient Blackswan.
Chakravarti, Sudhir. 1985. Krishnanagarer Mrtshilpa o Mrtshilpi Samaj. Cal-
cutta: K. P. Bagchi and Co. for The Centre for Studies in Social Sciences,
Calcutta.
cal Criticism. Delhi: Oxford University Press.
Eck, Diana L. 1998. Darśan. New York: Columbia University Press.
Ghosh, Prodyot. 2011. Banglar Lokshilpa. Kolkata: Pustak Bipani.
Ghosha, Pratapacandra. 1871. Durga Puja. Calcutta: Hindoo Patriot.
Goldblatt, Beth. 1979. ‘The Image Makers of Kumortuli: The Transformation
of a Caste-Based Industry in a Slum Quarter of Calcutta,’ unpublished Ph.D.
dissertation, University of Manchester.
Guha-Thakurta, Tapati. 1992. The Making of A New ‘Indian’ Art: Artists /
Aesthetics and Nationalism in Bengal/c 1850–1920. Cambridge: Cambridge
University Press.
Guha, Ranajit and Gayatri Chakravorty Spivak (eds). 1988. Selected Subal-
tern Studies. Delhi: Oxford University Press.
238
Gupta, Dipankar. 2004. ‘Introduction: The Certitudes of Caste: When Identity

Trumps Hierarchy’, Contributions to Indian Sociology, 38(1and2): v–xv.
Heierstad, Geir. 2009. ‘Images of Kumortuli Kumars – The Image-makers of
Kolkata: Changing Notions of Caste and Modernity during the Last Cen-
tury’, unpublished Ph.D. dissertation, University of Oslo.
Jain, Jyotindra. 1999. Kalighat Painting: Images from a Changing World.
Ahmedabad: Mapin Publishing.
Kapur, Geeta. 2000. When Was Modernism. New Delhi: Tulika.
Mitter, Partha. 1994. Art and Nationalism in Colonial India, 1850–1922:
Occidental Orientations. Cambridge: Cambridge University Press.
Mukharji, T. N. 1888. Art-manufacturers of India: Specially Compiled for the
Glasgow International Exhibition, 1888. South Africa: Government Printer.
Nochlin, Linda. 1971. Realism. Harmondsworth: Penguin Books.
Pal, Gautam. 1997. ‘Krishnanagarer Mritshilpa o Bhaskajyer Katha’, Paschim
Banga, 17–21(31): 183–86.
Pinney, Christopher. 2002. ‘A Secret of Their Own Country: Or, How Indian
Nationalism Made Itself Irrefutable,’ Contributions to Indian Sociology,
36(1and2): 113–50.
Pinney, Christopher. 2004. Photos of the Gods: The Printed Image and Politi-
cal Struggle in India. London: Reaktion.
Pollock, Sheldon. 1985. ‘The Theory of Practice and the Practice of Theory
in Indian Intellectual History,’ Journal of the American Oriental Society,
105(3): 499–519.
Rancière, Jacques. 2006. The Politics of Aesthetics. New York: Continuum
International.
Risley, Herbert Hope. 1892. The Tribes and Castes of Bengal. Vol. 1. Printed
at the Bengal Secretariat Press.
Robinson, James Danter. 1983. ‘The Worship of Clay Images in Bengal’,
unpublished Ph.D. dissertation, University of Oxford.
Sarkar, Smritikumar. 1998. ‘Indian Craft-Technology: Static or Changing – a
Case Study of the Kansari’s Craft in Bengal, 16th to 18th Centuries,’ Indian
Journal of History of Science, 33: 131–42.
Sengupta, Ratnottama. 2013. ‘Indian Idol but Not an Art Icon’. Times of
India, 12 January.
Sennett, Richard. 2008. The Craftsman. New Haven: Yale University Press.
Varma, K. M. 1970. The Indian Technique of Clay Modelling. Santiniketan:
Proddu.
239
11
THE COMMODIFICATION
OF CASTE AND POLITICS IN
KOLKATA’S KUMARTULI
Geir Heierstad
The Bengali Kumars or Kumbhakars belong to the pan-Indian potter

caste (jati). They have discarded the potter’s wheel since few want
their products. While most have entered into other occupations, some
still work with clay. In parts of Bengal, they make a living making
murtis, statues, of unbaked clay. And in Kolkata, the Kumars of
Kumartuli1 (neighbourhood of potters) form the last of the city’s larger
caste-based neighbourhoods. Its narrow streets are packed with seem-
ingly traditional artisans employing age-old techniques to create
sacred statues integral to ancient religious traditions. Thus, they might
appear like survivors of a long-lost tradition untouched by political,
cultural and economic changes within the society at large.
This is not the case. In many ways, Kumartuli is a hub of moder-
nity and commercial attitudes tightly connected to Bengali politics
and global markets. The Kumars are hard-working business people
claiming an identity as artists in an economically strained market. And
politically, the Kumars attempt to achieve development goals through
caste committees, as well as electoral politics. While old caste catego-
ries and stories are hardly remembered, being a potter by caste is their
selling point, their market advantage and their brand.
In this chapter, I propose that the modernity of the Kolkata Kumars
modifies the very notion of caste, as belonging to the caste of Kumars is
transformed from regulating marriage, commensality and labour, into
an emblem under which products are sold and political battles fought.
1 The empirical material presented is based on a fieldwork among the Kumartuli

Kumars in 2006 and 2007, as well as sporadic revisits, of which the last took place in
the autumn of 2013.
240
C O M M O D I F I C AT I O N O F C A S T E A N D P O L I T I C S
This, I argue, represents the commodification of caste – a process

through which caste affiliation is turned into an economic benefit or
good. Further, the politics of caste that the contemporary Kumars rep-
resent is mainly based on the continuing process of commodification.
Caste matters in West Bengal and Kolkata well beyond regulating
marriage. While occupational affiliation develops with changing mar-
kets and the emergence of new ones, caste regulates, to a certain extent,
access to networks that provide job opportunities or the financial secu-
rity to establish something new. Caste affiliation unites and is regu-
larly used, admittedly with various successes, to fight political battles
or secure access to power. The upper echelons of Bengali politics are
undoubtedly dominated by people hailing from castes that tradition-
ally have been better off, status-wise and economically. At the same
time, people belonging to castes that traditionally have been treated
with contempt and violence, try to use their numbers and unity within
the framework of participatory democracy to gain access to power.
Among the Kumars of Kumartuli, occupation, market opportunities
and political influence are also tightly related to their caste affiliation.
Still, they do not use this affiliation to simply get access to markets
or act as a political unit in their constituency in order to install their
man/woman in a position of power. Caste is not simply used as a
wide-ranging network, as a societal group of supportive peers or as an
identity to be evoked for political ends. Instead, they have transformed
their caste affiliation into an emblem and a brand, flagging the Kumars
as the true and traditional clay workers and makers of thakur proti-
mas (religious statues) in the state. As such, the Kumars are important
as keepers of the history and tradition of all Bengali Hindus. And
Kumartuli, their neighbourhood in Kolkata, is the central space where
this valued heritage is kept alive. It is with reference to their caste
affiliation and Kumartuli location that they enter into local and global
markets and engage in political efforts to ensure development.
Thus, I argue that how we perceive and analyse caste must be modi-
fied. Further, we have to appreciate that the ways through which caste
connects to economy and politics varies enormously. Empirically
grounded analysis that is sensitive to context and historical depth will,
as occasionally stressed (Searle-Chatterjee and Sharma 1994; Dirks
2001; Gupta 2004; Béteille 2005), provide an understanding of caste
that undermines all-encompassing, catch-all ideological understand-
ings, and in turn, strengthen a much more practice-oriented approach
in which caste implies forms of behaviour that differ according to
place, time and situation.
241
G E I R H E I E R S TA D
The argument of this chapter is divided into two sections built

around separate, but still deeply connected, aspects of life and work in
Kumartuli. The sections connect cases concerning everyday practices
of economy and politics. First, I will show how the meaning of caste is
transformed through the Kumars’ marketing strategies. These strate-
gies are important in commodifying caste and making caste affilia-
tion into a brand. Second, through branding caste, the Kumars work
through various committees in order to ensure development from
local politicians, irrespective of party background.2 Development
in this context implies better sewage, stable power supply, targeted
health services and better business/tourist facilities. In order to gain
recognition for their important role in the society at large, Kumartuli
has been turned into a concoction of production space, business loca-
tion, residential area and theme park. Before I enter into the everyday
economy and politics, however, the Kumars and Kumartuli require a
presentation.
Caste, classes and a neighbourhood

