Laura Dudley Jenkins

Identity and Identification in India
This book examines the interplay between state identifications of groups and
their own sense of politicized identity. The quota system for disadvantaged
groups in government employment, higher education and legislative bodies fails
to reflect the complex interactions of caste, class, religion and gender. This book
seeks to address this by contrasting official classifications with social identities
as articulated by protest groups.
Using empirical data including court decisions, caste certificates, census
categories and contemporary interviews, the author challenges theories of
identity construction and illuminates the impact of colonial and contemporary
policies on identity politics. Jenkins assesses the impact of the dynamic
processes of intermarriage, religious conversion and migration on the process of
official classification.
This in-depth study, combining primary research with theories of identity from
a number of different fields in humanities and social sciences, will appeal to
scholars interested in identity politics, Indian politics and Asian studies.
Laura Dudley Jenkins is Assistant Professor of Political Science, University
of Cincinnati, USA.
Identity and Identification in
India
Defining the disadvantaged
Laura Dudley Jenkins
LONDON AND NEW YORK

First published 2003
by RoutledgeCurzon
11 New Fetter Lane, London EC4P 4EE
Simultaneously published in the USA and Canada
by RoutledgeCurzon
29 West 35th Street, New York, NY 10001
RoutledgeCurzon is an imprint of the Taylor & Francis Group
This edition published in the Taylor & Francis e-Library, 2005.
“To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection
of thousands of eBooks please go to www.eBookstore.tandf.co.uk.”
© 2003 Laura Dudley Jenkins
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.
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
Dudley Jenkins, Laura.
Identity and identification in India: defining the
disadvantaged/Laura Dudley Jenkins.
p.cm.
Includes bibliographical references and index.
1. Social classes-India. 2. Caste-India. 3. Group identity-India.
I. Title.
HN690.Z9 S6352 2002
305.5 122 0954-dc21 2002031698
ISBN 0-203-40193-X Master e-book ISBN
ISBN 0-203-40837-3 (Adobe eReader Format)

ISBN 0-415-29680-3 (Print Edition)
Contents
Acknowledgments v
1 Identity and identification 1
PART I State simplifications 21

2 Adjudicating identities 22
3 Official anthropology 40
4 Caste certificates and lists 66
5 Categorizing and counting on the census 87
PART II Political complications 106

6 "Backward" Muslims and "Scheduled Caste" Christians 107
7 Hindu nationalism and selective inclusion 122
8 Class, classification and creamy layers 136
9 Women's reservations and representation 150
10 Conclusions 169
Appendices
Appendix I: Government documents 182
Appendix II: Interviewees 209
Notes 213
Bibliography 232
Index 252
Acknowledgments
In 1990, I was headed to Delhi when my train screeched to a halt. The

explanation gradually circulated through the train: Students were lying on the
tracks ahead, protesting against job quotas for lower castes. Accustomed to a
culture of student ennui, I decided that any policy that sparked such a reaction
was worth looking into. Given ongoing American controversies over affirmative
action, I was particularly interested in the insights the rest of the world might
glean from India’s longstanding policies for its various disadvantaged citizens.
Thanks to the many people I can only mention here, I was able to return to India
several years later to carry out the research that resulted in this book.
Of course all remaining errors are mine alone, but I have many people to thank
for making this project possible. I am most grateful to the administrators,
politicians, and activists who shared their time, thoughts and tea with me during
my research in India. I thank my academic advisors, here and in India, especially
M.Crawford Young, Robert Frykenberg, Ashish Bose, S.K.Thorat, Richard
Merelman, Leigh Payne, and Marc Galanter. Others whose comments made me
look at my research in new ways include Nandu Ram, A.Gajendran, Bhagvan
Das, Barbara Ramusack, Theodore Wright and Sumit Guha. For their enthusiasm
and suggestions, I am most appreciative of the anonymous reviewers of this
manuscript and the editors of Curzon Press. For their monetary support and
belief in my project, I thank the Fulbright-Hays program, the United States
Institute of Peace, the MacArthur Foundation, and the University of Cincinnati,
particularly the Taft Fund, Political Science Department, and Women’s Studies
Program. For logistical and research assistance, I am grateful to the staff of the
United States Educational Foundation in India and the many librarians who
helped me, especially those at the Parliamentary Library, the Supreme Court
Library and the Nehru Memorial Library.
Fellow panelists and audience members at academic conferences have helped
me to refine my thoughts, especially at the annual meetings of the Association
for Asian Studies, the annual conferences on South Asia at the University of
Wisconsin-Madison, and two smaller workshops at the University of Victoria
and Brown University. For their support in so many ways, ranging from warm
vi
meals to dynamic ideas, I owe a huge debt to Kristy Bright, Sushma Sharma,
Jeanette Dorner, Brian Axel, Maribeth Kobza, Durba Ghosh, Mona Siegel,
Melissa Brown, Elizabeth Frierson, Annulla Linders, Mira Seghal, Anne
Caldwell, Brenda Allwardt, Scott Kloek-Jenson, Kate Graney, Manu Bhagavan,
Chenyu Sun, Haimanti Roy, Sister Gemma, Chandra Mallampalli, David, Karen,
Bill, Vera, Jim and Michelle Dudley, Kim and Erin Jenkins and many others.
Above all, I want to thank Chris Jenkins, who helped in innumerable ways
through the whole project, and little Isabelle Maya, who showed up part way
through.
1
Identity and identification
Where am I to be classified? Or, if you prefer, tucked away?

(Frantz Fanon from Black Skin, White Masks 1967:113)
I had two surprises in store for me on 27 November 1996, as I stopped in central

New Delhi to attend a political rally. The first was the unexpected opportunity to
hear hundreds of men, women and children joining in a heartfelt rendition of
“We Shall Overcome,” sung in Hindi. The second, even more puzzling, was the
protesters’ demand to be classified as a low caste group, despite the fact that
none among them was Hindu. Why would a group of people protest for a low
caste status? A female student, while jotting the song lyrics into my notebook,
told me that the song, familiar to me as an anthem of the United States civil
rights movement, was also associated with rights activism in India. This rally
over the “right” to be recognized as a low caste can tell us much about state
classifications and group identities.
Many current battles over rights and recognition are disputes over the state’s
identification of groups in society. Reservation policies in India are a system of
quotas for disadvantaged groups in government employment, higher education
and legislative bodies. Like affirmative action policies in other countries,
reservations aim to increase opportunities for under-represented groups. Such
policies are targeted at particular groups identified by the government, but the
official classifications used often fail to reflect the complex interactions of caste,
race, class, religion, gender and other aspects of social identity. The protesters I
saw were Christians who wanted the government to recognize that caste
discrimination persists in their religious community and to include them in the
lower caste category eligible for reservations. Other groups I studied argued for
reservations on the basis of religion, gender or class.
Can a state empower its citizens by classifying them? Reservations are based
on the premise that recognizing group distinctions in society is necessary to
subvert those distinctions. At the same time, the official identification of
citizens, on the basis of caste, for example, has unintended side effects on
2 IDENTITY AND IDENTIFICATION IN INDIA
identity politics. It is important to distinguish between categories, as externally

defined, and groups, as self-defined entities; yet these outsider and insider
definitions do have an impact on each other. In his work on ethnic and racial
consciousness, Michael Banton points out that “[g]roups are created by processes
of assignment, both self-assignment and the actions of others in categorizing
persons as parties to relationships that confer rights and define obligations to
fellow members” (Banton 1997:13). Not only do categories influence groups,
but groups may also organize to reshape official categories.
In India, the most diverse democracy in the world, reservation policies for
disadvantaged castes, tribes and classes have created a new arena of conflict over
the official categories used to identify beneficiaries. These policies reserve
government jobs, university admissions and legislative seats for citizens
officially classified as “backward,” that is, members of the “Scheduled Castes”
(SCs) and “Scheduled Tribes” (STs); in addition, the “Other Backward Classes”
(OBCs), who previously had reservations in some states, have more recently
been granted a quota of central government jobs.1 The boundaries of official
backwardness remain ambiguous and contested in spite of attempts to neatly list
eligible groups in schedules, an effort initiated during the British colonial period
and continued after Indian independence in 1947 (Galanter 1984:121–87,
Irschick 1969).
The public controversy over these policies centers not just on their
redistributive effects but also on the difficulty of defining the pertinent groups
based on elusive concepts such as caste. The different branches of government
add further complexity, resulting in a variety of approaches to defining groups,
ranging from legal and administrative definitions to official anthropological
classifications. Even if categorization were simple, the desirability of officially
demarcating such groupings is an open question. As affirmative action programs
worldwide come under fire from people advocating group-blind approaches
(Jenkins 1998), I argue that what is at stake in debates over group-based policies
in India and comparable policies elsewhere is both equity and identity. Whether
people are identified and how they are identified by their governments have an
impact on the allocation of opportunities and also on the politicization of identity-
based groups. For example, the low caste Christian protesters I saw not only
were demanding increased access to jobs, education and political office but also
were asserting that they have a caste identity, a controversial claim that divides
the Christian community and encroaches on a primarily Hindu category.
Based on my study of the implementation of reservation policy categories and
the political reactions they inspire, I address the central controversy over group-
based policies: Do reservation policies reinforce the very categories they are
meant to undermine? This question has both theoretical and practical implications.
Theoretically, I bridge the gap between those who reject state classifications of
complex identities, regardless of their political utility, and those who emphasize
IDENTITY AND IDENTIFICATION 3
the imperatives of group-based mobilizations and policies, regardless of

concerns about fragmenting multifaceted identities. In policy terms, I expose
some excessive enforcement of official categories by a government, while
recognizing the social and political gains facilitated by group-based strategies.
To answer my question, I consider both the institutions of the state and the
political groups in society which construct and reconstruct reservation policy
categories.2 This study does not purport to be a psychological examination of
individuals’ identities or a comprehensive or quantitative analysis of the
economic and social effects of reservations.3 Rather, it is a case-based analysis of
the impact of these policies on governmental practices of identification and
identity-based political activism.
In Part I of the book, State simplifications, I examine the monitoring and
enforcement of the classifications used for reservation policies, focusing on the
various state institutions which patrol the official boundaries between categories.
These institutions include the courts, official anthropological surveys, caste lists
and certificates, and the census. Sometimes people fall between the cracks of
these rather staid categories due to dynamic processes such as intermarriage,
religious conversion, migration or other forms of social mobility, making it
difficult for administrators, judges or ethnographers to classify them.
Nevertheless, the state’s role as the arbiter of identity for these policies means
that officials must struggle to sort out ambiguous cases. Documents and
interviews with officials implementing reservations demonstrate that various
government institutions continue to cram people into the ill-fitting boxes of
oversimplified and often static classification schemes in ways reminiscent of
colonial policies, even for the arguably benevolent purpose of advancing the
disadvantaged. These policies sometimes give great weight to boundaries that are
becoming increasingly blurred in practice, an unintended and negative side
effect. Many policies and practices associated with implementing reservations do
seem to officially reinforce the group distinctions they are meant to undermine.
Yet, my examination of the protest groups mobilizing over reservations, the
focus of Part II, Political complications, shows that these policies also contribute
to the blurring of boundaries. In addition to the social, economic, and political
advances by members of disadvantaged groups attributed to reservations,4 group-
based policies have complex political repercussions. They do not just mold
complex and fluid identities into an official grid, like so much batter in a waffle
maker (Guha 2001). Rather, my interviews with political activists and
government officials and my review of protest group literature all show that the
categories themselves have become political catalysts, sparking challenges and
counter challenges to the definitions of the beneficiary groups. Through case
studies of identity-based protest groups, I show how caste, religion, class, and
gender intersect to create ambiguous or competing claims for reservation
benefits. In addition to the social mobility of the individual beneficiaries of
reservations, cross-cutting groups are demanding changes to governmental

categories or their boundaries, further undermining the old identity grid. Like the
low caste Christian protesters, various Muslim, Hindu and women’s
organizations are mobilizing over reservation categories, and the policies
themselves are becoming increasingly complicated as individual economic
criteria are layered over group-based qualifications. These continuing
challenges, particularly the various politicized groups demanding categorical
revisions, demonstrate that these policies do not simply reinforce the categories
they are meant to undermine.
All these controversies over policy categories could lead one to conclude that
these policies are hopelessly flawed and futile. Yet, in the context of a policy
meant to break down group disparities, the confusion over appropriate groupings
is also an indication of success. In her discussion of legal disputes over racial
categories in the United States, Adrienne D.Davis writes, “Categorical confusion
creates ruptures in the security of our racial taxonomic structure, calling into
question the practices by which we identify and label people” (Davis 1996:717).
Reservation policies throw many old categories into question, as administrators
and judges recognize the difficulties of classifying migrants, adoptees, and
intercaste families. The various individuals and groups demanding to be added to
the ranks of officially backward citizens raise valid concerns about the
opportunities left for the most disadvantaged, but they also rupture the security
of traditional categories.
In short, I draw on case studies of the state institutions involved in classifying
citizens to argue that governments undercut some of the positive effects of
reservations or similar policies by using overly simplistic or inert classification
schemes, which can reinforce, in an official sense, the boundaries between
groups. Nevertheless, I argue that such official identifications alone do not reify
or solidify identities, and they can even precipitate new challenges to ossified
categories, as my case studies of protest groups demanding changes demonstrate.
Bridging a rift in the theoretical literature, I build on theories critical of colonial
and post-colonial state simplifications of social identities, while recognizing the
utility of identity politics for disadvantaged groups. In short, I keep one foot in
the hermeneutics of identity and another in the politics of alliances. This
theoretical synthesis emerges from my empirical contribution, namely,
documenting interactions between official identifications and identity-based
mobilizations. The fact that people are challenging official categories shows that
the state oversimplifies society. However, the fact that many groups facing
discrimination focus their efforts on adjusting rather than eliminating official
categories indicates that such categories, although inevitably imperfect, remain
an amendable and valuable tool for undermining longstanding hierarchies based
on caste or other aspects of identity.
CATEGORIES AND IDENTITIES

I combine multidisciplinary theories about the layered, fluid nature of identity
with contrasting theories about the utility of highlighting a single identity for
purposes of political mobilization or public policy. My constructivist theoretical
approach allows me not only to take into account the complexity of identity,
which thwarts attempts to categorize people, but also to recognize the
instrumental uses of categories for disadvantaged groups. A focus on the process
of social construction means a recognition that identities are not fixed, but rather
are constantly being defined and redefined through interactions at all levels of
states and societies (Young 1993:21–5, Tilley 1997, Green 2002). Often group
identities are constructed through contrasting a notion of “self” with an “other,”
as in dichotomous categories of colonizers and colonized, the West and the
Orient, whites and blacks, or upper and lower castes. The scholars whom I
subsume under a constructivist banner emphasize the multilayered and
intersubjective nature of group identities and criticize the notion that such social
dichotomies and categories are natural, essential, or fixed.
This general approach to identity can be found in a wide variety of literature in
the social sciences and humanities.5 Although the idea of social construction has
become quite fashionable, appearing in various “new” approaches such as
postmodernism, the central premise—that people interactively create meanings
and identities which then shape their lives—has long been recognized by
anthropologists and other social theorists. As Clifford Geertz wrote, drawing on
Max Weber, “man is an animal suspended in webs of significance he himself has
spun” (Geertz 1973:5). While there is now fairly wide agreement on the fact that
caste and race are constructed (or “spun”) and not natural social divisions,
scholars differ in their analyses of these constructions. Some, focused on the
multiplicity of identity, reject such oversimplified categories as inherently
disempowering. Others retort that dismissing these categories may disable
potentially empowering identity-based politics.
The former school of thought, emphasizing the danger of categories, can
appeal to those who want to reject categories entirely, break down the “words
that build walls by labeling race and ethnicity” and allow people “to stand aside,
narrate, and debate the terms that others assign them” (Heath 1995:45).
According to this approach, the historically oppressive dichotomies of upper
caste/lower caste or white/black must be rejected in favor of an emphasis on
multiple and overlapping identities: Each individual’s identity is an amalgam
including gender, age, class, caste, race, nation, region, ethnicity, sexual
orientation, religion, and countless other facets that cannot be slotted into a
particular category. A few examples of this perspective include the work of
anthropologists and historians whose research underscores the fluidity and
complexity of social identity and problems of imposed categories (Daniel 1984,
Metcalf 1995); other examples are certain feminist and postmodern theorists who
focus on “hybridity” and the impossibility of unitary categories (Friedman 1995,
Bhabha 1994, Essed 2001) and literary figures whose personal experiences and
writings attest to the dangers of classification (Fanon 1967, Rushdie 1995).
Some scholars are particularly wary of categories when they are used by
governments in public policies. Arjun Appadurai is critical of the “officially
enforced labeling activities” associated with the colonial and postcolonial Indian
state’s policies. Eschewing the classification and counting of peoples, he argues
that “statistics are to bodies and social types what maps are to territories: they
flatten and enclose” (Appadurai 1993:326, 334). James C.Scott, examining
various ways states have organized societies schematically and treated people
according to these categories, discusses the disasters associated with even well-
intended “state simplifications” in authoritarian situations, ranging from
scientific forestry to compulsory resettlement (Scott 1998).6 Such critical work
raises corollary questions: What are the outcomes when similar simplifications
are used in colonial situations and, later, under democratic conditions, as in the
case of reservations in India? When simplified categories used to demean or
segregate become tools to undo the effects of that history, do the negative
consequences of using those categories evaporate or linger? To those skeptical of
official classifications, state categories seem unlikely vehicles for justice or
emancipation. My research challenges blanket assumptions that such categories
are dangerous but is attuned to the potential limitations and pitfalls of
categorization.
A second, and equally varied, group of scholars focuses on the instrumental
constructions of identities, illuminating how classifications not only can be a tool
of violence or oppression but also can be turned to the advantage of oppressed
groups. Since a focus on the multiplicity of identities fragments any given social
group, students of social movements have censured theoretical emphasis for
disabling transformative politics (Handler 1992). For example, in the United
States some African-American scholars criticize what has been called the
“postmodern conspiracy to explode racial identity” by constantly drawing
attention to the diversity within racial groupings (Fletcher 1994). Social theorists
have criticized the tendency of some constructivists to give equal weight to very
different axes of identity—all being constructions—and thus underemphasize the
particular oppression and lack of choice associated with racism or casteism.
The problem arises…when all “identities,” of whatever form, are treated as

of equal social validity, so personal lifestyle preferences, such as “musical
styles,” physical attributes such as “disability,” and social products such as
“race” and “class” are seen as being of the same moment… The result is
that fundamental social relations such as racial oppression become reduced
to lifestyle choices.
(Malik 1996:9)
Such a reduction downplays the unique social or political problems associated

with each identity.
Moreover, a single-minded emphasis on the fragmented nature of identity
ignores the usefulness of reconstructing and using categories such as gender, race
and caste. In order to develop grounds for resistance on the basis of gender, some
feminist scholars have resurrected categories rejected by other feminists:
Although feminists contend strongly among themselves as to whether the

concept of woman constitutes a universal category, they must for some
purposes and at some levels continue to act as if such a category indeed
exists, precisely for the reason that the world continues to behave and treat
women as though one does.
(O’Hanlon and Washbrook 1992:154)
This approach is echoed in former Indian Prime Minister V.P. Singh’s defense of
a reservation policy based on caste: “[I]f there is discrimination by birth, then in
delivering the remedy, identification of victims of such an order can be only
done by birth. So the remedy will also have to refer to birth, not because caste
has to be sanctified, but…there is a practical need to refer to birth” (Singh
interview 20 November 1996). In the United States, Supreme Court Justice
Blackmun made a similar argument: “In order to get beyond racism, we must
first take account of race. There is no other way.” Regents of the University of
California v. Bakke, 438 US 265, 407 (1978), Blackmun, J., concurring. Those
who advocate this approach contribute to the debate on categories by retaining a
practical and progressive concern with social and political outcomes, but at times
they fail to fully appreciate the implications of adopting and using certain
categories.
Blending these theoretical approaches, I accept the potential utility of
categories but also take into account the complexity of identity and dangers of
oversimplification. For example, I find that even previously degraded and
internally diverse caste and racial categories have been embraced as potent tools
of empowerment. In India, “[t]oday something quite different is happening: the
very sufferers from the system (including the caste system) are invoking caste
identity and claims” (Kothari 1994:1589). A poem of lower caste unity, “Hum
Dalit” (We the Oppressed), shared with me by a lower caste woman, transforms
the categories of oppressor and oppressed into the compensator and the
compensated.
You will have to compensate

for the cost of every thing…

Having opened falsehood’s curtain
for the truth
We
will disrobe
your policies.8
Thus any category, even caste, may be turned into a tool of empowerment, but
categories often make blunt tools. By emphasizing her caste identity in this poem,
the author downplays her gender identity and distinct experience of oppression
as a Dalit woman. Adopting a category such as “caste” to fight against caste-
based discrimination can subsume other significant identities and result in
contradictions.
In India, where people are, ironically, claiming to be backward (at least in the
official sense) in order to benefit from reservations, such contradictions abound.
Anthropologist Dorinne Kondo notes that people may “simultaneously resist and
reproduce, challenging and reappropriating meanings as they also undermine
those challenges” (Kondo 1990:221). Such seeming paradoxes—embracing
backwardness or recycling older categories—are inherent parts of reservation
policies and politics, which reproduce yet also reconstruct certain categories of
identity in the name of ending oppression.
POLICIES OF IDENTIFICATION
The theoretical and practical conundrum I address through my research is that
people have multilayered identities; yet those who have faced discrimination
may choose to emphasize precisely those disparaged identities in order to
subvert more invidious distinctions through group-based organizations and
policies. I find that such organizations often contest boundaries even as they rally
around categories of identity. My findings challenge assumptions that group-
based policies are divisive and suggest that it is possible to develop policies that
both recognize disadvantaged groups and reflect some of the complexity of
identity. These findings about the potential and the perils of state classifications
have practical implications for the many countries using social categories for
public policies.
Policies of affirmative action, promoting educational and employment
opportunities for disadvantaged groups, can be found in many countries, ranging
from India to Northern Ireland (on the basis of religion) and the United States
(on the basis of race and sex) (Jenkins 1998, Wyzan 1990, Nesiah 2000). In
addition to affirmative action, several other policies to accommodate cultural
diversity depend on the official use of social categories such as ethnicity, race,
religion, gender or caste. Electoral policies drawing on such categories include

separate electorates, separate districts, or proportional representation systems.
Such categories can also be the basis for various forms of limited political, legal,
or cultural autonomy or can be used to monitor antidiscrimination laws. Often
group-based policies depend on counting people by categories in a national
census for purposes of monitoring and enforcement. Two major criticisms of
such official social classifications frequently arise. First, contested categories can
become a new terrain for tensions. Second, the use of such categories in policies
may give new permanence and political momentum to the group divisions they
are meant to ameliorate.
Like those who argue that affirmative action in the United States is divisive,
critics of reservations in India have generally characterized these policies as an
impediment to their vision of a unified Indian nation.9 Some criticized the
“Mandalization” or increasing “caste consciousness” in the wake of the
extension of central government reservations to more disadvantaged groups in
1990, as recommended by the Mandal Commission (Singh and Sharma 1995,
Kumar 1992). Others predicted that quotas would “only aggravate social tension
by deepening caste division” (Hindustan Times 9 March 1996). The Hindu
nationalist Bharatiya Janata Party sought to limit certain reservations by
extending them only to groups embracing indigenous, “Indian” religions, namely
Hinduism, Buddhism and Sikhism. The tradition of opposition to group-based
policies extends back to Mohandas Gandhi, who, in spite of his support for some
reforms of the caste system, opposed separate electorates and group-based
policies for untouchables, in part due to fears that separating these groups from
the Hindu fold would weaken the independence movement and the emerging
Indian nation. Today groups such as the “Gandhi Caste Society” perpetuate this
philosophy by criticizing both the caste system and caste-based policies (Sajwan
interview 17 September 1996).10 The long-standing debate and dilemma
continues: India’s group-based preferential policies have been criticized for
entrenching group identities such as caste; yet they also empower members of
disadvantaged communities, diminishing the limits of caste.
Michael Walzer adds a post modern twist to the policy debate in his work on
“toleration” and possible models of government for multicultural societies. He
describes a “post-modern project” in which people “have begun to experience…a
life without clear boundaries and without secure or singular identities” and
respond “with resignation, indifference, stoicism, curiosity and enthusiasm to the
tics and foibles” of their “post-modern fellows” (Walzer 1997:87). Philomena
Essed likewise proposes moving “toward multiple, nonessentialist politics of
identity” (Essed 2001:498). These are pretty pictures, yet those facing insecurity
and intolerance are wary of the theoretical abandonment of boundaries in a world
where certain social boundaries, even if blurry, still cut them off from
opportunities. Arguments that group-based policies reinforce categories that are
not “real” tend to confound race or caste with racism or casteism.11 It is racism
and casteism that affirmative action or reservation policies are supposed to
counteract, and these are all too real. Acknowledging this may necessitate the use
of categories for a while longer, yet must we give up on eventually achieving
“life without clear boundaries”? Can policies recognize both continuing
discrimination and the complexity of identity?
I contemplate revisions to group-based policies by considering the variety of
categories emerging from different institutions of government and by studying
protest groups that both utilize these categories and challenge their boundaries.
For example, some groups favor what they consider to be more scientific or
objective policies based on class or economic criteria. Policy-makers sometimes
combine different categories to better reflect society, as in the combinations of
caste and class criteria in India. Such attempts can be difficult, as in the
designation of the controversial lists of Other Backward Classes in India. Recent
efforts to exclude a so-called “creamy layer,” or well off members, from the
Other Backward Classes, have shown some promise, although a lack of reliable
economic data complicates enforcement. India also has experience with different
combinations of caste and religious categories and is facing competing demands
to combine caste, class and gender considerations in legislative reservations. The
periodic re-evaluation of policies and categories in India also helps to reflect
social changes, but there is a fine line between adjusting policies and descending
into a massive proliferation of categories for each and every possible
combination of identities. Moreover, the corruption that thwarts the
administration of increasingly complex policies and the political momentum that
hinders attempts to revise lists are challenges that should not be underestimated.
Despite these potential pitfalls, some proposals and innovations in India offer
models from which the rest of the world might learn.
COLONIAL AND POSTCOLONIAL STATE-SOCIETY

INTERACTIONS
Through historically grounded analysis of contemporary cases, I consider
whether the categories previously used to exploit disadvantaged groups can be
used to undermine discrimination. Building on a rich body of research on the
tendency of colonial states to reinforce or even reinvent various ethnic identities,
I examine the contemporary Indian government’s surprisingly similar
proclivities. The literature on colonial states’ involvement in identity politics
documents how colonial recruitment, education, missionary work, cartography,
legal codification, and ethnographic classification contributed to the hardening of
ethnic, caste or racial fault lines (Horowitz 1985:149–66, Young 1994:228–36,
Brass 1985, Fredrickson 1981, Marx 1998). These processes recorded identities
and also reinforced them by changing the stakes associated with certain
identities. For example, in colonial India, the recording of castes in the census
contributed to the formation of interest groups that lobbied census
commissioners to try to improve their official caste rankings (Rudolph and
Rudolph 1967).
Colonial legacies shape contemporary identity politics. Fascinating accounts
trace conflicts between communities back to the legacies of colonial states or
discuss the wide-ranging after-effects of colonial rule on cultures and identities
(Pandey 1992, Prakash 1995). Building on this literature, I demonstrate in the
following chapters that some colonial practices persist in the postcolonial era.
One example is a court ruling that evidence of tribal affiliation from the colonial
era has more value than contemporary proof. Another example is that of
administrators and ethnographers who still refer to lists of “castes” made in
colonial times to codify two distinct and relatively fluid concepts, jati and
varna.12 By documenting the continuing influence of historical assumptions,
rules and practices, I build on criticisms of colonial states by demonstrating the
continuing relevance of such criticisms in the postcolonial era.
Contemporary states continue to shape social identities. Sociologist Ali
Rattansi describes the social construction of identities as a three-pronged
process, “involving processes of ‘self-identification’ as well as formation by
disciplinary agencies such as the state, and including the involvement of the
social sciences, given their incorporation in the categorization and distributive
activities of the state” (Rattansi 1995:257). Colonial anthropologists and census
takers were not the last data collectors to spark political responses. Michael
White and Sharon Sassler, in the field of population studies, draw attention to
how “issues of ethnic identification and assimilation are intertwined with the
data collection mechanisms used by official agencies” (White and Sassler 1995:
470). These contemporary analyses suggest that state identifications, whether
legal, administrative or scientific, continue to interact with and influence identity
claims.
Recognizing that reservation policies, and the group boundaries they depend
on, had precedents in India well before independence allows me to consider
whether current classifications for more progressive purposes are as divisive in
practice as colonial classifications, which have been characterized as “efforts to
render fluid and confusing social and political relationships into categories
sufficiently static and reified and thereby useful to colonial understanding and
control” (Stoler and Cooper 1996:11). Drawing a stark contrast between
oppressive colonial polices and benevolent postcolonial policies would ignore
the multiple actors and multiple motives at play in both periods. The imperatives
of state building in colonial and postcolonial times may be more similar than is
commonly assumed, and the Indian state’s colonial and contemporary impulses
to categorize and record identities have been remarkably resilient. Crispin Bates,
writing on racial theory in India, notes that “its applications were not uniquely
imperial but characteristic, much more generally, of the modus operandi of the
modern, centralized, bureaucratic state” (Bates 1995:222). Whether inspired by a
desire for control or for equity, or by a complex combination of these motives,
official classification often goes hand in hand with centralization; simplification
accompanies administration. That said, the shift to an independent, democratic
Indian state in 1947 resulted in new dynamics and dilemmas as the government
decided not only to use some colonial-era categories but also to create a modified
system of reservation policies.
When a government identifies certain groups of citizens as the targets of a
policy, state identification and social identity become intertwined. Charles
Taylor recognizes this “dialogical” nature of identity construction: “[O]ur
identity is partly shaped by recognition or its absence, often by the
misrecognition of others.” He argues that “the supposedly fair and difference-
blind society is not only inhuman (because suppressing identities) but also, in a
subtle and unconscious way, itself highly discriminatory” (Taylor 1994:34, 25,
43). According to this line of argument, states that use social categories for
policies like reservations risk “misrecognizing” some people, but the alternative,
not recognizing any particular groups, may be even more damaging to groups
and their identities. My research shows that groups who feel they have been
misrecognized (or inaccurately targeted by state policies) often protest against
the boundaries of state categories but still prefer the use of the categories to
nonrecognition. This finding highlights the continuing importance of state
recognition in constructions of group identities.
My interest in such state-society interactions through time also inspires my
focus on social resistance to state definitions, both historical and contemporary.
Even in colonial states, “the novel communal partitioning of society was not
simply implanted from above and beyond. An intricate dialectic unfolded”
(Young 1994:234). Various societal voices have been overlooked in the rush to
describe state, particularly colonial state, constructions of societies. Historian
Robert Eric Frykenberg points out that many scholars:
give too much credit to Europeans and too little to hosts of Native Indians
(mainly Brahmans and others imbued with Brahmanical world views; but
also Muslims imbued with Islamic world views) for the cultural
constructions (and reconstructions) of India. These Indian elites did as
much to inculcate their own views into the administrative machinery and
the cultural framework of the Indian Empire.
(Frykenberg 1993:534)
In addition to the importance of recognizing elite members of colonial and

postcolonial societies and their contributions to notions of identity, other scholars
have drawn attention to the contributions of social movements and political
activists to identity politics, as in the “new social movements” literature or

Benjamin Marquez’s work on the various constructions of racial identities by
Mexican-American organizations (Marquez 2001, Larana, Johnston and Gusfield
1994). New attention to the influence or “agency” of disadvantaged groups in
society is a notable contribution of the “subaltern school,”13 which recognizes
the importance of including in historical accounts the “small voices which are
drowned in the noise of statist commands”:
For they have many stories to tell—stories which for their complexity are
unequaled by statist discourse and indeed opposed to its abstract and
oversimplifying modes.
(Guha 1996:3)
My historical and contemporary focus on individuals and groups who resist the
state classification schemes gives me a unique view of the “meeting point of
state and society” (Skocpol 1985:27), where simplified state identifications and
complex social identities coincide or clash.
THE CATEGORIES
“Caste” has historically been used as a rough translation of the indigenous term
jati, referring to countless “birth groups” that vary depending on context and
region, or of another term, varna, which literally means “color” and refers to an
idealized hierarchy of brahmins, kshatriyas, vaishyas, shudras, and, below all of
these, the avarna (castes outside the varna system), sometimes referred to as
“untouchables.”14 These divisions, codified in ancient writings of the
subcontinent such as the Vedas (1500–1000 BC) and the Manavadharmasastra
(first century AD), are associated with different occupations and accompanied by
rules of behavior and ideas of purity. In reality such divisions are more
ambiguous and regionally varied than the codifications suggest; nevertheless,
caste continues to play a major role in the lives of many Indians, often having a
profound effect on opportunities in terms of residence, education, occupation,
social interaction and marriage (Bayly 1999:8–10, Searle-Chatterjee and Sharma
1994, Quigley 1993).15
The official Scheduled Caste category encompasses the “untouchables” or
Dalits, who are considered to be at the bottom of the caste system. Dalit, which
means “oppressed” or “ground down,” is the name currently preferred by
many.16 Previously known as the “depressed classes,” the Scheduled Caste
category was created by the colonial government in 1936 in order to implement
the 1935 Government of India Act. This act gave special electoral representation
to certain minority groups, including untouchables. After independence the
Scheduled Caste list was re-enacted with the Scheduled Caste Order of 1950,
prepared for the purpose of reservations.
The Scheduled Tribes, also known in parts of India as adivasis, are “those
groups distinguished by ‘tribal characteristics’ and by their spacial and cultural
isolation from the bulk of the population” (Galanter 1984: 147). In many cases,
Scheduled Tribes are hardly as isolated as many official and scholarly
descriptions make them out to be (Guha 1999). Historically, several groups have
crossed the line between tribe and caste, as tribes became absorbed into their
local caste hierarchies, generally in the lower ranks (Roy 1994). Although the
term “tribe” evokes a “problematic legacy of evolutionary anthropology,” tribes
in some cases have “transformed it from a stigmatized label into a political asset
and collective identity” (Karlsson 2001:37, n. 4). In spite of such instances of
social interaction and mobility, the groups dubbed Scheduled Tribes are among
the most socially and economically disadvantaged groups in India (Karlsson
2001:11). Survey data confirms that “Scheduled Castes and Tribes are more
likely to be among the most deprived of India as compared to the upper castes”
(Mitra and Singh 1999:196).17 The Scheduled Tribe category was also listed and
included as a protected minority in the 1935 Government of India Act and later
recognized in the Indian constitution for policy purposes including
reservations.18
No method is specified in the constitution to define a third category in India,
the Other Backward Classes (OBCs), except that the President should appoint a
Backward Classes Commission (Article 340). Several states within India have
had such commissions from time to time, and some have long histories of special
policies for the Other Backward Classes at the state level (Brass 1994:253–64,
Bayly 1999).19 Prior to the current National Backward Classes Commission,
which was recently institutionalized as a permanent government office to
monitor the lists of backward classes and policies for them, there were two other
national level commissions which submitted reports in 1955 and 1980. The
latter, known as the Mandal Commission Report, served as a basis for extending
reservations in central government jobs to the Other Backward Classes in 1990.
The term Backward Classes has had a variety of local usages, but notably, in
spite of the use of the word “class,” the category has generally not been defined
by applying solely economic criteria to individuals. The constitutional debate
suggests that Backward Classes were to be a list of castes or communities, rather
than lower classes in general (Kumar 1994:1). The Supreme Court in Indra
Sawhney v. Union of India (1992) ruled that both caste and poverty should be
considered when determining the backwardness of groups (Faundez 1994:23–
4).20 Generally speaking, the Other Backward Classes are economically
and socially depressed castes or communities, such as lower castes that are not
considered untouchables or other similarly disadvantaged non-Hindu
communities.
RESERVATION POLICIES IN INDIA

The Indian constitution explicitly permits reservations for backward classes
(Indian Constitution articles 15 and 16, Galanter 1984:164–5, Nesiah 2000).
Reservations for Scheduled Castes and Scheduled Tribes include reserved posts
in government service and public sector undertakings, university admissions
quotas, and reserved legislative seats.21 Based roughly on their percentage of the
population,22 the Scheduled Caste quota is 15 per cent and the Scheduled Tribe
quota is 7½ per cent in government service, although these quotas are not always
met. Reserved legislative seats for Scheduled Castes and Tribes in the national
Lok Sabha (House of the People) and in the state legislative assemblies are based
on the percentage of their population in each state. Reservations for Other
Backward Classes at the national level are not as extensive as the policies for
Scheduled Castes or Tribes. Preferences for Other Backward Classes in some
state services and educational institutions have been in place in parts of the south
for long enough that some groups have become quite powerful (Irschick 1969,
Galanter 1984, Parikh 1990); but reservations for Other Backward Classes in the
central government services are a newer development, only recently operational
in some states. Reservations for Other Backward Classes are legally limited to 27
per cent by a Supreme Court ruling holding that the combined total of quotas (for
Scheduled Castes, Scheduled Tribes and Other Backward Classes) may not
exceed 50 per cent. This results in disproportionately low quotas for Other
Backward Classes, who constitute far more than 27 per cent in several states.23
These policies have been effective, although inequalities persist. In spite of the
ambiguities of these categories, government data on the groups targeted by
reservations can give us some idea of the progress and pitfalls associated with
reservations. Government figures show that quotas in lower status jobs tend to be
filled, while quotas in the higher ranks of the civil service are sometimes left
unfilled (Planning Commission 1997–2002: Table 3.9.6, data from a 1995
Department of Administrative Reforms and Public Grievances report). For
example, while the percentage of Scheduled Castes in government services
increased to almost 17 per cent by 1994 (slightly more than their percentage in
the population), their representation in the lowest positions, Group D jobs, was
double their representation in Group A posts where important decision-making
occurs. Still, even that limited, 10 per cent Scheduled Caste presence in the
highest ranking jobs in 1994 was a notable improvement over their 3 per cent
representation 20 years earlier. The percentage of Scheduled Tribes in
government services also increased from 1974 to 1994 but still did not reach
their quota or their percentage in the population. Scheduled Tribe members in
government services are also unevenly distributed, with higher percentages in
lower ranking positions. Other statistical analyses show that Scheduled Castes
and Tribes have made gains in public employment but not enough to close the
gap between them and the general population (Sudarsen 1994).
Reservations also assure Scheduled Castes and Tribes representation in
legislatures and higher educational institutions. In the 1996 elections, for
example, over 19 per cent of seats in the Lok Sabha were held by members of
Scheduled Castes or Tribes (Planning Commission 1997–2002:3.9.23). In the
educational sphere, the Ministry of Education requested all states to reserve 15
per cent of seats for Scheduled Castes and 5 per cent for Scheduled Tribes in
their universities, and such quotas have been widely adopted although, again, not
always effectively implemented (Galanter 1984:63, Dushkin 1979). While
Scheduled Castes and Tribes have increased in general courses, their proportion
in professional courses such as medicine and engineering has lagged and, at
times, decreased (Chanana 1993). Other Backward Classes, on the other hand,
have substantial reservations in professional schools in a number of states
(Galanter 1984). Quantitative generalizations about the effect of reservations on
the Other Backward Classes are complicated by the disparate policies in
different states, the gradual implementation of new policies in recent years, and
the lack of census data on this category, all issues to be addressed in the
following chapters.
Many evaluations of these policies implicitly or explicitly point to the
difficulties of defining which sort of groups are the most appropriate beneficiaries
of reservations. For example, some argue that women of the Scheduled Tribes or
other targeted groups remain “doubly disadvantaged” (Dunn 1993:53, Chanana
1993). Although female literacy rates have risen, a significant gap remains
between the literacy rate of women of all communities (39 per cent) compared to
Scheduled Caste women (24 per cent) and Scheduled Tribe women (18 per cent),
making reservations of university admissions or of higher level government jobs
irrelevant for many (Planning Commission 1997–2002: Table 3.9.2, data from a
1995 Department of Education study). On the basis of a survey of low caste
elites, some scholars argue for economically based reservations, pointing to those
who benefit under the current system and use their new status “as a spring board
for further advances leaving behind not only their recollected history but also
their unfortunate community” (Roy and Singh 1987:142, 152). Subgroups may
be left behind, but others point to the positive effects of even a section within
each category advancing. “[T]here is a sizable section of these groups who can
utilize these opportunities and confer advantages on their children; their
concerns are firmly placed on the political agenda and cannot readily be
dislodged” (Galanter 1986:139). In short, the effectiveness of these policies
varies within the categorized groups, a situation which has sparked some of the
protests and proposed policy revisions featured in the following chapters.
OVERVIEW OF CHAPTERS
Part I:
State simplifications
The chapters of Part I document how states try to fit complex social groups into
standardized categories. Drawing on government documents and interviews with
state officials, including administrators, judges, and anthropologists, I examine
how these authorities execute reservation policies. The oversimplifications of
society which have accompanied reservations bear some resemblance to their
historical precedents, as they sometimes “freeze” identities. State authority can
add longevity or clout to categories, even when officials themselves often
recognize that they are oversimplified.
The courts are the focus of Chapter 2. I consider three contemporary court
cases featuring people making ambiguous claims for reserved opportunities. The
cases involve two sisters admitted into a college due to dubious certification as
members of a Scheduled Tribe, a woman in an intercaste marriage employed in a
reserved job on the basis of her husband’s status as a member of a Backward
Class, and a convert to Hinduism who claimed his Christian parents’ former
Scheduled Caste status. The circumstances leading to these cases illustrate the
malleability of some identities. The decisions demonstrate the legal imperative to
fit individuals into the official policy categories. Even while trying to assure a
progressive outcome for the most disadvantaged citizens, judges’ decisions over
reservation categories have had conservative outcomes, including legal
reinforcement of the notion that caste is determined at birth.
In Chapter 3, I examine legacies of colonial anthropology for current
government classifications. A comparison of Herbert Risley’s turn of the century
People of India and the contemporary “People of India” project carried out by
the Anthropological Survey of India demonstrates that both projects emphasized
caste classification, blurred the line between anthropology and administration
and became embroiled in political controversy. Whether colonial or postcolonial,
these projects served the needs of the state in similar ways. The reaction to these
projects differed, however, since the new study used some of the administrative
categories associated with reservations. In response, various groups complained
of imposter communities in the Scheduled Caste volume, criticized reservations
on the basis of the project’s findings, and even sponsored alternative studies in
order to demonstrate their own qualifications for backward status.
Chapter 4 turns to bureaucrats and regulations involved in codifying status in
the form of caste certificates and lists. Getting an individual caste certificate is a
necessary step to qualify for reservations. Interviews with bureaucrats and a
review of their rules demonstrate the risks of labeling and stigmatizing
individuals in the process of implementing reservation policies and the
challenges of untangling identities in applications involving migration, intercaste