Encircled by multi-storeyed concrete buildings on three sides and the
river ghats on the fourth is the collection of seemingly rundown work-
shops and low buildings that make up Kumartuli. There is a semi-
official entrance to the neighbourhood on the Chitpur road, with a
signboard and a small office that charges money from those who want
to take photographs of the Kumars. It is a slum, it is a residential area,
but most importantly, it is a business area, with buildings constructed
for the needs of the ongoing work. The constructions are a combina-
tion of kaca buildings, made from traditional materials such as mud,
clay tiles and bamboo and paka buildings, made from ‘modern’ cement
and bricks with corrugated tin roofs. All have electricity, but only a few
have running water. Water is available from communal pumps. A cou-
ple of the thoroughfares are asphalt, but most of the narrow lanes are
kaca. The covered, but not piped, sewer system is locally called paka.
The image-makers inhabiting the small neighbourhood constitute
several different classes. When it comes to the maliks, the workshop
2 Emphasising one’s caste as important to the society is not something entirely new. It
is an old model to achieve economic and political aims. However, it is empirically
underrepresented in attempts to understand how caste is practiced in contemporary
India, not to mention West Bengal.
242
owners, there is a difference between those running a single work-

shop and those who run a small-scale art industry with a number of
workshops and storehouses within and outside Kumartuli. While they
all are dependent on loans in advance of the pujas to finance the con-
struction work, the latter often have the possibility of acquiring a sur-
plus, which they can reinvest in the business, their children’s education
and consumer goods.
The malik, as well as their wage labourers, place their ‘original’
home, adibari, in either Bangladesh or in Nadia District in West Ben-
gal.3 Thus, the notion of adibari links the jati or caste to either West
Bengal (India) or East Bengal (present Bangladesh). The maliks have
organised themselves into two samitis (associations), dominated by
East and West Bengalis, respectively. The hereditary occupation, which
the contemporary Kumars have more or less actively chosen to pur-
sue, is only one aspect of their identity, and equally important is their
experience of being connected to a geographical area outside Kolkata.
Regarding jati affiliation, most of the people engaged in the clay
production are Kumbhakars, but there is a small minority of people
from other jatis as well. Every malik and labourer is aware of his or
her jati affiliation. As for the majority, they are potters, Kumars or
Kumbhakars. It is they who make the unbaked clay thakur murtis and
protimas, while their forefathers used to work at the potter’s wheel,
making earthenware and their foremothers made clay putuls (dolls
and miniatures). The caste knowledge seldom extends further than
to this jati awareness. Knowledge of sreni and gotra (two kinds of
sub-groups, clans, often used interchangeably) is lacking among the
great majority. Most have a rudimentary knowledge of their jati’s cre-
ation history, which provide them an important place in society. In the
words of senior artist Gour:
It is a story. . . that when Mahadeb (Shiva) married Sati

(Durga) . . . I do not know whether this is true or not, but
this is the story . . . From one of the beads of the garland of
rudraksha he [Mahadeb] used to wear, he created this caste
of Rudrapals . . . So, from rudraksha he created this caste to
3 Moumita Sen, in this volume, traces some of the more successful mritshilpis that more
frequently than others also make profane murtis. While Sen emphasises internal hier-
archies and power relations, this chapter investigates the Kumars more as a societal
group within society at large.
243
make all the ingredients like kitchen utensils and pots that we
usually use in our pujas . . . Earlier everything was made of
clay, these utensils. . .
Another rendition of the story also confirms the importance of the

profession by linking it to marriage:
When Shiva and Durga were married . . . Durga means the

daughter of Giriraja, he is Himalaya, King of the Mountain . . .
In our marriages we need ghat4 . . . Then there were no one
to make these clay pots. Nobody at that time used to do the
Kumars’ job.
Of all the pujas that the Kumars make images for, Durga puja is the
single most important. The present-day Bengali Durga puja tradition
goes back to the rich peoples’ quest to imitate aristocratic religious fes-
tivals of bygone days (Banerjee 2004: 43). They sought social prestige,
rather than to maintain a link to the common people – like that between
the rajas and jamidars (land owners) and their tenants (ibid.). While the
nouveaux riches, the babus, opened their houses for every class, there
was no guarantee of full participation by local residents, as for example,
at Savabazar Rajbari (palace), which entertained the public for free for
12 days, but restricted entry during the three most important nights.
Then, the British were invited over for ‘some sherry, champagne, brandy
[and] biscuit’ (Shripanto cited in Banerjee 2004: 44).
As Durga puja turned more and more into ‘business entertainment’
(Chaliha and Gupta 1990: 332), the common people experienced that
their space was shrinking. This paved the way for pujas arranged by
‘common’ people, the sarbajanin pujas. It started in the districts in the
latter parts of the 18th century and moved slowly towards Kolkata.
The first community-based puja in Kolkata today took place in 1910
(Chaliha and Gupta 1990: 332; Banerjee 2004).
With the democratisation of the puja in Kolkata, the number of pujas
increased (Goldblatt 1979; Banerjee 2004). For the Kumars of Kumar-
tuli, this meant more work. While some families from Nadia already
had settled in the capital, the custom of seasonal travel to Kumartuli in
4 Ritual pot; little clay, earthen or metal pitcher or pot; symbol of purity and bless-
ings from God. Managal ghat means blessings from God; the place where God is
residing.
244
order to make murtis ended after the establishment of sarbajanin pujas.

Now, the Kumars moved their families to Kumartuli and became regu-
lar residents. Moreover, they got an entirely new market to cater for,
and started a bazaar sale of more-or-less ready-made images.
While the democratisation of the pujas led to permanent settlement
in Kumartuli, it was the increased competition between Kumars to
attract new clients and individual Kumars with artistic ambitions that
reinvented the craft.
But the change did not stop with the influence of Kumars from else-
where than Nadia. From the early 1960s and onwards, individuals
without any inherited affiliation to the artistic work with clay started
to design images, as well as more spectacular pandals. These indi-
viduals, of whom many had an education from art colleges, started to
dominate the market from the late 1980s – in particular, among the
more prestigious ‘theme pujas’. Such pujas made a concept out of the
images, and especially, their make-shift shelter, the pandals, creating
impressive scenes from everything – from an Odhisa fisher village to
Egyptian pyramids. Artists from outside Kumartuli were able to earn
more money as theme or concept designers than most Kumars. Only
in the late 1990s did the Kumars start to challenge the outsiders as
designers of theme pujas.
As a competitive move to counter the outsiders, the Kumbhakars
often emphasise their jati affiliation as a guarantee of their artistic
abilities, claiming ‘they have it in their blood’. To purchase an image
from Kumartuli is, as will be described in the next section, advocated
as a purchase of an image with a genealogy going back to the ‘old’ his-
tory of the Bengali people.
Economics: developing the brand

Numerous trajectories are at work in Kumartuli. External influences
such as the establishment and increasing prevalence of sarbajanin puja
and internal forces such as the competition between Kumars are but
two examples. As competition grew, the need among the Kumars to
stand out, either through artistic skills and/or innovative appearance
of the thakurs they merchandise also became more imperative. Within
Kumartuli, several events became important in forging how to
approach the competitive environment of murti making and trading.
First, a local Kumar went artistic and broke with the traditional
style of sculpturing. This created unprecedented attention and oth-
ers soon learned that innovation was a good marketing strategy. The
245
artist in question was Gopeshwar Pal (1896–1952). Gopeshwar was

one of the persons from Nadia who travelled to Kolkata in search of a
full-time job. The workshop he established is named after him and is
today run by Byomkesh, who is related to Gopeshwar through his sis-
ter, Jamuna, who was married to Gopeshwar’s son, Siddeshwar. While
still working in Nadia, Gopeshwar made regular journeys to Kolkata
in order to work as a portrait and architecture painter (Chakraborty
1985). In 1924, his life took a new turn. Byomkesh says that in that
year, Gopeshwar found an advertisement for the British Empire Exhi-
bition, a colonial exhibition held at Wembley, London in 1924–25.
What made the journey possible is uncertain. Byomkesh says he just
applied and was selected to go together with the renowned dancer,
Udayshankar. A recent newspaper article states that it was Percy
Brown,5 then principal of Government Art College (during the period
1909–27), who selected Gopeshwar to represent India at the exhibi-
tion at Wembley, due to his extraordinary skills of sculpting the like-
ness of any person in a matter of minutes (Roy 2005). Yet another
story claims that the contemporary Bengali Governor Lord Carmi-
chael6 saw Gopeshwar’s work in Krishnanagar during an exhibition
there in 1915. Still, the fact remains that he left for England that year – a
journey that his family did not support.
In England, he showed his talent as a clay artist in various ways.
Byomkesh says he made a horse’s head in 45 seconds. George V and
other members of the British royalty are said to have visited the exhibi-
tion (Roy 2005). Due to the reputation he gained in England, Gopesh-
war was asked to embark on a tour around the world to exhibit his
talent. He turned the offer down and decided to go to Italy, where he
attended a sculpting course (ibid.). There, he learned to make sculp-
tures of stone and bronze.
When he came back to Bengal, Gopeshwar took two decisions: to
establish a workshop in Kolkata’s Kumartuli, not in Krishnanagar;
and to work with stone and bronze, not clay. Gopeshwar’s reputa-
tion as an eminent sculptor grew in Kolkata, and in 1932–33, he was
requested to make the Durga idol for the Kumartuli sarbojonin puja – an
offer he accepted. Gopeshwar abandoned the ekchala (single frame)
image and separated Durga and her children (Agnihotri 2001; Baner-
jee 2004). In addition, he wanted to capture the action of Durga
5 Best known as the author of Indian Architecture (1942–1943).