families, adoption, and conversion. The process of applying to be included on the
national list of Other Backward Classes necessitates group petitions, group data,
and group rankings. These processes of certification and listing illustrate the
administrative simplification of identities and the risks of reinforcing groups and
gradations.
Chapter 5 focuses on the census as a tool for implementing reservations. In the
wake of the controversial caste classifications associated with the colonial
census, the postcolonial Indian census enumerators counted only those
communities which were the focus of national-level reservations: the Scheduled
Castes and Tribes. With the extension of central government reservations to
Other Backward Classes in 1990, a debate ensued over whether to count more
castes, all castes, or no castes on the next census. Collecting data on certain
disadvantaged minorities can facilitate the implementation of reservation
policies; yet this practice involves monitoring some groups more than others, and
official census classifications have tended to remain static. Census enumerations,
moreover, continue to inspire many groups to prioritize the maintenance of group
numbers over the alleviation of group divisions. Nevertheless, census categories
are more permeable than other state simplifications, as the infusion of people into
the Scheduled Tribe category demonstrates.
Part II:
Political complications
Various protest groups are refusing to be “simplified” by the state. Based on my
interviews with political activists, officials, and politicians as well as my review
of protest group literature, I compare diverse attempts to reconstruct reservation
policy categories by disrupting the previous boundaries of eligible groups.
Various groups challenge the existing policies, and each other, by prioritizing
caste, religion, class or gender. These overlapping identities splinter both the
official categories and the political groups calling them into question.
The intersection of religion and caste has resulted in competing demands,
which are the topic of Chapter 6. I focus on protest groups from two minority
religions. First, some Muslims are asserting that all Indian Muslims should be
eligible for reservations, while others argue that class and even caste distinctions
within the Muslim community must be the basis for reservation categories.
Second, Dalit Christians are demanding recognition as Scheduled Castes in spite
of their doctrinally caste-free religion. These “Scheduled Caste Christians” face
challenges from other Christians opposed to an official caste distinction within
their community and from other Scheduled Castes opposed to an increase in the
number of competitors for reserved opportunities. These cases demonstrate how
overlapping identities result in competing perspectives on reservation categories

and thus undermine the reification of identities.
Chapter 7 turns to an increasingly important force in Indian politics, the Hindu
nationalists. Based on their constructions of national identity, Hindu politicians
have strategically included Scheduled Castes and certain minority religions
within the Hindu fold. They draw a line, however, to exclude Muslims and
Christians and oppose their demands for reservations. Current debates between
Hindu nationalists and advocates of more reservations parallel the historical
tension between nationalist leader Mohandas Gandhi, who felt that inclusive
Hinduism was the answer to problems of discrimination, and Dalit leader
B.R.Ambedkar, who felt, in contrast, that Hinduism was the root of the problem
and reservations were the answer. Hindu constructions of Muslims and
Christians as national outsiders have influenced the minority movements for
more reservations by impeding cooperation between them. Although Hindu
nationalists have strong political organizations and ideologies, the cultural
diversity that fragments the Muslims and Christians also fragments the Hindus,
inspiring some Hindu nationalist politicians to reconsider their past opposition to
certain reservations.
Chapter 8 turns to the intersection of class and classification, in particular, how
economic considerations complicate definitions of the Other Backward Classes.
The extension of national level reservations to Other Backward Classes in the
1990s renewed debates about the role of class in reservation policies. Various
critics of this policy, including leaders of student protests and members of the
Gandhi Caste Society, prefer economic definitions of backwardness. Further
policy adjustments, such as measures to exclude the so-called “creamy layer” (or
socio-economically advanced individuals) from the Other Backward Classes and
thus from eligibility for reserved jobs, assuaged some groups in the north but
also sparked new kinds of resistance in the south. Reservation policy changes that
brought new attention to class criteria provoked quite varied reactions from north
to south due to distinct policy histories. This analysis of political demands and
policy changes on the basis of economic criteria suggests that, at times, it is the
government that changes the categories and political activists that prefer static
classifications.
The contemporary bill to reserve one-third of seats in Parliament for women is
featured in Chapter 9. Demands to subdivide the category of “women” to include
distinct quotas for Muslim women and women of the Other Backward Classes
have repeatedly stalemated this effort. The legitimacy of women as a group for
reservations in their own right is a recurring theme in historical discussions of
reservation policies. The debates over women’s reservations and rights in the
final years of the British Raj demonstrate that religious and caste minorities were
a political priority whereas women were a “minor minority.” In their 1974
report, Towards Equality, the Committee on the Status of Women characterized
women as a “category” as opposed to a “community” in their consideration of

reservations for women. Women constitute a sizeable part of all other reservation
categories, yet they face gender discrimination as well. Historical and
contemporary debates over women and representation in India thus address most
directly the political dilemmas and policy quandaries posed by multiple,
overlapping identities.
The final chapter summarizes my conclusions about the relationship between
identity and identification. The conclusion offers some practical suggestions for
reservations or affirmative action policies and highlights the theoretical insights
gleaned from this area of study. By examining both state simplifications and
political complications, I synthesize the insights of two theoretical approaches to
classifications and identities.
Both advocates and critics of reservations are involved in the debates over how
to identify various groups and fit them into the state’s schemes. When the
categories are contested, it is not necessarily a sign that the policies are flawed
and therefore ineffective; in fact, such challenges can be a sign that they are
working well enough to throw the old categories into question. Such confusion
creates space for individuals and groups to challenge traditional categories. My
research shows that disadvantaged groups who embrace the social categories
embedded in public policies such as reservations can simultaneously use these
categories and contest their official boundaries. Competing political pressures
can bring about adjustments to simplified state categories, at times to reflect the
interests of powerful groups but, at other times, to better reflect the needs of the
truly disadvantaged in a complex and changing society.
Part I
State simplifications
2
Adjudicating identities
They used to make pickles, squashes, jams, curry powders and

canned pineapples. And banana jam (illegally) after the FPO (Food
Products Organization) banned it because according to their
specifications it was neither jam nor jelly. Too thin for jelly and too
thick for jam. An ambiguous, unclassifiable consistency, they said.
As per their books.
Looking back now, to Rahel it seemed as though this difficulty that
their family had with classification ran much deeper than the jam-
jelly question.
(Arundhati Roy from The God of Small Things 1997:30–1)
In the novel The God of Small Things, the protagonist Rahel grows up between
communities. Having a Hindu father from north India and a Christian mother
from the south makes her unclassifiable; an intercaste relationship that defies
classifications underlies the central tragedy of her story. Like Rahel and her jam,
some people do not fit into a clearcut schemata provided by the state. Scholars may
carry on open-ended debates over the complex and contingent meanings of
religious, regional, or caste-based identities, but when such terms are used in
public policies, government officials—and, in particular, judges—are often
forced to draw boundaries.
Reservation policies based on categories defined by caste, tribe or class have
resulted in much litigation involving people with identities that are difficult to
categorize. Three recent Supreme Court cases illustrate the challenges of legally
defining the disadvantaged in India. One case involves a family that cannot
“prove” that it is in a certain category; a second features a person who falls
between categories due to her intercaste marriage; and a third considers a person
who changes categories through religious conversion. These borderline cases
challenge a policy framework dependent upon social classifications that, often
necessarily, oversimplify society. As Christopher A.Ford discusses, one of the
STATE SIMPLIFICATIONS 23
challenges of adjudicating and administering identity is applying these

oversimplified categories in complex social situations:
Judges and administrators are less fortunate than social scientists: they
must at some point draw lines between rival claimants, rewarding one and
sending the other home empty handed. However analytically “soft” a
particular classification may be, making it a centerpiece of governmental
resource-allocation will require that it be “hardened” dramatically.
(Ford 1994:1234)
The judges to be discussed here thus participate, perhaps unwittingly, in a

process that threatens to legally harden social distinctions, even while enforcing
policies designed to offset the inequalities associated with these distinctions.
Historically, “hard” state schemata have often resulted from less than
benevolent intentions. Much has been written about colonial states, including
their legal systems, contributing to the freezing of social identities that were
previously more fluid and overlapping (Galanter 1992, Anderson 1991, Jalali
1993, Pant 1987, Young 1993, 1994). Thomas Metcalf notes the political and
intellectual objectives behind the British “ordering of difference”: “India was
‘known’ in ways that would sustain a system of colonial authority, and through
categories that made it fundamentally different from Europe” (Metcalf 1995:
113). Gyanendra Pandey emphasizes that “the primacy accorded to caste…was
related directly to the problems of identifying the centres of productive (hence,
revenue-generating) capacity and of maintaining law and order” (Pandey 1992:
68). The colonial legal system in India emphasized not only social order but also
orderly social categories. Legal decisions classifying religious converts, for
example, suggest that colonial courts gave more weight to a view of modernity
as “the authority of institutions to establish criteria for membership” than to a
view of modernity as “the capacity for change” (Viswanathan 1998:77). The
legal hardening of caste or religious classifications is not surprising given the
colonial objectives behind them.
In contrast, many contemporary governments are resorting to group-based
policies in order to combat severe social and economic stratification. Can such
policies avoid the pitfalls of reifying identities? Even in noncolonial situations,
James C.Scott warns, “State officials can often make their categories stick and
impose their simplifications, because the state, of all institutions, is best equipped
to insist on treating people according to its schemata” (Scott 1998:82). In no
place is this power of the state clearer than in the courts. When disputes over
eligibility for reservations arise, the courts adjudicate various identity claims.
Despite their postcolonial context, contemporary Indian Supreme Court
decisions demonstrate the continuing legal reinforcement of official identity
categories. In cases involving individuals trying to benefit from reservations,
court’s decisions mandate strict state surveillance of identity claims. In addition,

the court in these decisions relies on colonialera “proof” of identity and
reinforces the notion that caste and even religious identities are determined at
birth. Although the situations which led to the litigation in each case are
examples of the permeability of some social boundaries, the legal decisions
themselves tend to harden group boundaries, in the very process of implementing
policies meant to reduce group-based inequalities. In some cases, contemporary
judges are even more insistent than their colonial-era predecessors that the
government must fit people into official categories.
The first case involves a family’s disputed claim to be in the Scheduled Tribes
category rather than that of the Other Backward Classes. The central question in
the second case is whether a woman in an intercaste marriage should benefit from
reservations due to her husband’s backward status. The third case is about
reconversion from Christianity back to Hinduism, and whether a family’s
previous low caste status can be regained in order to qualify for reservations. The
following discussion of these contemporary Supreme Court cases focuses, first,
on the ambiguous identity claims people are making in order to benefit from
reservations and the forms of social fluidity these claims and their circumstances
illustrate. By social fluidity, I mean that various social processes can undermine
caste- or tribe-based inequalities within Indian societies. Examples of these
processes, which will be further illustrated by the cases, include sanskritization
(or the movement of entire caste groups up the social hierarchy), intercaste
marriage, and conversions to religions that do not, as a matter of doctrine,
recognize caste. By changing the stakes associated with being in certain groups,
reservations and the resulting legal definitions of groups affect these processes in
some cases, resulting in “desanskritization” and in different incentives for
intercaste marriage or religious conversion (Jayaram 1996:79–80, Karanth 1996:
94–5, 101, Webster 1994:171–2).
This chapter also includes analysis of the legal responses to these claims,
highlighting the court’s rather conservative reinforcement of traditional identities
in the process of implementing an arguably progressive policy. Legal restrictions
on reservations for individuals with shifting or unclear identities protect the
claims of the most unambiguously disadvantaged. On the other hand, such
restrictions could also discourage those who would bend the boundaries of
longstanding group-based hierarchies.
A ªSPURIOUSº TRI BE?

The Supreme Court of India’s 1994 decision in Kumari Madhuri Patil v.
Additional Commissioner Tribal Development centers on two sisters in the state
of Maharashtra, Suchita and Madhuri Patil, who were accused of being admitted
to a medical college on the basis of a “false Social Status Certificate.”1 The case
arose due to a rule that, to benefit from reservations, people must apply to the
government for certificates verifying that they are members of a Scheduled
Caste, Scheduled Tribe or Other Backward Class. The sisters in question had in
fact applied for and received Scheduled Tribe certificates as members of the
“Mahadeo Koli” tribe and were admitted into the college, but they still had to
apply to the state government’s “verification committee” (also referred to as a
“scrutiny committee”) to confirm their status as members of a Scheduled Tribe.
The verification committee eventually decided that the sisters were actually
“Koli” rather than “Mahadeo Koli.” In their state, Maharashtra, the “Mahadeo
Koli” are an officially recognized Scheduled Tribe, but the “Koli” are an Other
Backward Class (208). The verification committee “canceled and confiscated”
their Scheduled Tribe certificates, throwing their admissions under the Scheduled
Tribe reservation into question (208). In this case the sisters appealed to the
courts in order to continue their studies.
The Supreme Court ruled that the sisters were indeed Kolis and members of the
Other Backward Classes, a category which is considered less backward than the
Scheduled Tribes and which, in this case, would not have been as beneficial for
gaining college admission. Although Other Backward Classes do benefit from
reservations in higher education in many states, the Scheduled Tribes reservations
more often go unfilled, thus making this a potentially more useful designation
for admission purposes. For example, although the number of Scheduled Tribe
university students has increased, their numbers in certain subjects, particularly
medicine, is “too small and insignificant” (Chanana 1993:136). Only 2101
Scheduled Tribe students were studying medicine in India in 1988–9, and only
526 of these were Scheduled Tribe women, who made up less than 1 per cent of
all medical students (Chanana 1993:136). Thus it was easier for the Patil sisters
to get into their medical college as members of a Scheduled Tribe.
The court agreed with the verification committee that the sisters’ Scheduled
Tribe status should be revoked, although they allowed one sister, who had by the
time of their decision almost completed her studies, to sit for the final year
examination. The court ruled that the second sister could continue her studies if
she was eligible for admission as a “general candidate,” that is, without the
benefit of the seats reserved for Scheduled Tribes (218). The court concluded that
such identity claims “must be judged on [a] legal and ethnological basis. Spurious
tribes have become a threat to the genuine tribals and the present case is a typical
example of [how] reservation of benefits given to the genuine claimants have
been snatched away by spurious tribes” (213). The decision goes on to spell out
in great detail the proper procedure for verifying identities for the purpose of
social status certificates.
The court’s suspicion regarding “spurious” claims to Scheduled Tribe status is
a response to a perceived trend associated with reservations, the proliferation of
people claiming to be in one of the various backward categories. In contrast to
the process which eminent sociologist M.N. Srinivas called sanskritization (the
process of lower caste groups raising their status by emulating the practices of
upper castes), some individuals and groups are engaging in “desanskritization”
by purposefully making claims to belong to lower status caste or tribal groups
(Srinivas 1966:6, 1989:20, 56–7, Karanth 1996:94–5).2 Ironically, there are now
benefits to gaining one of the Scheduled or Backward labels, at least in a legal
sense. Such a designation carries potential opportunities, such as preferential
admission to college or quotas for government jobs, which may motivate some
individuals and even entire groups to try to claim a backward status, despite the
continuing social stigma of such a label. For example, whereas without the
incentive of reservations, the Patil sisters might have aspired to claim the Koli
status (still backward but arguably less so), in this case they claimed the more
backward Mahadeo Koli status.
Policy incentives to claim a previously degraded social category as one’s own
could conceivably have some broader, positive social and cultural implications,
perhaps by gradually reducing the stigma of such a category or at least blurring
the boundaries. On the other hand, at a practical level, such claims could
undermine reservations by taking economic and educational opportunities away
from what the court calls “genuine” Scheduled Castes or Tribes. Yet the term
“genuine” implies a clarity that many identities do not have.
Although this is not addressed in the court’s decision, it is notable that the
Koli of Maharashtra have organized as the Adivasi Koli Mahasangh (AKM) to
demand Scheduled Tribe status on the grounds that “real tribals” should get
reservation benefits; moreover, the state home minister told a crowd of 35,000
that “their demands are genuinely legal and correct” (Times of India, 27 July
2001).3 This development not only provides an example of group-level
desanskritization but also highlights the ambiguity clouding the very category
claimed by the sisters, which precludes definitive conclusions about their
“genuine” or “spurious” identities.
In the Patil sisters’ case, the court admonished people attempting to stretch the
boundaries of the Scheduled Tribe category, characterizing them as
“unscrupulous persons who come forward to obtain the benefit of such
reservations posing themselves as persons entitled to such status… The case in
hand is a clear instance of such pseudo status” (211). In the process of protecting
the Scheduled Tribes, however, the court reinforced their boundaries to such an
extent that even the so-called “genuine” group members may feel trapped. It did
so in two ways. First, the court said that group boundaries are to be policed
through a rigorous social status verification process. Second, the court treated
identities as static attributes by relying on the idea that caste or tribal
membership corresponds with genetic and cultural traits passed on from
generation to generation and by giving more weight to colonial-era evidence
about identities.
The Kumari Madhuri Patil decision includes a detailed description of how the
states should go about policing identity boundaries. Revenue officers originally
issue social status certificates to Scheduled Caste and Tribe members after “due
verification,” but this is followed by further scrutiny (211). In this case, for
example, the social status verification committee called on the medical students’
father to “furnish in the prescribed form the detailed information regarding his
family background, ancestry, and anthropology of ‘Mahadeo Koli,’ Scheduled
Tribe, to verify the veracity of his claim of status as S.T.” (208). This “scrutiny
committee” relied on “a report of an expert committee which had gone into the
sociological (sic), anthropology, and ethnology of the Scheduled Tribes
including Mahadeo Koli” (209). This report was the basis for a “questionnaire
prepared by the government and…given to and answered by the father of the
appellants,” as a sort of quiz on his knowledge of his claimed identity (209). The
court noted that the father “failed to satisfy the crucial affinity test” (212).
M.N.Srinivas’s concept of sanskritization is again illuminating, for he pointed out
that caste or tribal practices often change rather than remain static. Thus the
father’s practices may not correspond to a test based on older research on his
community, due to a prior history of sanskritization or emulation of groups with
higher status. This legal ruling, then, not only scrutinizes change in the form of
desanskritization but also sanctions sanskritization.
After describing the verification procedures followed in this case, the court
recommended several general guidelines to best scrutinize social status: In cases
of Scheduled Tribes, for example, the scrutiny committee should include a
research officer who has “intimate knowledge in the identification of the
specified Tribes” (213). An additional “vigilance cell” should include “police
Inspectors to investigate social status claims” by going to each person’s place of
residence and birthplace to “verify and collect all the facts of the social status
claimed” (215). In addition to examining birth registrations and school records,
the “vigilance officer” is to “examine the parent, guardian or the candidate in
relation to their caste etc. or such other persons who have knowledge of the social
status of the candidate” (215). The officer should then submit a report including,
in cases of Scheduled Tribe claims, information “relating to their particular
anthropological and ethnological traits, deity, rituals, customs, mode of
marriage, death ceremonies, method of burial of dead bodies etc” (215).
If the director reads the report and finds the claim “‘not genuine’ or ‘doubtful’
or spurious or falsely or wrongly claimed,” a notice and the vigilance officer’s
report are to be sent to the candidate “or through the head of the concerned
educational institution in which the candidate is studying or employed” (215).
The candidate may demand a hearing to present more evidence, and a “public
notice by beat of drum or any other convenient mode may be published in the
village or locality and if any person or association opposes such a claim, an
opportunity to adduce evidence may be given to him/it” (215). After additional
inquiry, if “the certificate obtained or social status claimed are found to be false,
the parent/guardian/the candidate should be prosecuted for making false claim. If
the prosecution ends in a conviction and sentence of the accused, it could be
regarded as an offense involving moral turpitude,” which would disqualify that
person from elective posts (216). (The Patil sisters’ case was not a criminal
prosecution, so this question was not decided; rather the sisters were the appellants,
arguing to be allowed to continue their studies.) By laying down these guidelines,
the court sought to uphold constitutional objectives for “the genuine Scheduled
Castes/Scheduled Tribes or backward classes,” which it perceived as threatened
by “unscrupulous persons” (216). These guidelines have served as a model in
subsequent litigation.4
Such protective measures for “true” beneficiaries may be well intentioned, but
this case and other cases about reservations illuminate the stringent demands
placed on anyone trying to benefit from these policies, since the burden of proof
of social status rests on them.5 “Mere recitals in documents” that an individual is
in a particular caste, if not “supported by independent corroborative material”
cannot form a basis for a caste certificate.6 Despite the elaborate verification
processes that have developed, committees have overlooked important
documents submitted to them and denied caste certificates to members of
Scheduled Castes or Tribes, depriving them of their rights under the Constitution
and resulting in appeals and lost time, money and opportunities for applicants.7
Moreover, continuing scrutiny and repeated inquiries into the social status of
beneficiaries throughout their careers have been oppressive enough to result in
claims of harassment.8
This pattern of repeated surveillance and investigation of citizens’ identity
claims is one way the state may reinforce caste or tribal distinctions in
communities and workplaces, in the very process of implementing policies
meant to undermine discrimination. Police inquiries in hometowns, letters sent to
universities or employers, and announcements calling for public comment about
the identities of people applying for certificates both publicize and tarnish their
status. The use of police to investigate at the local level could intimidate
applicants for certificates and imply to their neighbors that they are not
trustworthy. Moreover, the policing of identities is largely limited to the
disadvantaged communities, as only those who hope to benefit from reservations
need social status certificates.
In addition to police power, the state uses the power of “expert” knowledge, with
particular attention to anthropology, to classify Scheduled Tribes. The expert
committee on ethnology and the cultural affinity test illustrate this aspect of the
government’s power to classify; moreover, the court’s faith in anthropological
“proof” reveals an assumption that identities, particularly in the case of the
Scheduled Tribes, are unchanging.
In the Patil sisters’ case, the court ultimately portrayed tribes and castes as
static social groupings. First, the court held that “[t]he caste of the person…is
determined on the basis of the caste of their parents, basically for the reasons
[sic] that caste is acquired by birth” (213). This seemingly straightforward rule
relies on an assumption about the reproduction of caste or tribe from generation
to generation, an assumption that is challenged in all three of the cases discussed
in this chapter, as cultural change, intermarriage and religious conversion each,
in their own way, thwart identity replication. The court, nevertheless, discussed
Scheduled Tribe identity in terms of “genetical traits” passed on from one
generation to the next. This term evokes notions of biological inevitability, but,
in this decision, the court also used it to refer to cultural continuity, arguing that
cultural traits are passed on like genetic traits. For example, responding to the claim
that social mobility or modernization could explain the failure of the father to
pass the “affinity test,” or government questionnaire about the culture of the tribe
in question, the court argued that certain traits and customs will inevitably
continue: The “argument of social mobility and modernization often alluringly
put forth to obviate the need to pass the affinity test is only a convenient plea to
get over the crux of the question. Despite the cultural advancement, the genetical
traits pass on from generation to generation and no one could escape or forget”
(209).
In a later passage decrying the father’s lack of knowledge of the culture of the
Mahadeo Koli, this genetic metaphor for cultural continuity resurfaces: “His
feigned ignorance of the ancestry is too hard to believe…. The anthropological
moorings and ethnological kinship…gets genetically ingrained in the blood and
no one would shake off from [the] past, in particular, when one is conscious of
the need of preserving its relevance to seek the status of Scheduled Tribe or
Scheduled caste” (212). This passage reiterates the idea that identity is
determined by birth or blood, while also noting that reservations themselves may
further reinforce cultural continuity by the very processes of identity verification
laid down in this case. Both points illustrate how the legal implementation of
reservations, meant to change the status quo, can simultaneously reinforce the
notion that identities are permanent.
Using another static approach, the court credited older evidence of tribal
membership over newer evidence. The certificates of membership issued by a
contemporary “caste association” did not, according to the court, bear any value
as evidence, but a school certificate of the students’ father, “being pre-
independence period, it bears `great probative value' ” (208, 213 emphasis in
original). Contemporary evidence is arguably politicized, sometimes coming
from the very groups or associations that are seeking Scheduled Tribe status.
Thus, in its arguments, the court referred to official government classifications of
the tribe from as far back as 1933, and painted a picture of the Scheduled Tribe’s
“traditional moorings and customary beliefs and practices” as its defining

features (210).
The court relied on historical data and interpreted tribes as essentially static
entities, passing on cultural traits, as though they were genetic traits, from
generation to generation. In part, this artificial freezing of boundaries is a
response to the increasing challenges to these boundaries stemming from
desanskritization inspired by reservations. The court concluded: “The ingrained
Tribal traits peculiar to each tribe and anthropological features all the more
become relevant when the social status is in acute controversy and needs
decision” (212). This case is indicative of two ironic processes: First, citizens are
applying to be declared backward in order to move forward in society. Second,
the court, by monitoring identities and relying on historical ideas and evidence
about castes and tribes, is freezing the social categories reservation policy was
designed to weaken.
INTERCASTE MARRIAGE
In the case of Valsamma Paul v. Cochin University (Supreme Court of India
1996), a rejected applicant for a lecturer position in the Cochin University
department of law challenged Valsamma Paul, who had claimed her backward
husband’s status to successfully apply for that same position, which was reserved
for a member of a Backward Class.9 Valsamma Paul, described as “Syrian
Catholic (a Forward Class),” was married to a man described as “Latin Catholic
(Backward Class Fisherman)” (546). Assuming her husband’s status, she applied
for and was appointed as a lecturer in the law department of Cochin University,
in a position that was reserved for Latin Catholics.
Although Christians are not eligible for Scheduled Caste status (a critical point
in the next case on conversion), caste-based stigma persists in some Christian
communities. Therefore, some Christian castes, including that of Valsamma
Paul’s husband, have been recognized as Other Backward Classes and are
eligible for various central or state level reservations. Other Backward Classes,
as discussed in Chapter 1, are actually lists of disadvantaged castes or
communities. Thus, despite the fact that the couple at the center of this case is
Christian, the court quite matter-of-factly discussed them in terms of their
“castes.” The court focused on whether the lecturer became a member of her
husband’s caste and, if so, whether she should be counted as a Backward Class
member eligible for the reserved job.
The court concluded that, upon marriage, a wife becomes a member of her
husband’s family. She also becomes a member of her husband’s caste.
Nevertheless, she is not entitled to claim the benefits of reservations using her
new caste identity. While recognizing and even giving a legal nod to patrimonial
notions of identity based on the husband’s status, the court did not permit
reservations for women in intercaste marriages on the basis of their new status.
This defendant and her husband are part of a small but growing group of
Indians willing to ignore strong socio-cultural taboos and marry across caste
lines. Caste endogamy (“marriage within the same or allied castes”) means that
marriages occur within a limited “marriage circle,” and such social stratification
continues to influence marriage practices in contemporary Indian society
(Banerjee 1999:650–1). Intercaste marriages are being facilitated by the
increasingly diverse universities and workplaces resulting from reservation
policies. A study of attitudes toward intercaste marriage, based on interviews
with post-graduate students at Calcutta University, suggests that attitudes are
changing, as students are becoming more accepting of intercaste marriage than
their parents (Kundu and Sherif 1982:324). Yet, one should not exaggerate this
trend. A study of under-graduate and post-graduate students at Rajastan
University found that one in four women would not oppose their own arranged
marriage even if it was contrary to their wishes and indicates that their reasons
still include “because of tradition” and “because it would not be liked by caste
people” (Upreti and Upreti 1982:249–50). Despite “increased opportunities for
young people of both sexes from different castes to socially interact and fall in
love, the number of intermarriages…is still very small” (Saroja 1999:186–7).10
Reservations could serve to ameliorate the particularly strong social stigma
associated with women in intercaste marriages. The study of Calcutta University
students and their parents found that whereas 68 per cent of parents would accept
intercaste marriage for their sons, only 45 per cent would accept it for their
daughters (Kundu and Sherif 1982:324). Women marrying, “down” is more of a
taboo than men marrying lower caste women. Even the Valsamma Paul decision
noted that, historically, marriages such as hers, between upper caste females and
lower caste males, have been considered “invalid,” even when the opposite
combination was recognized as legitimate (560). The court explicitly asserted the
importance of social trends toward more intercaste marriage, noting the potential
for the institution of marriage to encourage “harmony and integration” as well as
“national unity and integrity” (547). Although recognizing the benefits of
intercaste marriage, the court ultimately did little to encourage this trend by
denying the woman the right to claim her husband’s status.
The court justified its position by pointing to the advantages such a woman
would have had in her early life; she lacked the extensive experience of
indignities and sufferings comparable to the Backward Classes. In addition to
these discussions of identity as defined by early childhood socialization and
experiences, the court also referred to biologically determined identity, resting its
decision on the troubling ground that caste is determined at birth. The court cited
a key case regarding reservations for the Backward Classes, in which caste is
described as a “socially homogeneous class.” “One is born into it. Its
membership is involuntary… Endogamy is its main characteristic” (554).11 None

of these descriptions fit this case of a women who became a member of a
backward class by marriage rather than birth. The court rejected the argument of
Valsamma Paul’s attorney that “birth by itself is not a determinative factor,” that
rather, “due to her marriage, she had subjected herself and suffered all the
environmental disabilities to which her husband…was subjected and to which all
other members of Backward Class[es] in the region are subjected” (551). The
court ultimately agreed with the opposing argument that “persons who by birth
belong to Scheduled Castes, Scheduled Tribes or Backward Classes alone are
entitled to the benefit… By marriage, adoption or any other device, viz., by
procuring false social status certificates, they are not eligible” (552). Legally
lumping together all means of gaining a caste status aside from birth, this
decision implies that all voluntary routes to backwardness are equally dubious,
whether it is by marriage or by obtaining a false certificate.
In this light, any evidence of choice in the matter of identity casts doubt on an
identity claim. The court faced a dilemma: Social mobility and voluntary
changes of identity through intercommunity marriage are to be welcomed, yet
the benefits aimed at those born Backward could be diluted by those who
become Backward. Intercaste marriage might, in some cases, be a means of
getting oneself or one’s children into a more politically or economically
expedient social category; yet by defining intercaste marriage as one such
“device,” and equating it with fraud, the court categorically rejected volition,
whatever its form, as a legitimate means of social mobility (Jenkins 2001b). The
decision concluded, “[a]cquisition of the status of Scheduled Caste etc. by
voluntary mobility into these categories would play fraud on the Constitution and
would frustrate the benign constitutional policy” (547).
A decision to deny reservations to a woman married to a backward man could
still have challenged the status quo, namely the patrimonial notion that upon
marriage the bride acquires the family, and caste or tribe, of her husband. Yet,
this court’s decision explicitly reinforces patrimonial marriage by insisting that
the defendant is, socially and legally, still a member of her husband’s family and
caste except with regard to her eligibility for reservations. In that instance she is
categorized by the higher caste status of her father. For reservations, it
seems, caste is determined at birth. An exception to this rule can be found in a
different case, that of a lower status woman marrying a higher status man. In this
combination, the woman took on her husband’s status and his category for the
purposes of reservations. Her new—in this case higher—status disqualified her
from previous benefits.12 These decisions embed gender discrimination in
policies meant to counteract other forms of social discrimination.
In the Valsamma Paul case, the court, on the grounds that reservations should
help those who have been disadvantaged from childhood, ended up reinforcing
an ideology central to the caste system, the notion of caste as a status determined
by birth. At the same time, it reinforced a contradictory ideology about a

woman’s status being determined by her husband’s status for most purposes.
Moreover, at a practical level, by denying the benefit of reservations in this case,
the court’s ruling potentially discourages intercaste marriage by denying one of
the advantages of such marriages for high-caste women, while reinforcing the
social and legal disadvantages. Thus the decision has the potential to reduce
intercaste marriages, which are one possible solution to caste stratification.
Some language in the Valsamma Paul decision hints at a nuanced
interpretation of identity, including a comment that people have “several
identities which constantly intersect and overlap” (548). The court even included
long references to equality for women and sang the praises of the United Nations
Convention on the Elimination of Discrimination Against Women (CEDAW)
(549, 563). Yet the decision itself embeds a notion of caste based on blood in the
legal interpretations of reservations, particularly penalizing women in the most
stigmatized of unions, that with a man of a lower status group. Historically,
social marriage rules have been applied “selectively so as to maximize the
marriage chances of higher castes and men, and very seldom operated in favor of
lower caste men and women” (Banerjee 1999:659). The Valsamma Paul decision
continues in this tradition.
This case shows that the categories associated with reservation policies can be
implemented in conservative ways, especially when legal classifications reaffirm
primordial or patrimonial notions of identity. Despite the policy goal of
overcoming caste-based disadvantages, this ruling on reservation categories
could be self-defeating by discouraging intercaste marriages, arguably one of the
best ways to overcome caste discrimination. Many young people still rely upon
their castes for social and economic opportunities, including but not limited to
caste-based reservations, and “they may try to keep their caste affiliation
unsullied by strictly following caste endogamy” (Saroja 1999:190). The very
reservation policies which are supposed to get beyond a system of status
associated with birth groups have resulted in case law holding that birth is the
only legitimate basis for an identity claim.
CONVERSION
In the case of S.Swvigaradoss v. Zonal Manager (Supreme Court of India 1996),
the petitioner’s parents had at one time been Hindus of the Adi-Dravida caste, a
recognized Scheduled Caste in the state of Tamil Nadu.13 Prior to
Swvigaradoss’s birth, his parents had converted to Christianity. According to
Swvigaradoss, he converted to Hinduism at age 14 and became a member of his
parents’ former caste. Subsequently he was employed by the Food Corporation
of India. The next year, however, he married “according to Christian rites in a
church,” and “[on] these facts, notice was given to the petitioner to show cause
how the petitioner would be entitled to benefits and privileges extended to the
Scheduled Caste candidates in the future” (101). Swvigaradoss filed a suit
arguing that he was baptized as a minor and that “[a]fter he became a major, he
is continuing as an Adi-Dravida” (101). Thus he wanted to be recognized as a
Scheduled Caste member, eligible for Scheduled Caste reservations. Unlike the
Other Backward Classes category, which can include some Christians, the
Scheduled Caste category does not include any Christians. Thus, the act of
conversion is a key element in this case. The court ruled that regardless of the
conversion, “the petitioner was born of Christian parents” and thus was not born
into Hinduism or the Adi-Dravida caste (103). Since a “Christian is not a
Scheduled Caste,” they denied his suit and declared that “the petitioner cannot
claim to be a Scheduled Caste” (103).
This case is an example of another form of social fluidity, conversions and,
occasionally, reconversions among lower castes, sometimes due to religious
convictions, sometimes for political or material motivations, and often for a
combination of reasons. Lower caste members may seek nominally caste-free
religions such as Islam or Christianity, and, historically, entire groups of lower
castes have converted from Hinduism. Such conversions, whether individual or
mass, are arguably not only spiritual decisions but also a form of political
protest. The mass conversion of untouchables to Buddhism in 1956, led by
B.R.Ambedkar, continues to inspire mass conversions to that religion (Zelliot
1996, Omvedt 1994: 247–9, BBC News 4–5 November 2001). Mass conversions
to other religions, such as Christianity in the late nineteenth century and Islam in
1981–2, were also in part protest strategies (Pickett 1933, Oddie 1997, Mujahid
1989).
Conversion involves complex and varied motivations, as a list of reasons
given in survey of Indian converts to Christianity illustrates: “To escape from
cholera… Because land owners oppressed us… Because our missionary helped
us against the Brahmans and the Rajputs… Because the love of Jesus won me…
To get a wife for my younger brother… To be saved from forced labor…
Because I wanted to know God… Because the wise men of my caste said I
should” (Pickett 1933:159–60). Adding to such divergent calculations,
contemporary reservation policies provide another material motivation for
conversion, the possible advantages of joining one of the religious communities
which include legally recognized Scheduled Castes, namely Hindus, Sikhs and
Buddhists.14
Fearing a spate of conversions inspired by reservations, the government has
taken steps. For example, a government circular in Tamil Nadu states that
members of Scheduled Castes who convert to Christianity, revert to Hinduism
(on the basis of which they obtain jobs in government service), and finally
reconvert back to Christianity, can have their employment and scheduled caste
status revoked.15 When litigation involves reservations and conversions, judges
find themselves in the difficult position of trying to assess converts’ motivations.

This is not only tricky but also troubling. As instrumentally motivated as some of
these conversions may be, the act of choosing another religious identity has
historically been one possible escape route from an undesirable caste identity.
Although these choices could help to undermine a system of caste-based
inequalities, judges implementing reservations have at times scrutinized and
delegitimized such choices.
In the case under consideration, a son of converts reconverted. As in the
intercaste marriage case, the court relied on an ideology of caste identity being
determined at birth, in this case going even further by assuming that religion is
also determined at birth. The decision’s conclusion was that a “Christian is not a
Scheduled Caste under the notification issued by the President. In view of the
admitted position that the petitioner was born of Christian parents and his parents
were also converted prior to his birth... petitioner cannot claim to be a scheduled
caste” (103). The court added yet another restriction to the fluidity of identity,
namely a rule that caste membership is to be determined by the government
rather than by the castes themselves.
Building upon the caste certificate procedure laid out in the Kumari Madhuri
Patil case, the Valsamma Paul court embarked upon a generalized crackdown on
identity fraud, directly relevant to the Swvigaradoss case at hand. The court held
that in the process of scrutinizing an identity, “the officer concerned should also
verify, as a fact, whether a convert has totally abjured his old faith and adopted,
as a fact, the new faith; whether he suffered all the handicaps as a Dalit or Tribe;
whether conversion is only a ruse to gain constitutional benefits.”16 Thus, under
Valsamma Paul the vigilance officers have the added responsibility of verifying
religious beliefs. One of the factors to be considered by these officers is whether
a convert’s new community recognizes the convert as a member. However, the
Swvigaradoss decision, issued later the same month in 1996 as Valsamma Paul,
emphasizes birth and pays little heed to social acceptance as an indicator of a
legally recognized identity, a shift from previous cases.
In a 1975 case the Supreme Court noted that the law since 1886 had
consistently held that people who reconvert to Hinduism can again become
members of their former castes, the key criteria being whether the other members
of the caste accept the convert back into the fold.17 For example in 1915, noted
jurist Ganapathi Iyer wrote that people “are born in their respective castes or
sects. It cannot be said, however, that membership by caste is determined only by
birth and not by anything else.”18 Another jurist used the analogy of membership
in a club to describe caste, rejecting the idea of caste as a birth group in favor of
caste as a voluntary, if sometimes exclusive, organization: “It is within the power
of a caste to admit into its fold men not born in it as it is within the power of a
club to admit anyone it likes as its member. To hold that membership of a caste
is determined by birth is to hold that the caste cannot, if it likes, mix with another
caste and form both into one caste. That would be striking at the very root of
caste autonomy.”19 In 1976, the Supreme Court of India adopted this rationale in
a case which, like Swvigaradoss, involved the offspring of converts by holding
that conversion to Hinduism would not automatically give a person membership
in his parents’ former caste “but he would become such member, if the other
members of the caste accept him as a member.”20 This decision reflected the
reasoning of a 1940 ruling that “the caste itself is the supreme judge.”21 Not all
court decisions in cases involving reservations and the offspring of converts have
been this sensitive to community sentiment (Galanter 1984:120, n. 60), yet in
1996 the Swvigaradoss court did not even address the communities in question.
In this case, Swvigaradoss’ community was hardly mentioned, except for his
possible ties to the Christian community, implied by the reference to his
marriage ceremony. The court did not discuss the opinion of any Adi-Dravidas
regarding his claim to be one of them. Rather, the conclusion emphasized birth
status and the government’s lists. Not only was the petitioner “born of Christian
parents” but also his parents were “converted prior to his birth and no longer
remained to be Adi-Dravida, a Scheduled Caste for the purpose of Tirunelveli
District in Tamil Nadu as notified by the President;” therefore the “petitioner
cannot claim to be a Scheduled Caste” (103). The court repeatedly emphasized
that it did not have the power to change the lists of Scheduled Castes, which
were constitutionally issued by the President and subsequently amended by
Parliament. However, the petitioner did not ask the court to change the lists;
rather he changed his identity and asked the court to find that he fit within the
current lists. Perhaps the court’s emphasis on the lists was a response to activists
trying to expand the definition of the Scheduled Castes to include Christians.
However, notably, their demand is quite different from Swvigaradoss’ argument
that, by converting from Christianity to Hinduism, he fit into the existing
category.
This decision applied an “identity as birth group” standard not only to caste
identity but also, more surprisingly, to religious identity. Perhaps the difficulties
of “proving” in legal terms who is a genuine Hindu contributed to this rather cut
and dried demarcation. Characterized by historian Robert Frykenberg as “[t]his
soft concept, this jumble of inner contradictions,” Hinduism is particularly
susceptible to open interpretations and porous boundaries (Frykenberg 1993:
523). In a reservations case that necessitated categorizing a Hindu-born convert
to Buddhism, the Supreme Court of India commented: “Hinduism is so tolerant
and Hindu religious practices so varied and eclectic that one would find it difficult
to say whether one is practicing or professing Hindu religion or not” (Galanter
1984: 309).22 Perhaps it was this ambiguity that inspired the court in
Swvigaradoss to consider religion and caste only at birth, rather than heeding
choices made later.
There are other possible explanations for the legal shift from a more open
interpretation of caste or religious identity to one in which birth groups and the
state’s official lists of castes reign supreme. One is the government’s long held
but increasing concern over the influence of foreign missionaries on Indian
citizens and its reluctance to back any policies that might facilitate or encourage
conversions (Sarkar 1999). Another explanation for the court’s inflexibility in
such cases is concern about conversions inspired by the reservation policies
themselves. The court is, in other words, increasingly preoccupied with
distinguishing between conversions of convenience and conversions of
conviction (Viswanathan 1998). The complex relationship between religious and
caste identities makes cases involving conversion particularly ambiguous.
Whether intergenerational conversions can or should allow a person to re-enter a
caste is a challenging question for a court trained to interpret the law. In a
psychological sense, the court would have enormous difficulty gauging
Swvigaradoss’s motivations for conversion. In a sociological sense, whether his
parents ever left their caste upon conversion is unclear; the question of whether he
rejoined it is even more complex. Growing more suspicious of conversions and
facing increasingly ambiguous claims for reservations, the court shifted from
reliance on self-definitions of castes to a top-down, birth-based approach to
categorizing citizens, which could undercut any existing flexibility in caste or
religious boundaries.
CONCLUSION
These three recent court cases bring to light the variety of social processes which
are helping to undermine identity-based inequalities, and they also illustrate how
the courts have responded to a variety of liminal identities. To summarize, the
case of the “spurious” tribe shows how the court, with the aim of securing social
justice for “genuine” beneficiaries, contributed to the construction of tribes as
timeless and unchanging categories and ordered strict policing of identity claims.
The decision on intercaste marriage reinforced the notion of caste as a category
determined at birth when reservations are at stake, with no recourse to voluntary
change through any means, including marriage (with the notable exception of a
state court ruling denying a woman reservations after marrying “up”). The third
case involving conversion suggests that in the court’s eyes, religion, like caste, is
determined by birth, and the state rather than the community in question is the
ultimate arbiter of identity.
The social circumstances highlighted in these three cases reveal some of the
ways reservation policies contribute to the blurring of social boundaries by
changing the incentives for desanskritization, intermarriage and conversion. At
the same time the legal decisions show the state’s tendency to react to these
changes by reinforcing group boundaries, even in the act of trying to reduce
group disparities. In the third case, the court’s decision on caste membership was
even more top-down and inflexible than some pre-independence court decisions.
These cases demonstrate that in spite of a postcolonial shift in the objectives of
state classifications, reservation policies, being dependent on the categorization
of social groups, may have unintended side effects, such as a strict state scrutiny
of identity claims.
These are the dilemmas of a group-based solution applied to individuals. At a
practical level, “spurious” claims could hurt the individuals who would
otherwise benefit from reservations. At the same time, incentives to claim
membership in a Scheduled Caste or Scheduled Tribe could reduce the stigma
associated with these categories. In the Patil sisters’ case, the court credited
precolonial evidence and precluded a dynamic interpretation of caste identities in
the process of attempting to protect the rights of the “true” beneficiaries of
reservations. By uncritically adopting colonial-era categories, the court
emphasized supposedly immutable characteristics of categorizable peoples. The
intercaste marriage decision, denying reservation benefits based on claims to be
Backward through an intercaste marriage, could keep benefits from women of
privileged backgrounds, but did nothing to encourage such marriages. In the third
case, the court tried to squelch conversions to more expedient religious
identities, but in the process, it shifted from precedents supporting community
self-definition toward state definition of identity.
Should we judge a policy by how it deals with exceptions? Arguably many
other people fit into the state categories more easily. Moreover, imperfect or
poorly implemented reservation categories may be better than no categories at all
for most of the disadvantaged citizens in India. The legal treatment of anomalous
or exceptional cases is important, however, because, given the goals of
reservations, such cases are not necessarily “problems” but rather signs of social
change. As articulated in the three cases discussed here, such cases have inspired
an unfortunate legal distinction between “genuine” and “not genuine” identities.
If the courts acknowledged that such cases are ambiguous, rather than assuming
fraud, the categories could be implemented in a way that recognizes and fosters
changes and choices already occurring in society. A more nuanced adjudication
of identities might recognize that even “tribal” cultures can alter over time, that
women in intercaste marriages may be stigmatized enough to become Backward,
or that religious and caste communities may accept new or returning members. Yet
courts are somewhat limited by the categories embedded in the policies
themselves. Legislative changes could put into practice at the national level a
category of reservations for people who have an intercaste marriage, one
alternative model that would not penalize people for stepping across social
boundaries (Paswan interview 29 December 1996).
Group-based policies, by upping the stakes attached to certain group
identities, draw attention to these divisions but also inspire challenges to the
boundaries between them. As seen in the court cases discussed here, the legal
response has been, at times, conservative and static rather than flexible, but this
state crackdown on categorization is not the whole story. Although judges may
try to force an unwieldy social structure into neat legal boxes, these group-based
policies inspire further struggles over the gray areas between categories.
3
Official anthropology
In a bookstore in Delhi, a salesperson, apprised of my interest in lower caste

politics, handed me a tome about the officially listed untouchable groups, The
Scheduled Castes (Singh 1995).1 The first thing to strike me was the cover, a
glossy photograph of a presumably Scheduled Caste woman with her back
against a tall stone wall, surrounded by her four grubby kids. She is beaming.
The second thing to strike me was the title of this new series, of which this was
the second volume. The series, by the central government’s Anthropological
Survey of India, was called the “People of India,” a name that had been used for
several rather notorious colonial ethnographic projects. Intrigued, I began to
examine this most recent avatar of the “People of India.”
From J.Forbes Watson and J.W.Kaye’s photographic collections to
J.D.Anderson and Herbert Risley’s ethnographies, colonial portrayals entitled the
“People of India” have classified, categorized and mapped the people under
study (Watson 1868–75, Anderson 1982 (orig. 1913), Risley 1915). Historians,
anthropologists and political scientists have since critiqued the colonial state’s
“ordering of difference” on the basis of caste, ethnicity, region, race or religion
in such studies (Metcalf 1995, Pandey 1992, Cohn 1987, 1996, Young 1994).
Despite the tainted past of such ethnographies, the Indian government’s
anthropological department, the Anthropological Survey of India, is engaged in a
project also called the “People of India.” According to the initial circular about
it, “[t]his will be a project on the People of India by the people of India,” a
phrase ringing with nationalism; yet, the goal of this national project is to
generate a profile of each “community” in India, largely defined in terms of
caste. Purportedly a work of apolitical anthropology, this endeavor is
nevertheless sponsored by the state and uses the administrative categories of
Scheduled Castes and Scheduled Tribes. Due to India’s reservation policies for
these disadvantaged groups, controversy rages over their boundaries, numbers
and social conditions (Galanter 1984, Jenkins 1998, 1999, 2001a). Although
castes are a major unit of analysis for the People of India projects, both past and
present, the latest project superimposes the new theme of national unity, a
politically useful focus for an ethnography sponsored by the central government

of India.
I argue that the resumption of this official caste-based anthropology, like
earlier colonial projects, reinforces caste. By underplaying continuing disparities
in the pursuit of conclusions emphasizing national unity, the contemporary
project both reiterates caste-based distinctions and reduces their legitimacy as a
tool for reservation policies. While such a study might be used to undermine
caste distinctions by providing data to refine reservation policies, the People of
India project’s conclusions have often been used to simply undermine the policies.
The People of India projects, colonial and postcolonial, and the varied identity
claims made about them, in them, and through them demonstrate the intertwined
nature of social identities and state identifications.
In this chapter I compare the current People of India project led by Kumar
Suresh Singh with the most famous colonial ethnography of that name authored
by Herbert Risley (1915). First I discuss how this comparison contributes to
critiques of colonial forms of knowledge by highlighting contemporary legacies
of colonial anthropology, particularly the continuing emphasis on caste and the
relationship between official anthropological identifications and politicized
identities. I then compare the colonial and postcolonial People of India projects
in terms of their tendencies toward classification, administration and
politicization. Both are preoccupied with classification, particularly of castes;
both blur the line between anthropology and administration; and, consequently,
both accounts have become politicized. Responses to the project in the context of
reservation policies highlight the political impact of such official identifications.
Finally I turn to the major point of contrast between these studies, namely, their
different conclusions about Indian nationalism. In spite of its more optimistic
conclusions that caste has not impeded national unity, the contemporary study
seems at times startlingly familiar; moreover, whether colonial or post-colonial,
these official ethnographies serve the needs of the state in similar ways.
It is important to preface the following by noting that the contemporary focus
of this chapter is by necessity limited in scope to one anthropological project
headed by Indian anthropologists in administrative rather than academic
positions; the approach discussed here is neither unique to nor generalizable to
Indian anthropologists. As Ronald Inden points out, many academics outside
India continue to harbor quite essentialist views of caste (Inden 1990:83). Many
anthropologists and social scientists in India, some quoted later in this article, are
reflective about colonial practices and critical of the People of India project,
contributing to ongoing debates about the relationship between the social
sciences and governments, both colonial and contemporary.
ANTHROPOLOGY, COLONIALISM AND THE STATE

Colonial administrators in India “darted across the subcontinent mapping,
tabulating and classifying” (Khilnani 1997:155). Caste categories were reified by
colonial administrators concerned about affirming their own superiority and
authority, and postcolonial India has been hard pressed to escape their
authoritative power. Although the administrators and power relations have
changed, the troubling relationship between anthropology and politics has not.
In many contexts, “[c]olonial anthropology supplied a library of ethnicity,”
writes M.Crawford Young in The African Colonial State in Comparative
Perspective. “The traits of a community closely examined through the classical
participant-observation methodology were projected onto a much larger
ethnically labeled collectivity, portrayed as a once and future reality through the
premise of the ethnographic present” (Young 1994b:233). The portrayal, in some
cases, contributed to the reinvigoration or reinforcement of indigenous
categories. Young observes that “the standard presumption was of discrete,
bounded groups,” and that “[i]n the extreme case, the colonial state veritably
breathed life into quite novel categories of identity” (Young 1985:74). Colonial
anthropology influenced the societies under study, including their
anthropological norms and practices, which, in turn, continue to influence these
societies. Categories of identity, once authenticated by “science,” the state, or
both, can take on a life of their own, as can be seen in a variety of contexts in
Asia and elsewhere (Brown 2001, Gladney 1996, Bremen and Shimizu 1999).2
In the Indian context, Gyan Prakash and Nicholas Dirks have been influential
critics of the use of the categories of caste, tribe and race by colonial
ethnographers. Dirks points out that:
Caste was just one category out of many others, one way of organizing and
representing identity. Moreover, caste was not a single category or even
logic of categorization… To read and organize social difference and
deference—pervasive features of Indian society—solely in terms of caste
thus required a striking disregard for ethnographic specificity, as well as
systematic denial of the political mechanisms that selected different kinds
of social units as most significant at different times.
(Dirks 1992a:60)
In the process, colonial ethnographers themselves became one such political

mechanism, selecting caste as a key unit of analysis and imbuing it with new
significance in the political context of colonial rule.
Prakash’s work on colonial ethnography and museums illustrates the “link
between classification and colonialism” (Prakash 1992:156). One extreme
example, from 1866, is a proposal, never fully implemented, for an all-India
ethnological congress. It was to be a sort of living exhibition, in which people,

“classified according to races and tribes, should sit each in his own stall, should
receive and converse with the Public, and submit to being photographed, printed,
and taken off in casts, and otherwise reasonably dealt with, in the interests of
science” (from the Proceedings of the Asiatic Society 1866:90, as quoted in
Prakash 1992:158). Crispin Bates points out that this idea was actually taken up
and implemented on a smaller scale in an 1866–7 exhibition in Jabalpur, at
which human “specimens” were displayed and a report produced. “By far the
bulk of the report was devoted to measurements: of height, length of upper arm,
lower arm, thigh, leg, breadth of chest and body, colour of skin, eyes, pupils,
beard and moustache, length or other peculiarity of heel, any other physical
peculiarities, and diet” (Bates 1995:239).3 Prakash observes that ambiguity often
thwarted such exercises in classification by the British, for whom “[t]o know was
to name, identify and compare” (Prakash 1992:155). This is not to say that the
colonial rulers invented caste out of the blue.4 Caste, and even caste studies,
preceded colonialism, but the Raj’s geographic scope and power in combination
with its propensity to categorize were unprecedented.
Of the many who have written about the power relations and politics behind
colonial anthropology in India, some have considered the implications of such
policies for the present. For example, Arjun Appadurai’s critique of “officially
enforced labeling activities” (Appadurai 1993:326) suggests that such practices
can have a far-reaching impact beyond the colonial period; yet more needs to be
written about official anthropology in contemporary India.5 Building on historical
work about the role of Indians themselves in the development of colonial
ethnographies (Irschick 1994, Bayly 1995:183), I take this line of study into the
present day, re-examining the telling critiques of colonial-era anthropology in
light of current developments.
Both before and after independence, administrators in India have tried to
simplify and codify society. Historian Thomas Metcalf precedes his discussion
of Risley’s People of India with the observation that “[a]s time went on, Indian
ethnography asserted ever more rigorously its scientific claims. Its categories,
embedded in censuses, gazetteers, and revenue books, became ever more closely
tied to the administrative concerns of the state. At the heart of this ethnography
remained always the study of caste” (Metcalf 1995:119). Colonial
anthropological studies were driven not only by the desire to contribute to
scientific progress but also by the imperative to monitor and control a diverse
land. These very motives persist in the contemporary, nationalist desire for
scientific achievement and unity in diversity. Thus the postcolonial People of India
project retains many of the problems of its colonial precedents, even with a
nationalist agenda.
Risley wrote that the “peculiar” institution of caste would most likely prove to
be a challenge for the development of Indian nationality: “So long as a regime of
caste persists, it is difficult to see how the sentiment of unity and solidarity can
penetrate and inspire all classes of the community” (Risley 1915:286, 293). The
new People of India project is an attempt to move beyond the colonial emphasis
on India as a place distinct from the West and divided internally. Yet, this
project embodies the ongoing tension between caste and nation, as it
simultaneously categorizes people as did Risley and, whenever possible,
emphasizes national unity and linkages.
In the following comparison, I find several continuities. Even post-colonial
anthropologists espousing national unity can be haunted by the ethnic and racial
categories bequeathed by colonialism (Young 1985). The current People of India
project breaks from the past in its optimism about Indian nationalism, yet its
nationalist conclusions are rooted in some much older ideas and practices. The
current project justifies the notion of a unified People of India by drawing on
methodologies such as nasal indexes and trait counting, as well as colonial era
assumptions, such as the idea that India is essentially communal in nature. The
project both classifies the people of India and includes optimistic arguments that,
in the end, these various categories are really quite similar. Such conclusions
reflect the difficult position of diverse postcolonial states, as the “emergent civil
societies of the terminal colonial era were simultaneously becoming one and
many” (Young 1994:39). The danger of this official, nationalist anthropology is
that categories such as the Scheduled Castes and Tribes get both a scientific and
an official stamp of approval, while some persistent disparities between these
categories get swept under the carpet.
RISLEY'S PEOPLE OF INDIA
The rise of caste as the single most important trope for colonial Indian
society, and the complicity of Indian anthropology in the project of colonial
state formation, is documented in a great many texts, perhaps nowhere
more fully, though complexly, than in Risley’s classic work, The People of
India.
(Dirks 1992a:68)
Sir Herbert Risley’s book, The People of India, is based on the extensive report
on caste he produced as Census Commissioner for the 1901 census. In the wake
of the mutiny by Indian soldiers of the colonial army in 1857, more and more
information was collected about the inhabitants of India. Anthropologist
G.G.Raheja argues that “the colonial imagination had seized upon caste
identities as a means of understanding and controlling the Indian population after
the blow to administrative complacency occasioned in 1857” (Raheja 1996:
495).6 The first nation-wide Indian census occurred in 1872, and subsequent
censuses, particularly Risley’s, tried to record additional information about the

various groups within India, their practices, and their “character.” This was no
easy task.
Classification
Regarding caste, Risley lamented that “no column on the census schedule
displays a more bewildering variety of entries, or gives so much trouble to the
enumerating and testing staff and to the central offices which compile the
results.” Nevertheless he described a process of “sorting, referencing and cross
referencing and corresponding with local authorities, which ultimately result in
the compilation of a table” (Risley 1915:109–10). He presented his theory that the
caste system originated in the interactions between racial types, offering his
contribution to the broader “scientific” study of race, including racial typologies
and anthropometry. He felt India’s caste system provided an ideal setting for
such physical measurements since “it seemed that the restrictions on
intermarriage, which are peculiar to the Indian social system, would favour this
method of observation, and would enable it to yield particularly clear and
instructive results” (Risley 1915:20). Most notoriously, he commented on the:
curiously close correspondence between the gradations of racial type

indicated by the nasal index and certain of the social data ascertained by
independent inquiry. If we take a series of castes…and arrange them in order
of the average nasal index, so that the caste with the finest nose shall be at
the top and that with the coarsest at the bottom of the list, it will be found
that this order substantially corresponds with the accepted order of social
precedence.
(Risley 1915:29)
Applying racial theories to his study of caste, Risley devoted much of his study
to detailed measurements and classifications of “physical types.”
In addition to physical types, Risley included many more chapters describing
caste in terms of “social types,” “proverbs,” marriage practices, religion, and the
origins of caste.8 During the colonial era, the search for order that dominated
science became enmeshed with the search for justifications for imperial rule
(Metcalf 1995:66–8, Fernandez-Armesto 1995:433–62). Studying racial
classifications and phenomena such as India’s caste system helped to distinguish
colonies from the West in order to “justify” foreign rule. Whether designing a
racial hierarchy or documenting practices seen as “uncivilized,” such studies
could undergird claims about the right to colonize. The obsession with
classification, however, was not only for academic interest and colonial
justification but also for the more practical goal of administration.
Administration
Anthropologist/administrators, a rather common hybrid in India, were aware of
the administrative uses to which their work could be put. One of Risley’s last
public addresses noted the value of teaching Indian anthropology to the men in
the Indian services (Risley 1969:xiv). Clearly, British classifications of castes
and tribes in India were inspired in part by their administrative uses. Notably,
Risley was an administrator before he was an anthropologist. Educated in law
and modern history at Oxford, he gained his experience in anthropology after
joining the Indian civil service (Risley 1969:xi-xii, Bates 1995:241–2).
Nevertheless his work in India allowed him to become president of the Royal
Anthropological Institute in 1910. He held a variety of administrative positions
throughout his career, serving as a member of a commission studying the
working of the Indian police and as the honorary director of the Ethnological
Survey of the Indian Empire. These posts are not as far afield as they may seem.
The labeling and registering of the so-called “criminal tribes” in colonial India
is a particularly striking example of the nexus of anthropology and
administration. In the People of India, Risley described one tribe as “habitual
thieves and burglars,” another as “the most treacherous and aggressive of all the
North-Eastern tribes,” and a third as “hunter, blackmailer, and highway robber”
(Risley 1915:Appendix Plates II and XXXV, 139). As part of what Bernard
Cohn calls their “surveillance modality,” the British developed criminal
ethnography archives and aspired to have systematic anthropometric methods to
identify criminals (Cohn 1996:10–11, Major 1999: 657). Criminal categories
were not the only ones of administrative interest: “Other categories of caste such
as money lending, agricultural or ‘martial’ were used as a basis for legislation
controlling land transfers, the grant of proprietary rights, and the regulation of
rents, as well as a basis for distinguishing between the loyal and the disloyal, and
for recruiting to the armed forces” (Bates 1995:228). In the post-Mutiny context,
administrators, based on their beliefs in the power of primordial ties and “martial
races,” wanted group studies, including evaluations of which groups were loyal
fighters and which were potential threats to order, as part of their efforts to avoid
such revolts in the future (Raheja 1996:504, Des Chene 1999:121–36, Metcalf
1995:122–8). In short, the line between administration and ethnography was
thin.
Politicization
Politicization of these studies was an inevitable result of the overriding
administrative concerns of those carrying them out. As described above, Risley
ushered the census process from a “bewildering variety of entries” on caste to a
final table, which could be easily consulted by administrators. Risley threw this
whole process into the political limelight with his decision to try to arrange the
castes in order of “social precedence” in the 1901 census, generating, in the
words of Ronald Inden, “a mountain of petitions and polemical literature
concerning caste standing” (Inden 1990:59). Various caste associations emerged
to attempt to increase their official status (Caroll 1978). As Risley himself
reflected, “The best evidence of the general success of the experiment, and
incidentally of the remarkable vitality of caste at the present day, is to be found
in the great number of petitions and memorials to which it gave rise” (Risley
1915:112). Optimistically, Risley went on to argue: “If the principle on which
the classification was based had not appealed to the usages and traditions of the
great mass of Hindus, it is inconceivable that so many people should have taken
so much trouble and incurred substantial expenditure with the object of securing
its application in a particular way” (Risley 1915:112). Whether the reaction was
due to his perceived accuracy or inaccuracy, clearly Risley’s attempt to arrange
castes in a hierarchy inspired many to try to interact with and influence the
colonial government.
One of the many classifications Risley drew on was the varna system. One
possible translation of the term “caste,” varna, as discussed in Chapter 1, refers
to an idealized hierarchy of brahmins, kshatriyas, vaishyas, and shudras,
corresponding with occupation and status. This type of codification has come
under criticism, both by some of Risley’s contemporaries as well as current
scholars, since variations in understandings of the varna system have always
confounded attempts to come up with such a classification scheme for all of
India (Cohn 1987:243, Quigley 1993:16, Pant 1987:155). Although recognition
of this system and its meaning varies tremendously, Risley saw it as a potentially
neat, if imperfect, typology to aid in the organization of status rankings, which
were carried out at the provincial level (Risley 1915:114). In response, petitions
for higher status were often attempts to claim a different varna status in the
census records, as in the Khatris’ objection to being classified as vaishyas: “A
meeting of protest was held…and a great array of authorities was marshaled to
prove that the Khatris are lineally descended from the Kshatriyas of Hindu
mythology,” considered a more forward caste (Risley 1915:112).
Colonial categorizations of caste for anthropological and administrative
purposes led to caste associations and caste-based organizing and lobbying to
increase official caste standing (Rudolph and Rudolph 1967). The jockeying for
position of Risley’s era takes an ironic twist in contemporary Indian politics,
when membership in an officially Backward Class can make one eligible for
reservations. The ongoing People of India project continues the traditions of
classification, administration and politicization, although the current politics of
reservations draws more attention to the question of who may become
“backward” than who may become “forward.”
THE NEW PEOPLE OF INDIA PROJECT
I trust this series on the People of India, which is based on a

comprehensive anthropological survey of the country, will be found useful
by all sections of our people, including students, researchers, teachers,
social activists, administrators and political leaders.
(K.S.Singh, India's Communities, People of India Vol. 4 1998:xii)
The current People of India project is a national ethnographic survey of “all

communities of India,” being undertaken by the Anthropological Survey of India,
an organization that emerged from the anthropological section of the Zoological
Survey of India (Singh 2000). The Anthropological Survey of India has been
called the “brainchild” of Herbert Risley (Bayly 1999:275–6), although it was not
founded until 1945. An initial impetus for the new People of India project may
also be attributed to Risley. When Prime Minister Indira Gandhi asked the
Director-General of her Anthropological Survey, K.S.Singh, for an
anthropological introduction to India, “Singh could only suggest H.H.Risley’s
The People of India…. Thus was born a new People of India” (Pinney 1994:22).
The project, headed by K.S.Singh, officially began in 1985, the year after Indira
Gandhi’s death, and the data was gradually published throughout the 1990s.
The People of India study received funding from the Indian Planning
Commission, but the political context of this study is a period in which
developmental zeal, especially the idea that caste would simply disappear with
modernity, has faded. Previous assessments of the imminent decline of caste,
characteristic of immediate post-independence optimism and the ideologies of
modernization theory, have been challenged by the persistence of caste as an
organizing category in political and social life, despite—or, some would argue,
because of—reservation policies for low castes (Randall and Theobald 1998,
Jalali 1993). In a context of continuing violence between different castes,
growing questions about exactly who should benefit from reservations, and
increasing mobilizations and political parties organized along caste lines
(Jaffrelot 2000), the government undertook a new, comprehensive study of the
“communities” of India.
Classification
Six hundred investigators participated in the study of 4693 communities in all of
the states and union territories. They interviewed 24,951 people, out of whom
4981 were women. They based their findings about each community on an
average of five key informants—described as “informed informants”—and spent
an average of 5.5 days researching each community (Singh 1998:x).9 Most
fieldwork was done between 1985 and 1994, followed by additional fieldwork
between 1994 and 1996, and the expected 43 volumes are gradually being
released. These volumes include, in addition to separate volumes on the
Scheduled Castes and Scheduled Tribes, national and state-by-state volumes on
“all” communities, as well as volumes on languages, biological variations and an
anthropological atlas. Former Director-General of the Anthropological Survey of
India from 1984 until 1993, K.S.Singh has “spearheaded, authored and edited”
the People of India project (Singh 1995: cover). Like Risley, Singh is both an
administrator and anthropologist. Singh’s discussion of this project and his own
previous administrative and scholarly work with Scheduled Tribes show a
genuine concern for the plight of disadvantaged communities (Singh interview
11 April 1996, Singh 1985:ix-xi). Yet the legacies of colonial anthropology haunt
aspects of the new project.
Colonial traditions of classifying castes and tribes—and even some of the
same classifications—persist in the current project. Lists of castes were drawn
from census and ethnographic surveys from as far back as 1806; although attempts
were made to update such lists, some characteristics of the previous People of
India project continued. For example, informants were asked to try to place
themselves in the four-fold hierarchical varna system, reminiscent of Risley’s
similar efforts to arrange castes in order of social precedence (Singh 1995:6,
1992a, 1994:7, Risley 1915:114). Singh writes, “There is a widespread
awareness of the varna system among Hindu communities (68.5 per cent), about
half of whom recognize their place in it (52.6 per cent)” (Singh 1998:xvii). A
telling finding is that “Most of the communities (62 per cent) construct their
identity from ethnographic accounts” (Singh 1998:xv), leaving readers of the new
People of India project to wonder whether repeated anthropological inquiries and
accounts about varna have contributed to the public’s “awareness” of such
categories. In spite of the inquiries about varna, the national volumes are
organized alphabetically rather than by social precedence. Notably, however, two
lower status categories, the Scheduled Castes and Scheduled Tribes, almost
always considered outside the varna system, have been singled out and
scrutinized in their own, more detailed, volumes.10
A form of classification that bears a strong resemblance to Risley’s project is
the use of biological data, including cephalic and nasal indexes and
“dermatoglyphics.” The organization of this data is also reminiscent of colonial
categories:
The attributes of physical diversity are discussed in this volume by pooling

the samples under seven categories of presumably related groups of
people, namely other (caste) communities (general population), Scheduled
Castes, Scheduled Tribes, Muslims, Christians, Parsis and others.
(Singh et al. 1994:7)
This list of categories is interesting for several reasons. The authors’ desire to
conduct anthropometric measurements on these “presumably related groups”
echoes Risley’s enthusiasm for India as an ideal laboratory for such methods due
to community endogamy. The labeling of the categories is also notable, signaling
which are the dominant groups and adopting administrative categories into an
ethnographic study. Upper caste Hindu “others” are also known as the “general
population” as opposed to other “others,” presumably sundry minority
communities. The lower castes, tribes and related minorities singled out by
colonial anthropologists and administrators prior to independence are also
singled out here. The organizing assumption seems to remain that religious, caste
and even administrative distinctions are the logical categories to use when
looking for actual physical differences, although the findings of this study
ultimately confound this assumption. The use of the government’s officially
listed, or “scheduled,” categories in this anthropological study brings us to the
administrative tendencies shaping the People of India project, another continuity
with colonial traditions.
Administration
The government of India continues to focus special research efforts on the
“empowerment of the socially disadvantaged groups,” including the Scheduled
Castes and Scheduled Tribes. In the Ninth Five Year Plan, the government’s
“research, evaluation and monitoring” needs included studies of “the problem
areas, the problem groups” as well as “special problems related to frontiers/
forest dwelling tribals; malnutrition and alcoholism among SCs and STs; the
emerging problems like the drug abuse/drug addiction” (Planning Commission
1997–2002:3.9.93). The focus on “problem” groups and behaviors is a thread
running from Risley’s project to the present. Even couched in terms of
empowerment, the assumed association of scheduled groups with alcoholism and
criminality continues in the present day. For example, alcohol consumption
practices are more consistently noted in the People of India volumes on
Scheduled Castes and Scheduled Tribes than in the volumes on all communities.
The most striking parallel to the dark side of colonial administrative
anthropology, particularly its focus on “criminals,” is another contemporary
Anthropological Survey of India project called the “Portrait Building System.”
This was an attempt to aid in apprehending criminals through collecting photos of
different population or “ethnic” groups. This initiative took place after the
assassination of Rajiv Gandhi and was a joint endeavor of the Bureau of Police
Research and Development and the Anthropological Survey of India. A total of
698 “ethnic groups” from all over India were photographed for this project.
Different features were clipped, classified and stored: “Hairline was classified
into 7, forehead 6, eye 11, nose 9 and chin into 9 classes” (Basu 1990:114). This
effort epitomizes the imperative to classify and administer, which guided both
colonial and postcolonial official anthropology.
In order to have diversity of features, 2234 normal adult persons were

photographed from Police and Detective Training Schools, post-graduate
and under-graduate college hostels, semi-desert and mountainous areas,
industrial establishments, market places of tribal areas, villages, tea
gardens, slums of big cities, etc. covering great distances on all types of
transport wherever necessary and occasionally on foot… For example,
number S234L is a segment that indicates that it is from South Indian
Material, its serial number is 234 and it is lip.
(Basu 1990:114)
The People of India project also included a collection of 21,362 photographs; the
goal was to take at least five photographs of each community. Although there is
no official connection between the two projects, the portrait building system
shows that a recent project involving the same organization, the Anthropological
Survey of India, was motivated by administrative concerns about criminality.
Materials collected for the People of India project could be useful for the portrait
building system; even if they are not used, the portrait project illustrates the
effect of administrative concerns on official anthropological agendas.11
Another crossover between anthropology and administration is the People of
India project’s use of the administrative categories of Scheduled Castes and
Tribes and subsequent entanglement with reservation policies. I asked the head
of the current project, K.S.Singh, whether the Anthropological Survey of India
had any influence on the schedules, or lists, of groups eligible for reservations,
given their official capacity as the anthropological advisor to the government of
India. Singh responded that “anthropologists of the Anthropological Survey of
India are serving on the committee of the government of India which
recommends the listing of particular communities as Scheduled Tribe or
Scheduled Caste” (Singh interview 11 April 1996). Administrators of the
Ministry of Welfare who had dealt with group petitions for Scheduled Caste
status had used some Anthropological Survey of India materials, sought advice
from the Survey, and included someone from the Survey on a nonpermanent
committee that initially went through such petitions (Choudhary and Khan
interviews 17 December 1996). A census administrator who, in his previous job
in the Census’ Social Studies Division, had responded to queries from the
Ministry of Welfare on the makeup of Scheduled Caste lists noted that the
Anthropological Survey of India had not generally been associated with this
work of submitting recommendations about lists but had more recently been
included in meetings on the subject (Chakravorty interview 16 December 1996).
Due to the use of official categories and the benefits at stake, this
anthropological project’s commitment to honor the self-identification of groups
was compromised. In an early description of the project, Singh wrote, “We decided
to go ultimately by how a community identifies itself in practice” (Singh 1987:
241). Yet later Singh wrote, in many cases, “a community, attracted by the
facilities extended to the SC or the ST, twists ethnographic accounts in its
endeavor to identify itself with either of the constitutional categories” (Singh
1995:25), accounts which, it seems, he felt he must untwist. The Scheduled
Castes volume was “guided by government of India notifications,” followed by
the “perceptions of the people” (Singh 1995:2). Although initially preferring self-
identification and continuing to recognize the “dynamic” nature of the caste
system, Singh faced an administrative reality: “The problem is the scheduled
castes and tribes categor[ies] have been listed in India. The list is there” (Singh
interview 11 April 1996).
In 1990, while the People of India project was in full swing, Prime Minister
V.P.Singh announced his intention to expand the reservations program for
central government jobs to benefit not just the Scheduled Castes and Scheduled
Tribes, but also the Other Backward Classes. D.L.Sheth, a sociologist and
member of the National Commission for Backward Classes, set up to help
implement these new policies, said the commission sometimes used People of
India data in its work, which dealt with disputes over which groups were really
Other Backward Classes (Sheth interview 6 September 1996). A former head of
the commission said he had no interaction with the project but that such a study
might be useful to them (Prasad interview 18 September 1996). Aside from its
utility for the administration of reservations, the project became involved in the
public controversy over the extension of central government job reservations to
this wider range of disadvantaged groups. As the name “Other Backward
Classes” suggests, this is an ambiguous category; moreover, some people were
adamantly opposed to increasing the percentage of reserved jobs. The People of
India project soon became a weapon in the debate.
Politicization
K.S.Singh downplayed any direct involvement between his scholarly endeavor
and the debate over reservations, yet in his writing on the People of India
project, Singh has commented critically on the definitions of the Other Backward
Classes used for reservation policies, classifications which at times must rely on
census data from 1931, the last time caste was comprehensively recorded (Singh
1987:240, 244). The official ceremony releasing the People of India data, in
October 1990, coincided with protests over Prime Minister V.P.Singh’s
announcement that he supported the 1980 Mandal Commission’s report on the
Other Backward Classes and that he wanted to extend 27 per cent reservations in
central government jobs to these groups. This was one of the most controversial
decisions of the V.P.Singh government and contributed to its fall from power
shortly thereafter. As upper caste students engaged in demonstrations, including
several self-immolations, the People of India data were released. The new study
was cited in critiques of the policy, since the People of India project listed a
smaller total number of backward classes than the older Mandal report, which
was used by the government:
As the row over the number of backward communities listed in the Mandal
Commission’s report continues a just-concluded study of the
Anthropological Survey of India has identified 1051 backward classes in
the country… The Mandal Commission has listed more then 3000
communities as socially and educationally backward.
(The Statesman, 1 October 1990)
This report was followed the next day by a similar report, entitled “Mandal’s
many mistakes,” in the Indian Express:
The findings of the People of India Project study of all castes and
communities in the country, conducted by the Anthropological Survey of
India, are a[t] great variance with the Mandal Commission’s report in
terms of the number of other backward classes… According to the study,
the most exhaustive and definitive work on communities in India so far,
there are 1051 backward communities whereas the Mandal Commission
has recorded 3743 OBCs… Investigators and scholars who participated in
the five-year project found that in all states and Union territories, the number
of SC, ST and OBC communities was actually far lower than recorded
previously.
(Indian Express-New Delhi, 2 October 1990)
The numerical disparity between the People of India project and the Mandal
report can be partially explained by repetitions of the same community in the
latter, due to synonyms and alternate spellings, yet such newspaper headlines and
articles implied that Mandal advocated including a large number of illegitimate
beneficiary groups on the reserved lists.
Entangled in the protests against reservations, the People of India project
became overtly politicized. Delhi University sociologist Andre Beteille argues
that this study, organized along caste lines, “cannot be passed off as a
disinterested piece of research, because it has the stamp of the authority of the
Government of India and becomes, therefore, in some sense an officially
acknowledged if not an officially affirmed classification” (Beteille interview 9
September 1996). Not only is a government bureaucracy behind the project but
government categories also organize it. B.K.Roy Burman, an anthropologist and

head of the Mandal Commission’s “technical committee” (although he stepped
down in protest over the commission’s use of data), has been saying since the
1960s that the categories of Scheduled Caste and Scheduled Tribe are “policy”
categories rather than “scientific” categories (Burman interview 31 October
1996). Thus their place in an anthropological study is dubious.
In part due to its official ramifications, this study, like Risley’s before it, has
generated petitions from groups anxious to be included or concerned about their
own entries or the entries of other groups. Whereas people petitioning Risley
hoped for higher status, contemporary reservation policies put scrutiny on the
lower status groups. For example, the second edition of the Scheduled Castes
volume includes a new appendix, a “restudy” of the Beda Jangam, “following a
complaint that the Jangam, the uppercrust of the Lingayat, had passed
themselves off as Beda Jangam in order to corner the benefits of the scheduled
castes” (Singh 1995:1366). The revised entry describes this group as “infiltrated”
and notes that the population rose very quickly between 1961 and 1981. Whereas
the initial entry reports, “In Kannada ‘beda’ means begger,” the appendix entry
hotly amends this: “These Jangam even claim that the Kannada word Beda…
means to beg, and that seeking alms is their age-old practice. This goes against
the traditional meaning…hunter” (Singh 1995:205, 1368). The new entry goes so
far as to scold the infiltrators: “They treat some communities as untouchables,
thus perpetrating caste divisions. Yet the irony is that some of them pass
themselves off as Beda Jangam in order to avail the benefits of protective
discrimination given to the scheduled castes” (Singh 1995:1368). This restudy
illustrates several continuities with the previous People of India project:
Informants still try to manipulate official anthropology to their advantage. People
still petition the government over the construction of official identities. Official
anthropological accounts still unabashedly address administrative concerns.
In addition to petitioners, there are numerous examples of groups citing the
People of India study to advance various causes. To give an example of the way
official categories and community claims interact, a Christian community in Goa
was fighting for Other Backward Class status, but eschewed the untouchable
Scheduled Caste category. A related concern was that they did not want their
street renamed for Dr B.R.Ambedkar, a Dalit leader and India’s first Law
Minister, due to concerns about the potential stigma associated with such an
address. In their efforts, they cited their entry in the People of India project as
scientific proof that they are not untouchable Mahars (Deccan Herald-Bangalore
26 January 1996).
In addition to those citing the People of India study for various political
purposes, other groups, unsatisfied with such official accounts, have sponsored
community studies of their own, which are often used to argue that they too are
“backward.” Let my People Go: Scheduled Caste People Group Profiles,
describes disadvantages of untouchable groups, including Christians, countering

official ethnography which must follow the government’s current policy that
Christians are not eligible for Scheduled Caste status (India Missions Association
1995). The People of India volume on Scheduled Castes, for example, mentions
the demand by some Christians and Muslims to be included as Scheduled Castes,
but only includes the officially recognized Hindu, Sikh and Buddhist Scheduled
Castes, another example of government policies overriding self-identifications
(Singh 1995:10). In contrast, Let My People Go attempts to undermine such a
classification by making an “anthropological” argument that Scheduled Castes
never have been Hindus:
Dr Vidhyarti, the famous anthropologist studying the religion of

untouchables, says they follow their own native festivals… The recent
opening of some Hindu temples for the Scheduled Castes is a purely
political gimmick since 1920 to get their votes… To put it more exactly, it
was out of the greatest concern to keep the two faiths separate, that
untouchability and isolation were stringently maintained. Therefore to call
a person as a Hindu Scheduled Caste is as absurd as calling for a hot ice
cream.
(India Missions Association 1995:40)
Another publication likewise argues, on “anthropological” grounds, that the

government’s policy of excluding Christians from Scheduled Caste benefits is
flawed. The following passage is part of a critique of the Presidential order of
1950 that “No person who professes a religion different than Hinduism shall be
deemed to be a member of a Scheduled Caste.” The passage is part of an
argument that this order makes no sense because no Scheduled Castes are really
Hindus.
Reply the Christian Dalit anthropologists “every word of it is wrong. First

‘caste’ is not a Hindu word… To say that Hinduism created us the Dalit
peoples is as absurd as to say that Nazism created the Jews… We the
Dalits predate the Hindus in India by 1000 years… On what basis [do] you
call us converts from Hinduism?”
(National Coordination Committee for SC Christians 1995:2)
Groups in society are objecting to their categorization by the government and,

recognizing the political weight given to anthropological data, are trying to fight
fire with fire.
An advocate of reservations for Muslims pointed out to me a Muslim
organization that sponsored its own study to highlight Muslim
disadvantages. Their volume is an attempt to challenge the official “social
science of backwardness,” which, it is claimed, does not do enough to compare

the situation of Muslim communities vis-à-vis Hindus, due to the government
treating such comparisons as sensitive and thus “classified” (Ahmad 1995:9).12
These counter-studies demonstrate that “anthropological” and other social
scientific justifications are seen as potentially valuable tools or weapons in
political arguments over the proper classifications for reservations.
The fact that the government has relied on anthropology to make and justify
administrative decisions in the past and continues to do so inspires groups to
make anthropological arguments to legitimize their own claims. Because of its
caste classifications and administrative uses, the new People of India has become
as politicized as the old, although the arguments over status have become
inverted, with various groups today arguing for the “backward” status their
ancestors desperately sought to shed. The two studies of the People of India also
draw different conclusions about Indian nationalism.
NATIONALISM AND THE PEOPLE OF INDIA