6 Governor from 1912 to 1917.
246
killing the demon, and thus, did away with the iconic representations
of the figures. He designed Durga as a battle-hungry goddess with her
trident raised to kill the demon Mahishasur as it was pursued by the
lion, while Mahishasur ‘had rippling muscles to accentuate the glory
of Durga’s victory over him’ (Banerjee 2004: 49). When Jagadish Pal,
who was to do the actual sculpting work, was told how to proceed,
he quit, feigning illness. Gopeshwar had to make the image himself.
The new form of Durga was appreciated by the people, but the pun-
dits were annoyed and initially refused to worship the image (Agni-
hotri 2001). However, the favour granted by the people made the style
widespread. This was the first art-er pratima (lit. art image), and the
beginning of a new era.
Second, refugee Kumars from East Pakistan also introduced new
techniques as they started to compete with the local Kumars of Kumar-
tuli. Providing something new in the market gave them a flying start
that the locals tried to counter.
Third, as the artistic styles no longer were bound by convention,
educated outsiders started, as mentioned above, to make images and
pandals. They captured many of the more prestigious pujas. Again,
the locals – consisting now of East and West Kumars – sought to pro-
vide art education to their heirs in order to compete with the outsiders,
but with limited success.
Thus, as the work of the image makers turned more competitive,
new strategies to survive in the market were applied. The artists made
more innovative images where the exhibition qualities overshadowed
those of worship. A few were able to send their children to art colleges.
Simultaneously, but less overtly, the Kumar artists started to explore
their jati background and its explicit connection to the neighbourhood
as a competitive advantage and tool. Thus, while a jati’s connection
to a given profession does not provide any monopoly or exclusiv-
ity whatsoever, the Kumartuli Kumars attempted to mould a bond
between consumers and their profession as image makers, wherein
emotions of tradition and the natural are played out.
The attempts of recapturing the markets through using their caste
background and their caste’s location as a brand consists of multiple
choices and practices made at both individual as well as collective lev-
els. The most important of these is connected to an understanding of
inherited experience providing both artisan skills and artistic origi-
nality. Further, and in a seeming contradiction to the inherited caste
traits, the Kumars sought the re-establishment of the importance in
the market through formalised education and public exhibition spaces.
247
Combined, these practices result in what I label the commodification

of caste.
Among the Kumars, it is common to explain their connection and
expertise in murti making as an inherited attribute. As G. Pal states:
‘Making murtis is in our blood, it runs in our family.’ It is in their
blood because they, the Kumars, were made to work with clay, to
make pots and murtis. And to the majority of the Kumars, there is
truth hidden in their creation story:
We are the Rudrapals. In Bengal there are so many other castes

that use Pal as their title, but Kumbhakar are the Rudrapals.
We are Rudrapals as we are created from Rudraksha.
It is an old story. It is our old history, just as we have a story
of Ramayana. Yes, there must be some truth in it. Who cre-
ated the story of Ramayana? It was Valmiki . . . So this is the
story of the origin of the Kumbhakar community. This is the
puranic story, just as the Ramayana is a puranic story.
While they learn the craft by spending their childhood in the work-
shop or being accepted as trainees by a renowned artist, the level of
competency some achieve is a source of pride. Stories elaborating their
natural take on the complex techniques are common:
Once I was making an idol – Durga was looking like a Bengali

girl and the demon like a British man. The demon was just
getting out of his disguise as the buffalo and trying to throw
the buffalo at Durga. So in the structure, the dead body of
the buffalo was held by the demon in his hands. Suddenly, a
group of foreigners with one interpreter [guide] entered my
workshop. They became surprised by the structure – because
it seemed quite impossible to make a [clay] structure in which
a buffalo was held by only two hands. Then they asked if we
became artists after learning engineering. Because that type of
structure can be made only by engineers. Then the guide
told them that these people are only artists, but they have
their own brains – though they have never achieved any
degree from any college. And they were surprised by our
art work.
Besides the actual craft of binding straws to the wooden structures

and applying multiple layers of clay, the inherited traits also includes
248
innovative abilities. In this process, the name of Gopeshwar again

becomes important in order to create a common genealogy for all
the Kumar clay artists. In a conversation with an elderly artist, here
named Gour, and who arrived from East Bengal to Kumartuli in the
1940s, about his family, he suddenly started to talk about famous art-
ists not part of his family:
We have heard about N. C. Pal [Nitai Chandra Pal] – he was

still living that day, but Gopeshwar Pal was no more . . . They
were big artists. They raised the name of Kumartuli.
Yes, there were many others, but these people were famous.
Gopeshwar Pal went to bilat7 (a western country), he made a
statue of George the Fifth within 15 minutes . . . Such was his
power (ksamata). His ancestral home was in Krishnanagar.
And N. C. Pal was also one of the great artists of India.
We used to listen to these people’s names. And I have seen
N. C. Pal – his house was there. But his sons were not able
to continue the glory of his work. Mani Pal is the nephew of
Gopeshwar Pal. Mani Pal is also a great artist . . . Gopeshwar
Pal was the greatest artist we ever came to know about – there
is still much of his handiwork in Kolkata – like in Dalhousie,
in Rupbani (Hatibagan) . . . He is the best. Before we came –
Gopeshwar Pal used to make the idol of the Kumartuli sarba-
janin durgutsab. Once the idol was damaged in a fire – within
one night he had created it again!
The fact that Gour mentions Gopeshwar when asked about his fam-
ily’s background and their arrival in Kolkata is typical for a successful
artist such as himself. An ‘outsider’ from East Bengal, he nevertheless
places himself within a genealogy back to the master craftsman from
Nadia. Thus, he places his family within the historical development
that pre-dates their arrival. The heritage of Gopeshwar belongs to all
the modern Kumars, be they from the Kumbhakar jati or others. He
is a common forefather in terms of working as an image-maker, and
is incorporated by most individuals in their rendition of their family’s
history. Such genealogies cannot be called mythical or invented, as
they are not told to explain actual (biological) ancestry. Moreover,
7 England or Europe and also America; loosely translates into ‘a western country
overseas’.
249
Gour, as a successful Kumar, places himself within a tradition of the

Krishnanagar image-makers – something that the people from Krish-
nanagar regularly argue against.
Through creating a genealogy back to master artists such as Gopesh-
war, most Kumars emphasise their capability for innovation. Gour’s
statement, from the same conversation where he mentions Gopeshwar,
is exemplary in this regard:
Many of my works are in the collection of every artist – pho-

tos, Nataraj and many other works – I have discovered many
new forms of art work (shilpi kaj).
The hair they now use for the idol . . . earlier its quality was
not very good. We [Gour and his brothers] and our father used
to think about how we can make it better. Then we discovered a
totally new thing. We coloured those jute fibres and tried it . . .
and then we put oil in it and then passed a comb through the
jute in order to arrange it. But now you can see this type of hair
everywhere . . . people were able to duplicate it. Now thousands
of people are using this method in order to earn an income. Now
you can buy it everywhere.
After Gopeshwar, exhibition value became requested in addition to

cult value by the clients, in a totally different manner than before.
This changed the Kumars’ approach to their work and themselves – a
change which can be analytically best understood as a modernistic
turn, or the emergence of a site- and time-specific modernity (Heier-
stad, forthcoming). Gopeshwar can, thus, be said to have facilitated
an alternative route to modernity through his break with common
image-making. He opens up a space, which breaks most of the bound-
aries concerning how Durga can be modelled. Now, it becomes the
responsibility of individual Kumars to go out into the streets and cin-
ema halls and seek inspiration in order to create something ‘new’. And
it is all done within the generation-old market of sarbajanin durgut-
sab, the capitalist heir of the patronage of earlier times. This specific
example of the ‘individuation of tradition’ (Errington and Gewertz
1996) is a signifier that modernity is approaching, a time when repu-
tation is fragile and the past does no longer provide all the answers
(Miller 1994: 321–22).
As a somewhat newer addition to re-establish the importance of
Kumartuli, and as a partial response to competition from outsiders, the
250
Kumars started to emphasise the importance of formalised education in

order to succeed as a clay artist. Among the younger generation, P. Pal is
one who has chosen this path; in 2001, he got his MA degree from The
Government College of Art and Craft in Kolkata. His reason was straight-
forward: ‘We make traditional murtis, to make it different in some way
and to make sculptures – to think in a different way – I choose to go to
an art college.’ And a little later in our conversation, he added:
In fact, if we have an art college background this can help us

in this business. We can get good orders. Yes, we have many
young boys who have, or study to get, an art college back-
ground. And other boys from Kumartuli . . . they are now
mainly graduates. Many boys are getting an education. . .
I think we should take control over the entire process of
making the theme, including the murti. Then it becomes a total
artwork. Earlier, people just ordered Durga murtis [from us]
and they got for instance people in Midnapur, Nadia, to make
the decoration and the pandal – that was very popular. And
they got people from Chandranagar to arrange the lighting.
But now, if we take control of the entire process . . . like if we
want to make Odhisa the theme, then we can take one man
from Odhisa . . . one or two artists from Odhisa, then they
can work here with us . . . making the murti and the pandal.
So, then we can arrange for some local Odhisa music to be
played. That is a total artwork . . . the kind of work we should
do . . . When I was in Art College, we did not manage to make
much investment [necessary in order to make larger images],
we [the family] used to make small art works. Now we have a
big budget, making theme pujas – so now we can do whatever
we want to in order to make our concepts clear and beautiful.
Thus, being an artist is partly reinvented as a modern profession