“India is no more one country than Europe—indeed very much less,” wrote
Herbert Risley in his early twentieth-century report, a conclusion that the
contemporary People of India project thoroughly rejects (Risely 1915:114).
Risley’s final chapter focused on “caste and nationality” and the question of
whether they are compatible or not (Risley 1915: 278–301). This chapter
contained prescient projections regarding the use of caste by political machines
and the prospects for Hindu nationalism in India, yet his conclusions were
pessimistic about the future of Indian nationalism. Emphasizing the many social
divisions that he spent the rest of the book classifying, such as language, religion
and caste, Risley’s conclusions foreshadowed colonial policies based on
assumptions of religious incompatibility, which culminated in Partition. Risley
argued that “while community of religion strengthens and consolidates national
sentiment, religious differences create distinct types within a nation and tend to
perpetuate separate and antagonistic interests” (Risley 1915:291).
Risley suggested that religious and caste distinctions boded ill for Indian
nationalism, although he prudently noted that discounting Indian nationalism
completely would be premature. “Caste seems at first sight to be absolutely
incompatible with the idea of nationality…but a caste or group of castes might
harden into a nation and…the caste organization itself might be employed to
bring about such a consummation” (Risley 1915:300). At the same time, in
keeping with his “scientific” study of caste, he likened a relaxation of the ties of
caste with the “withdrawal of some elemental force like gravitation or molecular
attraction” (Risley 1915:278). Risley’s skepticism of nationalism is a striking
example of what Nicholas Dirks has called the “relentless anthropologizing of
India, which served to misrecognize the social and historical possibilities for the
nationalist awakening, even as it worked to reify categories of social

classification” (Dirks 1997:184). Whether caste persisted or not, Risley’s final
chapter leaves one expecting disintegration rather than an Indian nation.
In contrast, the new People of India project’s director, K.S.Singh, emphasized
that his was the first truly “national” People of India project. When asked why he
thought a project encompassing all of India was so important, Singh responded,
“India has been a cultural area. India has been a civilizational area” (Singh
interview 11 April 1996). Alluding to ancient texts of the subcontinent, Singh
continued:
In Arthashastra and Mahabharata, they mention communities which are

spread from Kanyakumari to Kashmir, from Assam to Gujarat. So India
has been an ethnographic universe, and to that extent we have done the
survey…the first complete survey.
(Singh interview 11 April 1996)
Evoking ancient references to the most far-flung regions of the modern Indian
state, in all four directions, Singh implied that India had an “ethnographic”
coherence as a civilization or cultural unit in the past that justifies its status as a
political unit in the present. Given the relatively recent development of the Indian
nation with its existing boundaries, these allusions to ancient unity are
historically dubious, yet imagining India in such a way is useful to the modern
Indian state funding the project and trying to keep a diverse country intact. The
difficult role of nationalist anthropologists is epitomized in the explanation of
K.S.Singh, who uses ancient textual references to justify novel political
boundaries largely inherited from colonial rule.
Singh noted that although Risley made some comments about “national
character” in the Indian context, “colonial ethnography generally ignored
linkages of communities” (Singh 1987:244). Singh emphasized his own focus on
linkages to counter the argument that a study based on communities might, like
previous colonial studies, accentuate community distinctions or competition. In
the study’s early stages, Singh noted that there was as yet no anthropological
explanation for India’s unity: “It is with this integrational interaction…that this
project is largely concerned” (Singh 1987:246, 248). At the official release of
quantitative data generated by the project, Arjun Singh, then Minister for Human
Resource Development, felt that this “mapping of the human surface of India”
might be used to “strike an authentic chord for national integration” (Singh 1990:
100).13 Thus, in spite of its organization along the lines of subnational
communities, this project emphasizes national unity wherever possible.
The national scope and nationalist sentiment of the project are reasons for the
choice of the word “community,” rather than caste or jati, to label the units of
analysis. Rather than emphasizing divisions, community implies fellowship;
moreover the term could be applied all over India to countless types of
categories. An article from the project’s early years includes a discussion of this
decision:
The word community or samudaya as it is called in some states like Kerala

could be a more appropriate concept for an all India reference than caste
with its various local names. Scholars…have expressed their consensus that
community (samudaya) is less “offensive” and sounds more meaningful
and befitting than jati. The community, is a larger social entity; jati is a
closed structure.
(Singh 1987:238–9)
In contrast, the “old” ethnographers “seemed to take a peculiar delight in

describing…the divisions and subdivisions” (Singh 1987:238). An emphasis on
community, and unity, is “befitting” a national project, but it is hard to gloss
over the key role of caste as a unit of analysis, particularly in the studies of
Scheduled Castes. Although the English term community (rarely samudaya) has
been used extensively in the various volumes, the volumes on the Scheduled
Castes and Tribes, in particular, also refer to these groups as castes and tribes,
most obviously in their titles, but also in some textual passages.
In the most striking departure from Risley’s chapter on caste and nationality,
the current project specifically downplays caste as an impediment to nationality.
It does this through several types of arguments, emphasizing cultural and
physical similarities, the fluidity of identity and, above all, progress. One method
used to analyze national linkages is trait counting, a rather problematic
methodology (Moerman 1965). By counting a list of traits, defined as “a cultural
value, which is unique” (Singh 1992:98), anthropologists involved in the People
of India project could conclude that the communities in India really have a lot in
common. Volume seven is devoted in part to “Linkages: a Quantitative Profile”
and is one of the volumes Singh described as “the soul of the project” (Singh
interview 11 April 1996).14 Again drawing a contrast between colonial
ethnography and his project’s emphasis on processes of interaction and linkage,
Singh discussed the quantitative analysis of traits:
Integration is sharing, and this is, precisely, what this project is about. We
have tried to identify those traits which are shared by our people.
According to our findings, out of 776 traits that we have identified, as
many as 250 traits are shared by all of us. It shows that we are by and large
a cohesive society.
(Singh 1990:100–1)
Based on this analysis of traits, Singh argued that “there is a very high
correlation of traits between the SC and the ST, between ST and Hindus,
between the Hindu and Sikh, the Hindu and Muslims (which is very high
indeed)” (Singh 1990:100–1). Given ongoing tensions between these groups,
including some secessionist movements, this conclusion seems like an attempt to
“scientifically” justify Indian unity.
“Bio-anthropological parameters,” such as head measurements, are still cited
in the current study, not to make an argument about the racial origins of caste as
in Risley’s book but, quite the contrary, to make arguments about inter-caste
similarities: “The head shape of the dolichocephalic type is predominantly
distributed in most of the populations, including scheduled castes…there is less
variation at inter-group [caste] level than at inter-regional level” (Singh 1995:4).
More modern techniques produced similar arguments: “regional variations in
gene frequencies seem to be greater than the variation between castes or
religious groups” (Singh 1992:99). The message that physical differences are more
geographic than caste-based was, however, less clear in the People of India
project’s anthropological atlas (volume 11), which demarcated ten
“physiographic” zones, corresponding to both geographic and physiological
indicators, yet included the observation that “there is a tendency for relatively
broad noses in the scheduled castes in comparison with other communities.”
Such “rogue data” on noses bears a striking similarity to Risley’s infamous caste-
nose correlations (Pinney 1994:22). Although the continuity of the tradition of
taking head measurements is troubling, most major conclusions, at least, were
different.
In addition to anthropometric and biological analyses to reinforce the idea that
caste is no impediment to national unity, Singh also recognized some fluidity of
identity and lower caste progress, even evoking the constructivist notion of the
Other.
If you look at traditional ethnography, some people are called degraded. Or

menials. Or there are derogatory references to people… Now…we are a
democratic country. We are normatively politically equal. In fact one of
the merits of the project is that we have reflected the growing concern with
political equality of all the communities of India. It is not that
untouchability has gone…but there is movement towards equality… There
is movement towards acceptance of the Other. On equal terms. It is not
that all the ugly and vicious and savage characteristics of inequality have
disappeared, but…lots of communities have changed their names. They do
not want to be called by the names the ethnographic literature has given
them…a change in nomenclature is indicative of a change in their
perception of themselves.
Such change was reflected by noting name changes by communities rejecting a

name seen as derogatory to adopt a new one. The project was guided by
“sensitivity tests,” an attempt to use the most current names, although former
names now considered derogatory were still included in “the larger write up” as
a reference, according to Singh.15
Fluidity and lower caste progress were recurring themes in my interview with
Singh and in his project. Evidence of such progress is a welcome result, yet the
summaries of the Scheduled Caste data tended to emphasize the “glass half-full”
and avoid comparisons with higher castes; for example, the study summary
mentioned that Scheduled Caste literacy doubled between 1961–81 (to 21.38 per
cent) but did not include figures that would show that this literacy rate is still
much lower than the upper castes (Singh 1995:12). Similarly, Singh played up
increased opportunities while downplaying lingering barriers:
In this project on people of India, two things which have come out very
sharply in terms of change are, number one, people’s own aspirations,
people’s own perceptions of themselves and their relations with others, and
secondly, the changing economy, the diversification of occupations which
has taken place…. Better opportunities are opening up; migrations are
taking place; so these are the factors which make these communities or
caste[s] more dynamic than we imagined.
A book reviewer of the initial People of India volumes rejoiced that the “very
popular concept of ‘Indian unity in diversity’ has been perfectly manifested here
not through any sentimental view point but by way of the application of
scientific methodology” (Sarkar 1992:366). The contemporary People of India
project imbued some colonial era methods and classifications with a new spin, the
viability and progress of the Indian nation.
In spite of these nationalist conclusions, however, the overall structuring of the
project along caste lines offers political ammunition to those who choose to
emphasize their caste identity. A Dalit publication, for example, rather than
challenging the caste distinctions, has taken up the Anthropological Survey of
India’s recognition of distinct communities. An article in Dalit Voice, a rather
polemic publication for Dalit solidarity, reports that “The ASI verdict is that the
Indian society continues to be a collection of castes and communities…
scheduled castes have 450 jatis;” whereas “Upper castes want to destroy our
caste consciousness by promoting national consciousness. In other words, they
want to destroy the national consciousness of each jati and get them assimilated
in the Hindu nationalism…‘The People of India Project’ has warned that such a
Hinduisation of SC/ST/BCs has not worked and it will not work” (Dalit Voice
16–31 March 1996). One example of the project’s recognition of subnational
cultures is the poem by untouchable saint-poet Cokhamela which opens the

Scheduled Castes volume. This was, perhaps, an effort to give Scheduled Castes
a creative voice in the volume, although, notably, poetry about inequality from
the fourteenth century does not preclude the emphasis on present day progress in
later pages. In spite of the project’s focus on national unity, it is still a study
which gives subnational communities official, and politically useful, recognition.
The official reinforcement of community lines is what worries sociologist
Andre Beteille about the new People of India: “Risley’s exercise could have been
justified given the time when it was done, given the needs of colonial
administrators in a different kind of society; whereas this merely stops a natural
process of the dissolution of these categories by reconfirming them once again”
(Beteille interview 9 September 1996). For example, Prime Minister Narasimha
Rao’s comments at the release of the introductory volume of the People of India
reconfirmed distinct categories and illustrated the conflicting agendas of the
People of India. He lauded the “scientific,” “standard” and “systematic” study of
all of India. At the same time, his comments on the utility of the project for
determining the special medical needs of tribals exposed beliefs in the deep, even
biological, differences between communities: “You cannot treat tribals as you
treat a person of the plains area” (Rao 1993:2).
Not just categorized but also singled out into their own volumes, the
Scheduled Castes and Tribes face a predicament in the People of India’s dual
emphasis on community and nation. As noted, Dalit Voice has cited the project
to support a distinct Dalit identity, but others concerned about the plight of these
disadvantaged groups may find any utility associated with this official
recognition of their categories undermined by the overall emphasis on progress
and linkages. Those Scheduled Castes and Tribes who would use these
categories to argue for relief find little support in many of the project’s
conclusions, although some of the data could be useful. For example, the
introduction to the Scheduled Tribes volume bluntly states that “The tribes do
not suffer from any social stigma” (Singh 1994:13); yet, Scheduled Tribes are
largely perceived as “low status” according to the project’s own data on
community rankings, in the tradition of Risley.16
Public commentaries on the project use its themes of social mobility and
progress in attacks on reservations. A media account of the People of India
project at the height of the upper caste backlash against reservations in 1990
emphasized that:
[A] major finding is the tremendous diversification of traditional

occupations that has occurred. According to sociologists, this is a
significant indicator that backward classes are being integrated into the
mainstream… The study also shows that the largest area of employment
among the SC and ST communities is the Government Services. Many SC/
ST members are becoming teachers, engineers, doctors and village-level

workers.
(Indian Express-New Delhi, 1 October 1990)
Actually, 54 per cent of the reserved quota of central government jobs for
Scheduled Castes go unfilled, according to the National Commission for
Scheduled Castes and Scheduled Tribes (Human Rights Watch 2001). Moreover,
“the predominant occupation of the Untouchables is agricultural work… In
agriculture Untouchables are almost always paid labourers or sharecroppers,
rather than self-employed landowners” (Mendelsohn and Vicziany 1994:79).
Although the majority of the scheduled communities are stuck in the lowest
levels of employment, the People of India project provided fodder for the anti-
reservation agitation. The project’s positive spin on the data and the media’s
idyllic pictures of professional opportunities for Scheduled Castes and Tribes do
challenge stereotypes of disadvantaged communities. Yet, the project’s official
commemoration of communities, in combination with such exuberant
commentary, both reinforces caste and tribal categories and undermines the
political utility of the categories for those in them. Perhaps due to fears of the
clash between caste and nation, a specter raised by Risley’s earlier work, this
reincarnation of the People of India tries to gloss over caste disparities while it
continues to record caste for posterity.
CONCLUSION
The head of the current project, K.S.Singh, feels “we have laid the firm
foundation of postcolonial ethnography in our country” (Singh 1990:101). Yet the
People of India project fails to fully address the problems of the colonial
ethnographic tradition, especially the classification of peoples, the use of these
classifications for administrative purposes, and the subsequent politicization of
anthropology and the categories themselves. In some ways, the new project does
move beyond the work of Risley, particularly when there are attempts to reflect
changes, linkages and the development of a nation. Nationalism characterizes
this “postcolonial” ethnography, reflecting the changing needs of the state, which
shifted from an old concern with maintaining colonial rule by reinforcing social
divisions to a new preoccupation with maintaining national unity and integrity in
an independent and diverse country.
Persistent parallels between colonial and postcolonial anthropology can be
traced to some ongoing imperatives of governing. Rajni Kothari’s (1968)
description of governmental impulses to collect information in the “‘information
and communications explosion’ of our times,” was as applicable in Risley’s time
as our own:
Governments are in possession of information and intelligence over their

peoples…to an unprecedented degree, with an ability to probe deeply,
sometimes regions which were hitherto sanctified as “private and
personal.” A variety of impulses and skills join in this enterprise—
humanitarian motives, genuinely scientific pursuits, methodological
compulsions, the insecurity of those in positions of power.
(Kothari 1968:14–19)
Despite dramatically changing interpretations of the Indian nation, the Indian

state’s colonial and contemporary impulses to categorize and record the people
of India have been remarkably resilient. Whether in pursuit of domination or
development, administrators need to make societies “legible” and, in the process,
may reinforce the categories they only meant to record (Scott 1998).
Postcolonial states, with remnants of colonial administrative institutions and
traditions, seem particularly prone to these tendencies, as the reinvented People
of India project demonstrates.
The People of India projects—old and new—take on political significance
simply by being anthropology with an official stamp. Indeed, there is a
substantial dialogue afoot addressing the academic discipline’s relationship with
the government.17 Another debate surrounds the adoption of Western categories,
methods or modes of thinking into Indian social sciences. In a forum 30 years
ago on “academic colonialism,” Yogendra Singh warned against the dangers of
becoming “more imitative than innovative,” and this debate continues (quoted in
Uberoi 2000). Against the background of these broader disciplinary debates, the
People of India project, with its explicit links to the government as well as to
Risley, is controversial in academic circles.
Turning from academic politics to national politics, official anthropology in
India continues to be guided by administrative considerations, but this does not
mean that the People of India project always served the needs of particular
governments, as demonstrated by the uproar it inflamed when Prime Minister
V.P.Singh instituted new reservations for Other Backward Classes. K.S.Singh
argued that the Anthropological Survey of India, while guided by the policies
and plans of governments, maintains “considerable autonomy” in matters of
research: “It is not correct to assert that it is sarkari [governmental] anthropology
all the way” (Singh 2000). A desire for knowledge of a far-flung population and
a concern with national integrity and legitimacy are at the forefront of the
analysis. A variety of political parties have found these concerns compelling
enough to keep the project alive from 1985 until today, spanning a period of
quite rapidly changing governments. Although started under the Congress Party,
the nationalist bent of some of the project’s analysis seems to complement the
Hindu nationalist Bharatiya Janata Party’s ideology as well. The project’s
accounting of national cultural “linkages” reinforces this party’s emphasis on
“civilizational identity” and “cultural nationalism” (Bharatiya Janata Party

1998). There is tension between the project’s nationalist bent and the community
profile data, however, as illustrated by the Dalit Voice article featuring the
People of India project in a criticism of Hindu nationalism. In another striking
example of this tension, a Christian mission organization felt that these
community profiles would be so helpful in their conversion efforts that they
actually declared the People of India project a sign from God, much to the
consternation of Hindu nationalist Arun Shourie (Shourie 1999).18 The project,
although shaped by administrative concerns, is not simply a tool of the
government in power.
Postcolonial anthropology by official agencies is a fascinating amalgam of old
practices and new imperatives. The studies produced and public reactions to them
provide unique snapshots of the pressures and priorities of particular political
moments. For example, the use of similar forms of data to make opposite
arguments about the prospects for Indian nationalism illustrates the power, and
limits, of nationalist ideology when older practices and categories linger. Despite
the continuity of the central categories, the shift in public attention and
governmental scrutiny from the “forward” to the “backward” castes
demonstrates the influence of reservations on both official identifications and
social identity claims.
For example, People of India researchers had to neglect their plan to rely on
group “self-identification” when it conflicts with the official boundaries of the
Scheduled Caste category, as in the revised entry about the faux Beda Jangams
or the exclusion of Christians from the Scheduled Caste volume. Clearly the
postcolonial state continues to monitor the boundaries of its official categories. At
the same time, various social groups, including Dalits and religious minorities,
have utilized or challenged such categorizations, by reinterpreting them to
bolster their own political arguments or by putting forth their own studies to
portray alternative visions of the “People of India.” Official anthropology and
reactions to it provide telling examples of the give and take between analytic or
policy simplifications, on the one hand, and politicized identities, on the other.
In its more analytical modes, the new People of India project’s recognition of
the fluidity of identity provides a welcome addition to colonial classifications, yet
legacies of colonial anthropology persist in the bulk of the volumes’ capsule
summaries of communities. Even the more nuanced recognition of caste mobility
can be taken too far. The current People of India project’s sunny conclusions
about caste barriers breaking down—justified with a dizzying combination of
colonial era anthropometry, genetic research, trait counting and constructivist
notions of identity—could justify state inaction in spite of ongoing
discrimination. As I continue to ponder the Scheduled Caste volume I purchased
at the Delhi book-store, I must conclude that recognizing the continuing salience
of caste may be better than ignoring it, but this recognition could have been cast
in many ways. Concern for national unity has inspired a “rose colored glasses”
approach to analyzing the still limited progress of low castes, an approach
epitomized in the People of India project by the grubby yet happy cover girl on
volume two.
4
Caste certificates and lists
The administration of reservation policies, particularly the definition of

Scheduled Castes, Scheduled Tribes and Other Backward Classes through caste
certificates and lists, contributes to ongoing constructions of these castes, tribes
and classes. The administrative enforcement of categories shapes the incentive
structure within which individuals and groups claim or mobilize around particular
group identities. Building on political science theories about the social
construction of populations targeted by public policies, I argue that the
substantial benefits of reservations come at a cost to eligible groups. The
processes of identity verification associated with caste certificates, in particular,
tend to burden and stigmatize disadvantaged groups and to reinforce assumptions
that these groups have clear boundaries. In other words, my examination of the
administration of reservations shows that, although the policies might be
empowering, the process is distinctly not. The administrative processes used to
identify groups for inclusion on the lists of Backward Classes necessitate group
organization, group petitions and, ultimately, group rankings. In short, although
reservations offer opportunities to members of low status groups, the procedures
associated with official lists reinforce group-based activism and hierarchies.
Several scholars studying policy administration in complex social systems
note that states simplify societies and then structure those societies by
implementing their simplified schemes:
The simplified facts of the administrator are necessarily of a certain type—

static… What I want to argue about state simplification is that the modern
state, through its official attempts and with varying success, creates a
population with those standardized characteristics because it will be easier
to monitor, count, assess and manage.
(Scott 1995:35)
Administrators are forced to view societies as a finite number of standardized

groups in order to do what administrators do, implement policies. These policies,
in turn, can have an impact on group identities. In the words of political scientist
Theda Skocpol, an important area of study is “the ways in which the structures
and activities of states unintentionally influence the formation of groups and the
political capacities, ideas, and demands of various sectors of society” (Skocpol
1985:21). In other words, the “constant redefinition of group identities in the
political arena is an effect of politics and policy as much as a cause” (Lieberman
1995:441). The following discussion of the administration of caste lists and
certificates focuses on how bureaucrats disentangle and reassemble identities for
policy purposes. Such administrative practices can reiterate older status
distinctions and structure group identity claims.
Anne Schneider and Helen Ingram’s research highlights the links between
public policies and social constructions of the groups targeted by these policies,
which they call “target populations” (Schneider and Ingram 1993, 1995). An
example would be welfare policies and “welfare mothers” in the United States.
Government officials tend to impose burdens on less powerful or negatively
constructed policy targets (Schneider and Ingram 1993:337–41). Often policies
are based on the assumption that their target populations are readily definable
(Schneider and Ingram 1993:335). Depending on their policy experiences,
members of the targeted population may either favor collective action or
individual interaction with the government (Schneider and Ingram 1993:341).
Other scholars have applied some of these ideas in studies of policies and the social
constructions of groups as varied as landlords, the elderly, immigrants, and
people with AIDS (Hunter and Nixon 1999, Hudson 1999, Coutin 1998, Drass et
al. 1997). The various officially backward groups in India constitute striking
examples of target populations. Applying ideas inspired by this literature in the
quite different setting of India can increase our understanding of the
ramifications of group-based policies.
The administration of reservation policies in an exceedingly complex society
is a feat not well understood, especially since scholars have tended to focus on the
development, legalities, and political fallout of India’s reservation policies
(Galanter 1984, Mitra 1994, Nesiah 1997, Parikh 1997). Here I address several
aspects of the relationship between administration and identity, including
identifying and labeling, untangling identities, and individual and group
processes. Several conclusions emerge: Bureaucratic procedures associated with
reservations tend to burden, label and stigmatize the very groups they are meant
to help, particularly the least powerful groups. Reservation policies are based on
an assumption that target populations have objective parameters, even as
administrators struggle to define those parameters. Reservation policy processes
sometimes focus on individuals, as in the case of caste certificates, but the
process of officially listing Other Backward Classes necessitates that aspiring
groups must be somewhat organized, gather aggregate data, and compare
themselves to groups above and below them. Such policies and procedures
increase the efficacy of collective rather than individual interactions with the
government and administratively reinforce social groups and gradations.
IDENTIFYING AND LABELING BENEFICIARIES: A

TANGLE OF ADVANTAGE AND OPPRESSION
Every person who claims to belong to a Scheduled Caste or Scheduled [T]

ribe has to produce a certificate to the appointing authority as sufficient
proof in support of the claim so as to make him eligible for the various
relaxations and concessions.
(Muthuswamy and Brinda 1996:6)
Reservations in India are an extraordinary example of a constitutional

commitment to uplift the less powerful; however, the administration of these
policies also places substantial burdens on those very beneficiaries. The
verification of group identity is one example of the burdens imposed on less
powerful policy targets, such as Scheduled Castes, Scheduled Tribes and Other
Backward Classes. Not only is the process quite comprehensive and personal,
but it also can involve overt labeling and ongoing scrutiny for members of the
least powerful groups.
The Ministry of Home Affairs has “emphasized the need for a proper
verification being made by the certificate issuing authorities before they actually
issue a certificate” (Ministry of Personnel 1993:247). Such verification may be
based on such data as “revenue records and if need be, through reliable inquiries”
(Ministry of Personnel 1993:398). Verifications of Scheduled Caste membership
in particular can be personally invasive since, for this category, religious beliefs
and practices are scrutinized to exclude Christians and Muslims. According to an
organization protesting the denial of Scheduled Caste status to Christians, “The
SC applicants for caste certificates are humiliated and insulted by officials, if a
symbol related to Christian religion is noticed during the search of the
applicants, their houses or their parents” (National Coordination Committee for
SC Christians 1995b).
The courts have upheld and even encouraged administrative procedures to
monitor less powerful groups, as in the Supreme Court decision discussed in
Chapter 2 regarding two students alleged to be from a spurious tribe. Kumari
Madhuri Patil v. Additional Commissioner Tribal Development (Supreme Court
of India 1994).1 In that case, the court ruled that the process of getting a caste
certificate should include not only an initial application, but also further
verification by a “scrutiny committee,” police involvement in local inquiries, and,
in some cases, “affinity tests” to verify applicants’ knowledge of the
anthropology of the group they claim as their own. Leila Dushkin’s observation
about policies for the Scheduled Castes is apt: In practice, “the so-called
beneficiaries have to have a great deal of stamina and political expertise” (Dushkin
1979:665).
Other generally more powerful or more positively viewed groups who benefit
from special “concessions” in government employment policy, such as ex-
servicemen or sportsmen, also have to furnish certificates and document their
eligibility, but they have not inspired the copious memos, rulings and court cases
over how best to scrutinize their claims (Muthuswamy and Brinda 1996, Nabhi
2001). Moreover, in the case of the less powerful and negatively constructed
backward classes, even after caste certificates have been issued, repeated
verifications throughout their careers can haunt beneficiaries. These verifications
involve patrolling the parameters of reservation categories and implementing
procedures that identify but also label and stigmatize beneficiaries.
The sheer number of names used over the course of history to label the
Scheduled Castes indicates how difficult it is to give the many so-called
untouchable groups a common label and how easily such labels fall from favor
as they come to be seen as inaccurate, stigmatizing or simply out of date. As
Michael Banton points out, “The doubts about what are the best names for
groups, and where the boundaries are to be drawn, show that the groups people
recognize in everyday life are often multidimensional” (Banton 1997:13). A list
compiled by Simon Charsley and paraphrased here hints at the scope of the
countless labels used over the years to refer to Scheduled Castes. Some names
emerged from social or political movements and others were coined or at least
passed into common usage by administrators: Aprishya Sudras (used by Census
Commissioner Herbert Risley); impure castes, avarna, outcastes (outside of
varna or caste); Pariah (a jati name that became generalized); Pancham Bandum,
the Fifth Caste, Panchama (all meaning a fifth group outside of the “touchable”
categories); AdDharm, Adi-Karnataka, Adi-Dravida, Adi-Andhra (all emerging
from various regional movements trying to establish untouchables as distinct
religious or regional groups seen as “original”); untouchables, ex-untouchables,
Depressed Classes, suppressed classes, exterior castes, Scheduled Castes;
Harijans (coined by Gandhi), Dalits (coined by Ambedkarites), and Naaga
Holeya (in the Mysore area, “a neo-traditional term associated with a Buddhist
challenge to converted Christians to reunite with them” (Charsley 1996:17).
Of these labels, those which were adopted for administrative uses include
“Depressed Classes,” which first appeared around 1880 and which the British
used for the 1921 Indian census. J.H.Hutton, the Census Commissioner in 1931,
however, saw fit to replace what he called an “unfortunate and depressing label”
with another, namely, “exterior castes” (Charsley 1996:14). In the 1935
Government of India Act, this label in turn was replaced by a purely
administrative term, Scheduled Castes. In addition to the Scheduled Castes, the
Backward Classes label also extends back to the colonial era and was later used
in the 1950 constitution. The term “backward,” in contrast to the more neutral
term “scheduled,” evokes inferiority and inevitable comparisons to its opposite,
the “forward” citizens of India. The Other Backward Classes category has been
sporadically altered in recent years to Socially and Economically Backward
Classes (SEBCs), not much of an improvement. Labeling people as backward, in
the context of a policy to fight inequalities, simultaneously reinforces hegemonic
ideologies about group status and rankings.
Categorizing and labeling groups on the basis of untouchability or
backwardness in this way seems inevitably stigmatizing. Simon Charsley argues
that by subsuming diverse castes, dichotomizing society, and defining certain
categories of people as victims, such labels “refer to nothing those labeled do or
are, merely what others negatively do to them: they are excluded” (Charsley
1996:13). Moreover, the resilience of caste or jati distinctions within these
broader categories inhibits the growth of a wider political unity among those so
categorized (Charsley 1996, Omvedt 1993). Surprisingly then, in the political
realm, labels such as Scheduled Castes and Other Backward Classes have in some
cases become a political resource rather than a stigmatizing liability. Some
scholars trace the gradual growth of a pan-Scheduled Caste or pan-Dalit identity
and related political movements (Jalali 1993, Kothari 1994, Zelliot 1996). Others
note the increasing political power of the Other Backward Classes (Frankel
1988, Jaffrelot 2000).
As I will demonstrate in later chapters, the eagerness of various groups to be
endowed with scheduled or backward labels belies the assumption that these are
simply stigmatizing terms. Nevertheless, even if many groups are prepared to
swallow the stigma to get the useful label, or to accept the burdens to get the
benefits, being given these labels is a double-edged sword. The clearest case of
labeling and stigmatizing the groups targeted by reservations is the procedure of
literally pasting labels on their employee files:
The appointing authorities should verify the caste status of a Scheduled

Caste/Tribe officer at the time of initial appointment and promotion
against a vacancy reserved for Scheduled Caste/Tribe. For this purpose, the
caste and the community to which a SC/ST person belongs, his place of
residence and the name of the state, should be pasted on the top of the
service book, personal file or any other relevant document covering its
employee to facilitate such verification.
(Ministry of Personnel 1993:248–9, emphasis added)
Such labels make it difficult for employees in reserved jobs to choose to hide
their caste or tribal affiliation at work.
A survey of 55 Scheduled Caste elites, including bureaucrats, suggests that
caste anonymity is an important way to avoid being treated as an untouchable.
Thirty-four respondents reported that untouchability was observed at their

workplaces, substantially more than the numbers reporting the observance of
untouchability in places where it is harder to discover someone’s caste, such as
buses, trains, post offices, banks, or restaurants (Roy and Singh 1987:68).2 One
of those surveyed, a government official, responded, “When I meet a stranger he
starts interacting with me on equal terms but as soon as he discovers my caste his
behavior changes drastically and [he] starts maintaining distance” (quoted in Roy
and Singh 1987:69). This survey indicates that labeling groups in an attempt to
counter discriminatory practices can have negative effects, as such labels can
facilitate further caste discrimination.
A government official appointing a candidate to a reserved job may engage in
further identity verification. For example, “An appointing authority, if it
considers it necessary for any reason, [sic] verify the claim of a candidate
through the District Magistrate of the place where the candidate and/or his family
ordinarily resides” (Ministry of Personnel 1993:248). Hiring authorities are
instructed to include the following clause in written offers of employment:
The appointment is provisional and is subject to the caste/tribe certificates

being verified through the proper channels and if the verification reveals that
the claim to belong to Scheduled Caste or Scheduled Tribe, as the case
may be, is false, the services will be terminated forthwith without assigning
any further reasons and without prejudice to such further action as may be
taken under the provisions of the Indian Penal Code for production of false
certificates.
(Muthuswamy and Brinda 1996:6)
Even when job candidates have caste certificates, their appointments are to be
provisional until their identity is further scrutinized.
While clearly onerous, the process includes some protections for the applicant.
Once a certificate has been issued, the burden of proof shifts away from the
beneficiary; the courts have held that “after a valid certificate is issued, the onus
is on the authorities to establish that the certificate was erroneous” (Muthuswamy
and Brinda 1996:248). The proper procedure at that point is for the employer to
initiate a “discreet inquiry” by a local official to “find out the genuineness of the
community claimed by the individual” (Muthuswamy and Brinda 1996:248).
Such inquiries occasionally reveal “false” certificates, for reasons ranging from
carelessness, to corruption to ambiguity. Contemporary headlines, “Fake
Certificate Racket Busted” and “Caste Certificate Racket Lands Official in
Soup,” illustrate the value of these certificates and the temptation to falsify
identity for material gain (Hindustan Times-New Delhi, 16 July 1996, Indian
Express, 23 June 1997). In such cases individuals are to be given a “reasonable
opportunity” to justify their claims before their certificates are canceled
(Muthuswamy and Brinda 1996:248). If, instead, the certificate is duly verified
by local authorities, and “the report is received that the certificate is genuine,
thereafter the certificate holder cannot be further harassed to prove his caste/
community in any other manner” (Muthuswamy and Brinda 1996:247).
In spite of such rules, cases of repeated harassment have occurred.3 Moreover,
identity verifications, albeit less invasive, continue throughout the careers of
persons in reserved jobs, especially at times of promotion. Occasional changes in
the schedules or lists of eligible groups are the impetus for this practice of
periodically checking to see if employees are still eligible for their reserved jobs.
It may be mentioned that a Scheduled Caste Person, whose caste has been
descheduled after his initial appointment as a Scheduled Caste, is no longer
entitled to enjoy the benefit of reservation in promotions. This verification
of caste-status at every important up-turn of employee's career is
necessary so that the benefit of reservation and other scheme of
concessions etc. meant for SC/ST should go only to the rightful claimants
and not those who become disentitled to them.
(Ministry of Personnel 1993:249, emphasis added)
Even if employees’ group affiliations remain unchanged, the descheduling of

groups could disqualify them from their jobs. Points of promotion or “every
important up-turn of [the] employee’s career” are usually occasions for kudos,
but for the officially backward these are the points at which the employee’s
backward status is to be checked. From the time their status is pasted to the cover
of their personnel files and throughout repeated examinations of their status,
members of the groups targeted by reservation policies face several potentially
embarrassing bureaucratic procedures.4
To sum up, administrative procedures to identify those eligible for
reservations label and stigmatize the target populations. Calling someone
backward is stigmatizing in itself; checking to see if that person is “really”
backward can be even more so. One is either considered low (genuinely
backward) or a fraud (pretending to be backward). In other less bureaucratic
realms, sensitivity to the potential embarrassment of drawing attention to a
person’s official category is evident and occasionally leads to efforts to protect
target populations from public identification. A case in point is a conversation I
had with teachers at a central school, which, since it is set up for the offspring of
central government employees, has a sizeable percentage of officially backward
students. The teachers told me that according to a directive to teachers, they are
not to draw attention publicly to the special status of Scheduled Caste or Tribe
students or indicate in any way—during an attendance roll call, for example—
who in their classes is a member of an officially backward community
(conversation with New Delhi Central School teachers 17 July 1996). Although
commitment to this practice no doubt varies, such rules in the realm of education
are an attempt to avoid labeling and stigmatizing the youngest members of groups
eligible for reservations.
COMPLICATING PARAMETERS: MIGRATION,

MARRIAGE, ADOPTION AND RELIGIOUS (RE)
CONVERSION
Bureaucratic memos, forms, rules and regulations pertaining to reservations rest
on the assumption that an empirical categorization of citizens is possible and can
be carried out in an objective manner. However, as when the state adjudicates
identity claims or designs an ethnographic study, bureaucrats issuing caste
certificates are often faced with unclear boundaries. These men and women face
particular difficulties classifying those persons who seem to fall between
categories, whether due to migration, intercaste marriage, adoption, or
conversion. Such cases challenge the notion that the categories themselves can
be considered stable and identifiable; yet bureaucrats gamely implement rules to
carry out the policies. Those administering reservations, unlike policy-makers,
see ambiguous cases in their day-to-day work. Cognizant of such challenges,
these bureaucrats occasionally admit that standard responses to some situations
are impossible.
In order to be eligible for reservations, members of Scheduled Castes,
Scheduled Tribes or Other Backward Classes must be issued certificates by
bureaucrats with the authority to do so.5 Those issuing certificates must verify
that applicants and their parents belong to the community in which they claim
membership; that their particular communities are included in the Scheduled
Castes, Scheduled Tribes or Other Backward Classes lists in their state; that
applicants are really from the state, or area within a state, in which their group is
listed; and, finally, that applicants claiming to be Scheduled Castes are either
Hindu, Sikh or Buddhist (Ministry of Personnel 1993:397). After all of this
verification, the caste certificate officers may still face ambiguity. Additional
regulations define the boundaries in cases involving migration, marriage,
adoption and conversion.
Migration
Geographic boundaries are one of several types of boundary issues facing the
bureaucrats who issue caste certificates. Membership in the groups eligible for
reservations is determined in part by location. The lists, or schedules, of
communities vary from state to state, and, in some cases, communities are included
in the lists only if they are from certain areas within states. The Ministry of
Home Affairs has issued directives reminding bureaucrats to carefully consider

migrants applying for caste certificates:
I am directed to say that many instances have come to the notice of this
Ministry wherein certificates of belonging to a particular Scheduled Caste/
Tribe have not been issued strictly in accordance with principles governing
the issue of such certificates…it is possible that two persons belonging to
the same caste but residing in different States/U.T.s [Union Territories]
may not both be treated to belong to Scheduled Caste/Tribe or vice versa.
Thus the residence of a particular person in a particular locality assumes
special significance.
(Ministry of Home Affairs Memorandum to the State
Governments 1977) (A directive consulted
by Ministry of Welfare officials as of 1996)
In cases of migration, people remain members of a scheduled category even if

they move to a region within their state in which their caste or community is not
scheduled. In cases of migration between states, a person can only claim
Scheduled Caste or Tribe status according to the lists of “the State to which he
originally belonged” (Ministry of Personnel 1993:397). The Supreme Court
reinforced this bureaucratic principle, ruling that migrants who were considered
members of Scheduled Castes or Tribes before they moved can only derive
reservation benefits from their original state. Action Committee on Issue of Caste
Certificate to Scheduled Castes and Scheduled Tribes in the State of
Maharashtra and Another v. Union of India (1994).6 Thus the application form
for Scheduled Caste Certificates in Delhi, where there are many migrants, asks
each applicant for the duration of his or her stay in Delhi, the complete address
of his or her father, and even the “State to which ancestors belong” (Application
Form for Scheduled Caste Certificate).7
Who should issue a migrant’s certificate? This question is potentially
confusing for both bureaucrats and applicants. An administrative rule states that
the authority issuing a certificate should be “the one concerned with the locality
in which the person applying for the certificate had as his place of permanent
abode at the time of the notification of the relevant Presidential Order,” namely,
the order which lists his caste or community as a Scheduled Caste or Scheduled
Tribe (Ministry of Personnel 1993:398). Such a rule can inconvenience migrants
seeking certificates, another example of the burdens placed on less powerful
target populations, although the identity verification of migrants is sometimes
carried out by mail (Mahesh interview 29 November 1996).
In short, in order to verify group boundaries, administrators tie applicants’
official identities to their “original” place. This can make it more difficult for
migrants to get caste certificates, and, even if they manage to get them, the
certificates cannot be used to obtain jobs in their new states. Resentment of

migrants from other states taking over local jobs was the impetus for another
form of job reservations in parts of India, namely ethnic reservations for “sons of
the soil” (Weiner and Katzenstein 1981). The rule disqualifying inter-state
migrants from reservations for Scheduled Castes or Tribes is a similar example
of policies favoring people with “ancestral” ties to a place.
Marriage and adoption

A “guiding principle” issued by the Ministry of Home Affairs states that claims
of Scheduled Caste or Tribe status due to intercommunity marriage should be
decided based on the caste of a person at birth.8 Non-SC/ST members married to
SC/ST members do not become members of their spouses’ categories. Also, SC/
ST members retain their status even after marriage to a non-SC/ST. The Ministry
of Home Affairs issued this “guiding principle” to authorities empowered to
issue certificates (Ministry of Home Affairs circular letter 1975). Notably, court
decisions have not always been consistent with these bureaucratic principles in
cases involving the categorization of women. As mentioned in Chapter 2, in a
case in which a lower-status woman married a higher-status man, the woman did
not get to keep her officially Backward status after marriage. Atul Chandra
Adhikari v. State of Orissa, Orissa High Court (1995). However, in Valsamma
Paul, a subsequent mixed marriage case discussed in-depth in Chapter 2, the
Supreme Court held that caste, for the purpose of reservations, was determined
at birth, in keeping with the guiding principle of the executive branch.
The offspring of intercaste marriages, not surprisingly, present additional
challenges to administrators. The guidelines used by an official of the Ministry
of Welfare stipulate that “the crucial test [is] to determine whether a child born
out of such a wedlock has been accepted by the Scheduled Caste community as a
member of their community and has been brought up in that surrounding”
(Ministry of Home Affairs Memorandum to States 1977) (a directive consulted
by Ministry of Welfare officials as of 1996). According to the same
memorandum, community recognition determines Scheduled Tribe status: “[I]t is
the recognition and acceptance by the society of the children born out of a
marriage between a member of a Scheduled Tribe with an outsider, which is the
main determining factor irrespective of whether the tribe is matriarchal or
patriarchal.” A letter from the Ministry of Welfare reiterated the need for
administrative “verification of the acceptance given by the members” of the
community in question before classifying offspring of intercommunity marriages
(Ministry of Welfare letter 1994). This overrode a rule, which was more
empowering for the target population, that children could be classified
“according to the declaration of the parents regarding the way of life in which the
children are brought up” (Government of India, Social Welfare Department
1975).9 Now the administrator must determine whether the father or mother’s
community is most accepting of the child.
This, obviously, may not be a clear call, and one bureaucrat I spoke with told
me that in most cases, he would categorize children with their fathers (Mahesh
interview 29 November 1996). A “frequently asked questions” page on the
website of the Ministry of Social Justice and Empowerment reiterates the
unofficial standard that a father’s classification generally determines that of his
children.10
Question: What shall be the status of the offspring of a couple one of whom
is a Scheduled Caste/Scheduled Tribe?
Answer:…If the child has been accepted by the Scheduled Caste or
Scheduled Tribe Community and has been brought up in the surrounding
of Scheduled Caste or Scheduled Tribe spouse then, the child would be
treated as Scheduled Caste or Scheduled Tribe, as the case may be.
However, each case is to be examined on its merit. However, in general,
the following illustrations are made:
Example I Father—Scheduled Caste/Scheduled Tribe

Mother—non-Scheduled Caste/non-Scheduled
Tribe Child—shall be Scheduled Caste/
Scheduled Tribe
Example II Mother—Scheduled Caste/Scheduled Tribe
Father—non-Scheduled Caste/non-Scheduled
Tribe Child—shall be non-Scheduled Caste/
non-Scheduled Tribe
Note: each case will be decided on case to case basis as per its merit.11
The tension between the general “illustrations” and the case specific “note” in
this answer illustrates the quandaries of bureaucrats trying to manufacture
standard responses to specific situations. Since only these two examples are
given, the father’s status appears to be the default answer to questions about
categorizing intercommunity offspring, unless an industrious bureaucrat
determines that the mother’s community was more accepting. These cases of
intercommunity offspring illustrate bureaucratic standardization and the lack of
power of target populations. Intercommunity couples have little say in the
classification of their children, and even cases involving matriarchal
communities are likely to be standardized in a patriarchal way, as the
presumption is that the father’s identity trumps the mother’s.
“Illegitimate” children are an exception, according to a rule based on
additional assumptions about gender roles. In such cases, the government
assumes that “the illegitimate children are generally brought up by the mother in
her own surroundings. Therefore, if the mother belongs to the Scheduled Caste
and brings up the child within a Scheduled Caste community, the child can be
taken as a member of the Scheduled Caste community” (Ministry of Home
Affairs Memorandum to States 1977). This memorandum also addresses the
status of the offspring when both parents are members of Scheduled Castes or
Scheduled Tribes but each belongs to a “different sub-caste or subtribe.” In such
a case, the children can be presumed to be members of the Scheduled Castes or
Tribes and usually to be members of the father’s subcaste or tribe, although this
rule is not absolute in cases of desertion or divorce. A similar rule applies in
cases of offspring when one parent is Scheduled Caste and the other is Scheduled
Tribe.
Notably, these rules regarding children of mixed marriages leave much up to
the administrator, who must determine something as subjective as the
“acceptance” of a person by a group. Moreover, each of these specific rules are
followed by equivocation, or recognition of the possibility of ambiguous
boundaries in spite of the above regulations for so many combinations of identity.
As in the Ministry of Social Justice and Empowerment website, each part of the
memo on inter-category marriages closes with a phrase such as: “The above are
general observations, however, each case has to be examined in the light of the
circumstances prevalent in that case and final decision has to be taken ther[e]of,”
or, “[t]he general position of the law has been stated above. However, each
individual case will have to be examined in the light of existing facts and
circumstances in such cases” (Ministry of Home Affairs Memorandum to States
1977).
Cases of adoption have also generated bureaucratic rules to clarify and verify
the identities of adoptees. When people apply for certificates on the grounds that
they were adopted by a Scheduled Caste family, for example, the burden of
proof rests on them. They must substantiate the validity of their adoptions.
Generally, adoptions after a person has turned 15 years old or has married are
suspect. In such cases, a special inquiry by the District Magistrate is necessary in
order to determine whether such adoptions are permitted by the customs
particular to the relevant community. If so, the Magistrate must make a special
note of that on the certificate. Certificates are denied to those who are not living
with and supported by their adoptive parents or those who have ties to or are
receiving financial help from birth parents who are not in a group eligible for the
certificate (Ministry of Personnel 1993:397).
To summarize, most people are officially stuck with their “original” caste or
tribe in spite of intercommunity marriage or adoption, unless they can meet the
stringent burden of proof for legitimate adoptions. This set of rules echoes the
rules regarding migrants, who retain their classification in their original home.
Bureaucrats and their memoranda mention the need to consider each individual
case, particularly in the classification of the offspring of mixed marriages.
However, even in such cases, standard rules of thumb have developed that
generally give such children the status of their fathers. Moreover, Ministry of
Welfare directives squelched procedures that once empowered parents in
intercommunity marriages to declare their children’s community, in favor of
bureaucratic verification of community acceptance. Since communities may be
quite ambivalent about intercommunity couples and their children, this is another
example of administrative burdens on the less powerful, burdens which do little
to encourage such marriages.
Conversion and reconversion