‘based on formal qualifications that can in practice be secured by
only some and not all members of the society [through] high levels
of education . . . and . . . the expenditure at least of time and usually
also of money’ (Béteille 2005: 305).
However, it is the quality of having the knowledge in your blood,
for instance, mastering the craft of straw binding, applying clay and
decoration, combined with formal education that are marketed as
the forte of the Kumars. Art education alone, thus, does not help,
251
according to the Kumars. In the words of a senior Kumar, Gaur, when

talking about outsiders:
He is reputed enough and has passed out from Art College. He

even does not know how to bind the straw. In our Government
Art College they never teach how to bind the straw (laughing).
But there should be a department for straw binding.
Caste Kumars are able to combine both, as Nimai states: ‘Of course
it is good to create a new thing, but not by the destruction of the old. I
have to accept my father first, only then can I accept my son.’
To sum up, in Kumartuli, the individual Kumars are responsible
for promoting their work and themselves in a way that attracts cus-
tomers. The first person to do so, to fashion himself as an artist, was
Gopeshwar. He utilised the new possibilities created by the sarbajanin
durgutsab. While the traditional image was made in a patron–client
relationship using a given client’s (household’s) special mould to shape
the face of Durga in the same way as it always had been, the new one
demanded a new mould every year. The individual Kumar must spend
time investigating how to create a Durga that will be appreciated by
committees and the audience. He must mould with reference to pre-
vailing fashion, popular culture and contemporary events. He has to
seek inspiration from tradition and scriptures, as well as from Bolly-
wood and Hollywood movies.
When the Kumar artisan threw away the moulds, he turned into an
artist. This was not done without regret, as there is a sense of decline
in the devotional value and of denying their heritage. Increased expo-
sure to the market provides opportunities for the innovative Kumars
who are able to sell their products as both a traditional artisan and a
modern artist.
As such, in the eyes of the Kumartuli Kumars, they are tradition-
bearers adapted to the demands of contemporary consumers and cli-
ents. And this unique combination of blood-based know-how and
formal education is their selling point, which is what provide the cli-
ents with images of both devotional and exhibition value. Further, as
the Kumartuli Kumars are the original image makers of Kolkata, pur-
chasing images from them also provides the clients with genuine, tra-
ditional objects that connect the consumers with their heritage, their
history. Buying images from the Kumartuli Kumars certifies authentic-
ity and genuineness. However, that is a message not every potential
client either realises or accepts.
252
Challenges in re-establishing more of the former monopoly of image

production and capturing the theme-puja market is not the end of the
Kumars’ continuous drive to succeed. A more recent attempt to re-
build the neighbourhood as an air-conditioned mall for the artists – a
murti mall – is in the pipeline.8 At present, this is the culmination of
the attempts of the branding of Kumartuli and the commodification
of caste among the Kumars. Moreover, the plans also showcase how
caste as a brand and commodity is used to achieve political goals, such
as attracting funds for urban development.
Politics: using the commodity

Being an iconic neighbourhood known to most inhabitants of greater
Kolkata and beyond, Kumartuli appears neglected: neglected by the
inhabitants, the workshop owners and the Municipal Corporation. To
a certain extent, it is exactly this negligence that provides the neigh-
bourhood with its aura of being a part of the past, gathering for the
present, a resource for those who want to connect with tradition, both
sacred and profane.
However, among the Kumars, development, such as better infra-
structure, is a shared aspiration to which they turn to politicians and
government institutions to attain. Achieving development is difficult as
the artists are not a homogenous group of people and they are not rep-
resented by any autonomous umbrella organisation. Numerous organ-
isations are at work in the neighbourhood, ranging from local labour
unions to regular political parties, not to mention various NGOs.
In terms of the presence of the political parties, the Kumars of
Kumartuli do not represent any exception to the overall political land-
scape of Kolkata. In the post-independence period, both the Congress
and the CPI(M) have had a rather equal support from the artists, while
the Trinamul Congress (TMC) captured a large portion from the late
1990s onwards. However, participation in party politics and elec-
tions has not been an important strategy in order to achieve political
benefits, such as development funds and schemes. Both for the wage
workers and the workshop owners within Kumartuli, the various
organisations established by them and uniting different groups have
played a much more significant part.
8 Since the smaller and less economical successful Kumars would have problems buying
into the mall, this re-building would in effect also imply gentrification.
253
The wage labours’ union was established in 1974 in order to regu-

late working hours. Its main opponents are the two samitis represent-
ing the workshop owners. The first was established by East Kumars
in the 1950s with the main objective of providing guarantees for bank
loans. They wanted it to represent everyone, but some West Kumars
opposed them and started a competing samiti. The fights for funds
and development for their members have been fields of high competi-
tion. Only during the past few years has there been some kind of col-
laboration, and then, mainly concerning the possibility of developing
a murti mall.
The Kumars work through their samitis to develop the neighbour-
hood. Within the samitis, they are discussing their selling point and
they want to strengthen the Kumars’ position in the market through
economic means (bank loans), infrastructure (electricity, drainage,
working space) and providing the members with combined work-
shop and outlet spaces. Again, they want to raise the awareness and
position of the Kumartuli Kumars in society at large. In the battle to
attract development from the authorities, the samitis use their now
commodified caste background and Kumartuli’s history in order fight
the necessary political battles.
Unnayan and its English equivalent, ‘development’, are much-used
words in Kumartuli. The feeling of living and working in a slum with
the health hazards and lack of prestige that this implies is widespread.
According to one of the younger Kumars, who has entered the internet
to sell his products mainly abroad, there are a number of challenges
facing the neighbourhood:
The number one problem is electricity. The electricity problem

in my workshop is the DC-line. In the rest of Kolkata, the DC-
line is gone, but in Kumartuli there is still a DC-line. There are
two types of lines, AC-line and DC-line, you know. The line
here is old; there has been no modernisation. I cannot use my
computer when there is a DC-line here. The DC-line is a very
old system. This system is very old, from the time of the Raj.
This is a big problem.
The second problem is moisture, the damp weather. The
environment is not healthy . . . Sunlight is not reaching the
shops.
And the third problem is the clay, from the Ganga . . . eh . . .
how to say it? The hospital people say the clay is not good, dan-
gerous clay for . . . [pauses] There are some problems with Ganga
254
water, and we use clay from the Ganga for our work. With the
clay from Ganga, we get some disease.
And as Mintu Pal, the joint secretary of the Kumartuli Mritshilpi

Samiti, officially described the situation in an interview: ‘There is a
serious space crunch, filthy environment and no civic amenities in
terms of road infrastructure and drainage’ (Bhabani 2007).
They state they are underdeveloped and that the government does
not do anything to help them. When asked to be specific about what
kind of development they want, most point to the dilapidated work-
shops; they mention the lack of water and medical facilities, the need
for an exhibition space and dependable electricity. However, the
concept of unnayan is almost always used to refer to ‘development’
provided from the outside, be it an NGO providing health checks or
government-funded and implemented rehabilitation.
The samitis had success in the late 1970s and got the Kolkata
Municipal Corporation (KMC) to provide a covered sewage system.
Besides this, little has been done in the eyes of the Kumars. From the
samitis’ perspective, their poor success rate seemed to improve around
2004. Then, under the Left Front, the Kolkata Metropolitan Develop-
ment Authority (KMDA) initiated a plan to build a multi-storeyed
building with exhibition rooms and spaces for workshops. According
to the joint secretary Mintu Pal:
We want the Kumartuli development project to be imple-

mented. It will not only be for the betterment of existing
artisans working in the area but for rejuvenating the art of
ancient Bengal . . . We think there will be a huge employment
opportunity if the Kumartuli rejuvenation project is done in
a proper manner.
(Bhabani 2007)
Another interviewed image-maker, Somnath Pal, adds: ‘We are now

in talks with the state government and KMDA officials regarding plot-
related matters. A location near Kashipur in north Kolkata is likely
to be chosen for shifting artisans temporarily in November this year’
(ibid.). As of 2015, nothing has happened besides that the few families
that actually moved out of Kumartuli awaiting the rejuvenation have
returned.
To navigate the political landscape of Kolkata is difficult, with the
KMDA – under the administrative control of the Urban Development
255
Department of the Government of West Bengal – and the KMC hav-

ing overlapping responsibilities. However, when it comes to larger
infrastructure development in the Kolkata Metropolitan Area, it is the
state-controlled KMDA one need to address.
It was a Lok Sabha MP from the Calcutta North West constitu-
ency, hailing from CPI(M), Sudhangsu Sil, who initially led the com-
mittee which was set to administer Kumartuli’s makeover in 2007.
The redevelopment project was part of the Jawaharlal Nehru Urban
Renewal Mission. The KMDA decided to construct a three-storeyed
building for the artisans. It would cover 3.22 acres and 298 shops,
of which 166 are clay workshops, 51 workshops for decorative
material, 81 other establishments and 524 residential flats (KMDA
2007). The project should include, in addition to workshops, com-
munity facilities such as an exhibition hall, community hall, health
centre, post office, banks and an art gallery. The total project cost
was set to be INR 307 million (ibid.). From the KMDA’s perspec-
tive, the motivation for initiating such a large development project
was to provide the artisans with a better life as they ‘have kept alive
a traditional art which has withstood the onslaught of modernity’
(Ghosh 2013).
The murti mall was the state’s response to the Kumars’ politi-
cal demand for unnayan. It was a catch-all answer to demands for
improved sanitation, modern power distribution, occupational safety
and health, as well as facilities to promote their traditional craft and
trade. It was a proposition that a majority of the Kumars came to
accept and support, even as it was not what they had imagined as a
solution to their problems.
Thus, from petitions to political parties, mainly CPI(M) represen-
tatives, and civic bodies such as KMDA, the two samitis achieved
a whole lot more than they had ever asked for. But in the plans of
the KMDA and the politicians involved, the murti mall is presented
as the answer to secure the livelihoods of the artisans who manage
an important part of Bengali tradition. Even as the murti mall will
combine residential areas with workshops and exhibition spaces and
health facilities, it is portrayed as a heritage centre, as a place in which
to rejuvenate an ancient tradition. As the government of West Bengal
states in an overview of places of interest in Kolkata: ‘At Kumartuli
clay images of Gods and Goddesses are fashioned by potters. It is
a place of Art and Craftsmanship of Bengal School of Art heritage’
(BanglarMukh n.d.). Being part of most heritage walks arranged in
256
Kolkata, the heritage status of Kumartuli and the work of the clay art-
ists are firmly accepted and recognised by society at large.
Thus, the involved officials and the general public echo the language
of the Kumars as they fight for their economic survival and domina-
tion over the murti market. The development of Kumartuli is seen
and portrayed as a heritage project, accepting the special status of the
neighbourhood and its artists even though they never have had any
exclusive control over the production of religious and profane images.
Despite the outcome of the murti mall project, the Kumars have been
successful in keeping and capturing the consensual understanding of
Kumartuli as the original home of the original makers of images.
In Kumartuli, with its open workshops and the busy lanes lined
with statues in varied degrees of completion, there are always groups
of gazing and photographing sightseers around. Even as the KDMA
have not been able to build the samiti-supported murti mall as a heri-
tage centre, Kumartuli figures in most guide books as a place not to be
missed. Thus, Kumartuli are perceived by most of the inhabitants
as having a market value precisely because it is a neighbourhood
dominated by members from a single jati. Parimal Pal captures this
when he states, in a discussion over the samitis’ support for the
KDMA project:
It is very difficult to communicate with all the people of

Kumartuli, because there are different types of people living
here. For instance, we have our own home here and our work-
shops. Here in Kumartuli there are also some people who only
have their homes here and who are involved in other profes-
sions. And there are also some who live in rented houses in
this area . . . The government never discusses such problems
with us common people. We get information from the samiti
and newspapers, and this makes a lot of confusion. Once they
arranged a meeting here, which was cancelled because of all
the confusion. . .
In fact, the concept is not clear to us and also for me it is
very hazy. They have published many things, like it will be a
multi-storied building with a swimming pool and so on . . .
which is only a story. . .
In addition, the government does not have a clear idea
about our business and the kind of place that we need for our
business. We do not need a big concrete building; actually it is
257
not good at all for our use. A government engineer can never
build the kind of place that we need. Because it is best for our
profession to have a mud floor, it can absorb the dust and such
things . . . We do not need big concrete walls, we need such a
room through which enough air and light can pass. . .
The government officials are not very conscious about these
things; they are only thinking that we need a very clean place.
That that will be good for our business, but that is their idea,
not ours. You need different types of space, workshops, for
different kinds of businesses. Such as when I went to visit one
place that worked with acid industry, they have mud floors –
so it is important.
As a realisation of the potential the existing Kumartuli has gained,

one of the samitis has made a semi-official entrance and collects a
nominal fee for visitors who want to take photos of their picturesque
neighbourhood. Others have attempted to create a common show-
room where individuals could display their work and sell smaller
sculptures, artefacts and souvenirs.
Conclusion
Through moving from image-making entangled in a patron–client
economy, the Kumars entered the open market in order to supply the
growing demand for protimas created by the community pujas. In the
new, more capitalist-like bazaar-economy, it became an asset to be
inventive in order to attract clients. The stress on exhibition value and
the competition from outsiders increased the Kumars’ awareness of
their background and trade. They used this perception of tradition
connected to caste in order to promote themselves, their location and,
as an extension, their products as original and authentic. In other
words, they started to commodify their caste affiliation. As keepers of
tradition and authenticity, they were able to attract political recogni-
tion and be singled out as an important group and place in need of
development.
Kumartuli as a heritage site and the Kumars as keepers of an
authentic craft is a modern construction born out of necessity in a
highly competitive market. As a heritage site, Kumartuli is a reminder
of the caste-segregated black town of old Kolkata. Not only is it a
reminder, it is also the sole survivor among the larger caste-based
258
neighbourhoods. Spatial aspects of caste segregation and how they

have changed through times is not an unknown territory (Kramer and
Douglas 1992; Waldrop 2008; Davis et al. 2009; Vithayathil and Singh
2012). West Bengal and Kolkata are no exceptions; cultural notions
of purity and the enforcement of economic and political privilege have
dominated the specialisation of caste in colonial Kolkata. Today, new
forces, including the state government and national funds, add to the
continued autonomy of Kumartuli as a neighbourhood dominated by
a single caste.
As keepers of an authentic craft the Kumartuli Kumars have a
jati affiliation that used to and still concerns occupation, besides the
more widespread matrimonial and religious-ritual aspects. But due
to a strained and competitive market for their products, their actual
occupation as clay artisans has changed a lot, starting with the intro-
duction of community pujas. The continual reinvention of their trade
and craft as they adjust to market demands has also transformed how
they perceive their background and jati affiliation. It has made them
aware of their role and history as artists of sacred images – a self-
asserted (self-proclaimed) authenticity they use in promoting their
products.
The processes through which the clay artists use caste politically
to gain development in terms of economic and social leverage coin-
cide with transformative processes in which the sense of belonging
to a given caste, and what such a belonging entails, changes. Caste
as a cultural categorisation, an ideology or a set of practices, changes
in relation to inter-caste self-understanding and societal change. At
this level, caste is something that is practiced, not solely an ideology.
As such, the clay artists try to use their caste affiliation as an asset
economically in order to regain market dominance, and politically to
attract development.
To what extent they have succeeded or will succeed in their attempt
is hard to evaluate. However, they have made the general public and
relevant officials use their narratives of the Kumars as keepers of an
art intrinsic to the Bengali heritage, and of Kumartuli as the authen-
tic place of this craft. As such, Kumars of Kumartuli have worked
to ‘brand their otherness, to profit from what makes them different’
(Comaroff and Comaroff 2009: 24), not only with a reference to their
artistic skills, but also – and with more vigour – through emphasising
their caste. And caste affiliation is, like ethnic belonging, an inherited
trait that, given its widespread social acceptance, with ease lends itself
to ‘brand otherness’. The Kumartuli artists’ politics of caste have thus
259
been successful in terms of being acknowledged by the political estab-