For the Scheduled Caste category, bureaucrats must determine the religion of caste
certificate applicants, since only Hindus, Sikhs and Buddhists are eligible. Two
types of liminal cases arise: first, converts to another religion, such as Islam or
Christianity, who reconvert back to one of these eligible religions and, second,
converts’ offspring, who reconvert to an eligible religion. The “points of
guidance” for both types of cases are to “establish that such a convert has been
accepted by the members of the caste claimed as one among themselves”
(Ministry of Personnel 1993:397). If so, reconverted applicants can be
considered members of their own or their parents’ original Scheduled Caste.
Notably, the directive does not recognize those who simply convert (rather than
reconvert) to one of the eligible religions. If they or their parents were never
Hindus, Sikhs or Buddhists, they cannot get a certificate. Moreover, Buddhists,
to be eligible, must be “neo-Buddhists,” formerly members of a Hindu lower
caste. A Dalit Buddhist lawyer, Bhagwan Das, describes the government’s
decision to include such neo-Buddhists as Scheduled Castes: “They said, yes, we
will include them subject to [the] condition that they should embrace Buddhism
from such and such caste. So when we applied for a certificate, you have to
mention that I am embracing this since this date and I am a Buddhist convert” (Das
interview 30 April 1996).
In order to monitor the boundaries of the slippery category of religion,
government authorities continue to pay attention to the religion of Scheduled
Castes even after the caste certificates are issued and the beneficiaries of
reservations are employed. According to a directive to government employers,
“the appointing authorities should stipulate in the letter of appointment issued to
Scheduled Caste candidates that they should inform about the change, if any, in
their religion to their appointing/administrative authority immediately after such
a change” (Ministry of Personnel 1993:248). It seems that employees who could
lose benefits would be unlikely to voluntarily reveal their conversion, but the
fact that the demand is made at all signals that the government is keen to
continue monitoring the boundaries of their target populations.
In brief, these rules hold that, as long as they do not convert yet again,
applicants may receive Scheduled Caste certificates by reconverting to Hinduism,
Buddhism or Sikhism if they are accepted by the community in question. A key
rule is that they must be reconverting to their own or their parents’ “original”
religion. The emphasis on an applicant’s original place, original caste and
original religion in the caste certificate process makes the applicants seem “stuck”
in their original communities. In the case of religion, however, conversion away
from one of the designated religions could immediately expel a certificate holder
from Scheduled Caste status. The categorization of converts is a particularly
thorny issue in Indian politics, and there are some tensions between judicial
decisions and administrative rulings in this area. A Supreme Court decision
involving converts, discussed in Chapter 2, disregarded whether members of a
Scheduled Caste accepted a convert to Hinduism back into his parents’ former
caste, ruling against the convert’s claim to be Scheduled Caste because he was
not born in that caste. S.Swvigaradoss v. Zonal Manager, Supreme Court of
India (1996).12
These examples involving migration, marriage, adoption and conversion
illustrate the rules by which bureaucrats assign people particular identities for the
purpose of receiving caste certificates. The examples confirm James Scott’s
observation that many people “defy easy categorization; they must be sorted into
one bin or the other because the exercise requires that the population be divided
into these categories” (Scott 1995:35). In other words, administrative rules tend
to render even flexible boundaries static, in spite of changes in location, family
or religion. According to the bureaucratic regulations summarized here, official
identification and status are tied to a migrant’s original place, to a married man’s
original caste, or to a convert’s reconversion to their original religion.
The static portrayal of status may be the most feasible rather than the most
accurate means of boundary verification. Although reservation policies may be
based on the premise that there are clear cut categories, the bureaucrats writing
and implementing boundary rules are forced to grapple with the ambiguity of
identities, occasionally noting, in the examples above, the need to proceed on a
case-by-case basis. Questions facing administrators include whether a person has
been accepted by a caste or community after conversion or whether the adoption
of older children is permitted by community custom. Bureaucrats at the Ministry
of Welfare dealing with Scheduled Caste and Scheduled Tribe issues argued that
cases must be seen in context and that bureaucrats must respect community
sensibilities in ambiguous cases. On the basis of their own experience, these
administrators pointed out that the Scheduled Caste and Tribe categories are
internally complex, due to their intersection with other categories such as women
and the disabled (A.K.Choudhary and Aziz Khan interviews 17 December
1996). The myriad procedures involved in implementing these policies reinforce
the assumption that there are identifiable categories, although, in practice,
bureaucrats seem to recognize that the boundaries could be interpreted in a

variety of ways.
INDIVIDUAL AND GROUP PROCESSES: MAKING

THE LIST
The administration of reservation policies shapes the incentive structure within
which people choose to interact with their government as individuals or as
groups. To illustrate, I will contrast the experiences of individuals trying to get
Other Backward Class certificates with the experiences of groups trying to get on
Other Backward Class lists. Although the certificate process may affect more
people directly and is a more individualized process, an increasing number of
groups are petitioning for Other Backward Class status, a process which
reinforces group-based strategies.
Official decisions about issuing caste certificates focus on the question of
which individuals qualify as members of a scheduled or backward group.
Decisions over the lists of eligible groups, in contrast, focus on the question of
which groups qualify to be listed. Thus the process of petition is different. In the
first case, an individual goes to a local bureaucrat and fills out a form providing
personal and family information. In the latter case, a group submits data about
the group as a whole to a state or national Backward Classes commission.13 The
bureaucratic procedures associated with reservations vary in their tendencies to
reinforce individual or group-level political activity.14
The lists of Other Backward Classes are in flux due to the relatively recent
extension in the 1990s of reservations to these groups at the national level. This
development led to a new administrative unit, the National Backward Classes
Commission, the first permanent commission of this kind, although it has a
changing membership. When a person applies for an Other Backward Class
certificate, the process is similar to the process already described for Scheduled
Caste and Scheduled Tribe certificates. K.Mahesh, the Executive Magistrate in
charge of issuing caste certificates for part of Delhi, and his associate, Ramesh
Kumar, described the process as follows (Mahesh and Kumar interviews 29
November 1996): An individual fills out a form, after which the administrators
“send a person for verification,” generally a bailiff or inspector. Mahesh noted
that since you find “black, brown and fair” among all castes and cannot make a
visual verification, the inspector goes to the individual’s place of residence and
makes inquiries, basically by talking to the neighbors. Documents that may be
required of the applicant include an ID card from the Election Commission, a
ration card or school certificate to prove identity and residence, or, most useful, a
caste or tribe certificate of a parent (OBC Certificate Application Form,
Annexure I: Supporting Documents Required).15 Since Other Backward Classes
in some states are a new category, older relatives’ certificates may not be
available; in such cases local inquiry is especially necessary. Mahesh notes that
the OBC category is quite “anthropologically complex,” a composite of caste,
class, and profession, further blurred by “mixing” and “migration;” thus he tends
to eschew “anthropological” data on the groups in question and to rely instead on
an individual’s official documents and the accounts of locals (Mahesh interview
20 November 1996).
Unlike the Scheduled Castes and Scheduled Tribes, Other Backward Class
applicants also go through a second process to verify that they are not a member
of the “creamy layer” of the Other Backward Classes.16 The official “creamy
layer” criteria identify individuals who have risen to the top of their community
in socioeconomic terms. The creamy layer of OBCs are not eligible for
reservations. This effort to exclude the “forward backwards” population from
benefits is the topic of Chapter 8, but the procedure is relevant to the present
discussion of individual-level administrative processes. Due to the exclusion of
more well-off people from Other Backward Class benefits, the application form
from the government of Delhi goes beyond group details such as caste, sub-
caste, occupational group, and “[s]erial number of the caste in the Central List of
OBCs,” to ask for information pertaining to an individual applicant’s “creamy
layer” status. This part of the application includes questions about the applicant’s
parents or husband—the assumption, evidently, is that a wife could not lift her
husband into the creamy layer—including employment, land, and income
(“Application Form for a Certificate for Eligibility for Reservation of Jobs for
Other Backward Classes in Civil Posts and Services Under Government of
Delhi” and “Form of Certificate to Be Produced By Other Backwards Classes
Applying for Appointment to Posts Under the Government of India”).17
Documentation to be submitted includes such items as a salary slip and tax
records, if any. Currently, once an OBC certificate is issued, it is issued for life,
regardless of a subsequent change in status, but the next generation might be
excluded from the category.
Applying for an Other Backward Class certificate is an individual act,
although certainly not solitary, as can be seen from the neighborhood inquiries
involved.18 Moreover, most of the questions and documentation require personal
papers and immediate family data rather than group-level statistics or
anthropological data about one’s caste. People in this process are hardly
atomized, but the quest for a certificate is significantly more individualized than
the collective process needed to get on the list itself.
In contrast to the individual certificate process, the process of applying to the
National Commission for Backward Classes in order to be listed as an Other
Backward Class reinforces group cooperation. One of the mandates of the
Commission, formed in 1993, is to standardize the various lists of Other
Backward Classes prepared by state government commissions and the central list
prepared by the previous, temporary National Backward Classes Commission,
known as the Mandal Commission. The commission was “enacted on the

direction of the Supreme Court to set up a permanent body for entertaining,
examining and recommending upon requests for inclusion and complaints of
over-inclusion and under-inclusion in the central lists of Other Backward
Classes” for central government reservations.19
In the words of sociologist D.L.Sheth, a “social science member” of the
National Commission for Backward Classes, the commission’s central question
is “Who are the OBCs?” (Sheth interview 6 September 1996).20 The five
member commission “has to monitor this policy of inclusion and exclusion on
the affirmative action list for backward classes” (Sheth interview 6 September
1996). The Commission deals with the Other Backward Classes, and potential
Other Backward Classes, as groups only, in contrast to the individualized
bureaucracy of caste certificates. The Commission initially compiled a central list
made up of those groups that were on both the Mandal list and the state lists.
Then, as Sheth described, it turned to the consideration of more ambiguous
cases: “What about those who were not in the Mandal list but in the state list?”
and vice versa; “And about those who may not be on either but may still be
genuinely backward?” The Commission receives both applications from
communities trying to get on the lists and complaints against “those who have
been wrongfully included.” Another daunting task, “This commission also has to
review the entire list after ten years…and take out those communities which may
have ceased to be backward” (Sheth interview 6 September 1996).
According to M.L.Mathur and Hoshain Singh, of the Research Division of the
National Commission for Backward Classes (Mathur and Singh interviews 18
July 1996), the process starts when a group fills out an application in the form of
a questionnaire. After the Commission reads the questionnaire and collects
additional data from the state in question, they hold a hearing. Sheth described the
process: “We would hold public hearings, where all others can appear and
contest or support what the claimants have to say. And when we were in doubt,
we also institute[d] social inquiry, some kind of research on that community.”21
The “most important data are about their social backwardness” as a group (Sheth
interview 6 September 1996).
Such data can be difficult to collect. Thus Sheth and the Backward Classes
Commission supported counting Other Backward Classes on future censuses, a
contentious issue to be discussed in Chapter 5.
I think if there is policy of giving reservation then it is logical that they

should be counted, like SCs and STs are counted… Our report
recommended that that’s the most logical thing to do. Just by counting
caste, you don’t support caste… This is part of the policy. We must have
data.
(Sheth interview 6 September 1996).
Sheth also finds “very useful” anthropological data about groups, such as the
government’s recent People of India Project, which he characterizes as an
“ethnographic update of caste, communities and tribes in India… We also used…
that kind of material in the Commission, in our decision making” (Sheth
interview 6 September 1996). In contrast to the caste certificate officers, who
eschew anthropological data for more individualized inquiries and
documentation, the Commission tries to gather group level statistical and
ethnographic data and encourages the government to collect more.
Given the paucity of group-level data, compiling the data for the questionnaire
and preparing for a hearing require substantial group organization and resources
on the part of the petitioners (See Questionnaire for Consideration of Requests
for Inclusion and Complaints of Under-Inclusion in the Central List of Other
Backward Classes).22 First, the questionnaire demands data on the applicant
group as a unit, such as literacy data and percentages of the group involved in
various occupations. Second, the form includes inquiries into the history of the
group’s previous official classifications, if any, as a caste, community or tribe.
Third, the form requests that the group:
(A) Furnish the names of two castes/communities…at a level immediately

higher than the caste/community under consideration. Give reasons. (B)
Furnish the names of two designated backward castes/ communities in the
State, along with serial number in the State List, which are more or less at
the same level as the caste/community under consideration. Give reasons.
(Questionnaire for Consideration of Requests for Inclusion
and Complaints of Under-Inclusion in the Central
List of Other Backward Classes: 20)
Sounding a bit like colonial attempts to rank jatis into hierarchical lists, this step
requires that the group see itself as a unit in relation to other groups. It also
reinforces the idea that groups are not just categorized but also ranked. This
procedure has resulted in groups collecting and submitting data in an attempt to
demonstrate that they are more backward than a group already on the list. For
example, in the state of Rajastan, the “Rajput Reservation Front” leaders
compared Rajputs to Jats, who were already on the list, in terms of their
representation in political office and at various universities. They argued that
“the Rajputs are more politically and educationally backward than the Jats,” all
in an attempt to get Rajputs on the OBC list (The Statesman, “Rajastan’s creamy
layer vies for OBC status” 30 October 2001). It may be the less backward rather
than the more backward who have the resources to prove their own
backwardness and thus wage such a campaign successfully.
The listing process, in contrast to the caste certificate process, underscores that
applicants are part of a group, reinforcing the perception that engaging in politics
and interacting with the state is best done collectively. Organized and politically
powerful groups have more resources to come up with strong applications in
spite of the lack of publicly available data, such as census figures on such groups.
If a group is unified and numerous, with the potential to vote as a block, they can
pressure state or central governments to be included on the lists, making it
crucial that the state and national commissions are made up of professional
administrators rather than politicized appointees. Sheth noted that in some states
“there are some borderline cases of large communities which have acquired
political clout, but are peasant communities; and it may be disputable whether
they really are backward” (Sheth interview 6 September 1996). He also pointed
out that “communities see power in this unification for benefits.” In fact, seeing
such reinforcement of group identities has caused him to propose more
individualized criteria for Backward status:
After my experience on the commission, I think some kind of parental

social status index, or socio-occupational status index should serve as
criterion… In aggregate terms it will not be individualized; [the] same
communities will by and large get this [benefit]… But it will take out the
political possibility of OBC votes and unification and solidarities that go
with this.
(Sheth interview 6 September 1996)
In contrast to this more individualized proposal, the existing process of

constructing Other Backward Class lists encourages group actions. In pursuit of a
policy to undermine group distinctions and disparities, the procedure of listing
groups actually generates group petitions and rankings.
To summarize, exposure to two very different bureaucratic processes
associated with reservations reinforces people’s perceptions of their relationship
to the state in divergent ways. Getting an Other Backward Class certificate is a
more individualized although far from private procedure. In contrast, petitioning
to get on the Other Backward Classes list necessitates collective action to press
group claims on the state.
CONCLUSION
In India, the quite detailed attention to classification necessitated by caste
certificates and lists contributes to continuing constructions of the Scheduled
Castes, Scheduled Tribes and Other Backward Classes. The overt labeling of
beneficiaries, the reliance on “original” identities in ambiguous certificate cases,
and the emphasis on group-based claims and data in applications for lists can
harden group identities. This examination of administrative practices in India
reinforces arguments about preferential policies and target populations in the
United States, where racial and ethnic categories have remained administratively
softer:
[A] coherent system of group benefit-allocation requires more attention to

the process of classification than has hitherto been given, but this attention
may solidify the very social divisions beyond which preferential programs
were ostensibly designed to move us.
(Ford 1994:1285)
After serving on the National Backward Classes Commission, D.L. Sheth’s

recommendation to move toward an “individualized” process reflects his similar
concern about solidifying categories.
Reservation policies in India give substantial benefits to members of less
powerful groups. It is the administrative procedures associated with these
policies that at times constitute a burden. There is a trade-off at play: potential
social, economic or political benefits in exchange for regulations and procedures
that involve at times invasive identity verifications and labels. Many people and
groups are willing to go through these processes for the potential benefits, but the
procedures may be unnecessarily burdensome, particularly for the least powerful
citizens. The same degree of verification, labeling and social stigma are not
associated with policies that target more powerful groups, such as reservations
for sportsmen. In some places, the development of village computers and
interactive government web sites may be making it less burdensome to get caste
certificates; yet the digital divide means that the least powerful citizens will not
benefit from these advances any time soon (Dugger 2000, Luce 2001, Paswan
2000).23
The countless historical labels for today’s Scheduled Castes, labels that have
been used and discarded over the years, attest to the problems associated with
categorizing together several discrete communities on the basis of their poor
treatment by others rather than positive qualities of their own. Despite some
attempts to limit embarrassment or harassment of Scheduled Castes, Scheduled
Tribes and Other Backward Classes, procedures continue to haunt beneficiaries
of reservations. The necessity of identity verification through local officials and
“respectable persons of the locality,” a phrase used on a Scheduled Caste
certificate application, implies that the applicants themselves are less than
respectable and not to be trusted (Application Form for SC Certificate).24 One
striking procedure is literally labeling files of employees in reserved jobs with
their Backward status, a status which is to be reviewed at every “career upturn”
or promotion, ordinarily occasions of honor and respect. And yet, as we shall see
in later chapters, many groups are demanding such labels in spite of the obvious
stigma associated with a word like “backward.”
Official identifications tend to be based on standard and rather static

determinations of a migrant’s original place, a married person’s original caste, or
a convert’s original religion. Policy-makers might assume that such boundaries
can be empirically and objectively determined; yet, policy-implementers—the
bureaucrats actually trying to work out the details and standards and to issue
caste certificates—sometimes find they must consider each case individually.
This reinforces James Scott’s argument that “[t]he people who actually make the
assignments to these categories know the fictional and arbitrary quality
underlying each of these decisions and know that they hide a wealth of
problematic variation” (Scott 1995:35). In practice, bureaucrats often recognize
continuing ambiguity.
Moreover, the public continues to produce a plethora of petitions demanding
changes to the official lists. The lists may reinforce group-based strategies, but
official categories are often challenged, not simply accepted. Bureaucratic
attempts to codify and identify may solidify some identities but do not harden
them completely. This study of India highlights an important blindspot in the
“target population” literature: The term “policy target” does not capture the
agency of those targeted. “The potential beneficiaries of welfare measures are
not just passive ‘targets’: they think, choose and respond to policies” (The
Hindu, “Caste Count Revisited” 16 September 2000).
A comparison of two types of government petitions, an individual applying for
an OBC caste certificate and a group applying for OBC status, illustrates the
divergent effects of policy administration on people’s interactions with their
government. The caste certificate procedures, involving an individual petition
followed by collection and verification of personal rather than aggregate data,
reinforce a sense that individuals need to press their own claims—and hope that
their neighbors back them up. In contrast, the National Backward Classes
Commission’s procedures necessitate group level strategy, coordination, and
data, and they seem to reinforce, according to one Commission member,
“unification and solidarities” among Other Backward Classes and those aspiring
to become backward (Sheth interview 6 September 1996).
Administrators implement official identity parameters and procedures and can
impose bureaucratic burdens on the least powerful citizens. These powers of the
Indian bureaucracy, which is still dominated by higher status groups in Indian
society, compellingly illustrate the importance of having a representative
bureaucracy, one of the goals of job reservations (Planning Commission 1997–
2002: Table 3.6.9).25 Yetthe implementation of these potentially empowering
reservation policies has resulted in some significant procedural side effects:
burdensome labels, static classifications, and group petitions and rankings.
Administrative simplifications do influence group identity constructions, as when
individuals and groups in India must claim backwardness in order to get ahead.
5
Categorizing and counting on the census
In 2001, census enumerators asked India’s one billion citizens whether they were
members of the disadvantaged communities officially known as the Scheduled
Castes or Scheduled Tribes. To publicize the census, the Census Commissioner,
accompanied by television cameras, served as the enumerator for the President
of India. President K.R. Narayanan identified himself as a member of a
Scheduled Caste, but when he was asked to name his caste as part of the standard
verification procedure, his caste did not appear on the official list, much to his
consternation. His caste, on the SC lists in the state of his birth, Kerala, did not
appear in the lists for the Indian capital (The Hindu, “Blanks on the census form”
25 February 2001, Constable and Laxmi 2001:A12). This is just one prominent
example of the numerous challenges and controversies associated with census
classifications based on identities such as caste, tribe or religion.
Although not all castes are counted on the census, Scheduled Castes and
Tribes are. Census enumerators, to verify whether people should really be
counted as members of these categories, ask them to name their caste or tribe, so
their answers can be checked against official state-by-state schedules.1 In
addition, census enumerators record religion, which can also serve to disqualify
self-declared Scheduled Caste members from being recorded as SCs, since this
designation is not open to Muslims or Christians (Jenkins 2001a, Wright 1997,
Wyatt 1998). These census questions juxtapose personal identity and official
identification, which do not always neatly overlap, as in the case of the
President.
In spite of initial intentions to downplay caste in the postcolonial census and
ongoing definitional challenges, this limited accounting of castes persists and
may even be expanded into a comprehensive caste count in the future. After
colonial census administrators’ misguided yet meticulous attempts to rank and
record castes, the first post-independence Census Commissioner only collected
community data from the so-called “Special Groups,” including the Scheduled
Castes and Tribes, Other Backward Classes and Anglo-Indians. This decision
was motivated by the idea that past census taking had reinforced casteism,
dividing Indian society and facilitating colonial rule. Thus the 1951 census
included only those questions that were considered necessary to fulfill the
constitutional commitment to public policies such as reservations. Subsequently
the census count of Other Backward Classes stopped. However, the controversial
extension of reservations in central government jobs to the Other Backward
Classes (OBCs) in the 1990s caused some people to advocate resuming a
comprehensive count of caste so that these groups too could be counted.
Although the government ultimately rejected this proposal for the 2001 census,
the serious and ongoing consideration of a general caste question on the census is
somewhat puzzling. Given post-independence critiques of this colonial practice
by both politicians and academics, why is a caste census being seriously
considered in India? Recent proposals to resume a comprehensive caste count
not only signal a change in perspective from the independence-era optimism that
caste would fade away with modernity, but also reflect a continuity in the
process, namely, the persistent presence of caste on the census in spite of the
goal of removing it.
Postcolonial census classifications and enumerations have a mixed legacy for
identity politics in India. Census results serve as a policy tool for designing and
monitoring reservations to break down group disparities, yet, as I will
demonstrate, practices of classifying and enumerating Scheduled Castes,
Scheduled Tribes and religious communities simultaneously reinforce divisions
in several ways. Census classifications tend to maintain static identity categories
by failing to recognize some positive social changes. In addition, the state
monitors certain already oppressed groups more stringently than the general
population. Census enumerations continue to inspire some groups to prioritize
the maintenance of group numbers over the alleviation of group divisions. Also,
although the census is a potentially valuable source of quantitative data on the
social conditions of various groups, official tabulations and dissemination of data
sometimes emphasize group numbers more than their conditions.
In these ways, census classifications and enumerations reinforce official
categories; nevertheless, the boundaries remain somewhat pliable. Dramatic
jumps in the populations of certain Scheduled Tribes from one census to the next
illustrate changing identity claims in response to official designations and the
benefits associated with them. In an ironic reversal of the demands and petitions
of colonial subjects to increase their status in the census rankings, some groups
are now trying to be counted in the Scheduled categories. Those joining the
ranks of the Scheduled Tribes demonstrate that this census category is both
influential and permeable.
The relationship between the census and reservations rests on two practices:
classifying and counting. The official social categories used on the census match
those used for reservations for the Scheduled groups, although the methods of
determining who is in those categories vary. “Scheduled” identity on the census
is supposed to be self-declared to the enumerator, although answers are checked
against a state list and can also be disqualified due to religious affiliation. These
official “checks” on largely self-declared identities are somewhat akin to the
People of India project’s modified self-identification principle, discussed in
Chapter 3. In the context of adjudicating or administering reservations, the topics
of Chapters 2 and 4, Scheduled Caste or Tribe identity claims are even more
scrutinized. Census data on Scheduled Castes and Tribes factor into the reserved
quotas allocated for these groups.2 Continuing to collect social and economic
data about the scheduled categories and religious minorities can help the
government to monitor their progress and populations and adjust policies
accordingly. Thus decisions about whether and how to classify and count people
on the census can have an impact on the opportunities of disadvantaged groups.
The utility of the census data for achieving further opportunities for the
disadvantaged through reservations offsets some of the divisiveness associated
with counting caste; nevertheless some counterproductive practices might well
be reconsidered, particularly as an expanded caste-wise count is contemplated.
Analysis of modern India’s census experience can contribute to ongoing
debates worldwide about the use of identity categories on national censuses. The
United States, for example, revised the 2000 census to allow multiracial citizens
to check multiple races (Nobles 2000, Skerry 2000). South Africans after apartheid
made a rather heart wrenching decision to retain racial categories in the census,
with the new goal of monitoring progress toward equality.3 Britain recently
introduced a rather controversial ethnicity count (Dale and Holdsworth 1995,
White 1999), and the new censuses of the former Soviet Republics are sparking
controversies about national, ethnic, and religious distinctions (Kertzer and Arel
2002).
Given the lengthy history of census taking in India, Indian administrators,
academics and politicians have long been aware of the difficulties and downsides
of identity classifications. The following discussion of some divisive tendencies
of the Indian census is not meant as a criticism of census administrators, who,
although coding a plethora of answers into standardized matrices, articulated
some of the most nuanced discussions of India’s complex diversity I encountered
in interviews on this subject. Indeed the divisive dynamics of census categories
occur at multiple stages in the census process, from the design of categories and
questions, to data collection, tabulation, dissemination and politicization. Thus
while group reinforcement associated with the census is in part due to
administrative procedures, it is also due to political pressures, to media coverage,
or simply to the imperatives of any large scale accounting of a complex reality.
I first show that the arguments of post-independence politicians foreshadowed
current academic critiques that colonial census takers emphasized, simplified,
and reified social categories such as caste, tribe and religion. Despite these
longstanding concerns, such categories never did disappear from the census.
Current debates in India about resuming a comprehensive caste count in the
census illustrate not a new but a renewed emphasis on caste and community, in
large part due to growing administrative and political concerns about
reservations. I analyze some ongoing practices associated with classification and
enumeration and argue that these tend to reinforce group divisions despite the
goal of breaking them down. Finally, I turn to some recent population jumps due
to identity shifts. These shifts into the Scheduled Tribe category became
politicized due to the reservations at stake. In comparison with the court cases,
caste lists and certificates, and even the People of India project, discussed in
earlier chapters, census categories are more porous. They rely upon self-
identification, although the enumerator does a little checking as in the example
of the President and the 2001 census. I conclude that divisive administrative
practices and political dynamics have haunted group-based census categories
from colonial times to the present, with renewed vigor in recent years, yet some
people still manage to complicate such categories, if not always overcome them.
CRITICS OF THE COLONIAL CENSUS

The colonial census administrators were hardly the first to record data about
groups on the subcontinent, although their legacy has the most direct influence
on today’s census administrators. Historian Sumit Guha compellingly argues that
precolonial record-keeping and processes of identity definition are too often
ignored in historical work on the Indian census, and he discusses the
classifications and enumerations by the Mughals and the Marathas that preceded
the British and Indian censuses (Guha 2001). Before the British, Mughal writers
and rulers commented on group characteristics and tried to standardize skin color
classifications. Mughal surveys, such as the Ain-i-Akbari, were part of the
inspiration for the first Indian gazetteers in English, in the late eighteenth century.
In addition, the early colonial census administrators drew on revenue collection
records of earlier rulers, such as the Nayakas (Bayly 1999:103–5, 108, Barrier
1981).
The first colonial census of India was initially planned to coincide with the
1861 census in Britain, but in 1857 Indian soldiers in the colonial army revolted,
resulting in a change of plan. The rebellion both delayed the initial colonial
census and renewed the motivation of the British to gain more knowledge about
the population of the land that officially became their colony in the rebellion’s
aftermath.4 After several regional Indian censuses, a census of India with a
common schedule was carried out in 1871–72. India has had an unbroken series
of decennial censuses thereafter, although not all regions have been covered each
time. Attempts to record the variety of communities and castes in the census
contributed to caste-based mobilizations and associations, often with the aim of
increasing official status, particularly after Herbert Risley attempted a
hierarchical ordering in the 1901 census (Rudolph and Rudolph 1967: 116–19,
Charsley 1996). The 1931 census commissioner J.H. Hutton, author of Caste in
India, also tried to develop caste indices, but the “hyperpoliticization” of caste,
as well as the need for an abbreviated census during Second World War, meant
that comprehensive caste counts ceased after 1931 (Dirks 2001:221, 226, Hutton
1946).
The social and political side effects of the colonial census inspired much
criticism from Indian leaders in the early years of independence as well as from
several historians and other scholars in more recent years. Indian leaders at
independence, such as Jawaharlal Nehru, wanted to put certain colonial census
practices behind them, particularly caste tabulations (Bose interviews 19 July
1996 and 4 November 1996). Some of their comments foreshadowed current
academic criticisms; however, like many of their contemporaries, these early
postcolonial leaders spoke in what now seem to be idealized terms about the
stasis of tradition and the dynamism of the modern era they hoped to enter.5
Sardar Vallabhbhai Patel, Deputy Prime Minister, spoke in 1950 at a
conference about the census to be carried out the following year, contrasting it
with the colonial emphasis on India’s divisions:
The forthcoming census is the first census of a Free Republican India.

Formerly there used to be elaborate caste tables which were required in India
partly to satisfy the theory that it was a caste ridden country and partly to
meet the needs of administrative measures dependent of [sic] caste
divisions. In the forthcoming census this will no longer be a prominent
feature and we can devote our energies and attention to the collection and
formulation of basic economic data…of the individual and the state.
(Quoted in Natarajan 1972:266)
The shift in emphasis from groups to individuals and from caste classifications to
economic data signaled the optimism at independence about “modernization”
and the fading away of caste. At a 1959 census conference, Home Minister
Govind Pant echoed these ideologies of modernity: “In the olden days the
conditions were static… You were concerned mostly with matters pertaining to
caste, religion and so on, but now times have changed. We are on the move and
our society has become…dynamic” (quoted in Natarajan 1972:267). These
leaders argued that “traditional” identities like caste and religion were static,
particularly as portrayed in the colonial census, and they committed themselves
to new and progressive agendas for the postcolonial census operations.
These critics of the colonial census, anticipating the speedy demise of caste,
would be surprised at the current debates over resuming a caste count.
Years later, colonial census taking became the subject of much academic
scrutiny, which echoed these earlier concerns about the imposed divisiveness of
census categories and the opportunities for colonial manipulations. Inspiring
further research, Benedict Anderson’s seminal work, Imagined Communities,

notes census takers’ “intolerance of multiple, politically ‘transvestite,’ blurred, or
changing identifications. Hence the weird subcategory, under each racial group,
of ‘Others’—who, nonetheless, are absolutely not to be confused with other
‘Others’” (Anderson 1991:166). The practices of census taking slot novel or
ambiguous answers into standardized categories. India’s multifaceted diversity
makes it an extreme case with regard to this problem. Historian Thomas R.
Metcalf describes the enumeration of castes for the colonial census as a “project
of formidable difficulty… Constant efforts had to be made to reduce the
bewildering array of caste names returned by individuals to a consistent order,
and to fit all enumerated individuals properly into the assigned categories”
(Metcalf 1995:121). Personal identities and official identifications frequently did
not mesh neatly.
Moreover, the oversimplification of complex identities on the census did not
just record social categories but also had political and social effects on them.
While acknowledging that disputes over caste rankings preceded the British
census, historian Frank Conlon notes:
the enduring interest of the British in caste as a system which both divided
and ranked their Indian subjects, produced an extensive response among
those subjects, and also sometimes created new categories by statistical
sleight of hand or administrative fiat.
(Conlon 1981:104)
Anthropologist Arjun Appadurai argues in even stronger terms about the far-
reaching effects of the census on social and political life:
The Indian census, rather than being a passive instrument of data-gathering,

creates, by its practical logic and form, a new sense of category-identity in
India, which in turn creates the conditions for new strategies of mobility,
status politics, and electoral struggle.
(Appadurai 1993:316)
Appadurai’s description makes a census seem ill suited to be a mechanism for

social justice, which was the justification for continuing the caste count after
Independence. The social disputes and political mobilizations associated with
caste classifications are no surprise to those who argue that the colonial census
“served the cause of both science and imperialism” (Bose 1991:20) and was “an
important early apparatus of colonial rule” (Dirks 1997:209). Census taking
facilitated colonial goals, ranging from revenue collection to law and order.
Understanding and even exacerbating internal divisions could smooth the way
for continued rule.
These political leaders and scholars have severely criticized the colonial
census for reinforcing or politicizing distinctions on the basis of caste, tribe and
religion. Has the postcolonial Indian census avoided such divisive tendencies?
Can a census used to design and monitor policies to overcome community-based
disadvantages—in contrast to a colonial census used to gather revenue or
maintain order and control—avoid the pitfalls of reinforcing divisions? These are
the questions raised by recent proposals to resume a comprehensive caste count.
A RENEWED EMPHASIS ON CASTE

Recognizing the problems associated with the colonial state’s classifications,
policy-makers after independence made one attempt to count all of the backward
classes but then discontinued counting castes in the census except for the
Scheduled Castes and Tribes. Prime Minister V.P.Singh in 1990 announced his
controversial intention to extend central government reservations to the Other
Backward Classes. This decision contributed to the fall of his government,
sparked violent protests, and faced legal challenges (Dirks 2001, Engineer 1991,
Mandal Commission 1980, Parikh 1998). After the Supreme Court’s eventual
approval of these OBC reservations, with some limitations, the administrative
challenges of implementing this policy came into focus (Justice R.N.Prasad
interview 18 September 1996). Indira Sawhney v. Union of India (Supreme Court
of India 1993).6 Many argue that census enumerators should count Other
Backward Classes or even return to the practice of counting all castes (Singh
interview 20 November 1996, Sheth interview 6 September 1996). Otherwise,
state and national backward class commissions must continue to rely on their
own studies or extrapolations from 1931 census data to estimate populations of
the Other Backward Classes.7
A proposal to resume a caste count on the 2001 census received impetus from,
among others, the Ministry of Social Justice and Empowerment, which could use
the data in its efforts to monitor the empowerment of the disadvantaged; the
National Backward Classes Commission, to aid its attempts to identify the OBCs;
the Kerala High Court, which issued a direction to the central government to
consider a caste-wise count for 2001; and various political organizations
(Vijayanunni 1999, Shah 1998).8 For example, the National Backward Classes
Commission “recommended that that is the most logical thing to do. Just by
counting caste, you don’t support caste,” according to D.L.Sheth, a sociologist
who served on the commission. On the dearth of OBC data he lamented, “It
becomes very difficult to even judge whether they are adequately represented
in services, because we don’t have data, castewise data” (Sheth interview 6
September 1996). After initial support, the government eventually “decided
against the idea of a caste-wise Census, 2001” (Registrar General and Census
Commissioner, M.Vijayanunni quoted in Narayan 1999:1). The proposal, still a

possibility for a future census, drew mixed responses.
Registrar General Vijayanunni initially defended the idea of a caste-wise
census. His discussion highlighted the political context of this proposal, as the
increasing politicization of caste confounded the hopes of an earlier era that caste
would make a graceful exit:
After the post-Independence euphoria, when it was vainly thought that

caste had been cast off, the course of democratic India has seen the
resurgence of caste in the polity through the decades, with caste awareness
reaching a high in the post-Mandal 1990s. The first census since then, in
2001 is the right time to reintroduce the caste question in the census
questionnaire and fill the glaring sociological data gap.
(Vijayanunni 1999)
The increasing political and administrative importance of caste, the backward

castes in particular, set the context for this proposal (Chandra 2000, Jaffrelot
2000, Kothari 1994). The Registrar General both addressed and downplayed
fears that a caste census would result in massive protests, worse than the tragic
anti-Mandal demonstrations: “[M]ere collection of caste data can entail no such
opposition” (Vijayanunni 1999).
More attuned to parallels between this “mere” data collection and the oft-
critiqued colonial census, social scientist Ghanshyam Shah argued, “The
proposed census of caste and collection of information under the categories
created by experts, as done by colonial rulers in the past, would hardly enhance
our understanding of caste” (Shah 1998). He was particularly wary of official
classifications: “Once such categories are created, they will in course of time
emerge as ‘facts’… Do we learn from history or repeat the blunders of the
colonial state?” (Shah 1998).9 At a symposium of government officials and
policy-makers in Delhi on the issue of caste and the 2001 census, the most
widely cited argument for not counting caste was that it would “intensify divisive
caste identities,” and many felt that counting caste would revive the “divide and
rule” dynamics of British caste enumerations (Deshpande and Sundar 1998).
Others reflected on the differences between colonial and postcolonial census
operations in their arguments for resuming a caste-wise census. Former Prime
Minister V.P.Singh supported the idea of counting Other Backward Classes on
the census: “The census should take steps to know the quantitative dimensions of
the issue” (Singh interview 20 November 1996). Singh downplayed the dangers
of official group distinctions by contrasting contemporary classifications and
British policies: “There is a qualitative difference. British never intended sharing
power with the depressed classes. Here the aim is to share power” (V.P.Singh
interview 20 November 1996). As an enlarged caste count is contemplated, we
must consider whether changed motives are enough to change the outcomes of
administrative practices. Has the post colonial Indian state really steered clear of
the divisive tendencies of the colonial census, or has contemporary caste-based
data collection perpetuated some of the same problems as colonial census
classifications?
Given the persistence of caste in the postcolonial census, to some the proposed
expansion of the caste question seems to be a logical expansion of past practices.
Now is a crucial time to take a closer look at those practices. The following
discussion illustrates the continuing attention to caste in the postcolonial census
and, in particular, the continuing divisiveness associated with classifications and
enumerations.
CLASSIFYING: A CASTE BY ANY OTHER NAME¼

One objective of group data collection, according to some government accounts,
is to help “destroy” group discrimination associated with the caste system. A
national Backward Classes commission, set up after independence, concluded in
its report:
Before the disease of caste is destroyed all facts about it have to be noted
and classified in a scientific manner as in a clinical record… The Census
officers must have permanent ethnologists and sociologists… As long as
social welfare and relief have to be administered through castes, classes or
groups, full information about these groups should be obtained and
tabulated.
(Kalelkar Commission 1953:159)
The commission emphasized the importance of data collected and organized

along group lines. Although many of the commission’s recommendations were
largely discarded, they reflect the faith, at the time, in the possibility of a
systematic and scientific destruction of these group distinctions through group-
based policies. The Minister of Home Affairs worried about separatism due to
the report’s emphasis on caste yet requested state government surveys of
numbers of Backward Classes (Dirks 2001:226). Moreover, Scheduled groups
continue to be counted and studied in the national census. Thus, in spite of the
desire to downplay caste on the census, expressed above by census officials,
substantial inquiries into caste, as well as related census practices, persist.
Examination of postcolonial census documents demonstrates that, ironically,
in pursuit of the goal of social change through reservation schemes, census
classification practices reinforce static identity categories and monitor the
“special” groups much more than the general population. A practice going back
to the early British colonial census persists as the census enumerators consult
lists of castes to standardize the thousands of different responses they received.