lishment not as a run-down neighbourhood in need of rejuvenation,
but also as a caste-based group connected to a trade in need of protec-
tion and support.
Bibliography
Agnihotri, Anita. 2001. Kolkatar Protimashilpira. Kolkata: Ananda.
Banerjee, Sudeshna. 2004. Durgā Puja: Yesterday, Today & Tomorrow. New
Delhi: Rupa & Co.
BanglarMukh. N.d. ‘Places of Interest In and around Kolkata’, http://westbengal.
gov.in/BanglarMukh/Download?AlfrescoPath=WebContent/Departments/
Tourism/Other%20Documents&FileName=Places%20of%20Interests%20
in%20&%20around%20Kolkata.pdf (accessed 2 January 2014).
Béteille, André. 2005 (1996). ‘Caste in Contemporary India’, in Dipankar
Gupta (ed), Anti-Utopia: Essential Writings of André Béteille, pp. 150–79.
Oxford: Oxford University Press.
Bhabani, Soudhriti. 2007. ‘Potters’ Town Kumartuli May Finally Get Make-
over’, IANS/IndiaeNews.com, http://www.indiaenews.com/pdf/75147.pdf
(accessed 26 April 2008).
Chaliha, Jaya and Bunny Gupta. 1990. ‘Durga Puja in Calcutta’, in Sukanta
Chauduri (ed), Calcutta: The Living City Volume II, The Present and
Future, pp. 331–36. New Delhi: Oxford University Press.
Chakraborty, Sudhir. 1985. Krisnanagarer Mritsilpa o Mritsilpi Samaj. Kol-
kata: K. P. Bagchi Publishers, on behalf of the Centre for Studies in Social
Sciences.
Comaroff, John L., and Jean Comaroff. 2009. Ethnicity, Inc. Chicago: Uni-
versity of Chicago Press.
Davis, Allison, Burleigh Bradford Gardner and Mary R. Gardner. 2009. Deep
South: A Social Anthropological Study of Caste and Class. Columbia: Uni-
versity of South Carolina Press.
Errington, Fredrick and Deborah Gewertz. 1996. ‘The Individuation of Tradi-
tion in a Papua New Guinean Modernity’, American Anthropologist (new
series), 98(1): 114–26.
Ghosh, Sahana. 2013. ‘Artisan Trade of Kumartuli Faces Changing Times’,
Khabar South Asia, 9 September, http://khabarsouthasia.com/en_GB/
articles/apwi/articles/features/2013/09/09/feature-06?change_locale=true
Goldblatt, Beth. 1997. ‘The Image Makers of Kumartuli: The Transformation
of a Caste-Based Industry in a Slum Quarter of Calcutta’, unpublished PhD
dissertation, University of Manchester.
260
Gupta, Dipankar (ed). 2004. Caste in Question: Identity or Hierarchy? New

Delhi: Sage Publications.
Heierstad, Geir. Forthcoming. Illusion of Traditions: Branding of Caste in
Kolkata. London: Anthem Press.
KMDA. 2007. ‘The Fact and Figures of the Kumartuli Urban Renewal Projects’,
KMDA Online, http://www.kmdaonline.org/plans_stat-projects-kumartuli.
html (accessed 5 March 2008).
Kramer, Carol and John E. Douglas. 1992. ‘Ceramics, Caste, and Kin: Spatial
Relations in Rajasthan, India’, Journal of Anthropological Archaeology,
11(2): 187–201.
Miller, Daniel. 1994. Modernity: An Ethnographic Approach: Dualism and
Mass Consumption in Trinidad. Oxford: Berg Publishers.
Roy, Subhajoy. 2005. ‘Call from Kumartuli’, The Telegraph, 4 October, http://
www.telegraphindia.com/1051004/asp/calcutta/ story_5302722.asp (accessed
3 September 2007).
Searle-Chatterjee, Mary and Ursula Sharma (eds). 1994. Contextualising
Caste: Post-Dumontian Approaches. Oxford: Blackwell Publishers.
Vithayathil, Trina and Gayatri Singh. 2012. ‘Spaces of Discrimination’, Eco-
nomic and Political Weekly, 47(37): 60–66.
Waldrop, Anne. 2004. ‘Gating and Class Relations: The Case of a New Delhi
“Colony” ’, City & Society, 16(2): 93–116.
261
GLOSSARY
Adhiar Sharecropper
Adhikar Right
Adibari Original home
Bagdi Formerly untouchable caste
Bahujan (Oppressed) majority
Baidya Upper-caste community
Bargadar Sharecropper
Bauri Formerly untouchable caste
Bhadra Respectable
Bhadralok Genteel or respectable folk who usually, but not
exclusively, belonged to the three higher castes of
Bengal
Bhaskar Sculptor
Bhumi puja Hindu ceremony performed before commencing
the construction of a house or building
Bigha Approximately one third of an acre
Biri Small, cheap, hand-rolled cigarettes
Brahman Upper-caste community
Chanda Donation; subscription
Chandal Formerly untouchable caste
Chasi Farmer; owner-cultivator
Chasjomi malik Landowner
Chhotolok Small people; the lower orders
Dharma rashtra State founded upon the moral/religious principles
of the Hindus
Donkas; kashis Sound instruments
Ghat Earthen, ritual pot used in, for instance, marriage
ceremonies
263
G L O S S A RY
Gharana Schools or guilds. Originally, a term used in Indian

Classical music
Gotra Sub-caste group, clan
Gram baithak Informal village meeting
Gram panchayat Village council; the lowest tier of the panchayat
system
Jati Caste, endogamous group
Jotdar Large landholding peasant
Kaca ‘Traditional’ materials such as mud, clay tiles, and
bamboo
Karta Master; patriarch
Kayastha Upper-caste community
Khet majur Agricultural labourer
Khoini Chewing tobacco
Kumbhakar The potter caste of East India, primarily Bengal
and Bihar
Lathi Heavy wooden stick
Mahishya Caste group of intermediary status
Malik Owner
Mandir Temple
Mritshilpi Clay-modeller
Murti Statue
Namasudra Formerly untouchable caste
Panchayati raj Village local self-government
Pandal Makeshift structure housing religious statues
Para Neighbourhood; hamlet
Pracharak Preacher
Praja Tenant
Pratimashilpi Artist producing god-images; pratima = god-image,
shilpi = artist/artisan
Puja Religious ceremony; worship
Putul Doll
Rajbanshi Formerly untouchable caste
Samiti Group, association, society
Sanghadhipati Organisational head
Sarbajanin puja Community-based puja
Satyagraha Non-violent direct action method of agitation
initiated by Mahatma Gandhi
Shilpi kaj Artwork
Sonar Bangla Golden Bengal
264
G L O S S A RY
Sreni Sub-caste group, clan

Talukdar Small- or medium-sized landlord
Tebhaga Two-thirds of the share (of the agricultural produce)
Thakur protima Religious statue
Unnayan Development
Zamindar Revenue collector and landholder under British rule
Zilla parishad District-level council
265
INDEX
absent-minded casteism 103–20; Backward and Minority Communi-

caste question 104–9; Dalit and ties Employee Federation
Muslim alliance, possibilities (BAMCEF) 13, 114–15, 176, 187
for 118–20; Dalit discourse, Bagdi caste 44, 100, 132, 135, 143,
contemporary 113–18; 196, 197, 198, 199, 200–11, 213
reservations, of West Bengal Bagjola Camp 69, 71, 73, 74, 76
109–13 Bahujan Mukti Party 114
Academic Realism 226, 228, 231, Bandyopadhyay, Bibhutibhushan
236 195
Adibasi community 170 Bandyopadhyay, Sarbani 4, 5, 9, 11,
agonism 13, 175 126
agricultural transformation 133 Bandyopadhyay, Sekhar 3, 5, 10,
agriculturist refugees 98 105, 106, 149
alliance, Dalit and Muslim 118–20 Banerjee, Mamata 8, 9, 12, 128, 148
All India Depressed Classes Banerjee, Mukulika 7, 126
Association 40 Banerjee, Tarashankar 138–9, 195
All India Refugee Front 154–5 Bangiya Jana Sangha (BJS) 38–9
All India Scheduled Castes’ Bangla Dalit Sahitya Sanstha 114,
Federation (AISCF) 45, 52–5 159
Alpajan Banam Bahujan (The Few bargadar 49, 50, 51, 136
against the Many) 176 Barga-Kshatriyas 206
Ambedkar, B. R. 92, 109, 114, 159, baruni mela, festival 78, 154
187–8 bastabdharmi 231
amra Palera 221 Basu, Jyoti 148
Amrita Bazar Patrika 64, 65, 77, 91 Basu, Nagendranath 173
An Approach to Matuaism (Pande) Behal, Rana 28
152 Below Poverty Line (BPL) cards 133
anti-land acquisition struggle, Singur Bengal Pact 44
125–43; everyday life, caste Bengal Partition League 47
and class in 132–40; patronage, Bengal Provincial Scheduled Castes
hierarchy and organisation 140–3 Federation 45
Appadurai, Arjun 2, 19, 23, 125 Bepari, Saroj 162
artisanal labour 14, 218, 222–6 Beruniyas 29–30
Asur Memorial Day 170 bhadralok 1, 3, 4, 5, 7, 8, 13, 19,
atma-samman (soul-respect) 118 30–1, 35–58, 148
267
INDEX
Bharat Sevashram Sangha 39–42 Civil Disobedience movement 89