Historically, enumerators had consulted lists of terms that should and should not
be used. The administrative volume of the census of Bengal in 1901 included a
“list of vague and indefinite entries found in the census returns of 1891 that
should be carefully excluded from colom [sic] 8 (caste) of the census schedules,
except in the special cases where the term is said to indicate a true caste” (quoted
in Cohn 1996:224).
Up to the present day, enumerators continue to “correct” the responses of
those they are counting. (Even a response from the President of India!) Only
people who claim to be of castes that are included in the Scheduled Castes lists
are counted as true Scheduled Castes. These lists can only be amended by an act
of Parliament; thus the lists may be more static than the populations they
represent.10 The continuing necessity of these relatively static caste lists results in
the government’s failure to fully recognize shifting identities due to a concern
with tabulating “true” castes.
For example, partly because of the impact of reservation policy “schedules”
and partly because of social movements among the lower castes, some people
not only identify with their own caste or jati (out of the thousands of
geographically scattered and socially distinct jatis) but also sense a larger
identification with the lowest, “untouchable,” castes in general (Omvedt 1994).
Some claim to be “Harijans” (meaning “children of God,” Mohandas Gandhi’s
name for untouchables) or “Dalits” (meaning the “oppressed,” a term associated
with Dalit leader and father of the Indian constitution, B.Ambedkar) (Zelliot
1996, Gottschalk 2001). Many are even rallying around the government’s
category of Scheduled Castes by forming scheduled caste associations (Jalali
1993). This tendency toward embracing a pan-Dalit identity is a potentially
empowering social trend. The census may be encouraging this trend by
continuing to focus attention on Scheduled Castes and Scheduled Tribes as
broader categories. Yet enumerators must ascertain someone’s specific group
within the larger Scheduled categories to record that person.
The enumerator’s instructions for the 1961 census illustrate an attempt to put
the increasing number of people claiming broader identity categories back into
smaller caste categories:
Do not write the names of the Scheduled Castes in general terms as

“Harijan,” “Achhut.” You should ascertain the name of the caste when it is
returned and write it. If a person is negligent and insists on calling himself
merely “Harijan,” tell him that this description will not earn the person any
benefits under the Constitution permissible to Scheduled Castes and
Scheduled Tribes. This may persuade him to give out the correct name.
(Mitra 1961)
This policy discredits unity among lower castes by reinforcing separate caste
names as “correct.” The adoption of composite names, in contrast, is portrayed
as “negligence.” Thus the postcolonial census procedures continue to ignore some
changing identities and reinforce some divisions, particularly among
disadvantaged citizens.11
Not everyone’s caste was recorded for the census. In practice, however,
everyone had to be asked a caste question. For the first post-colonial census in
1951:
[T]he Government of India had already accepted the policy of official

discouragement of community distinctions based on caste. They decided
therefore, that no general Race, Caste or Tribe enquiries should be made
but that an enquiry should be made regarding Race, Caste or Tribe only to
the extent necessary for providing information relating to certain special
groups of people who are referred to in the constitution of India… The
relevant Census questions were re-framed so as to enquire, in relation to
every citizen, whether or not he was a member of a “Special Group.”
(Gopalaswami 1953b:1)
The necessity of looking at a list of caste names to determine whether an

individual was “special” or not meant that the census questions had to be much
more specific in practice. A 1951 census report addressed some common
difficulties with slotting citizens into identity categories. After a notation that for
the purposes of question 14 (sex) “Eunuchs and Hermaphrodites should be
treated as MALES” (emphasis in original), ambiguities pertaining to “special
group” status were discussed. The report noted, “In actual practice, they had to
ask every person his caste and it was only at the time of recording the answer
that they had to examine the list…to decide on the answer to be written”
(Gopalaswami 1951 vol. VII:415).12
In spite of the attempt to minimize the caste question, all respondents had to
be asked, reminding all of the continuing official importance of such
distinctions. Yet only lower caste members’ specific castes were written and
kept in the records of the government. By the next census, in 1961, the
government wanted more detailed tabulations on each Scheduled Caste and
Scheduled Tribe, including industrial classification, age, marital status,
education, religion, land, and mother tongue (Mitra 1961).13 Detailed tabulations
can be useful tools, but, as discussed below, the data on the disadvantaged are only
useful to them if their social conditions get as much attention as their fertility
rates.
Census-sponsored ethnographic studies also returned with the 1961 census
(Padmanabha 1978). That census included 400 village survey monographs, with
descriptive reports on caste. These postcolonial reports were reminiscent of the
ethnographic appendices about castes that often accompanied the colonial census,
as in the caste report by Herbert Risely for the 1901 census, which became the
basis for his People of India, the subject of Chapter 3 (Risley 1915). Colonial
ethnographers had an interest in painting populations as “barbaric” to attempt to
justify colonial rule, or in identifying “criminal tribes” to monitor or “martial
races” to employ (Cohn 1996, Des Chene 1999, Dirks 1997). In light of this
history, the 1971 census’ “ethnographic notes” contain some startling
descriptions of certain castes as “inveterate criminals” (Bayly 1999: 275). If a
caste count is carried out, socio-economic data could make this exercise
worthwhile, but analyses of various groups’ so-called criminal proclivities is not
the sort of supplementary data likely to help them advance. In India, census
classifications raise recurring and troubling issues, particularly the tendency to
retain static, subdivided categories and to impose more on the privacy of the
disadvantaged through additional state scrutiny and record keeping.
COUNTING: THE POWER OF NUMBERS
I remember an old village officer in Madras putting the first objective to

me as that we must catch every man. Just so.
(M.W.M.M.Yeatts, Census Commissioner for the 1941 census, quoted in
Natarajan 1972:144)
Can “catching” each person in the census be liberating? For disadvantaged

groups, the numbers tabulated from the census can aid in monitoring and
enforcing reservations; yet because these numbers are so used, the politics of
reservations has added new complications and controversies to the census
enumeration. For example, although their census answers do not “qualify” them
for reservations, many people associate the census with reservations, since they
are both government projects and use the same administrative categories.
Although the eligibility of individuals is not affected, their answers can cause
shifts in the relative percentages of Scheduled Castes or Tribes in the population,
which could affect the future allocation of reserved legislative seats, government
jobs and university admissions.
In spite of their utility for reservations, census numbers can, at the same time,
reinforce the divisions between existing social categories in two ways. First, the
association between population numbers and power, due to democratic electoral
politics as well as the quotas associated with reservations, means that people may
become more concerned about counting and maintaining group numbers than
about doing away with group distinctions. Second, the release and tabulation of
data sometimes emphasize group numbers rather than their conditions,
diminishing the value of the census data as a tool for policy refinement or social
empowerment.
Concern over population numbers has caused some people to retain categories
they might otherwise wish to discard. The demand to count the Other Backward
Classes is a case in point. One census scholar, in a striking reversal of
independence era statements on the modernity of moving away from a caste
count, argues:
If the modern census is conceived as a social document of a nation, it

would be imperative then to include caste questions (relating, of course, to
those which are identified as backward) in the subsequent census
questionnaires to throw light on their educational and economic status.
(Mohanty and Momin 1996:167)
Although the census critics discussed earlier suggest that the colonial counts, in
particular, had dramatic effects on the social order, subsequent efforts to
downplay caste on the census certainly did not make castes disappear in society.
The necessity of using archaic 1931 census data for current policies for Other
Backward Classes further complicates any claims about the “modernity” of
avoiding a caste-wise count and forces some who might otherwise eschew such a
count to advocate it simply to gather the numbers needed to administer existing
policies.
The power of numbers also sways groups to embrace rather than disavow their
backward status. Some even try to come up with their own population figures.
Census administrators suggested that the logistics and politics—of coordinating
comprehensive, official lists of castes and then counting them may be too much
for an already huge census operation. Such concerns may have factored into the
decision not to expand the caste count for 2001. One census administrator, while
acknowledging the utility of Other Backward Class numbers for government
planning, pointed out the challenge: “The demand may be genuine, but it is a
very…voluminous exercise” (Chakravorty interview 16 December 1996).
Undeterred, some states and, more tellingly, caste associations are trying to fill
the gap by sponsoring their own, unofficial Other Backward Class censuses
(Hasan 2000:173, n.16). In the words of sociologist A.M.Shah:
Democracy, it is said, [is] a game of counting heads… Since the

introduction of adult franchise, every caste is counting its heads… Many
castes now conduct their own censuses and present their population figures
to the government. This has forced the Backward Class commissions in
several states to conduct special censuses… The forty year old policy to
exclude caste from the census is thus slowly getting reversed.
(Shah 1989:13)
The censuses sponsored by the backward classes themselves are of dubious

impartiality, but their efforts show the perceived importance of maintaining and
recording group numbers and relative percentages in society, rather than
eschewing the backward label.
A second troubling dynamic associated with the official headcount is a
tendency for the media, politicians, and even official tabulations to emphasize
“minority” population numbers. Historian Robert E.Frykenberg traces the
dangerous development of the idea of a so-called permanent “majority
community” in Indian politics (Frykenberg 1987). A corollary development has
been the tendency to scrutinize the relative numbers in the various “minority”
communities, further shoring up these distinctions. When census statistics are
released, numerous newspaper articles appear expressing concern over the rate
of growth of the Scheduled Caste and the Muslim populations in particular.
These numbers often overshadow data on their social conditions.
This reference to numbers and to majority or minority status is exemplified by
the most controversial of all census figures—the relative Muslim and Hindu rates
of population growth. The religion question on the census could be useful for re-
evaluating reservation policies by showing whether Muslims (some already
receiving benefits as Other Backward Classes and others currently demanding
that right) are advancing in the educational or economic spheres (Jenkins 2001a).
Yet official census tabulations and media coverage have often stressed the
relative numbers over the relative conditions of disadvantaged groups.
The population growth rates of different religious groups, as well as the
scheduled groups, are published widely, and one of the most common refrains
from the politicians of the Hindu right is the rapid growth of religious minority
populations.14 English language newspaper reports on census results, accessible
by a largely elite audience, are particularly prone to emphasizing population
growth rather than other social or economic data about disadvantaged groups.
Under the headline, “Census shows massive influx of Muslims in N-E, Delhi,”
Shivaji Sarkar notes, “The Home Ministry officials, meanwhile, have taken steps
to accelerate the pace of fencing around the Bangladesh border” (Financial
Express-New Delhi 17 July 1995). In another account:
According to the latest report published by the Census Commissioner of

India, the population of Muslims in the country has shown an overall
growth of 32.76 per cent during the decade between 1981–91 which is
higher than the growth of Hindus…. According to another report published
recently, the population of Scheduled Castes in Delhi has been increasing
at a faster rate than the general population.
(Sahay 1995)
Inadvertently highlighting the failure of the census reports to shed light on

possible explanations for these figures, this newspaper article concludes, “Now
whatever be the actual causes of these higher growth rates of Muslims and SCs,
the phenomenon is sure to cause apprehension and suspicion in the minds of
other communities or castes and would certainly harm the country’s family
planning program.” The author then argues that Hindus, particularly upper castes,
would “shun” family planning in response to the census report (Sahay 1995).
This may well be the case if they read the following account by P.S. Sharma,
entitled “Danger Signal for Hindus.”
But Muslims have always been multiplying at a faster rate than the
Hindus, as evidenced by the censuses held from 1891 to 1991. The result is
that for the last over 100 years, the communal composition of the country
has been changing census after census in favour of Muslims… The rise of
Muslim population at a faster pace than the Hindu population strengthened
the Muslim demand for Pakistan…even in truncated India after Partition,
Muslim population is growing at a faster rate than the Hindu population. If
other things remain the same, this will in course of time ensure that Hindus
are reduced to a minority and India becomes an Islamic state.
(Free Press Journal 13 September 1995)
Abdul Malik Mujahid discusses news accounts that actually tried to project when
Hindus would become a minority in the state of Tamil Nadu (Mujahid 1989:93).
These reports occurred in the wake of mass conversions of untouchables to Islam
in that state in the early 1980s. These are extreme examples of the competitive
politics of numbers, which can be inspired by census data interpreted without
context or cross-tabulation.
Clearly, relative populations and growth rates of different groups are widely
circulated; yet, the Indian state postpones or simply does not tabulate data for
certain other illuminating tables, such as the Muslim literacy rate in comparison
with the Hindu literacy rate, on the grounds that these numbers are “sensitive”
information. “The census tables do not present such data though the questions on
religion and literacy were asked about every individual,” notes census scholar
Ashish Bose (quoted in Mohanty and Momin 1996:6, Bose interviews 19 July
1996 and 4 November 1996). Notably, another arguably more sensitive table,
on differential fertility by religion, has been included in recent census
publications. The possible relationship between fertility rates and illiteracy rates
remains hidden, and the misguided assumptions linking high fertility and Muslim
“culture,” particularly polygamy, pervade public discussion.15
In a published interview, the Registrar General of India, M.Vijayanunni, was
asked about the politicization of census figures and whether it is “warranted to
talk about the Hindu rate of growth or the Muslim rate of growth given the fact
that religious communities in India are so amorphous?” He responded, “The point

however is that Muslims have always registered a higher growth rate.” A follow
up question and answer reveal the Indian census reports’ recent emphasis on
population figures to the detriment of social indicators: “Since illiteracy and
poverty have always been reflected in high birth rates, could the high growth
rates indicate the social status of minority communities in India?” The Registrar
General responded, “This is merely a hypothesis. Unfortunately, we have no data
to corroborate this” (Indian Express 17 December 1995). Thus, the potential for
group-based data to combat intergroup inequities remains partially unfulfilled.
Part of the power of numbers lies in their publication and interpretation:
Which numbers are publicized? Which are compared? Like the data on Muslims,
data on the population growth of Scheduled Castes and Scheduled Tribes have
been released, while more detailed data on their social conditions have been
delayed. Former Deputy Registrar General of the census, B.K.Roy Burman notes
that “[f]or the 1981 Census, no all India Volume, giving data by individual
scheduled castes and tribes, was published. I am not sure that all the
contemplated tables in respect of individual SC/ST will be available before the
advent of the 21st century” (George 1999).16 Even if the government has not
declared this data “sensitive” and the root of the problem is the sheer volume of
data to be processed, the delayed release of the data weakens the justification for
asking about caste. In 2001, enumerators counted Scheduled Castes and Tribes
but did not collect income data, making quantitative economic analyses of the
Scheduled groups in comparison with the rest of the population difficult.17 In
short, census data about the social and economic conditions of groups could be
used to combat divisions, whereas their relative numbers can be used to provoke
fear and reinforce divisions.
SHIFTING IDENTITIES
I counted nine I had no right to count (But this was dreamy unofficial
counting).
(Robert Frost from “The Census Taker” 1995:165)
In the previous examples, the practices of official classification or the emphasis

on group numbers resulted in the reinforcement of the government’s categories.
In the following example, census categories were adopted, yet simultaneously
challenged, due to a trend demographer Ashish Bose calls “retribalization” (Bose
interview 4 November 1996). Retribalization refers to the dramatic leaps in
census figures for certain scheduled communities. In spite of all the dynamics
and mechanisms of the census that tend to reinforce existing classifications, the
reliance on self-declared identities means that there are instances in which people
still manage to cross state-drawn boundaries.
In 1981, census takers recorded a sudden jump in the number of people
claiming Scheduled Tribe status in the state of Maharashtra. This population
explosion could not be attributed to births alone, but rather was due in part to
people deciding to change categories.18 In the case of the Halba/Halbi tribe, the
1971 population of 7205 rose to 242,819 in 1981, in part due to members of a
Koshti sub-caste claiming to belong to this tribe. The fact that at one time the
Maharashtrian government had recognized this sub-caste as tribal members
demonstrates the unclear boundaries between tribals and non-tribals and makes
the case more ambiguous than conclusions about “fraudulent responses” imply
(Kulkarni 1991).
At the same time, reservations are one possible incentive for what historian
Sumit Guha has described as the “infiltration” of the Scheduled Tribes in
Maharashtra, whose population grew by “an implausible 50 per cent in 1971–81”
(Guha 1999:193 citing Gaikwad 1986). This could be seen as “ironically enough,
another indicator of progress,” notes Guha (1999:193). The voluntary movement
of individuals and groups into stigmatized categories arguably breaks down
boundaries, although administering group-based policies aimed at Scheduled
Tribes certainly becomes more complex, and their impact may be diluted. The
improbable enumerations of Scheduled Tribes in Maharashtra demonstrate that
in spite of the seemingly obsessive tendencies of governments to classify, people
occasionally throw a wrench into the administrative machinery of the census.
“The policing of these increasingly artificial and highly permeable social
boundaries by any state, however omniscient, will almost certainly become
increasingly difficult in the new millennium” (Guha 1999:198). Enumerators
have some authority to “check” census answers to make sure a group is on the
schedule but cannot verify whether individuals are “really” in the groups. As
discussed in previous chapters, administrative “scrutiny committees” and judicial
efforts to expose “spurious” groups or individuals, cast suspicion on voluntary
identity shifts (an unfortunate outcome), but even these more stringent processes
cannot eliminate such shifts (Jenkins 2001b). The colonial and postcolonial
governments’ role in structuring identities certainly can be overstated. Surges in
census populations show that the government is not able to stand guard at the
boundaries of official categories. Yet the fact that a group is trying to become a
Scheduled Tribe shows the government’s indirect influence on identity claims
through the construction of a particular menu of categories and a related
opportunity structure.
The persistence of shifting identities shows that census policies associated
with reservations may reinforce certain categories but cannot contain society
neatly within them. The reservations-related uses of census data about different
social categories tend to inspire suspicion of shifting identity claims rather than
relief over the flow of people in and out of the disadvantaged categories.
Nevertheless, these shifts in claimed identities show that people are not
imprisoned by the boundaries of official categories, although they are certainly
influenced by them.
CONCLUSION
Dilemmas over categorizing and counting various identity-based groups on the
census continue to plague the Indian government as well as many governments
worldwide. The census has had a particularly close relationship to identity
politics in Indian history. Census data on religion helped draw the boundaries
between India and Pakistan at Partition; census data on “mother tongue” aided in
the reorganization of states starting in 1956; census data on castes and tribes
figured into the percentage of reserved government jobs, university admissions
and legislative seats allocated to Scheduled Castes and Tribes.
Whether identity-based data collection exacerbates or quells group divisions
and tensions is an important question for culturally diverse countries. This
analysis of the recent history of the categories most useful for contemporary
reservation policies in India—caste, tribe and religion—demonstrates that some
aspects of classifying and counting are counterproductive. Intentions changed
dramatically at independence, but the postcolonial census has not made a clean
break from colonial practices; on the contrary, the growing push for a
comprehensive caste tabulation inspires a sense of deja vu. Data on castes and
religious minorities could be a political tool for the disadvantaged, but problems
persist in the census, including the government’s need for strict classifications
despite the goal of more social fluidity, the more intrusive scrutiny of certain
minorities, the continuing competition over group numbers, and the limited
dissemination of certain “sensitive” data. If caste and religious data could be
used more effectively to alleviate inequities and improve reservations and other
policies for the disadvantaged, this would help make up for the inherent
problems of official identity-based census categories. As it stands the politics of
the census and, in particular, the categories relevant to reservations are in danger
of doing more to reinforce divisions than to provide tools to fight against them.
Regarding the tensions between ambiguous groups and official categories,
reservations have raised the stakes attached to certain identity claims, motivating
some people to challenge their categorizations. Often, however, people have
pushed the boundaries without ultimately under-mining the state classification
schemes: Other Backward Classes want to be included in the count; some
members of various Scheduled Castes are identifying with the larger SC
category; the people who try to switch into the Scheduled Tribe category are
embracing the category itself, even while perforating its boundaries. Benedict
Anderson says of the colonial census: “It tried carefully to count the objects of
its feverish imagining” (Anderson 1991:169). Now groups are re-imagining their
own identities in part based on their identification in the census.
Part II
Political complications
6
ªBackwardº Muslims and ªScheduled
Casteº Christians
Society does not simply allow itself to be boxed up by the state. Social groups
challenge state categories, both reacting to and exploiting the inherent
complexity of overlapping identities. Thus we cannot stop at simply vilifying
official categories. State classifications do not entirely determine social fault
lines, and these classifications may be used to the advantage of subordinate
groups.
Although critically examining structures imposed by governments remains
important, recognizing the agency of members of society, their ability to act and
even challenge such structures, is a crucial part of any analysis (Giddens 1977).
Political scientists Lloyd and Susanne Rudolph, based on their work in India,
were among the first to point out the limits of colonial categories in their 1967
classic, the Modernity of Tradition. Although their work foreshadowed some
criticisms of imperial or governmental categories by scholars such as Edward
Said and Michel Foucault, they later noted that a Saidian or Foucaultian
approach can be taken too far: “We would challenge the claim that always and
necessarily power dominates knowledge and that knowledge serves domination”
(Rudolph and Rudolph 1967, 1996:5, 6, 8–9, Said 1978, Foucault 1979).
Rejecting oversimplified categories, especially those imposed or re-inforced
by Western social science or imperialism—race and caste being two prominent
examples—is an appealing idea. Yet such blanket generalizations can be
misguided. For example, consider sociologist Pierre Van den Berghe’s sweeping
conclusion about race:
In practice, social race is always a social stigma for the subordinate group,
and all attempts to pretend otherwise have been singularly unsuccessful.
Pragmatically, in terms of policy, it means that institutionalization of racial
categories, however innocuous or even benevolent it may appear, is
frequently noxious in its consequences. I am thinking of such measures as
racial questions on the census, race-based affirmative action and similar
measures, which have generally had the effect of reinforcing stigmatized

racial distinctions.
(Van den Berghe 1995:364–5)
In drawing this conclusion, he ignores any agency of the “subordinate group” in

the matter. As Adrienne Davis points out, “Racial categories are never solely the
possession of the dominant culture; they are internalized and, in part, created by
subordinate groups” (Davis 1996:720). In short, race, caste and other social
categories have at times been taken up by various subordinated groups as tools
of empowerment.
Moreover, by denying any legitimate use of such categories, one might be
challenging the categories but leaving the power structure intact. As Adrienne
Davis further explains, “[N]ot all statements that are counter-categorical are
likewise counter-hegemonic. A statement may expose the hierarchy but do
nothing to challenge or reform it. Some counter-categorical actions may even
reinforce the hierarchy” (Davis 1996:719). A blanket rejection of any
consideration of caste or race in public policy, based on the commonly held
assumption that consideration of such categories will inevitably reinforce them,
is an example of such an action. Disadvantaged groups may want to retain
categories—even those that have been used to oppress them—for their own
strategic purposes (Spivak 1987).1
In this and the following chapters, I examine protest groups that are using, but
also reconstructing, the government classifications discussed in the previous
chapters. In other words, although groups adopt or internalize some of the
categories used by the state in their efforts, they are also attempting to recreate
these categories by disrupting the official boundaries of the Scheduled Castes,
Scheduled Tribes and Other Backward Classes. The complexly interwoven
identities characteristic of India provide much fuel for such challenges, drawing
on all manner of social groupings and various combinations thereof, including
caste, religion, class and gender.2
This chapter demonstrates that Muslim and Christian demands to be included
in the officially backward categories have sparked dissent within these minority
communities in addition to the more publicized criticism from various Hindu
organizations. The resulting controversies over who should be allowed to
become backward illustrate ongoing political constructions of religious, caste,
and national identities. The growing number of citizens in India demanding to be
declared officially backward, an ironic outcome of reservation policies, is an
example of official categories influencing political activism.3 People’s assertions
of unofficial identities, in turn, challenge the official identification of citizens.
Based on case studies of two religious minority movements and their aspirations
for reservations, I argue that India’s cross cutting identities lead to frequent
disagreements over group-based policies but that these competing demands also
POLITICAL COMPLICATIONS 109
prevent the reification of state categories and the dangers of dichotomized

conflict.
The following case studies of recent Muslim and Christian efforts to join the
official categories used for reservation policies demonstrate that contemporary
group-based policies, like their colonial precedents, influence identity claims. At
the same time, minority activists’ demands push the boundaries of official
classifications. Many have criticized India’s past and present caste-based policies
for further dividing the nation along caste lines rather than alleviating caste
discrimination. Echoing a common conclusion drawn by journalists, politicians
and academics in contemporary accounts about reservations, an article in the
Hindustan Times included the following argument: “[W]ith more and more
clamoring to be categorized as backward…the quota system has come to acquire
a self-perpetuating character…widening the scope to cover communities that
hardly need such props is nothing but the height of inequity, which will only
aggravate social tension by deepening caste division” (Hindustan Times 9 March
1996).4 The following case studies demonstrate that communities demanding
reservations may indeed lead to some social tensions, yet they do not necessarily
deepen caste divisions. On the contrary, they often complicate the currently
recognized divisions.
Recent political movements to revise the beneficiary categories demonstrate
that state edicts alone cannot reify social identities. Rather, the countless
identities that characterize Indian society have resulted in a wide variety of
protest groups lobbying for revisions to their classifications. Although such
varied and competing demands have arguably contributed to some tensions
between and within groups, at the same time, they have complicated and thus
diffused potentially more serious conflicts between dichotomized sides. There
are some striking examples of this interplay between state and societal categories
in India. Some Muslims are demanding Other Backward Classes (OBC) status for
their entire religious community, and certain Christians are arguing that they
should be recognized as Scheduled Castes (SCs). Since Muslim and Christian
religious doctrines do not recognize caste, such claims are quite controversial
among current (predominantly Hindu) beneficiaries of reservations as well as
among factions within these minority religions. Low caste Hindus fear that their
benefits would be diluted by a larger pool of eligible beneficiaries. Some Muslims
and Christians are opposed to recognizing caste within their supposedly
egalitarian religions or are leery of categorizing themselves with backward
groups. Because these debates challenge both the existing distribution of
material benefits and widely held assumptions about the link between Hinduism
and the caste system, these policy debates are both politically and religiously
contentious.5
MUSLIM DEMANDS FOR BACKWARD STATUS

In the 1990s, Muslims renewed their demand for backward status for their entire
population in India. Muslims benefited from group policies during the colonial
period in the form of reserved legislative seats, provided in 1909, as well as
separate electorates and a quota of 25 per cent of civil service positions,
introduced in 1926. Although the Indian Constituent Assembly’s Advisory
Committee on Minorities and Fundamental Rights considered continuing the
reservation of legislative seats for Muslims after independence, Muslim
reservations ultimately were a casualty of Partition. As Theodore P.Wright
describes, “after the communal carnage following independence, neither were
Hindu representatives willing to continue this concession nor were the remaining
Muslim committee members prepared to press for it” (Wright 1997:852–8).
Partition, the ultimate reification of communal politics, also renewed the
commitment of Indian politicians such as Jawaharlal Nehru to the construction
of a secular state. This vision of secularism was incompatible with reservations
on the basis of religion. The end of Muslim reservations coincided, however,
with a renewed need to uplift the Muslim minority, now decreased in size and
decapitated of its elite, which had, in large numbers, left for Pakistan.
There is constitutional support for reservations for Scheduled Castes (a
category from which Muslims are legally disqualified), Scheduled Tribes (which
are rarely Muslim), and the Other Backward Classes (which can include
Muslims). Due, in part, to the vagueness of the category, reservations for the Other
Backward Classes were relegated to a back burner to be taken up, or not, on a state-
by-state basis. Reservations for the OBCs were put into practice at the national
level in the 1990s, after Prime Minister V.P.Singh dusted off the 1980 Mandal
Commission Report on Backward Classes and a somewhat modified version of
its recommendations was approved by the Supreme Court.
The Mandal Commission had declared over 80 Muslim groups to be backward.
According to the data they used, backward Muslims made up a little more than
half of the total 11.2 per cent of the Indian population that was Muslim; those
specified as backward included groups such as weavers, oil crushers, carpenters,
and dhobis (clothes washers) (Mandal 1980:60–1, Times of India 18 January
1996, Hamid interview 2 September 1996). Some southern states have gone
farther, as when Kerala classified all Muslims as backward for state-level
reservations purposes in 1994 (Bayly 1999). The current demand for national
level reservations for all Muslims was jump started in the mid-1990s, although
rumblings of such a demand had occurred from time to time in previous years
(Ahmad 1980). The revived demand has left different Muslim groups at odds
with each other, some pressing to be classified purely along religious community
lines and others demanding that class or caste categories be retained and
strengthened.
“In my view the entire Muslim community in the country forms a backward
class,” wrote Syed Shahabuddin, a founder of the Association for Promoting
Education and Employment of Muslims, in a letter to the Welfare Minister in
1995 (Muslim India February 1996:78).6 In conferences on the issue in Delhi and
Hyderabad, the Association demanded “the recognition of the Muslim
community, as a community, as a Backward Class…and for the consequent
extension of reservation to the community, in proportion to its population and
level of backwardness, both in higher and professional education as well as in
public employment” (Association for Promoting Education and Employment of
Muslims 1996, emphasis in the original). Based on his research on the leaders of
the renewed demand for Muslim reservations, political scientist Theodore
Wright argues, “they feared that if Muslims did not get on the backwardness
bandwagon, they would be left competing for an ever diminishing proportion of
open (unreserved) seats with an ever larger pool of Forward (twice-born Hindu)
rivals” (Wright 1997:854). Concerns about both Muslim political unity and
proportional opportunities for Muslims underlie this campaign.
Syed Hamid, the President of the Association, argued that the entire Muslim
community in India is socio-economically depressed and discriminated against,
so some positive action must be taken by the government (Hamid interview 2
September 1996). Shahabbudin emphasized the necessity of “cutting the cake”
not just horizontally by caste and class, as in current reservation policies, but also
vertically, by religion, in order to evenly distribute opportunities (Shahabuddin
interview 11 September 1996). The association’s preference is for a separate
quota for Muslims, rather than including them in the existing groups eligible for
reservations, so that the Muslim community can “enjoy the full benefit of their
rightful measure of reservation, free of all apprehensions of any encroachment by
other relatively advanced communities if bracketted with them” (Association for
Promoting Education and Employment of Muslim 1996:1). Shahabuddin, like his
association in their official resolution, argued that those Muslims already
declared backward would have “first claim” to benefits, but is this enough to
protect the more disadvantaged Muslims? (Shahabuddin interview 11 September
1996). Other Backward Class Muslim and Dalit Muslim groups are skeptical.
They have their own ideas on how the cake should be sliced.
Those Muslim groups already included in the lists of Other Backward Classes
have a vested interest in keeping the competition, including upper class or upper
caste Muslims, off the lists. “While Islam may be casteless, our society is divided
on the basis of castes,” argued Shabbir Ansari, President of the All India Muslim
OBC Organisation, at their first national convention in New Delhi in 1996 (The
Hindu 30 August 1996:3). One of their major questions, posed at a 2002 seminar
sponsored by this organization and Mumbai University, is “why are Indian
Muslims always projected as a single, monolithic group without cultural and
social variations?” (Minwalla 2002). A major goal of this organisation is to get
more arguably backward Muslim groups on the official lists of backward classes.
According to their estimates, over 90 per cent of Muslims should be considered
backward but not all Muslims. The current OBC lists contain several Muslim
communities. The head of the National Commission for Backward Classes,
R.N.Prasad, pointed out that, theoretically, non-Hindu religions take away the
caste basis of social stratification, but in reality there is a “hangover of caste
sticking to them” (Prasad interview 18 September 1996). In response to official
state and national level lists of backward groups, the All India Muslim OBC
Organisation has carried out surveys in order to compile its own lists of Muslim
backward classes, including additional Scheduled Castes and Scheduled Tribes
in addition to Other Backward Classes. For example, in the state of Maharashtra,
where the organisation started, it demanded that 42 Muslim groups be added to
the lists of the officially backward (Sonavane interview 2 December 1996).7
An activist in this organization, Vilas Sonavane argued that “Religion has
been used to suppress the basic contradictions in Indian society, which is [sic]
caste” (Sonavane interview 2 December 1996). In contrast to the Association for
Promoting Education and Employment for Muslims, which he said had accused
him of “dividing Muslims,” Sonavane felt that reservations should be an Other
Backward Classes issue, not a Muslim issue. Using caste and class categories to
undermine arguments for reservation categories based on religion, he described
this movement as “deconstructing the myth of religion in India,” namely the
perennial polarization of majority versus minority religions. Because he is
focusing on overlapping axes of identity, he characterized his movement as “the
first postmodernist movement in India” (Sonavane interview 2 December 1996).
Although “the philosophy of caste as a superstructure of social discrimination
is quite contradictory to the basic beliefs of Islam which implicitly emphasizes
equality and universal Muslim brotherhood,” caste-like stratification persists in
Muslim societies in India (Ansari 1960:27). The status distinctions between the
descendants of immigrants and descendants of converts, the pre-existing caste
distinctions of those who converted, as well as occupational hierarchies have
resulted in a complex system of categories internal to the “Muslim community”
(Ahmad 1978, Ansari 1960, Jenkins 2000, Mann 1992, Mondal 1996). The
Muslim OBC Organisation complicates the notion of a Muslim community and
“takes the stand that secular social structures and class/caste hierarchies
transcend and come prior to religious identities” (Bidwai 1996). The organization
tries to work across religious divides, drawing parallels between the plight of
lower status Muslims, Hindus and Christians in its arguments and agenda.
For example, in their efforts to get Scheduled Caste status for the Muslims
Dalits, the All India Muslim OBC Organisation aligns itself with the Dalit
Christian organizations making similar demands. The Scheduled Caste category,
originally only open to Hindus, has been expanded to include Sikhs and
Buddhists. In a memorandum to the Welfare Minister, the President of the All
India Muslim OBC Organisation, Shabbir Ahmed Ansari, pointedly asked why
the Christian and Muslim Dalits had been excluded from Scheduled Caste
benefits.8 This argument, to be considered in more depth in the next case study
focusing on Scheduled Caste Christians, suggests the potential for a pan-
religious lower class movement.
Their advocacy for the Scheduled Caste Muslim cause means that the All
India Muslim OBC Organisation is also pan-backward in approach, meaning
they are not just submitting arguments before the government on behalf of Other
Backward Classes. They have called for an end to “discrimination” against
“Muslim Dalits only on grounds of Scheduled Caste categories of India” (Ansari
1996). Since Scheduled Caste reservations, unlike those for Scheduled Tribes
and Other Backward Classes, are closed to Muslims, the “SC Muslim” demands
are the most controversial. Other groups have taken up this demand as well. For
example, the All India Backward Muslim Morcha (AIBMM) has likewise argued
that Muslims should be included in the Scheduled Caste category. Led by Aijaz
Ali, they have opposed the demand for a religiously based quota for all Muslims,
preferring to press the claims of the Dalits within their religion.9
Another class-based criticism of Muslim reservations comes from the opposite
end of the social spectrum. Although they have not formed an organization
actively pressing their point of view, some upper class Muslims are opposed to
declaring their whole religious community backward. For example, in a
newspaper article entitled “Muslims must aim higher than quota,” M.Yusuf Khan
argues that “unlike the Harijans, Muslims were never a socially disadvantaged
group” (Khan 1996). This opinion was expressed to me most strongly in
Hyderabad, at the third convention on the issue of reservations for Muslims,
sponsored by the Association for Promoting the Education and Employment of
Muslims. Although the official speakers and participants were largely supportive
of Muslim reservations, in discussions, some people in the predominantly elite,
Muslim audience referred to the princely history of the area and pride in their
community, which made them leary of being categorized with India’s lower
classes. In the conference’s more informal afternoon discussion, various
audience members made several objections and comments regarding Muslim
reservations, signaling some ambivalence about this approach: [Hindu] Dalits
and Brahmins will definitely put their own person forward, but our approach is
altogether different. We don’t want to be treated as backward, like SCs and STs.
A Hindu tailor may be treated as backward, but a Muslim tailor is not backward
(audience members at Hyderabad conference 18 August 1996, paraphrased).
In contrast to the Association for Promoting Education and Employment of
Muslims, with its purely Muslim agenda, the All India OBC Organisation
complicates simple religious categories with its critiques on the basis of class and
caste. Muslim Dalits and even some Muslim elites further garble efforts to
promote a reservation for Muslims. Although all of these views are responses to
the government’s classifications, none is entirely satisfied with the boundaries of

the official categories.
To put debates over the merits of a Muslim reservation in historical
perspective, it is useful to recall the comments of anthropologist Arjun Appadurai
on the “pathology” of colonial classification and enumeration on the basis of
religion:
The process by which separate Hindu and Muslim identities were

constructed at a macro-level, and transformed not just into imagined
communities but also into enumerated communities, is only the most
visible pathology of the transfer of the politics of numerical representation
to a society in which representation and group-identity had no special
numerical relation to the polity.
(Appadurai 1993:332).
Shahabuddin’s demand for a quota on the basis of the total Muslim population is
a contemporary example of this dynamic: “Theoretically, they [Muslims] may be
entitled (at the central level) to 4.2 per cent out of the 27 per cent quota for the
OBCs, 2 per cent out of the 15 per cent for the SCs and 1 per cent out of the 7.5
per cent for STs. And a fourth subquota in the quota for High Castes”
(Shahabuddin 1996:77). Shahabuddin stresses the importance of proportional
numerical representation through these figures but ultimately proposes a single
Muslim quota on the basis of the total Muslim population, rather than dividing
the Muslims into such subquotas.
Shahabuddin’s letter to the Welfare Minister is particularly adamant that a
policy differentiating Muslims would be detrimental to Muslim unity: “I do not
think you would like your Government and Party…accused of dividing the
Muslim community” (Shahabuddin 1996:78). Another advocate of Muslim
reservations on the basis of religion argues, “by giving reservation to certain
classes among Muslims the government will be creating casteism among Muslims
to divide their strength” (Khan 1996:226). Ironically, those advocating Muslim
reservations through such accusations of government plots to “divide and rule”
the Muslim minority are echoing criticisms of the original reservations for
Muslims, which have often been portrayed as a colonial effort to divide and rule
Indian society as a whole along religious lines (Appadurai 1993, Wright 1997:
852).
Does simply demanding to be counted and represented along different lines
subvert this dynamic of reified, enumerated communities? If enough groups raise
objections to official categories, they are not undermining categorization per se,
but they are exposing the fallacy of essentialized communities, the assumption
that those who share a religion or caste necessarily have shared interests or form
the basis for logical political units. Replacing rigid caste classifications by
returning to the rigid religious classifications of an earlier era may not be the
answer; yet the ongoing arguments within the Muslim community over the future
of reservations force people to re-examine facile assumptions stemming from
historically oversimplified categories.
CHRISTIAN DEMANDS FOR SCHEDULED CASTE

STATUS
Another religious minority in India, the Christians, are smaller in number, yet
this minority’s demands for reservation rights has received as much, if not more,
attention than the Muslim demands. Christians are about 2.5 per cent of the
population, and about “two-thirds of the 20 million Christians in India could be
described as dalits” (Wyatt 1998:16). Thus, some Christian organizations are
demanding that Dalit Christians be recognized as Scheduled Castes. The notion
of “Dalit Christians” gradually gained recognition in the 1980s, and since 1990
activists have pressured the government to amend the constitution to broaden the
definition of the Scheduled Castes to include them (Wyatt 1998). A rise in
violence against Christians in India, accompanied by increasing ideological
attacks by Hindu nationalists, has caused Christians to mute these demands
recently. Some Hindu nationalists, the subject of the next chapter, argue that
missionaries are “targeting weaker sections” for conversion and suggest that
granting reservations would reward this practice (Times of India “Missionaries
Targeting Weaker Sections: VHP” 31 July 2000). In spite of such opposition,
several Christian org nizations are committed to opening the Scheduled Castes
category and associated protections and benefits to Christian Dalits. Although
theirs is another purportedly caste-free religion, they argue that caste
discrimination persists and should be recognized.
The Scheduled Castes category, encompassing the lowest castes, or Dalits,
originated with the Simon Commission and was used in the Government of India
Act of 1935. In 1936, the Government of India (Scheduled Castes) Order made
clear that “no Indian Christian…should be deemed a member of a Scheduled
Caste” (quoted in Massey 1991:28). This decision faced few objections at the
time, for the purpose of the legislation was to assure special electoral
representation for certain minority categories, and Christians had these benefits
in their own right. This separate representation for religious minorities was
eliminated at independence, however, and the purpose of the Scheduled Castes
list shifted to include the administration of not only elections but also other
social and economic benefits and protections. Nevertheless, an Order of the
President (1950) retained the rule that “no person professing a religion different
from Hinduism shall be deemed a member of a Scheduled Caste” (Galanter 1984:
144).
In 1956 disadvantaged Sikh communities were included as Scheduled Castes,

followed by neo-Buddhists in 1990. These decisions reinforce an inclusive legal
notion of “Hinduism,” a category broadened to subsume such indigenous
religions, while still excluding Christianity and Islam. Legal scholar Marc
Galanter feels that the court’s continuing acceptance of a legal theory that caste
is coterminous with Hinduism (or, at most, Sikhism and Buddhism) “reflects the
continued force of a view of caste groups which sees them as units in an
overarching sacral order of Hinduism…From this view of caste derived the long-
standing reluctance of the courts to give legal effect to caste standing among non-
Hindu communities” (Galanter 1984:144). This selective inclusion of Scheduled
Castes into the Hindu category indicates a shift in post-independence reservation
policy from its more secular framing toward an increasingly Hindu nationalist
articulation, in both legal and political arenas. In response to competing
identifications of Scheduled Castes, Hindu nationalists have been more eager to
subsume Scheduled Castes of certain religions that they consider “Indian.” Thus
Sikhism, which developed from an egalitarian reformist response to Hindu
traditions, and Buddhism, the religion to which B.R.Ambedkar and his Dalit
movement converted in the 1950s, were legally reabsorbed as “Hinduism,”
whereas Islam and Christianity, constructed as “foreign,” remain politically
useful “Others.”10 This distinction will be further elaborated in Chapter 7.
Lobbying for Scheduled Caste Christian benefits extends back to 1950, the
date of the Presidential Order that made them ineligible, and a few states have
extended state-level benefits to this group similar to the benefits extended to
Scheduled Caste Hindus (Kananaikil 1983:15). In the 1990s the movement for
revision of the Scheduled Caste category at the national level has been the
subject of increasing political activity by Christian organizations, although
recently they have kept a somewhat lower profile.11 A controversial Scheduled
Caste Christian protest in conjunction with a prayer meeting attended by Mother
Theresa drew national attention to the cause in November of 1995 but elicited
criticism from Hindu nationalist organizations (Times of India 21 and 24
November 1995). One leader of the movement for Scheduled Caste Christians
lamented, “Until today only words are given, promises are given, assurances
made by the Prime Minister, Welfare Minister, and the government officials, but
in practice they are not doing anything. We don’t believe that they will do it”
(Lourduswamy interview 19 December 1995). The Common Minimum
Programme agreed on by the 1996 United Front coalition government included
Scheduled Caste status for Dalit Christians, but this promise, too, failed to
materialize. Subsequently, the Hindu nationalist Bharatiya Janata Party’s
formation of a coalition government at the center took even more wind out of the
movement’s sails.
In addition to the more publicized Hindu nationalist critiques of Scheduled
Caste Christian demands, to be discussed in the following chapter, controversy
has erupted within the church over officially recognizing low caste Christians.
Caste inequalities persist within Christian communities in India, where separate
seats, communion cups, burial grounds and even churches for lower castes persist
(Forrester 1980, Japhet 1998, Koshy 1968, Webster 1994). In addition to caste
discrimination among Christians, Dalit converts to Christianity also continue to
face the social and economic disadvantages associated with their castes in the
wider society, in regard to housing, education, employment and practices of
untouchability (Kananaikil 1990). Bishop M.Azariah, in Chennai, pointed out the
contrast between Christian doctrine and social practice in India: “The Christian
teaching demands that you practice equality,” yet “the caste feeling is so strong,”
still playing a role in marriage discrimination, for example (Azariah interview 10
July 1996).
Such inequities, then, are part of the life experiences of lower caste Christians,
an experience diametrically opposed to the egalitarian ideals of Christian
doctrine. Despite continuing inequities in practice, public policies such as
reservations were, in part, inspired by Christian egalitarian ideals taken up by
indigenous reformers. Historian Robert Frykenberg notes, “The critique of caste
begun by Protestant Christians from abroad was increasingly taken up in the
twentieth century by Indians, whether national secularist or national Christian in
ideology” (Frykenberg 1985). The tension between egalitarian Christian ideology
and stratified Christian society is at the root of the controversy over the status of
the Scheduled Caste Christians. Some Christians argue that “untouchable
Christians” is not a contradiction in terms and that SC Christians should be
officially recognized; others respond that such a stance is dangerous and will
communalize Christianity. In other words, it could divide the Christian
community along caste lines or push it further into the communalist politics
pitting Hindus against religious minorities in India by antagonizing Hindus.
The Scheduled Caste Christian activists are in the awkward position of having
to argue that their own religious community engages in caste discrimination. On
this basis they argue that excluding Christians from the Scheduled Caste category
is religious discrimination. Their emphasis on division within Christianity leads
them to pose an alternative vision, the unity of Scheduled Castes of all religions.
In the words of one activist, S.Lourduswamy, “Dalits of all religions are living
together. They are also equally undergoing all the disabilities—social,
educational and economic disabilities—due to the traditional practices of
untouchability” (Lourduswamy interview 19 December 1995). On this basis,
several organizations, such as the National Coordination Committee for SC
Christians, are trying to expand the Scheduled Caste category to include
Christians by lobbying politicians, holding seminars, and orchestrating protests,
ranging from Christian school closures and signature campaigns to relay hunger
fasts and mass rallies.12 A striking example of such activism was a dharna (protest)
held in New Delhi on 30 November 1995, where protesters staged the symbolic
crucifixion of a Christian Dalit (photograph from Network News June 1996).
This movement embraces the Scheduled Caste category at the same time as it
criticizes its boundaries, desiring the benefits it can confer while lamenting its
limited definition. In a pamphlet distributed to Members of Parliament, the
National Coordination Committee for SC Christians makes the following
argument about Christian Dalits:
Except for the (wrong) records in the revenue offices he is a Dalit in every
sense of the word viz. Ethnically, lineally, racially, socially, economically,
culturally, vocationally, geographically, relationally, contextually, and
emotionally. HE CONTINUES A FULL DALIT EVERYWHERE
EXCEPT IN THE IGNORANT MIND OF THE EXECUTIVE.
(National Coordination Committee for Dalit Christians 1996:8, emphasis in
original)
As one leader of the Committee described Dalit converts to Christianity, “their

conversion has not converted them from the caste system, caste mindedness,
caste consciousness” (Lourduswamy interview 19 December 1995).
Highlighting the divisions in the church, these activists turn to Scheduled
Caste or Dalit identity for unity, although some recognize that the SC category is
really a construction of the state rather than a community of people. For
example, Father Jose Kananaikil deconstructed the SC category even as he
advocated the inclusion of Christians within it: He recognized that Dalits are
separate groups “artificially brought together” and noted that “culturally they
have separate identities,” although some solidarity is developing (Kananaikil
interview 11 January 1996). Nevertheless, in their public demands, activists often
emphasize pan-religious SC or Dalit identity and downplay the material benefits
associated with official backwardness. A case in point is S.K. Chatterjee’s
Presidential Address at the National Convention of the All India Christian
People’s Forum in 1996. “The demand for SC status on equal terms with other
dalits is not [a] question of jobs or scholarships. It is a question of their identity
itself” (quoted in the All India Christian People’s Forum newsletter, Network
News, November 1996; Chatterjee interview 22 November 1996). A Dalit
Solidarity Programme, launched in 1992, engages leaders from Christianity,
Buddhism, Islam, Sikhism and Hinduism in meetings and discussions.13
The Dalit Solidarity Programme also brings Christian leaders in northern India
in contact with those in the south. Some Christian leaders in the southern state of
Tamil Nadu, for example, have taken a particular interest in Dalit Christian
issues and are at the forefront of Dalit “liberation theology.” Bishop Azariah of
Chennai emphasized “social realities” and “human realities” by listing six types
of gaps between Dalits (of all religions) and the rest of society due to caste
discrimination: economic, social, political, cultural, psychological, and physical

(Azariah interview 10 July 1996). In Madurai, a Jesuit priest, inspired by
learning of a forum of African-American Jesuit priests, started meetings for local
Dalit Jesuits (Jayaraj interview 6 July 1996). Dr V.Devasahayan, associated with
a Dalit Theology Department in Chennai, stressed the importance of making the
Dalit Christian movement inclusive. “When we talk about Dalit liberation, we
talk about the liberation of Dalits in all religions and in all parts of India… Our
concern is for the total Dalits (Devasahayan interview 10 July 1996).
Some Christians, however, are skeptical about demands for Scheduled Caste
status, as recognition of caste within Christianity not only goes against Christian
doctrine but, they fear, might further divide the church. Unlike the Association
for Promoting Education and Employment of Muslims, however, there is no
comparable movement advocating reservations for all Christians as a religious
minority. In fact, many who are opposing Scheduled Caste Christian reservations
also oppose reservations on a religious basis, due to fears that Christians are
becoming more “communalist” in this matter (Sharif interview 17 November
1996). Pastor of a Protestant (Church of North India) congregation, Salim Sharif
felt that, due to SC Christian demands, “We are becoming another class and
caste” (Sharif interview 17 November 1996). This is an interesting comment,
perhaps reflecting a fear that Christianity, too, will become absorbed as just one
more group within a broadly defined Hinduism. The remark also expresses
concern that Christians are becoming involved in the politics of division rather
than unification. Instead of demanding Scheduled Caste Christian status, he
argues that Christians should ask “unitedly” for help and reservations for all the
poor, whether they are Christian or non-Christian (Sharif interview 17 November
1996). Church Counselor Shakuntala David went beyond the multireligious Dalit
category advocated by some SC Christian organizers to suggest that the term
Dalit transcends both religion and caste: Dalit means downtrodden, she argued,
and it could even refer to a downtrodden Brahmin (David interview 17 November
1996). In this way class-based arguments further complicate religious or caste-
based claims for backward status.
These competing demands coming from factions of Christians and Muslims
are only the tip of the iceberg. Other demands for new or revised reservation
categories are flying fast and furious. In addition to those perspectives already
mentioned, some are proposing a separate reservation category for Dalit
Christians (DCs) as opposed to incorporating them in with the SCs (Indian
Express 26 September 1996, Network News November 1996). The next chapters
explore still more demands, based on religion, class and gender. Official
attempts to categorize the disadvantaged have spurred many groups into political
action. At the same time, claims and counter claims on the basis of religion,
caste, class and gender suggest that official and unofficial constructions of
identities may never coincide. What do all these contested categories say about
group-based policies for diverse democracies?
CONCLUSION
Policies to assuage ethnic tensions may spark them anew, but tension over
competing identity claims is not necessarily destructive. It is precisely the variety
and number of groups protesting their official classifications that may prevent
such classifications from “sticking” long after their purpose has been served. It is
the cross-cutting competition between and within groups that can prevent a large
scale mobilization of “us” versus “them.” These case studies of religious
minority arguments about reservations demonstrate that official categories do not
necessarily reinforce social cleavages. In fact the categories have sparked heated
debate over the very nature of caste discrimination and its relationship to religion.
Groups continue to point out when they think they have been misrecognized
by the state, and competing claims prevent the solidification of group
boundaries. The overlapping nature of identities in India makes it impossible for
any religious organization to present a united front on the issue of reservation
categories. Those that do demand revisions in the current official categories, such
as those wanting reservations for all Muslims or for SC Christians, continuously
problematize the official lists and keep them from becoming cemented into the
public consciousness. Others counter with criticism or different demands, such
as OBC Muslims who are demanding consideration of casteism or Christians
who are reluctant to recognize casteism. These arguments, in turn, debunk
assumptions about monolithic religious communities. In short, although these
movements challenging the current reservation schemes undeniably contribute to
some social and political tensions, taken as whole they are not simply reinforcing
community fault lines but rather proposing many alternative and often
conflicting categorizations of Indian society.
Donald Horowitz proposes several formulas to alleviate or avoid ethnic
conflict. One way he addresses is the “dispersal of conflict” or the creation of “a
new, lower layer of conflict laden issues.” He discusses this in the context of
federal systems that geographically disperse national level ethnic bifurcations
into the more complicated context of many states, resulting in “a more complex—
and therefore less tense—politics at the center.” (Horowitz 1985:604–5).
Ashutosh Varshney provides a thoughtful application of Horowitz’s discussion
of dispersed ethnic configurations in his assessment of the challenges such
dispersion poses for Hindu majoritarianism (Varshney 1998:42–6). The cases
examined in this chapter illustrate a variation of this principle. Rather than
dispersing the polarized politics of religion or caste in a geographic sense, the
above debates over reservation categories shatter dichotomized conflicts in the
realm of ideas by constantly drawing into question the boundaries of castes and
religions and their relationship to each other. In this sense, intergroup tensions do
not disappear but become certainly more complex and arguably less tense than
conflicts between clear cut and unambiguous groups. The ongoing debates over
who is in which category, as opposed to nation-wide fighting between clearly-
defined sides, is a lower, and generally less explosive, form of conflict over
reservations.
In some cases new alliances cut across traditional dividing lines of religion.
Some groups explicitly downplay religious distinctions, as in the Dalit Solidarity
Programme that attempts to bring together an interfaith group of similar castes. A
Muslim OBC organization demonstrates “the power that the concept of a
subaltern organization cutting across the lines of religion has come to acquire”
(Bidwai 1996:12). Some are trying to capitalize on even broader subaltern
alliances by emphasizing Dalit-Muslim or OBC-Muslim unity across religions
(Nath 1995, Dalit Voice 16–30 June 1996:21). In contrast to these examples of
cross-religious ties on the basis of caste or class, other Muslims and Christians
downplay class and caste divisions and try to emphasize unity within their
religious groups.
Contemporary Muslim and Christian protest groups are shaped by the
structure of the state reservation categories but also demonstrate agency by
challenging static classifications. The resulting melange of competing demands
helps to prevent the reification of such classifications. Moreover, these religious
minority debates over reservations are only the tip of the iceberg. Muslim and
Christian activists face a backlash from Hindu nationalists, who oppose
extending benefits if this would reward converts to those religions seen as
“foreign,” namely Christianity and Islam. Arguments for economic criteria for
reservations have led to new rules to exclude the “creamy layer” from the Other
Backward Classes. The demand for reserved seats for women in India’s
Parliament has failed repeatedly, largely due to objections that a gender-based
category does not recognize disadvantages on the basis of class or religious
minority status. Official attempts to categorize the disadvantaged have spurred
many groups into political action to press claims and counter claims on the basis
of religion, caste, class and gender. These ongoing clashes between various
official and unofficial constructions of the disadvantaged are the topics of the
chapters to come.
7
Hindu nationalism and selective inclusion
The Sangh Parivar views the demand by the Christian Church for extension
of reservation benefits to Dalit Christians as part of a “global strategy by
the Vatican to evangelize the world by the Jubilee Year 2000 AD”.
(Indian Express-Kochi 3 October 1996:1)
Hindu nationalist politicians and activists oppose those demanding reservations

for Dalit Christians and those demanding that the entire Muslim community be
declared backward. There are Muslim and Christian opponents as well, but the
most powerful and organized opponents are Hindu nationalists. Legal scholar
Adrienne Davis’s statement that “some counter-categorical actions may even
reinforce the hierarchy” is a salient critique of their approach (Davis 1996:719).
Hindu nationalists, while disparaging the divisive categories of reservations,
draw their own lines and hierarchies between religious groups in the name of
national identity and fail to adequately address caste or class stratification. In
light of the rise of Hindu nationalism, the fragmented and fluid nature of the
minority demands discussed in the previous chapter has a downside. If minorities
are beset with intersecting and competing subgroups, what hope have they of
standing up to Hindus who make up over 80 per cent of India’s population? Yet,
the complexities of India’s “ethnic configuration” may also be Hindu
nationalists’ biggest challenge. The Hindu nationalist Bharatiya Janata Party
struggles to appeal to voters beyond their northern, Hindi-speaking, upper caste
base and, in order to gain power at the center, has been forced by coalition
partners to give up the very planks of their platform that are most threatening to
minorities (Varshney 1998, Ali 1998).
Although social “hybridity” and “intersecting voices” are gaining much
scholarly attention these days, the complexity of identities is neither new nor
unique to India (Bhabha 1994, Young 1976, 1993). Georg Simmel’s classic work
on the “web of group affiliations” recognizes that many groups “intersect” in an
individual and that “the larger the number of groups to which an individual
belongs, the more improbable is it that the other persons will exhibit the same
combination of group affiliations, that these particular groups will ‘intersect’

once again” (Simmel 1955:140). The different arguments within the Indian
Muslim, Christian, or Hindu “communities” attest to the intersecting identities
(such as caste, class or religion) pulling group members in different directions.
These various affiliations are an antidote to extremism. Simmel notes that when
a single group affiliation influences someone in a “pervasive and one sided
manner,” affiliation with another group “quite apart from the nature of the
groups involved” is sufficient to instill a “stronger awareness of individuality in
general and at least to counteract the tendency of taking his initial group’s
affiliation for granted” (Simmel 1955:151). This suggests that although
reservation policies highlight certain groups, political activists emphasize various
other groups, and the people exposed to these debates are unlikely to take a
single identity for granted. Our “web of group-affiliations” means that political
complications can confound attempts at state simplifications.
This chapter focuses on the Hindu nationalist rejection of Christian and
Muslim demands for reservations. Building on a history of selectively including
some minority religions and lower castes into Hinduism while at the same time
excluding “foreign” religions, Hindu nationalists have their own spin on the
debate over reservation categories. Current tensions between Hindu nationalists
and advocates of minority reservations parallel an earlier dispute between
Mohandas Gandhi and B.R.Ambedkar prior to Independence. Gandhi advocated
a more inclusive Hinduism and rejected reservations, whereas Ambedkar
rejected Hinduism and advocated more reservations. Yet even inclusive Hindu
nationalism became selective. The contemporary Hindu nationalists’ rejection of
Muslim and Christian demands for reservations is an offshoot of their nationalist
ideology, in which Muslims and Christians are outsiders and political Others.
Moreover, the rising political power of Hindu nationalists has driven a wedge
between these two minority groups in spite of their complementary demands.
The organization and political savvy of the Hindu nationalists bodes ill for
Muslim and Christian reservations, although the varied group affiliations that
splinter these minority religious groups also cut across Hindus. Their desire to
head and maintain a diverse coalition government at the center has inspired some
Hindu nationalist politicians to tone down their opposition to various
reservations.
ABSORBING MINORITIES: EXPANSIVE

DEFINITIONS OF HINDUISM
By Hindu nationalists, I am referring to a group of three related organizations:
the Bharatiya Janata Party (BJP), a political party; the Vishwa Hindu Parishad
(VHP), the wing more involved in social and cultural issues, although not
apolitical; and an umbrella organization, the Rashtriya Swayamsevak Sangh
(RSS), again a purportedly cultural but often political organization. Known as

Hindu nationalists, the Hindu right, or the Sangh Parivar (blood family), these
organizations propagate a Hindu communal ideology known as Hindutva.1
According to the Bharatiya Janata Party’s 1996 election manifesto:
The BJP is committed to the concept of one nation, one people, one culture
—our nationalist vision is not merely bound by the geographical or
political identity of India, but defined by our ancient culture [sic] heritage.
From this belief flows our faith in “Cultural nationalism,” which is the core
of Hindutva.
(BJP Election Manifesto 1996)
An important part of this ideology is defining what is and what is not “Hindu” or
“national.” This process of definition occurs not only in political rhetoric but also
in laws and policies. Hindu nationalists have attempted to construct most
Scheduled Castes and certain minority religions as members of the “Hindu”
religious category. In contrast, they portray Muslims and Christians as non-
Hindu and “foreign,” although occasionally they describe these minorities as
former Hindus, who were led astray and converted.
Even Mohandas Gandhi, known for religious tolerance and staunchly opposed
to virulent Hindu nationalism (and, tellingly, opposed by the virulent Hindu
nationalist who assassinated him), had set ideas about the proper categorization
of Indian society and the relationship between caste and Hinduism. Gandhi
dubbed untouchables “Harijans,” or “children of god.” Notably, he chose a
Hindu name for god, Hari, implying that these groups were to be included within
the Hindu fold. His motivations were complex, but most likely included his
sincere desire for better treatment of untouchables, his practical fears of political
fragmentation during the nationalist struggle, and his distrust of conversions
away from Hinduism by discontented untouchables:
Having himself, by personal fiat, ‘converted’ all Untouchables into

Harijans and, thus, into co-religionist Hindus, he was resolute in his
opposition to what he saw as misguided, perverse, and politically
motivated conversion through “alien proselytization”.
Gandhi advocated reforming Hinduism to include rather than revile “Harijans”

and discouraged their exodus to different religions through conversion. Gandhi’s
grandson and biographer, Rajmohan Gandhi, points out that the context of the
freedom movement defined his approach, which was shaped at that time by his
overriding concerns about political divisions (Gandhi interview 20 September
1996).
Some so-called Harijans, including B.R.Ambedkar, opposed their involuntary

absorption into the very religion that had cast them off as untouchables. As
historian Robert E.Frykenberg points out:
[It was] this annexation of untouchables into Hinduism, this arbitrary co-
optation by definition, without leaving any freedom of choice to the
untouchables themselves, which was to be so profoundly offensive to
Ambedkar and other leaders of the untouchables. Indeed, all subsequent
obfuscation notwithstanding, this action by Gandhi has never been
forgotten by Untouchable leaders.
Gandhi’s action contributed to the continuing, close association between the

Harijan category—prior to 1935 known officially as the depressed classes, after
1935 as the Scheduled Castes—and Hinduism.
A related dispute in the political history of religious and caste categorization in
India is the divergent approaches to lower caste justice advocated by Gandhi and
Ambedkar. Whereas Gandhi preferred reform from within Hinduism to address
Harijan concerns and opposed separate policies for them, Ambedkar demanded
group rights. Ambedkar’s followers eventually popularized yet another name for
Harijans: Dalits, which means the “oppressed” or “crushed.” In defiant contrast
to the rather paternalistic and decidedly Hindu “children of Hari,” the term Dalit
evokes political solidarity and protest and is still widely used today. Dr
V.Devasahayam, a Dalit liberation theologian in Chennai, articulated the
meaning of the term Dalit:
They wanted to name themselves. The word Dalit simply means the
oppressed. So they wanted to take a name which also reflects their own
reality, because they were fooled by the gift of several other names which
did not reflect the reality. For example, they were given the name
Harijans… In reality they were not allowed to go inside the temple. Even
in the villages today the Dalits are not allowed to go inside the temple, but
still there are people who would like to call them as children of god. So
they say no, enough of this nonsense.
(Devasahayan 10 July 1996)
Ambedkar’s ultimate rejection of Hinduism occurred in 1956 when he, together

with many of his followers, converted to Buddhism.2 This was a powerful
protest against both their exclusion from Hindu society and their inclusion within
Hindu politics.
These ideas about the name and nature of the lowest caste category are
reflected in the views of Gandhi and Ambedkar about reservations in the form of
reserved legislative seats. Gandhi, although in his own way an advocate for the
rights of Harijans, so vehemently opposed separate electorates for them that he
started a hunger fast, fearing that such special provisions would prove divisive to
Hinduism and thus the incipient Indian nation (Das 2000, Pyarelal 1984).
Ambedkar was a leading advocate for such measures and, eventually, was largely
responsible for including references to the special policies for Scheduled Castes
and Scheduled Tribes in the Indian Constitution after Independence. Gandhi’s
opposition to reservations was in part a political calculus but also was a
reflection of his “guiding principle” and spiritual theme, the “oneness of
humanity” (Gandhi interview 20 September 1996). In a 1931 speech against
separate electorates and reservations for the Depressed Classes, Gandhi said:
I claim myself, in my own person, to represent the vast mass of the

“untouchables”… And I would work from one end of India to the other to
tell the “untouchables” that separate electorates and separate reservation is
not the way to remove this bar sinister… I say it is not a proper claim
which is registered by Dr Ambedkar, when he seeks to speak for the whole
of the “untouchables” in India. It will create a division in Hinduism which
I cannot possibly look forward to with any satisfaction whatsoever. I do
not mind the “untouchables” being converted to Islam or Christianity. I
should tolerate that, but I cannot possibly tolerate what is in store for
Hinduism… I will resist it with my life.
(Quoted in Pyarelal 1984:97–8)3
It is impossible to say whether Gandhi would have approved of the contemporary

Hindu nationalists’ heated opposition to extending job reservations to more
Muslims and Christians. Gandhi’s speech shows more tolerance of conversion
than one finds in the contemporary Bharatiya Janata Party, but his concern over
dividing Scheduled Castes from the Hindu category is echoed in current Hindu
nationalist rhetoric.
Gandhi became more critical of Christian conversions and missionary work as
the 1930s progressed (Harper 1992:152–3). These later critiques make him more
useful to today’s Hindu nationalists, although a congruence of opinion is not
necessary for Hindu nationalist politicians to find a popular historical figure
politically useful. Not only have Hindu nationalists attempted to appropriate
Gandhi, but they have also used the image and appeal of Ambedkar in
contemporary political campaigns, a development that would have, no doubt,
appalled him (Maclean 1999). As one scholar puts it, “from the late 19th century
to the present, Hindu communalism has constantly projected itself as a caring
response to the ‘pernicious’ caste divide” (Agrawal 1994:247). This brief history
of ideas about the relationship between Hinduism, untouchability and group-
based policies shows that the current debates over reservation categories have
precedents in major pre-independence debates over the nature of national,

religious and caste identities.
EXCLUSIVE INCLUSION: SETTING BOUNDARIES TO

HINDUISM AND CASTE
Some Hindu nationalists idealize a broadly defined Hindu identity to counter
social splintering, which they blame both on existing reservations and on the
minority groups who are proposing categories of their own design. Yet such
unifying pleas often contain hidden, or not so hidden, distinctions of their own.
The Hindu nationalist parties in India project an image of Indian unity, while at
the same time reinforcing caste and communal divisions. This nexus of casteism
and Hindu nationalism can be seen in Hindu nationalist responses to reservations,
especially the contemporary Christian and Muslim proposals to increase their
communities’ access to reservations.
Often this response involves contradictions. For example, “In accordance with
the overriding concept of Hindu brotherhood, caste is not recognized in RSS
[Rashtriya Swayamsevak Sangh] activities,” yet “[a]n occasional qualified
defense of the fourfold varna and caste system as a social order that in the past
had harmoniously held together Hindu society by organizing the division of
labor on the basis of individuals’ differing abilities is still expressed in RSS
publications” (Spitz 1993:249). The Hindu nationalist organizations are
dominated by upper castes and have, in the past, opposed reservations, calling
them a divisive policy. They were at the forefront of the uproar over Prime
Minister V.P.Singh’s 1990 decision to extend central government reservations to
the Other Backward Classes. Although they have since toned down their
opposition as this policy came into effect, they remain the most organized and
strongest critics of the Christian and Muslim proposals to further expand the
reservations beneficiary pool. They prefer a “gradualist strategy of persuasion
which emphasizes the unity of all Hindus” over group-based caste reform policies
(Spitz 1993:249). Advocates of a nation defined as “Hindu,” they have been
particularly critical of group-based policies that would benefit converts or even
long-term members of religions with no connection to Hinduism.
Legal precedent helps to explain the extension of the Scheduled Caste
category to include lower castes within Sikhism and Buddhism—both religions
doctrinally opposed to the caste system. Legal precedent also facilitates the
exclusion of Muslims and Christians from this category. Some Indian religions,
with the notable exceptions of Islam and Christianity, have been officially
subsumed under “Hinduism” for several legal purposes (Jenkins 2001b).
Buddhists, Jains and Sikhs have been considered “Hindus” in the arena of
“personal law,” or religious civil law, as in the 1955 Hindu Marriage Act. Such
an inclusive view of Hinduism is also included in the constitution itself, with
regard to temple entry law, in Article 25 on the freedom of religion. In the

constitutional clause allowing states to insist on opening Hindu religious
institutions to all Hindus, “the reference to Hindus shall be construed as
including a reference to persons professing the Sikh, Jaina or Buddhist religion”
(Bakshi 1996:47). Drawing on this progressive constitutional clause on temple
entry, an opponent of including Muslims or Christians in the Scheduled Caste
category argued in a Hindustan Times editorial: “The Scheduled Castes have
always been regarded as part of the Hindu community,” and “for the purposes of
social welfare and reform the Hindus include the Sikhs, Jains and Buddhists”
only (Deshpande 1990). Aizaz Rizvi, National President of the BJP Minority
Morcha, defined Sikhs as governed by Hindu law and thus not part of his
Minority Morcha, which includes only Muslims, Christians and Zoroastrians
(Rizvi interview 12 September 1996).4
At many points throughout history, Hinduism has been quite expansive, open
to multiple interpretations, and ready to absorb tribes as castes and local deities
as members of the Hindu pantheon (Roy 1994). Its doctrinal diversity makes
Hinduism seem ideally suited for tolerance. “How,” asks writer Shashi Tharoor,
“can such a religion lend itself to fundamentalism?” (Tharoor 1997:56). Despite
Hindu philosophy’s potential for open interpretations and Hindu history’s wealth
of examples of new groups being absorbed into the fold, a legal and ideological
line divides Hinduism, along with religions considered indigenous to India, from
Islam and Christianity. This pervasive logic even filters into the arguments of
non-Hindu opponents of Scheduled Caste Christian demands: Shakuntala David,
a Christian, argued that Buddhists and Sikhs receive reservations because they
are part of a larger Hindu community (David interview 17 November 1996).
Political calculations in a democracy also figure into the Hindu nationalists’
inclusive yet exclusive definition of Scheduled Castes. Years after the ideological
battle between Gandhi, with his notion of Hindu Harijans, and Ambedkar, with his
Dalit Buddhist conversion movement, “neo-Buddhists” were welcomed back into
the Scheduled Caste fold in 1992. Sikh Scheduled Castes had already been
similarly reabsorbed in 1956. Despite these precedents, the Hindu nationalists
remain unwilling to absorb the religious Others, Muslims and Christians, which
have been so politically useful as groups to mobilize against.
Although blurring the line to a certain degree over who is a Hindu, those
opposed to reservations for Muslims and Christians are adamant that “caste”—
and therefore the Scheduled Caste category—only extends to Hindus.5 For
example, a newspaper editorial entitled “What Does SC Mean?” included the
argument that the “true meaning of Scheduled Castes is indicated by the word
‘castes.’ Castes are a central feature of only Hinduism and not of Christianity or
Islam” (Deshpande 1990). Satyanarayan Jatiya, who later became the Minister of
Social Justice and Empowerment, was the National President of the BJP
Scheduled Caste and Scheduled Tribe Morcha when he made a similar argument
that the Scheduled Castes are only in the Hindu religion: The Muslims and
Christians did not have this inequality problem. Although there was economic
inequality, there was no social inequality or discrimination (Jatiya interview 12
September 1996, paraphrased). Satyanarayan Jatiya helped organize a Scheduled
Caste rally against reservations for “SC Christians,” on the grounds that the SC
reservation will be diluted if a new population is added (Jatiya interview 12
September 1996).
In short, Hindu nationalists have at times been quite broad when defining
Hindus, although whether they were including groups or subsuming them is an
important question. They have at times tried to include Scheduled Castes, Sikhs,
Buddhists and Jains within the Hindu category. In contrast, their definitions of
the caste system are not as loose, especially in their arguments denying the
existence of low caste Christians or Muslims. Hindu nationalists not only bar
Muslims and Christians from their “big tent” notion of Hinduism but also
exclude them from their conception of the Hindu nation.
THE BOUNDARIES OF THE NATION: MUSLIMS AND

CHRISTIANS AS OUTSIDERS
The Hindu nationalist opposition to Muslim and Christian demands for
reservations is rooted in historical associations between these religions and
invasion or imperialism. Historian Barbara Metcalf notes that a “striking
characteristic of public life in India has been an intensified use of historical
narratives to define the nature of India’s people and draw the boundaries of
citizenship.” She describes the “version of history” propagated by Hindu
nationalists: “There once was a Hindu Golden Age; Muslims came as foreign
invaders; Muslims were oppressors who ultimately ushered in a period of
decline” (Metcalf 1995:952). Muslims, she argues, have been given a role of a
“foil against which the unity of others—Hindus, the nation—can be constituted,
and injustices of class and wealth obscured. The history that identifies Indian
Muslims as aliens, destroyers, and crypto-Pakistanis, with its profound moral and
political implications for citizenship and entitlements, is critical in sustaining
that role” (Metcalf 1995:963). Such constructions of history certainly do have
implications for membership in the nation and for the definition of official
categories with their related entitlements. Hindu nationalist rhetoric commonly
links Muslim demands with the idea that they are “foreign,” either foreign
invaders of yore or those responsible for Partition and Pakistan.
For example, in 1994 when the demand for national level Muslim reservations
was renewed in an organized manner, Lal Krishna Advani, BJP President,
criticized the demand on the basis that it was divisive and might re-create
Partition-like conditions (Syed Hamid interview 2 September 1996). BJP Vice
President Madan Lal Khurana likewise suggested that reservations for Muslims
would only occur at the cost of Scheduled Castes and Scheduled Tribes and
would lead to the disintegration of the country (Times of India 21 July 1996:7).
In reaction to a government proposal in Andhara Pradesh to include all Muslims
as Backward Classes at the state level, the BJP “opposed the inclusion of Muslims
en masse as this would give a ‘double benefit’ to the Muslims” (Shatrugna 1994:
2400). How they would benefit doubly is unspecified; perhaps this is a reference
to those backward Muslim groups who are already included in the Other Backward
Classes. In any event, this is an example of a common refrain of the Hindu right,
the idea that Muslims have been appeased and pandered to.
Hindu nationalists also oppose extending reservations to all Muslims for fear
that it would encourage conversions. The mass conversion of untouchables to
Islam in Tamil Nadu during 1981–2 revived Hindu nationalist fears that this
minority was growing at their expense (Malik 1989). The Vishwa Hindu
Parishad (VHP) has started an “anti-conversion and reconversion programme,”
including proposals to train Hindu missionaries in a center in Ayodhya. This
location is symbolically significant, since, in perhaps the most striking example
of Hindu nationalists using “historical” tactics against the Muslims, Hindutva
forces in 1992 destroyed a mosque in Ayodhya, claiming that Muslims had built
it on the birthplace of the Hindu god Ram. The “anti-conversion” plan is to send
10,000 Hindu missionaries into “backward and tribal areas in the country” to
prevent conversions of disadvantaged citizens to non-Hindu religions and to
encourage reconversion to Hinduism (The Organizer 21 July 1996).
Conversion is an even stronger theme in Hindu nationalists’ arguments that
Christian Dalits should not benefit from reservations. The Hindu nationalists
emphasize the historical association between Christianity, colonialism and
missionaries, arguing that opening the Scheduled Caste category to Christians
would increase conversions to this religion as well. These themes have
influenced the broader SC Christian debate. For example, the Catholic Bishops
Conference of India’s directive to Christians to vote for parties supportive of SC
Christian demands inspired the following letter to the editor on the subject:
Conversion by Christian missionaries is said to be changing the very

demographic profile of the backward and tribal districts of the country. If
this religious imperialism continues, a day will come when the churches in
India will have control over the ballot boxes, and bishops will dictate state
policy. Secularism will then be a hollow slogan in India.
(Indian Express 22 March 1996:8)
The President of the Vishwa Hidu Parishad, Ashok Singhal, told a VHP
gathering that “the motive of Christian missionaries is not the welfare of people.
The main purpose is to bring maximum area under their influence” (Times of
India “Missionaries Targeting Weaker Sections: VHP” 31 July 2000). Criticisms
of missionary plans to target low castes feed into Hindu nationalist opposition to
recognizing Christian SCs. Pointing to potential material motivations rather than
spiritual motivations for conversion, such as access to church-based educational
or social services, Hindu nationalists do not want to make conversion easier by
allowing low caste converts to keep reservations. Ashok Singhal’s rhetoric
simultaneously excludes and criticizes Christians and Muslims: “Hindus never
believed in converting people of other faith[s] to their own religion” (Times of
India 31 July 2000).
In response to such charges, the All India Christian People’s Forum (AICPF)
and other SC Christian organizations spend much of their time asserting that
Christians are part of the Indian nation too. An AICPF resolution reads: “The
tendency to treat dalits belonging to Hindu and non-Hindu religions differently,
being motivated by revivalist and fundamentalist forces and to brand Christianity
as an ‘alien’ religion, is not warranted by history and is aimed at denying the
nativity of dalit Christians who are as indigenous as all other dalits” (Network
News November 1996:1).
When Mother Theresa participated in a prayer meeting in conjunction with an
Scheduled Caste Christian rally, Hindu nationalists led a public outcry against a
foreign Christian taking a political stand on reservations. BJP General Secretary
Sushma Swaraj reiterated the idea that the SC Christian demand is not only
emanating from foreign missionaries but is also dividing the nation: “Mother
Theresa has always spoken for unity and harmony of people, but the demand for
reservation for Christian dalits is by itself divisive in nature” (Times of India 21
November 1995). The Vishwa Hindu Parishad organized a demonstration, and
leaders, calling Mother Theresa a “Christian missionary carrying out conversions
of the innocent poor and destitute,” again criticized her support for SC Christians.
Mother Theresa ultimately issued a statement that she “never demanded
reservation for Dalit Christians” (Times of India 24 November 1995). This
statement is just one example of the powerful effects the Hindu nationalists are
having on the various Muslim and Christian movements discussed in the
previous chapter, including the way they categorize themselves.
THE IMPACT OF HINDU NATIONALISM ON MUSLIM

AND CHRISTIAN DEMANDS FOR RESERVATIONS
By reinforcing Hindu-Muslim rivalry, the Hindu nationalists, in one sense, add
momentum to the activists pushing for backward status for all Indian Muslims.
Hindu nationalism reinforces their reluctance to divide Muslims along class or
caste lines even for the purpose of reservations. Both the Hindu nationalists and
the activists working for Muslim reservations agree on something: They are
opposed to the Muslim OBC agenda. “The BJP…is just as hostile to this new
trend. It knows that the trans-religious subaltern solidarity could be the nemesis
of the religion-based identity politics it pursues” (Bidwai 1996:12). In addition to

entrenching religious competition between Hindus and Muslims, growing fears
about Hindu nationalism have hurt prospects for ChristianMuslim cooperation on
the issue of reservations.
Although a shared trepidation about conservative forces has inspired some
coordination among minority religions, such as the interfaith Dalit Solidarity
Programme, often fears of Hindu nationalist reactions have stymied such
collaborative efforts. Some leaders of the SC Christian movement are reluctant to
champion the similar cause of the Muslim Scheduled Castes. A Dalit theologian
committed to Dalit solidarity nevertheless discussed the strategic challenges that
could face Christian SC activists if they joined forces with Muslims: The
government “might consider giving Christians reservations because they are
apolitical. When you talk of Muslims then [the] BJP would be up in arms
because they are a substantial number and therefore a threat to the Hindu
community” (Devasahayan interview 10 July 1996). Some SC Christian activists
feel that the Muslim demand is even more unpalatable to those uncomfortable
with the Christian demand and would thus undermine their own goals.
Scheduled Caste groups too have expressed concern that including Christians
within their quota will lead to including Muslims too, a slippery slope that will
increase competition for the same number of opportunities. For example, the
leader of the Scheduled Castes and Scheduled Tribes Joint Action Council,
K.V.Kumaran, worries: “If Christianity is now included, Islam will also have to
be added” (The Indian Express-Kochi 3 October 1996:1). Many Christians say
that strategically it would be better to get their own demand passed by
Parliament and then try to work on Muslim issues. Father Jose Kananaikil noted
that since the BJP is especially opposed to reservations for Muslims, Christians are
saying “at least include us” and are “not pressing for Muslims” (Kananaikil
interview 11 January 1996). Although they “support” the Muslim demand to be
treated as OBCs, some SC Christian publications explicitly state that their
demand “should not be confused with the demand of various other groups and
communities to be included as Scheduled Castes. The demand for equal justice to
SC Christians should not also be mingled with that demand to treat the entire
Muslim community as OBCs” (Daniel 1996:9). This pragmatic distancing of
themselves from other similar movements undermines Christians’ rhetoric about
SC unity. So far, there seems to be more emphasis on symbolic unity with those
Dalits who already benefit from reservations, Hindus, Sikhs and Buddhists, than
actual cooperation with the Muslims.
The leader of the demand for backward status for all Indian Muslims, Syed
Shahabuddin, said he had supported but not cooperated with the Christian Dalit
movement, although he is against any religious test used to exclude people,
whether Muslim or Christian, from the current categories (Shahabuddin interview
11 September 1996). Another leader in this movement, Syed Hamid, when asked
about possible cooperation with the Christian Dalit movement, said that the
Muslim movement, in contrast to the Christian one, has been a more “sporadic
effort” than the Christian one, which had complicated coordination (Syed Hamid
interview 2 September 1996). Although limited by some SC Christians’
pragmatic reluctance to be associated too closely with the backward Muslim
cause, Aijaz Ali’s Backward Muslim Morcha has spoken positively about the SC
Christian movement, stating that the morcha would “join hands with Christians
in mobilizing public opinion” (Indian Express 24 November 1995). Lack of
interest and coordination on the part of some other Muslim leaders limits the
possibility of joining forces with Christians working on similar challenges to
current reservation policies. Yet perhaps the most significant limit on the
Scheduled Caste unity advocated by the SC Christian movement is their own
fear of greater Hindu opposition if Muslim reservations are also at issue.
The Hindu nationalist critique of Muslim and Christian demands is so
influential because of the political savvy and organizational strength of the Sangh
Parivar, in contrast to the much newer SC Christian organizations, led largely by
church leaders rather than politicians, and the even less institutionalized Muslim
reservation movement. Two meetings on the single evening of 4 September 1996
in New Delhi illustrate the contrast between the Hindu nationalist and SC
Christian organizations. Both meetings were devoted to the issue of the SC
Christian demands, but the first was organized by the National Coordination
Committee for SC Christians and the second by the BJP legal cell, the Adhivakta
Parishad. The NCCSCC meeting, held in order to present their arguments before
Members of Parliament, included most of the leaders of the movement in Delhi,
who outnumbered the few Members of Parliament who came—MPs who were
already supporters of their cause. Speakers cited Bible verses to support their
claims, hardly a method to appeal to those not already in agreement. Later that
evening at the Adhivakta Parishad’s well-attended meeting against SC Christian
demands, a BJP member (who had not attended the previous meeting) cited the
very same Bible verse being used by the Christians and had a counter argument
ready to deliver in his speech. The contrast between the two movements in terms
of their organizational power, awareness of their opponents’ arguments, and
political sophistication was striking.
THE WEB OF IDENTITIES AND THE MODERATION

OF HINDU NATIONALISM
Fighting against such a political organization, what chance do the splintered
minorities have? Yet the varied identities cutting across India also limit the
Hindu nationalists’ appeal, especially at the national level. The multiple,
overlapping identities of India mean that Hindu nationalists have been forced to
broaden their rhetoric to try to include the lower castes. More recently some have
even made attempts to appeal to Muslims, who were so marginalized in the past.
Hindu nationalists’ opposition to expanded Other Backward Class reservations in
1990 became muted as these policies became a reality. When they came to
power at the center, the BJP government even expanded the lists of OBCs, which
is one way to broaden their appeal to these numerous middle castes and potential
regional allies.6
In another attempt to widen their political net, the Bharatiya Janata Party,
traditionally an upper caste bastion, named a Dalit, Bangaru Laxman, as party
president. He not only tried to appeal to lower castes but also, in his first speech
as party President, argued that the party needed to make “sustained efforts to
reach out to Indian Muslims” (Rahman 2000). With Muslims at 12 per cent of
the population, and Christians at only 2.5 per cent, the latter are not an electoral
priority. In a published interview, when asked specifically about reservations for
Christians and Muslims, Laxman was more politic in his response than some of
his party members, who have harshly criticized the policies. Although he
followed his party’s line by not supporting policy changes that would make SC
Christians and all Muslims eligible for reservations, he softened his response. He
voiced support for literacy programs and pointed out that “[s]ome Muslims—
Ansaris, Qureshis—have already been covered under the Mandal Commission
recommendations [reserving government jobs for certain backward castes]. If a
better case was made with regard to the others, it would be examined” (Rahman
2000). Although silent in this interview on the issue of Christian reservations,
Laxman’s party-line stance was overlaid with a more moderate tone regarding
the issue of Muslim reservations. This reflects the BJP’s political need to bring
more groups into their fold while not alienating their original adherents.
CONCLUSION
The Muslims, Christians, and Hindus discussed in chapters 6 and 7 shuffle and
reshuffle categories based on their diverse conceptions of caste, religion and
nation and their distinct ideas about the relationships between these identities. In
their inclusive mode, Hindu nationalists have historically made some attempts to
incorporate lower castes and certain minority religions, those seen as indigenous,
into their conception of Hinduism. Yet a community is, in part, defined by what
it is not. In their exclusive mode, the Hindu nationalists have kept Muslims and
Christians at arms length, portraying them as foreign conquerors and
missionaries whittling away at the Hindu nation. Such ideologies feed into their
opposition to any expansion of reservations for Muslims and Christians.
The groups discussed in these chapters are motivated by a complex mixture of
ideas about identity and concerns about material benefits. The dynamics of
reservations as a means of resource distribution complicate the development of
pan-religious or cross-class alliances. The material stakes associated with
reservations lead groups to jockey for position. Those who are already officially
Other Backward Classes or Scheduled Castes might see backward Muslims and
SC Christians as fellow disadvantaged groups, but competition over limited
reserved opportunities causes many OBCs and SCs to oppose extending
reservations to additional Muslims or Christians. Hindu nationalists have tapped
into fears about both identities and interests to rally the largely Hindu Scheduled
Castes, for example, against SC Christian demands. Some SC Christian leaders
have been a bit reluctant to join forces with Muslims, for fear that their own
demands, if so linked, will never be met due to the particularly vehement
opposition to Muslim demands by many Hindu nationalists. Concerns about the
Hindu nationalists’ reactions have shaped the contours and alliances of the
Muslim and Christian reservation movements, at times dividing them from each
other.
Although very astute at manipulating interests and identities, the Hindu
nationalists are not immune from the dilemmas of balancing their ideology of
identity with their interest in gaining political power in a democracy. Hindu
nationalists, as they have tried to gain and hold on to power at the national level,
have had to compromise on some hardline ideas about Hindu identity or
Hindutva in order to deal with the practical problem of expanding their potential
pool of voters and coalition partners. Balancing the need to expand with the need
to retain their more hardline supporters, the Sangh Parivar has used its many
related organizations as well as a variety of political figures, ranging from the
moderate Bangaru Laxman to the militant Lal Krishna Advani, to put forth many
faces of Hindu nationalism. Recognizing the need to reach out to Scheduled
Castes, Other Backward Classes and even Muslims, their opposition to some
reservations has become more muted, yet opposition to Christian and Muslim
demands persists.
By recognizing political agency, we can appreciate the ability of groups to put
forth their own self-definitions rather than simply absorbing state definitions of
their identities. These self-defined identities are at times reigned in by judges and
administrators. Yet, official categories are not ossified, in large part due to their
political meaning. Categories are combined (and recombined) to make a variety
of strategic identity claims. At times these claims, such as the absorption of
Scheduled Castes and certain minority religions into Hinduism, perpetuate the
power of the majority. Yet, notably, lower castes have not always returned the
embrace of Hindu politicians, as demonstrated by Ambedkar’s conversion
movement in the 1950s and Kancha Ilaiah’s contemporary manifesto, Why I am
not a Hindu (Ilaiah 1996). Disadvantaged groups, such as lower castes and
religious minorities, have used both a positive and relatively “convertible”
category, religion, and a seemingly negative and more fixed category, low caste
status, to challenge the boundaries of the categories themselves and the
hierarchies they represent.
8
Class, classification and creamy layers
Economic status has proven a complicating factor when it comes to determining

who is eligible for reservations in India. Policy-makers, judges, and
disadvantaged groups all face difficult questions regarding how or if class should
be taken into consideration when identifying the legitimate recipients of
reservations. Should these boundaries be drawn along caste and religious lines,
or would economic criteria be a more appropriate basis for allocating benefits?
Should more complex or comprehensive definitions of disadvantage be
developed? Recent challenges to reservation policies have focused on the issue
of class. These challenges, like the religious disputes discussed in preceding
chapters, have thrown the very categories of group identity into question. This
chapter considers contemporary claims and policy revisions on the basis of class
or economic criteria.
Two policy changes, in particular, placed economic criteria in the political
limelight. First, in 1990 the central government, which formerly reserved jobs only
for the Scheduled Castes and Tribes, reserved 27 per cent of central government
jobs for the Other Backward Classes. Other Backward Class groups are less
socially degraded than the scheduled groups, so economic indicators are a key
factor in their definition as backward. These controversial national level policies
for Other Backward Classes thus renewed debates over economic definitions of
backwardness.
The second policy change to draw new attention to economic criteria was the
“creamy layer” ruling. In a decision on the constitutionality of national
reservations for Other Backward Classes, the Supreme Court mandated that the
government make further use of economic considerations in the name of helping
the “truly backward.”1 The court directed the government to develop criteria and
mechanisms to skim off the “creamy layer” of the Other Backward Classes, by
which it meant that the government must find a way to disqualify the more
advantaged individuals in these classes.2 As a result of the national level OBC
reservations and the creamy layer ruling, Indians have further developed
multifaceted reservation categories, which take into account not just caste but
also economic “backwardness” and individual economic mobility.
Policy changes giving economic considerations an increased role in

reservations demonstrate that official categories are not set in stone. Rather, the
government has occasionally revised reservation categories in response to social
changes and political demands. In this case, in contrast to the preceding chapters,
the government has instigated changes to the categories, whereas activists have,
at least in many cases, pushed for sustaining the old categories. State
simplifications are not inevitable; here a Prime Minister and Supreme Court were
willing to press for complex and changeable categories that would recognize the
forces of social mobility. Many social organizations and activists, in contrast,
protested against such changes.
For example, the new national level reservations for Other Backward Classes
provoked protests and demonstrations in north India by upper caste students and
others concerned about dwindling opportunities for the “forward” communities.
Although this protest has faded, an organization called the Gandhi Caste Society
continues to offer an alternative vision of a caste-blind society and caste-free
policies. The new creamy layer policy sparked resistance as well, particularly in
the south. Most strikingly, the southern state of Kerala simply refused to
implement it. Reactions to the changes in policy varied regionally, due to
different policy histories regarding the Other Backward Classes.
The Other Backward Classes category is defined by an elaborate combination
of social and economic criteria. Ultimately it is a list of groups, usually castes,
rather than individuals. Some states, particularly in the south, had long histories
of reservations for Backward Classes prior to the more recent extension of
reservations in central government jobs to these groups all over India (Parikh
1997, Galanter 1984, de Zwart 2000, Bhagavan 1994). For example, to counter
the domination of the administration by Brahmins, the princely state of Mysore
in 1921 started preferential policies for Backward Classes, which “included
everyone except the Brahmins and those whose mother tongue was English:
some 96 per cent of the population” (Dushkin 1979:661). In 1922 the Madras
presidency issued an order to government officials “to give priority in their
recruitment policy to non-Brahmans and other so-called ‘backward’
communities” (Irschick 1969: 219). This history meant that reservations for
various Backward Classes are much more widely accepted, even expected, in the
south. Thus societal reactions to the central government’s 27 per cent reservations
for Other Backward Classes and to the Supreme Court’s creamy layer limits on
Other Backward Classes varied by region. In the northern capital city of Delhi,
for example, the Prime Minister’s announcement of more OBC reservations
sparked deadly protests. In the southern state of Kerala, in contrast, the creamy
layer limits on Other Backward Classes provoked resistance in the form of
denials that a creamy layer existed in that state.
Economic or class-based arguments over reservation categories abound.
Previous chapters illustrate related concerns about upper class Muslims and
Christians taking over reservation benefits. The next chapter addresses how the
demand for a general category of women’s Parliamentary reservations has been
effectively squelched due to arguments in Parliament over the necessity of
including a subcategory for women of the Other Backward Classes. Some
political parties have proposed adding new reservations for lower class, high
caste individuals. A few people demand that the government not only add more
economic considerations to other types of categories but reject other categories
entirely in favor of economic criteria. For example, a symposium on the subject
of reservations by leading academics, published in the journal Seminar, included
the following call for “a new criterion”:
The principle of reservations should be followed, but not on a caste basis

as it is resulting in social disintegration. A shift in the criterion for
reservations from the orthodox caste principle to a scientific economic
class principle may perhaps change the very colour and philosophy of the
society.
(Tilak 1981:36)
Given the legal, bureaucratic and political momentum behind the current
categories, a shift to entirely economic criteria is unlikely; yet, in recent years,
reservation policies have taken an economic turn.
This chapter focuses on policy changes that have added more economic or
class considerations to reservation categories and on the political activists who
continue to challenge even these revised categories. Several major topics will
illustrate this interplay between state and society: The Mandal Commission’s
1980 report on Other Backward Classes; Prime Minister V.P.Singh’s 1990
decision to partially implement the Commission’s findings and reserve 27 per
cent of central government jobs for Other Backward Classes; the protests and
other forms of resistance sparked by these new reservations; the Supreme
Court’s 1992 decision about the legality of these reservations; a government
commission’s development of a creamy layer policy; and, finally, political
activists’ resistance to the creamy layer rules. Two major conclusions emerge:
First, these developments suggest that, at times, it is the government that pushes
for changing reservation categories and groups in society that prefer a more static
approach. Second, debates over the role of class in reservation policies
demonstrate that societal reactions to reservation categories, and the role of class
in particular, vary quite dramatically from north to south due to divergent policy
histories.
THE MANDAL COMMISSION REPORT ON OTHER