bichardharar lodai 177 Class I junglies 27
Bishwakosh, encyclopaedia 173 Cohn, Bernard 19
Biswas, Asima 165 communalism 43–4
Biswas, Ganapati 161, 164 communal violence 46
Biswas, Hemanta 71 Communist Party of India (CPI)
Biswas, Kanti 120 49–51, 54, 55, 56, 70
Biswas, Manohar Mouli 160 Communist Party of India (Marxist)
Biswas, Sukriti Ranjan 113, 114 (CPI(M)) 201–3; leadership,
Biswas, Susil Kumar 153 disintegration of 209–12
Biswas, Swapan 155, 163 contemporary caste mobility 100
Biswas, Upendranath 112 contemporary Dalit discourse
Bonilla-Silva, Eduardo 121 113–18
Bose, Nirmal Kumar 86 contestations, Dalit politics and
Bose, Sarat Chandra 91 46–9
Buddhism, rational religion 181 Cooper’s Camp 72, 73
Byapari, Manoranjan 113, 116–17, craft 216–38
149 craftsmen 220–3
cultural and social capital 36
Campbell, George 27, 28 cultural antagonism 98
caste-based artisanal community 238 Cuthbert, S. T. 22
caste-based social movements 1
caste capital 36 Daily Telegraph 229
caste, disappearance of 83–101 dalapati/dalanetri 154
caste domination 110 Dalit counter publics 6
caste domination, labour theory of Dalit discourse, contemporary
28 113–18
caste identity 212 Dalit literary movement 12
caste inequality 108 Dalit migration 63–8
casteism, absent-minded see absent- Dalit movement 111
minded casteism Dalit-Muslim alliance, breakdown
caste networks, Dalit politics and of 64, 67
46–9 Dalit political activism 149
caste question 104–9 Dalit political assertion 35–58
Cavell, Stanley 177 Dalit political elites 115
Chakravarti, Sudhir 223 Dalit politics: and caste networks
Chandals 152 46–9; class and 49–51; and
Chandra, Kanchan 212 contestations 46–9; CPI and
Chandra, Uday 4, 5 49–51; marginalisation and
chasi model 137 appropriation of 42–6; and
Chatterjee, Partha 4, 11, 49, 106–7, questions of class 51–6
125 Dalit refugees 61; Mandal and 71–5
Chatterji, Joya 106 Davis, Marvin 7
Chatterton, Eyre 23 debi murti 226
Chattopadhyay, Sanjeeb Chandra 30 Delhi Pact 65
Chaturtha Duniya 158–9, 160 democratisation 196
Chaudhury, Anasua Basu Ray 5 The Dependent Masses of
Citizenship Amendment Act (2003) Independent India see Swadeen
147, 161 Bharater Paradeen Bahujan
268
INDEX
Depressed Classes League 62 Hembrom, Lerka 169, 184

Desiya community 180 hierarchy 216–38
Desiya self-hood, contradictory Hindu dharma rashtra 39
affirmations of 172–5 Hindu Milan Mandir (HMM) 39,
Devi, Baroma Binapani 147, 153 40, 42–3
Dhanger Coles 21–2 Hindu-Muslim pact 88
Die Gossnersche Mission Unter den Hindu sangathan 39–42
Kols (Nottrott) 20 hundred days work programme
Dirks, Nicholas 19 210; see also National Rural
Dirty Swines 20–3 Employment Guarantee Act
discrimination 29 (NREGA)
divide-and-rule policy 188 hyper-visibility of caste 5–6
Durable Inequality (Tilly) 30
identity 216–38
early caste movements 46 idol-making community 223
East India Refugee Council (EIRC) illegal migrants 147
72 The Indian College of Arts and
East Pakistan Renaissance Society 92 Craftsmanship 230
Economic and Political Weekly Indian Journal of Economics 112
(Sinharay) 3 inter-sectoral diffusion, tool
economic change 193–214 techniques 227
egalitarian protocols 185 Islam, Nazrul 118–19
elections 12 Itibritte Chandal Jiban (Byapari)
employment, diversification of 204 116
eternal caste-disgust 117
ethnic segregation 31 Jaagaran 50, 56
exceptionalism, West Bengal 2–6 Jai Mulnivasi 13, 176
jati 2, 14, 193, 197, 198, 206, 207,
Fox, Richard 24 211, 212, 216, 220–3, 240, 243,
245, 247, 249, 257, 259
Ganashakti, newspaper 85 Jele Kaibartta 137
gate-keeping concepts 125 jungli Kols 25
Ghosh, Anjan 36
Ghosha, P. C. 224, 225 Kahar Bauri sub-caste 138–9
Ghosh, Pradyot 172–3 khet majur lok 129, 130
Ghosh, Prafulla Chandra 110 Khilafat movement 44, 88
gram baithak 129 Kisan Conference in Panjia 49
Guha, Ranajit 30 Kohli, Atul 171
Guha, Sumit 24 Koley, Gopal 139–40
Kolkata-centred upper-caste
Halder, Mahananda 153 dominance 151
harer mati 226 Kolkata Metropolitan Development
Harichand-Guruchand movement Authority (KMDA) 255, 256
147–67; building up 152–7 Kolkata Municipal Corporation
Harichand Seba Sangha organisation (KMC) 255, 256
153 Kols, tribe 9, 19–20; caste
Harijan sweepers 111 pyramid, position at 29;
Harisevak magazine 159 cheapness of labour 28;
Heierstad, Geir 14 colonial and missionary records,
269
INDEX
citing of 23; as coolies 23–8; Matua Mahasangha Patrika (Biswas)

description of 20–1; as labourers 153, 155, 158, 163
par excellence 22; market for Matua Mahasangha, politics of
aboriginality 27; migrants from 147–67; community organisation
Chotanagpur 22; scavengers of in 160–6; re-emergence of 149–52
Calcutta 20; in tea plantations Matua-ra Hindu Noy 159
26–7 Metcalfe, Charles 25
Kothari, Rajni 3 migration, of Dalit 63–8
Krishak Proja movement 89, 90 Mitter, Partha 230, 233
Krishnachandra, Maharaja 222, 223 mobilisational literature, of castes
Kumbhakar mritshilpis 43–4
(clay-modeller) 216–38 modern mritshilpi, tradition and
Kumbhakars, caste and politics westernisation 223–8
in 240–58; caste, classes and modern social institution 19
neighbourhood 242–5; economics, Mohapatra, Prabhu 28
brand developing 245–53; politics, Mollah, Abdur Rezzak 120, 191
using commodity 253–8 moneylending 200
Mookerji, Shyama Prasad 91
Laclau, Ernesto 171, 178 Mosse, David 30
landownership 136, 137, 140, 197 Mukharji, T. N. 231
late colonial challenge, caste and Mulnibasi, imaginary of 169–91;
85–91 affirming the universal 186–9;
Left Front hegemony 3 predicaments 179–81; struggle
liberalisation, privatisation and over meaning 181–6
globalisation (LPG) 183 Mulnibasi Samiti, organisation 170
Lievens, Constant 23 Muslim infiltrators 8
Lindberg’s argument 136 Musolmaner Koroniyo (Islam) 119
literacy rate 88, 208
literary activism 158 Nabanna, harvest festival 174
Locke, Henry Hover 230 Namasudra movement 92, 153
Lurka Coles 21–2 Namasudra refugees 61, 62, 64, 66,
69–70, 75
Mahishya caste 89, 90, 130–5, National Mulnibasi Organisation
137–41 13, 176
Mahishya Samaj 138 National Rural Employment
Majhi Pargana Gaonta, organisation Guarantee Act (NREGA) 184,
170 210, 211
Mandal, Jogendranath 50, 62, 71–5, Neo-Classical Realism 232
80, 106, 116, 120, 187 Nielsen, Kenneth Bo 4, 7, 12
mangsher mati 226 non-agricultural income 204, 205
Mani, Prem Kumar 170 non-Ashraf Muslims 45
The Marginal Nation (Samaddar) Non-Cooperation movement 44
162 normalisation, West Bengal politics
master-mritshilpi, bhadralok 6–9
artisthood and 229–35 North Para Sekhs 198, 204, 209–11
Matua Dharma O Darshan 159 Norval, Aletta 187
Matua literary movement 158–60 Nottrott, Alfred 20
Matua Mahasangha (MM) 3, 4, 12, NREGA see National Rural
13, 76, 78, 106 Employment Guarantee Act
270
INDEX
Pal, Dulal 229, 230 Rawls, John 178

Pal, Gopeshwar 229, 246, 248, 249, reciprocal nexus 132
250 refugee camp 68–9
Pal, Jadunath 232 refugee influx 105
Pal, Jagadish 247 refugee rehabilitation, policies of 97
Pal, Mintu 255 rehabilitation, struggle for 68–71
Pal, Pradip Rudra 233 reservations, of West Bengal 109–13
Pal, Ramesh Chandra 233 Risley, H. H. 222
Panchayat Act 151 ritual identity 214
panchayati raj 60 ritual purity 29
Pande, Debdas 152, 156 Roughsedge, Edward 21–2
partition 61, 63–8, 91–101, 188; Roy, Ananya 127
consequences of 93–7; violence Roy, Dayabati 6, 126
105 Roy, Haimanti 67
partition-enforced migration 6 Roy, Indrajit 13
Pather Panchali (Bandyopadhyay) Ruud, Arild 7, 13, 126, 137
195
patronage 140–3 Samaddar, Ranabir 4, 9, 100, 162
patron-client relationship 199 Sandel, Michael 178
Paundra-Kshatriya caste people 46 Sangha literature 43
peasant consciousness 107 Santal Hul 25
phakibad 139 Sanyal, Hitesranjan 90
Pinney, Christopher 223, 228, 236 Sara Bangla Bastuhara Samiti (SBBS)
Pirojpur Namasudra Conference 51 70
Polia community 172 sarbajanin durgutsab 250, 252
political culture 1 Sarkar, Shyamsundar 170, 180, 182,
politicisation of caste 3 189
Pollock, Sheldon 227 Sarkar, Smritikumar 227
Poona Pact 41 Sashtric knowledge 225–7
population, of West Bengal 109 Sasmal, Birendra Nath 89
post-communism 6–9 Scheduled Caste movement 10,
Pouchepedass, Jacques 25 48, 60–1; before 1947 61–3;
Praja Socialist Party (PSP) 70 Mandal and Dalit refugees
Pramatha Ranjan Prathamik 71–5; proportion of 87–8;
Bidyapith school 160 rehabilitation, struggle for
pratimashilpi 235–6 68–71; Thakurnagar and space
prejudice 98 reinvention 75–9
primitivism 23, 28 Seal, Anil 88
profit, maximisation of 225 self-consciousness 118
protibadi dharma 156 Sen, Amartya 178
Pune Pact 53 Sen, Dwaipayan 6, 11
Sen, Moumita 14
raiyats 25, 26 Sennett, Richard 229
Rajbansi community 61, 62 Sen, Uditi 6
Rajbansi jotdars 63 sharecropping 134, 135, 200
Rajogrihe Satobarsho 158 Sharma, Jayeeta 27
Rakshi Dal 39, 40, 42, 43, 44, 46 Shuddhi movement 41, 51
Rana, Santosh 85, 100 Simanta Bangla, weekly newspaper
Rancière, Jacques 228 164
271
INDEX
Singur Krishi Jomi Raksha Thakur, Harichand 76, 149, 152,

Committee (SKJRC) 127–32, 154, 159, 162
140–3, 141–3 Thakur, Kapil Krishna 78, 155
Singur movement 126, 127–32, 144 Thakur, Manjul Krishna 161, 163,
Sinharay, Praskanva 2–3, 5, 12 167
SKJRC see Singur Krishi Jomi Thakur, P. R. 62, 75, 153, 155, 160
Raksha Committee thakur protimas 241
social composition, of West Bengal tied people 199
109 Tilly, Charles 30
social counter-revolution 11 transformative politics 169–91
social discrimination 113 Transportation of Native Labourers
social identity, change of 193–214; Act (1863) 26, 28
clientilism to disengagement The Tribes and Castes of Bengal
196–200; CPI(M) 201–3 (Risley) 222
socio-economic mobility 219 true democracy 186
Sree Sree Harichand Mission 153,
159 United Central Refugee Committee
Sri Sri Harichand-Guruchand Seva 95
Sangha 165, 166 United Central Refugee Council
Suhrawardy, H. S. 62, 91 (UCRC) 70, 71, 73, 95, 150
Sutherland, Major J. 20, 22 upper-caste domination 83–5, 108
Swadeen Bharater Paradeen Bahujan upper-caste elite culture 100
(SBPB) 176, 178, 181, 182 Utkal Brahman contractors 101
Swadeshi movement 5, 10, 37, 57
Swadeshi nationalism 5 Verma, K. M. 224
Swadhinata 50 159 village politics, caste role in 206
village society, caste in 194–6
Tarakeswar Convention 63 voting patterns 29
Taylor, Charles 178
Tebhaga movement 49, 63 Weber, Max 31
Tenancy Act 200
Thakur, Debendralal Biswas 159, zamindars 24, 25, 50, 51, 47,
160 88–90, 93, 207, 227
Thakur, Guruchand 76, 79, 149, Zeuxis 232
152, 155, 158, 159, 164 zilla parishad 141
272

Politics of Caste in West Bengal

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Politics of Caste in West Bengal

Uploaded by

Copyright:

Available Formats

THE POLITICS OF CASTE IN

“This important volume imaginatively unravels heterogeneous, sub-

Uday Chandra is postdoctoral research fellow at the Max Planck

Kenneth Bo Nielsen is postdoctoral research fellow at the Depart-

Contributors: Sarbani Bandyopadhyay/Sekhar Bandyopadhyay/Anasua

Exploring the Political in South Asia is devoted to the publication

Also in this Series

THE VERNACULARISATION OF DEMOCRACY

RISE OF THE PLEBEIANS? THE CHANGING FACE OF

BROADENING AND DEEPENING DEMOCRACY

POWER AND INFLUENCE IN INDIA

DALITS IN NEOLIBERAL INDIA

WHY INDIA VOTES?

THE POLITICS OF CASTE IN WEST BENGAL

POLITICS, LANDLORDS AND ISLAM IN PAKISTAN

Edited by Uday Chandra,

ISBN: 978-1-138-92148-1 (hbk)

1 Kol, coolie, colonial subject: a hidden history of caste

2 Another history: Bhadralok responses to Dalit political

3 Partition, displacement, and the decline of

4 Partition and the mysterious disappearance of

5 An absent-minded casteism? 103

6 The politics of caste and class in Singur’s anti-land

7 Building up the Harichand-Guruchand movement:

8 Transformative politics: the imaginary of the Mulnibasi

9 From client to supporter: economic change and

10 Craft, identity, hierarchy: the Kumbhakars of Bengal 216

11 The commodification of caste and politics in

10.1 Pratima of Goddess Kali being finished at

4.1 Distribution of caste and religious groups as

‘Caste is not important in Bengal’ is an often-repeated phrase in discus-

historians first provide fine-grained case studies of particular castes

the more important variable for determining electoral fortunes than

This anthology grew out of a scholarly exchange on the political life of

AISCF All India Scheduled Castes Federation

Uday Chandra, Geir Heierstad, and

This anthology explores a much-neglected theme in South Asian his-

Situating West Bengal exceptionalism

the pages of Economic and Political Weekly in 2012. Sinharay, who

the Yadav-dominated Samajwadi Party competes with the Chamar-

states such as Tamil Nadu, Maharashtra or Bihar, where political par-

If caste in West Bengal was thus ‘hyper-visible’ right up to the time

‘Post-Communism’ and the normalisation of

the CPI(M). Similarly, Mukulika Banerjee’s (2010) case-study of the

Left Front rule; the CPI(M)’s heavy-handed land acquisitions in Singur

exercise of social agency. Sen’s chapter also focuses on contemporary

political position outside the ambit of bhadralokism, and by declaring

of caste relations and identities in a diversity of sociopolitical settings.

Michelutti, Lucia. 2008. The Vernacularisation of Democracy: Politics, Caste

Sinharay, Praskanva. 2012. ‘A New Politics of Caste’, Economic and Political

Historical anthropologists of modern India, such as Bernard Cohn

‘Dirty Swines’ in the imperial capital

1 See, for example, BL/IOC/APAC/P2553.

A race distinct from the great Hindoo family both in manners,

Not having any of the feelings of veneration for Bramins

To compare the Lurkas to the Dhangers, Roughsedge wrote, ‘they [the

5 Extract Bengal Political Consultations No. 38 of 3 June 1820, IOR/F/4/800/21438.

conditions at home’ (cited in Mahto 1971: 19–22). In a similar vein,

Kols were treated by higher caste landowners in the south-western

the Sunderbans and the tea plantations in Assam and sub-Himalayan

unsurprising is the fact that the fierce mid-19th-century debates over

By the end of the nineteenth century, Chotanagpur labourers

According to a newspaper report in The Times, between 700,000

labourers in western Bengal, though as pointed out earlier, it was not