BACKWARD CLASSES
The debate over the place of class in Indian reservation policies largely revolves
around the fate of the Other Backward Classes. Marc Galanter has compiled a
list of ten types of historical definitions of “backward classes,” noting that the
term extends at least as far back as 1880 (Galanter 1984:154–5). The 1950
constitution named the amorphous category, Other Backward Classes, meaning
those disadvantaged groups that are neither Scheduled Castes nor Scheduled
Tribes, but did not specify who they were. Several state and national Backward
Classes Commissions have since struggled to define these groups.3 These
commissions used caste or community as their unit of analysis and classified
castes and communities using a combination of social and economic criteria.4
Building on numerous state commissions and one previous national
commission on Backward Classes (the 1955 Kalelkar Commission), the Indian
government constituted a national Backward Classes Commission, commonly
known as the Mandal Commission. In its 1980 report, the Mandal Commission
found that “backwardness” persisted among communities outside of the
Scheduled Castes or Tribes, concluded that reservations were justified for such
Other Backward Classes, came up with criteria to define them, and listed those
Other Backward Classes that met the criteria. They concluded that “excluding
Scheduled Castes and Scheduled Tribes, Other Backward Classes constitute
nearly 52 per cent of the Indian population” (Mandal Commission Report 1980:
61).
One difficult question that the commission faced was how to define “classes”
(Shah 1996, Mukherjee 1999). The Mandal Commission, like the 1955 Kalelkar
Commission before it and the current National Backward Classes Commission
after it, continued to use a group-based system. Indeed members of the
commission said their task was to “lay down criteria for identifying recognizable
and persistent collectivities and not individuals,” noting that “in the Indian
context such collectivities can be castes or other hereditary groups” (Mandal
Commission Report 1980: 55–7). In this sense the Mandal Commission opted for
a definition of class that took into account more than just economic criteria. It
drew on a 1969 Supreme Court decision stating that “‘class’ means a
homogeneous section of the people grouped together because of certain
likenesses or common traits, and who are identifiable by some common
attributes such as status, rank, occupation, residence in a locality, race, religion
and the like.”5 With this definition of class, the Mandal Commission could
include in the Backward “classes” some groups that were not Hindu castes, but it
could not include a stratum of individuals who were all poor but otherwise
socially unconnected. Debates over the role of class and caste in Indian society
and the relationship between them continue to spark much academic as well as
policy debate (Mukherjee 1999, Sheth 1999). Post-Mandal quantitative analyses

comparing the Scheduled Castes and Tribes, Other Backward Classes and other
communities indicate strong correlations between these categories and class as
defined by factors such as occupation, assets, type of house, and income (Mitra
and Singh 1999: 188–9).6
The Mandal Commission engaged in class classifications with the knowledge
that government classifications are legal and do not violate the Indians’
constitutional right to equality, as long as the classification is not “arbitrary.” As
the Supreme Court ruled, government classification “must always rest on some
real and substantial distinction bearing a reasonable and just relation to the things
in respect of which the classification is made.”7 Moreover, the official
classification of backward classes is specifically protected by a constitutional
amendment. Article 15 of the constitution prohibits discrimination on grounds of
religion, race, caste, sex or place of birth, a potential limit on even compensatory
discrimination. The year after the constitution was adopted, two Supreme Court
decisions barred compensatory discrimination in arenas other than government
employment and limited eligible groups.8 In response, the Parliament passed a
constitutional amendment to Article 15 to protect broader reservations for
Backward Classes, including other areas such as education: “Nothing in this
article…shall prevent the State from making any special provision for the
advancement of any socially and educationally backward classes of citizens or for
the Scheduled Castes and the Scheduled Tribes.”9 The explicit constitutional
reference to “special provisions” protects reservation policies for the scheduled or
backward groups from legal challenges on the basis of suspect classifications or
reverse discrimination. The amendment refers to “socially and educationally
backward classes” without mentioning economics, an impetus for the subsequent
decisions of the courts and commissions to grant reservations to disadvantaged
social groups rather than to the poor in general.
After grappling with the meaning and legality of class classifications, the next
question the Mandal Commission faced in its delineation of the Backward
Classes was the definition of backwardness. The commission examined castes or
communities by looking at a wide variety of indicators of backwardness before
deciding whether to include them on the list of Other Backward Classes.
Through surveys of these groups, the commission evaluated such factors as the
type of labor done by group members, average age at marriage, educational
levels, the distance of households from drinking water, and the value of assets.10
Both economic and social standing entered into the calculation of backwardness,
although social factors ultimately counted more than purely economic factors
(Hansen 1999:143, Engineer 1991). In short, the national level designation of
backward classes in the Mandal Commission Report emphasized group status
while also taking into account economic factors.
The 1980 report, however, had little practical effect until a decade had passed
and a government supportive of their findings came to power. In 1990 Prime
Minister V.P.Singh announced a plan to extend reservations in central
government services to the Other Backward Classes, largely as defined by the
Mandal Commission.11 Reservations are legally capped at 50 per cent of total
positions or seats. The Scheduled Castes already had 15 per cent reservations and
the Scheduled Tribes had 7½ per cent reservations, so Other Backward Classes
received 27 per cent reservations in central government jobs. Half of central
government jobs were now off limits to the “forward” communities, and protests
erupted to try to stop this controversial new policy.
ANTI-MANDAL ACTIVISM
Urban, upper caste students were particularly active in dramatic demonstrations
against the proliferation of quotas, which they saw as a threat to their own
opportunities. Most of the protests were in north India, many in Delhi. In the
worst moments of a protest in Delhi, a student set himself on fire, and in the
weeks to come, several other students also immolated themselves.12
Three former student leaders of the Anti-Mandal Commission Forum at Delhi
University spoke to me in 1996 about the opinions that spurred them to protest in
1990. Hare Ram Singh was head of the Forum’s action committee; O.P.Singh
headed the legal committee, and Neeraj Kumar was also involved in the Forum
and the protests. As we sipped chai on the campus of the Delhi School of
Economics and began to discuss the 1990 protests, a young beggar approached
us. They seized the opportunity to ask, “How can this girl benefit from
reservation?” The policy emphasis, they felt, should be on infrastructure for
“genuine development,” such as free, good education for all, rather than on
reservations which “sharpen contradictions” and “balkanize on a caste basis.”
Insisting that they were not against the concept of reservations entirely, they
voiced their opposition to the perceived political motivation behind the Other
Backward Class reservations. They felt V.P.Singh and his Janata Dal-led
government granted OBC reservations in exchange for the political loyalty of the
OBC “vote bank.” These student leaders also expressed dissatisfaction with the
way Other Backward Class reservations were initially going to be implemented,
namely, that “economic criteria [were] not taken.” Too often, they felt, even
Scheduled Caste and Tribe reservations were taken by the upper classes. In
addition to wanting more attention to economic factors, they preferred a system
with the “individual as unit” as opposed to a group-based system (H.R.Singh,
Kumar, and O.P.Singh interviews 22 November 1996).
Since the dramatic student protests, other groups have remained more quietly
opposed to the increasing lists of officially backward citizens. For example, the
Gandhi Caste Society, or Gandhi Sangam, is an organization trying to do away
with caste distinctions in society and public policy alike. At a 1996 meeting in
Delhi, several speakers spoke out against “Mandalization” and advocated the use
of economic criteria for reservations. The Delhi President of this society, Leela
Sajwan, argued that reservations should be based on economics rather than caste
(Sajwan’s address at Gandhi Caste Society Meeting, Constitution Club, Delhi, 1
August 1996; Sajwan interview 18 September 1996). The Gandhi Caste Society
criticizes the use of caste labels in school admissions, hostel admissions,
employment, and politics. Its members encourage individuals to change
surnames that indicate caste or jati and to become, instead, a single “Gandhi
caste.” They are thus expanding on Gandhi’s approach to caste reform by relying
on society, particularly Hindu society, to reform itself. Like the historical
disputes discussed in Chapter 7 between Gandhi and Ambedkar, who advocated
group-based policies to rectify caste injustices, the Gandhi Caste Society opposes
the caste reservations advocated by the Mandal Commission and V.P.Singh.
One speaker at a meeting of this organization expressed the organization’s
ultimate goal of “a casteless, classless society” (address by a Professor of the
Indian Institute of Public Administration, 1 August 1996). Other speakers argued,
“Let us start with the children,” advocating an end to school forms which include
caste and religion and encouraging the younger generation to throw out their
surnames and stop asking other questions about caste (Gandhi Caste Society
Meeting, 1 August 1996). This movement, which holds to the optimistic belief of
an earlier generation that caste will fade through assimilation and integration, is
another form of anti-Mandal activism and criticism of both official and
unofficial caste categories.
THE CREAMY LAYER RULING AND THE NATIONAL

COMMISSION FOR BACKWARD CLASSES
In addition to resistance by activists, the new reservations for Other Backward
Classes also faced a legal challenge. The 1992 Supreme Court case Indra
Sawhney v. Union of India tested the constitutionality of such reservations. The
court’s decision addressed how economic factors should figure into definitions
of backwardness. The court upheld 27 per cent reservations in central
government services for Other Backward Classes, as long as the “creamy layer,”
or the more socially and economically advanced persons within these groups,
was excluded. In the same decision, the court also forbade reservations for
economically disadvantaged upper castes. In other words, the court ruling
suggested that to be a member of an Other Backward Class, it was not enough to
be in a lower caste if one’s father was a government minister. It was also not
enough to simply be poor if one was in a high caste.
The court directed that the central, state and union territory governments
constitute permanent commissions to deal with definitional issues, to make lists
of Other Backward Classes and to assess proposed additions to or subtractions

from their lists. In response, the central government instituted the National
Commission for Backward Classes, which was discussed in greater detail in
Chapter 4. Unlike the temporary commissions of the past, this is a permanent
institution, although its members have limited terms. This commission is
responsible for defining a national list of Other Backward Classes based on the
Mandal Commission’s list and on the various state level lists. This commission’s
task is both an administrative nightmare and a political minefield, since state lists
of backward classes vary widely, and many groups aspire to keep, or gain, that
designation.
In Indra Sawhney the Supreme Court also ruled that individuals are to be
excluded from eligible Other Backward Class groups through “creamy layer”
rules: “Within four months from today the Government of India shall specify the
bases, applying the relevant and requisite socio-economic criteria to exclude
socially advanced persons/sections (‘creamy layer’) from ‘Other Backward
Classes.’”13 States with OBC reservations at the state level were to do the same
within six months. Although the creamy layer ruling applied only to the Other
Backward Classes, not to the Scheduled Castes or Tribes, this was a
groundbreaking decision. Henceforth, both group and individual criteria were to
be taken into account when defining the Other Backward Classes.
DEFINING THE CREAMY LAYER: THE PRASAD

COMMITTEE
Justice R.N.Prasad, the first head of the National Commission for Backward
Classes established in 1993, also headed a new Creamy Layer Committee to come
up with criteria for national level Other Backward Class reservations. Prasad
noted that, following the Indra Sawhney decision, the rules allow for a situation
in which a group continues to be an Other Backward Class but individuals within
that group are excluded on the basis of advancement or prosperity (Prasad
interview 18 September 1996).
The creamy layer rules agreed upon by Prasad’s committee necessitate
examining applicants, their parents and, in the case of women, their husbands in
order to skim off those who hold high-ranking government or military positions
(or who have parents or husband in such a position), those who own a certain
amount of irrigated land, and those with an annual income level over 100,000
rupees.14 These rules are thus much more complex than a simple economic cut-
off. A long list of grounds for exclusion from the Other Backward Classes
includes a wide variety of considerations relating to the employment, property
and income of the applicant and the applicant’s immediate family.15 Such rules,
combining numerous kinds of socio-economic advantages, have created a number
of administrative challenges.
The administrative challenges are not limited to the complexity of the rules.
Prasad noted that at times non-OBC or creamy layer candidates try to
“influence” officials to get caste certificates. In response, he advocated more
“transparency” to flush out the cream, in the form of full inquiries and public
postings of registries of those applying for and getting certificates, perhaps in
panchayat (local council) headquarters. Such transparency, however, raises some
of the same issues touched on in Chapter 4 on caste certificates, namely privacy
and stigmatization. Yet Prasad defended the idea of making applications more
transparent, claiming that doing so will help solve another persistent problem in
the doling out of reservations: Officers in the administration, mostly upper castes,
sometimes hinder the granting of caste certificates to legitimate applicants by
claiming a lack of information. Were the process to be more transparent,
members of Other Backward Classes and their associations or advocates could
use the public paper trail to organize a protest, a response Prasad felt would be
quite appropriate (Prasad interview 18 September 1996).
The creamy layer rules have resulted in more classification issues and related
activism, as well as manipulation, from both sides of the administrator’s desk.
The complexity of the issue is multiplied by the fact that individual states come
up with their own creamy layer rules for state level reservations, although
several have adopted the Prasad Committee rules (Prasad interview 18
September 1996). Such state-level decisions provide examples of resistance to
and manipulation of the creamy layer categories.
RESISTANCE TO CREAMY LAYER RULES

By the time that the Supreme Court approved Mandal-style Other Backward
Class reservations and the government resumed their implementation, the
student protesters and political parties which stood against them only a couple
years earlier accepted them much more quietly than one might have expected.
Not only did V.P.Singh’s Janata Dal Party embrace the Other Backward Class
reservations but the Congress Party and Bharatiya Janata Party also eventually
included reservations for Other Backward Classes in their manifestos. This
gradual acceptance is in part due to the growing recognition, even in the north, of
the political power of the backward classes (Jaffrelot 2000). In some cases,
groups formerly opposed to OBC reservations have become Other Backward
Classes themselves (The Statesman, “Rajastan’s creamy layer vies for OBC status”
30 October 2001). At the same time, increasing economic liberalization is
gradually reducing young people’s single-minded focus on access to public sector
jobs: “‘People have by and large reconciled,’ says Rohit Sharma, then an anti-
Mandal agitationist and now a marketing executive” (Indian Express, “Mandal
Battlefields Barren Now” 23 September 1993). Another reason for the growing
acceptance, even among former opponents of Other Backward Class
reservations, is the Supreme Court’s decision: The court’s stamp of approval

gave these policies a sense of legitimacy, or at least inevitability, in the eyes of
former protesters. At the same time the court’s decision addressed one major
concern of protesters by insisting on creamy layer exclusions to attempt to
prevent the “forward backward” citizens from taking reserved opportunities from
others.
In contrast to the northern states, where the anti-Mandal demonstrations were
concentrated, much of the south has had a longer tradition of reservations for the
backward classes. There the new reservations at the central level were more
widely accepted from the outset. In fact in the south it has been the limits on
reservations posed by the creamy layer criteria that have become the subject of
controversy. Thus the creamy layer regulations that assuaged some of the
concerns in the north about Other Backward Class reservations generated some
new resistance, particularly although not exclusively in the south.
The politics of defining the creamy layer have been particularly contentious in
the southern state of Kerala, which claimed to have no such layer (Bose interview
4 November 1996). This area was formerly the princely states of Travancore and
Cochin, which like other areas in the south had a history of progressive policies
for backward communities during the colonial era. For example, in the mid-1930s
the government of Travancore appointed a Public Service Commission, which
allocated many posts in government service to a list of 14 communities on a
proportional basis (Ouwerkerk 1994:87–8). After independence, Kerala had three
state level commissions on Other Backward Classes and reservations for these
groups in technical and professional colleges and state government jobs, all prior
to the national Mandal Commission (Mandal Commission Report 1980:13–4).
Notably, Kerala’s state-level OBC policies have taken into account household-
level data, such as an income ceiling, when determining eligibility for
reservations (Sheth 2000:265, Mandal Commission Report 1980:14).
Nevertheless, the government of Kerala, with its long recognition of the Other
Backward Classes, resisted the nationally mandated creamy layer rule to trim the
ranks of the backward.
In response to the Supreme Court’s ruling, the government of Kerala refused
to set up a commission to identify a creamy layer, and it passed a resolution in
1995 denying that there was any creamy layer at all in the state, the Kerala State
Backward Classes (Reservation of Appointments or Posts in the Services Under
the State) Bill. In response, the Supreme Court insisted that the government of
Kerala find such a layer and actually appointed a commission to do so. The
Joseph Commission’s report raised hackles in Kerala by finding a creamy layer
and by including a controversial suggestion that Muslims be categorized as a
creamy layer and thus kept off backward classes lists (The Statesman, “Exclude
Muslims, says panel: Kerala Government in a fix over Job Quota Proposal” 21
September 1997). The Supreme Court nullified the state’s resolution denying the
existence of a creamy layer in Kerala, and in December 1999 the Joseph

committee criteria came into effect (Rediff.com, “Supreme Court strikes down
Kerala backward classes act” 13 December 1999).
In response, Kerala appointed its own commission, the Narendran
Commission, to come up with more lenient rules and thus shrink its creamy
layer. The commission was explicitly charged with providing “maximum
benefits permissible within the parameters of the December 13 Supreme Court
verdict to backward communities” (The Hindu, “Government Appoints Panel on
Creamy Layer” 13 January 2000). Options considered to achieve this end
included raising the annual income ceiling for the creamy layer category from
the one and one-half lakh (150,000) rupees recommended by the Joseph
Commission. Some groups in Kerala have suggested that individuals with
incomes falling under a ceiling as high as five lakh rupees should still be
included within the Other Backward Classes. In addition to challenging the
current income ceiling, other political groups have called for a recalibration of
the rules for calculating income from property, which have been criticized as
unfair to farmers. Others have called for excluding NRI (Non Resident Indian)
income, received from family members abroad, from the creamy calculus.
Education is yet another factor, which has resulted in a proposal that those
without primary education should never be considered creamy, regardless of
their income.
Yet the predominant suggestion to the Narendran Commission by the various
Backward communities was to try to exempt the state from the creamy layer
rule. “The commission had invited suggestions from most organizations of
backward communities and minorities to help it in its task. But rather than
presenting their case with documentary support for the quantum of increase in
the income ceiling, almost all organizations have taken the stand that there is no
creamy layer among the backward communities in Kerala” (Venugopal 2000).
The Izhava community, eligible for reserved government jobs since the
mid-1930s, have been particularly opposed to the creamy layer principle (Osella
and Osella 2000:211).
In a southern state with a large population already declared backward and a
regional tradition of Other Backward Class reservations, it has become politically
difficult to skim off constituencies, particularly the most powerful members of
those constituencies, through new creamy layer regulations. Kerala is not the
only state to contest the new boundaries of backwardness, but this case provides
a glimpse at the challenges of adding new criteria to existing categories. At times
political activists, whether anti-Mandal or anti-creamy layer, have not pressed
for changes to the official categories but rather for the status quo. The northern
anti-Mandal protesters and southern anti-creamy layer protesters have very
different attitudes toward reservations for Other Backward Classes; yet what
unites them is their preference for older policies. Student protesters in Delhi did
not want new Other Backward Class quotas and various activists in Kerala did
not want new restrictions on Other Backward Class quotas.
CONCLUSION
Advocates of class or economic criteria often espouse them as a more rational,
scientific or objective way to allocate reservations, one that also avoids the
problems of state complicity in caste or religious distinctions. It is true that a
scheme based largely on caste can be problematic. As administrators from
colonial times to the present have discovered, “caste” refers to a variety of types
of identity, which are by no means uniform from locality to locality (Pant 1987).
Grappling with the notion of caste can make economic factors, which may be
more quantifiable, seem like an appealing basis for a policy.
In India, however, economic criteria can be as difficult to administer as caste
criteria. D.L.Sheth, a sociologist and member of the National Commission for
Backward Classes, found, in his work evaluating whether groups are backward
or not, that “economic criteria are more difficult because there is a lot of
informality, still, in this economy, in [the] agricultural sector, and people’s
income…it is common that people may be earning 20,000 rupees but show it as
only 2000. So economic criteria are difficult to implement…occupational and
caste categories are in fact neater” (Sheth interview 6 September 1996). In
addition to complications arising from India’s informal economy, corruption
obfuscates economic criteria. People lie to avoid taxes, rendering income
statistics unreliable, and the Other Backward Class certificates, which involve
creamy layer verification, are sometimes falsified.16 Also, data from an
applicant’s extended family may be quite relevant to their economic and social
status but difficult for the government to monitor.17
The Indian government, rather than shifting to purely economic criteria,
initiated reservation policies with two layers of criteria, which allow the
government to consider a more comprehensive combination of factors, both
group and individual, both caste and economic class. The creamy layer criteria,
developed in the context of new Other Backward Class reservations, constitute a
policy shift from purely group-based categories toward a recognition of
individual economic factors. Both the new reservations and the new individual
level restrictions on them provoked social protest.
Like the previous chapters, political activism over the role of economic factors
in reservations demonstrates that social groups pose many challenges to state
classifications; in this case the competing challenges differed most dramatically
along regional lines. In contrast to the previous chapters, the trends toward
economically informed classifications discussed here show that state
classifications are occasionally flexible and that sometimes it is the political
protesters who demand the status quo. Social mobility, in part a result of
reservation policies, has helped create a well off “creamy layer” within
disadvantaged groups. Although unable to reach vast numbers of landless
laborers, reservations have resulted in an educated subset within the eligible
groups, employed in government service and a political force to be reckoned
with (Galanter 1984:551, Mandal Commission Report 1980, Roy and Singh
1987). Concern that an advanced subset benefits disproportionately from
reservations was one impetus for political protests against reservations in Delhi.
Ultimately, advantaged groups were skimmed off; yet the cream did not leave
quietly, particularly in Kerala, another example of resistance to official
categories.
Serious questions remain about the use of economic factors. Is caste-based
discrimination waning enough to justify a shift toward economically defined
categories? So far, the Indian Supreme Court seems to think not. In the Indra
Sawhney case, it ruled that economic backwardness alone is not enough to
qualify for reservations, reinforcing the original goal of reservations as a measure
to uplift degraded communities rather than to redistribute wealth and
opportunities to the poor.18 India’s application of Other Backward Class and
creamy layer categories demonstrate that it is possible to consider class in
addition to caste, rather than rejecting altogether a category that is still salient in
a society in which casteism persists. Observing the policy initiatives along these
lines in India will give other countries an indication of whether it is possible to
overcome the significant administrative and political challenges associated with
drawing affirmative action boundaries around multidimensional categories such
as the “backward” or the “disadvantaged.”
Are more multifaceted categories, including economic factors, necessarily
better tools to fight disadvantage? They may more accurately reflect complex
disparities; yet, the increasing use of economic criteria in India also threatens to
undermine the ability of reservations to affect change. Depending on how they
are applied, these criteria may either expand or restrict the beneficiary pool to
such an extent that the status quo is served. In some states, for example, the
Other Backward Classes category has been expanded to such an extent that the
vast majority of the population is eligible for reservations (Parikh 1997:175,
Srinivas 1997). Such generous definitions of disadvantage may water down the
effectiveness of reservations for the most disadvantaged. At the same time,
skimming off the creamy layer from disadvantaged groups threatens to leave
reserved seats at the highest levels unfilled (Pandian 2000). An alternative to
disqualifying perhaps the most qualified individuals is subdividing broad
categories at the group level. For example, some states, such as Bihar and Tamil
Nadu, have tried to take into account different levels of disadvantage by creating
a “Most Backward Classes” category. Yet such distinctions add new definitional
challenges demanding careful consideration. When the state government of Uttar
Pradesh sought to make a controversial distinction between the Backward, More
Backward and Most Backward Classes the year before an election, the Supreme
Court stayed the appointments arguing that “the entire exercise seems to have
been done in a hurry” (Deccan Herald, “Appointment of ‘most backward’ in UP
stayed” 22 January 2002). These continuing dilemmas mean the role of
economic criteria will be one of the most contested issues in debates over
reservations for years to come.
9
Women's reservations and representation
Introduction of the Women’s Reservation Bill was stalled in Lok Sabha on

Monday amid unprecedented scenes of snatching of papers from the
Speaker and the law minister and the virtual coming to blows of
members… As the shell shocked minister stood rooted to the spot, the
member tore the papers with relish and flung them in the air provoking
members from the treasury benches to storm the well. By this time, the
well of the Lok Sabha looked like a veritable battle field with members
from both sides preparing for a scuffle as the Speaker adjourned the House
for the day.1
What caused such a commotion in the lower house of India’s Parliament in July
of 1998? The Women’s Reservation Bill was an attempt to reserve 33 per cent of
seats in Parliament and state assemblies for women. The percentage of women in
India’s Lok Sabha reached 8 per cent in 1984 but has since stagnated (Narayan
et al. 2002).2 In a society characterized by many forms of stratification, demands
for subquotas within the category of “women” for other disadvantaged groups
have repeatedly squelched the bill’s progress. The Women’s Reservation Bill that
caused such a stir in the Parliament would reserve for women one-third of all
seats, including seats reserved for Scheduled Castes and Tribes, but it did not
differentiate between Other Backward Class or Muslim and more privileged
women. Defining which social categories should be eligible for such reservations
leads to heated disputes because these questions involve both emotional
commitments to group identities and material calculations of group interests.
Reservations in India apply not only to government jobs and university
admissions, but also to legislative seats. Scheduled Castes and Scheduled Tribes
currently have reserved seats in Parliament. For example, only members of
Scheduled Castes can run for the seats reserved for Scheduled Castes, although
all voters in the districts designated to fill the reserved seats can vote in the
election, whether they are a member of the Scheduled Castes or not. Women,
Other Backward Classes, and religious minorities do not currently have reserved
seats in Parliament. Prior to independence, some seats in colonial legislatures

were reserved for religious minorities, women and “depressed classes” (the
forerunners of contemporary Scheduled Castes), but reservations on the basis of
religion and sex ceased at independence. A 1993 constitutional amendment
included a reservation of one-third of local council seats for women, and the
Women’s Reservation Bill is an effort to extend this model to the level of the Indian
Parliament.
An unfortunate consequence of these policies to help disadvantaged groups in
Indian society is that competition over the allocation of reservations often
divides groups that might benefit from working together to fight inequality. A
case in point is the repeated downfall of the Parliamentary reservation proposal
for women due to demands for subquotas within the category of women for Other
Backward Classes and for Muslims. Each category is arguably made up of victims
of historical discrimination and inequality; yet claims and counter claims for
reservations on the basis of sex, caste, class and religion have resulted in a
political stalemate. Previous chapters have illustrated how overlapping identities
can prevent the reification of religious and other groups in India. This issue of
Parliamentary reservations illustrates a possible downside to this dynamic,
namely political gridlock due to competing arguments by various disadvantaged
groups. Notably, many of these groups, especially women, are both competing
and overlapping. The politics of competing inequalities can divide oppressed
groups, which, if consolidated, would have the potential to be numerically and
politically dominant.
The issue of reserved legislative seats for women in India has reappeared
without resolution at different historical moments. The following discussion
focuses on three such moments, the debates over constitutional reforms in the last
years of the British Raj, the disagreements over a major government report on
the status of women in 1974, and the demand for the Women’s Reservation Bill
in the 1990s. In each period, the key question concerns the extent to which
women constitute a legitimate group for reservations in their own right. In three
very different contexts, similar arguments resurfaced. A recurring theme was
that women are merely a “minor minority” or “category” within other
communities; some are oppressed and some are not. This conception of women
shaped the policy debates and the policy outcomes, so that women’s
reservations, if instituted at all, have been subdivided along the lines of other
reserved categories. Before turning to the three historical cases, I will introduce
the groups involved and place them in a theoretical context whereby their
respective fates can be understood.
GENDER IDENTITIES AND INTERESTS

When official identification becomes the basis for various material perks,
identities and interests are intertwined. The relationship between the politics of
recognition (of different identities) and the politics of redistribution (of material
resources) suggests that solidarities are constantly emerging and shifting on the
basis of both the emotional pull of certain identities and rational calculations of
interests (Hoskins and Rai: 1998). People have multilayered identities, which can
include gender, class, caste, race, religion, ethnicity as well as many others, but
material considerations may highlight one facet of identity. As Madhu Kishwar
writes, “a group or person may begin to assert a particular identity with greater
vigour if it provides greater access to power and opportunities” (Kishwar 1996a:
6).3 For example, Indian women at various times have asserted their identity as
women in order to demand reserved legislative seats. The political power
associated with these seats has important material dimensions, including the
opportunity to have a say in the distribution of state resources. Women in elected
office are in a position to try to “improve their access to the resources that count,
from education and credit to the ownership of land and housing” (Jaquette 1997).
As the lively debate over women’s reservations illustrates, identities and
interests do not neatly coincide, although group-based policies sometimes link
the two. Political scientist Shane Phelan has observed:
Interest talk may make sense if all the members of a group share every
“relevant” social characteristic or submerge difference(s) among
themselves, but this eventuality is increasingly unlikely. In modern
societies, where overlapping social movements and identities are
increasingly present, interest becomes as unstable as identity.
(Phelan 1995:338–9)
Can policies reflect these instabilities rather than reinforce the equation of
interests with single identity-based groups? Perhaps to truly advance the interests
of women, their multiple identities, including gender, caste, class and religion,
must all be taken into account. However, as noted in Chapter 1, even women
who would argue that “women” do not constitute a universal category,
sometimes choose for political purposes to “act as if such a category indeed
exists, precisely for the reason that the world continues to behave and treat
women as though one does” (O’Hanlon and Washbrook 1992:154). In other
words, gender discrimination persists in all strata of women, which is a major
impetus for the demand for an undivided women’s reservation.
Yet, in India, as elsewhere, the category of women is riddled by class,
religious, and countless other cultural divisions, which can result in different
degrees of disadvantage. Economists Jean Dreze and Amartya Sen note the
cumulative effect of layers of oppression, such as when “different sources of

disadvantage are combined (e.g. the handicap of being female is added to that of
belonging to a scheduled caste and living in a backward region)” (Dreze and Sen
1995:114–15). The current dilemma facing the Indian women demanding
reserved seats in Parliament is that their demand is a strong response to
continuing gender-based discrimination and disadvantages in politics, yet their
proposal may not adequately recognize other categories which are significant to
women. In practice, opening the Pandora’s box of subquotas has mired the
women’s reservation bill in endless debates over which groups of women should
receive their own categories of reservations.
A review of the overlapping, yet competing, categories of women, Other
Backward Classes, and Muslims and their eligibility for reservations will set the
stage for the following discussion. Although women had reserved seats prior to
independence and now have reservations in local councils or panchayats, they do
not currently have them in Parliament. The Other Backward Classes are eligible
for other types of reservations, but not for Parliamentary quotas. Muslims had
special rights under the British, including reserved legislative seats, but only
those Muslim communities classified as Other Backward Classes currently benefit
from national-level reservations, and those do not include Parliamentary seats.
The issue of women’s reservations illustrates how India’s intersecting
identities result in competing arguments that some groups are more unequal than
others. As seen in previous chapters, even in the face of state reinforcement of
distinct classifications, a myriad of social groups reassert transecting categories.
Yet, examining debates over women’s reservations shows that, historically, a
sort of hierarchy of disadvantage emerged. People proposing or making
reservation policies repeatedly prioritized “communities” over “categories” and
relegated women to “category” status.4 Women, half of virtually all communities,
seem particularly well positioned to shatter the discrete classifications embedded
in government policies. Women’s current demand for a reservation of their own
challenges some long-held ideas about the legitimacy of women as an
appropriate group for reserved representation.
MINOR MINORITIES: WOMEN'S RESERVATIONS IN

COLONIAL LEGISLATURES
The demand for women’s reservations in legislative bodies is not new; neither is
the controversy over the appropriate categories to use when allocating
reservations. Legislative reservations for women were under discussion during
debates in the 1920s and 1930s over constitutional reforms for India. At that
point, too, the category of women took a back seat to other categories, such as
religion and caste. One official went so far as to place “female suffrage and
legislative seats in the category of minor minorities (religious communities such
as Muslims and Sikhs and caste groups such as the so-called ‘untouchables’ were
considered the major minorities)” (Ramusack 1990:316). Such official
recognition of minorities, in part a response to the concerns and demands of
various groups in society, cleaved the population into major and minor
minorities; notably all of these minorities together are much more than half of
the Indian population and would constitute a formidable political force.
In the last decades of colonial rule, Britain granted Indians limited rights to
serve as representatives in legislative bodies. In part an effort to assuage
nationalists and expand the “circle of collaborators,” such policies also
contributed to “divide and rule” tactics by giving special electoral rights to
certain groups (Nair 1996:122). These policies include the 1909 Indian Councils
Act (based on the Minto-Morely Report), the 1919 Government of India Act
(based on the Montagu Chelmsford Proposals), and the 1935 Government of
India Act. This period also saw the first attempts at forming associations of
women across the entire nation, such as the Women’s India Association (WIA)
in 1917, followed by the National Council of Women in India (NCWI) in 1925
and the All India Women’s Conference (AIWC) in 1927. These major women’s
associations tried to influence the new policies by passing resolutions, sending
delegates to conferences, submitting memoranda and letters, and lobbying
various decision-makers. The provincial legislatures gradually extended suffrage
rights to women between 1920 and 1929, yet the franchise was so limited, by
property qualifications, for example, that only 14 per cent of men and 1 per cent
of women could actually vote. While the women’s associations’ primary
political goal was to have more women enfranchised, they also became involved
in the issue of reserved seats for women in legislatures. Although these major
Indian women’s organizations ultimately opposed reserved seats, reservations
were granted to women in 1935.
At that point some “major minorities,” such as Muslims, had already received
reserved seats as well as separate electorates. Reservations meant only Muslim
candidates could run for seats reserved for Muslims. Separate electorates meant
that only Muslim voters could cast ballots for those seats. Muslims received
separate electorates under the Government of India Act of 1909. The Government
of India Acts of 1919 and 1935 granted Muslims, Sikhs, and Christians reserved
seats and separate electorates. “Depressed Classes” (lower caste groups) were
provided a few nominated seats in 1919, more in 1925, and even more elected
seats in 1932, but they did not get separate electorates (Galanter 1984, Metcalf
1995). These proliferating categories, some nationalists argued, facilitated
continuing British control.
Concerned about divisions within the nationalist movement, the leading
nationalist organization, the Indian National Congress, objected to special
electoral rights for any of these groups. The major women’s organizations, in
turn, came to oppose similar proposals for women. Women associated with other
political organizations held a different position. Muslim activist Begum Jahanara

Shah Nawaz, for one, agreed with the Muslim League’s support for reserved
seats and special constituencies for Muslims and also, initially, supported
reservations for women. When the Indian National Congress boycotted the first
Round Table Conference on constitutional reform in 1930, the British
government nominated Nawaz and another non-Congress delegate, Radhabai
Subbarayan, to represent women’s interests at the conference (Visram 1992:35).
These two women submitted a memorandum in favor of reserved seats for
women, arguing that “we regard the phrase ‘a fair field and no favour’ at the
present time as an illusory one” (quoted in Pearson 1989:205). The major
women’s associations agreed with these delegates’ demand for adult suffrage, but
rejected the reservation of seats.
When the Congress decided to participate in the second Round Table
Conference in 1931, the president of the All India Women’s Conference
(AIWC), Sarojini Naidu and Begum Shah Nawaz introduced a joint
memorandum of the three major women’s associations, the AIWC, NCWI, and
WIA (Ali 2000:177). This memorandum stated that “to seek any form of
preferential treatment would be to violate the integrity of the universal demand
of Indian women for absolute equality of political status” (quoted in Pearson
1989:207). Although previously the WIA supported reservations for women as a
“transitional necessity,” they signed the joint resolution, thus allowing the
nationalist agenda to supersede concerns specific to women (Pearson 1989:206).5
Sarojini Naidu, influenced by Mohandas Gandhi and the Indian National
Congress, stressed national unity over group rights, speaking critically of “those
purely artificial questions of vulgar fractions, that arithmetic which divided a
power into little fractions for this community and that community” (quoted in
Ali 2000:177). Although Begum Shah Nawaz, like the Muslim League, no
longer supported women’s reservations, Radhabai Subbarayan returned to the
second Round Table Conference in 1931 as well, and she continued to support
women’s reservations.
Skeptical of the major women’s organizations’ emphasis on adult franchise,
Radhabai Subbarayan argued that the vast majority of Indian women were still in
“a state of civic inertia” having “not yet attained selfconfidence or political
consciousness.” She felt that reserved seats would be “one practical step” to ensure
women’s presence in politics (quoted in Ali 2000:176). But support for the cause
of reservations was dwindling among the national women’s associations, who
prioritized equal voting rights. The WIA’s Muthulakshmi Reddi subsequently
wrote in opposition to reservations, emphasizing the need for a “common
platform”: The “only way to bring the Brahmans, the women and the pariahs
together on a common platform is by enfranchising the women and the depressed
classes on equal terms with others” (quoted in Pearson 1989:208). In other
words, if women and other politically marginalized groups wanted equal rights to
vote, they should not demand special rights to seats.
After the second Round Table Conference, the government selected Radhabai
Subbarayan as a member of the Franchise Committee, established in 1931, to
make recommendations on franchise issues in India. The selection of Subbarayan
indicated that the government was not ready to grant India the full adult suffrage
advocated by the nationalist women’s organizations but was open to reserved seats
(Ali 2000:183–4). Indeed, the 1935 Act included limited suffrage rights with
qualifications that still prevented many men and even more women from voting.
Women could vote if they were literate, owned property or married men with
property. The Act enfranchised one woman for every five men enfranchised
(Visram 1992:38–9). Despite the stance of the major Indian women’s
organizations, the Government of India Act of 1935 granted women 41 reserved
seats in the provincial legislatures, as well as limited reservations in the central
legislature.6 Despite their dim view of such a “minor minority,” the colonial
government had added women to the list of groups with reserved seats. Yet, their
overriding concern with the major minorities affected even the women’s
reservations, which were subdivided on a religious basis. Various “communities”
had already been granted not only reserved seats but separate electorates. There
were several women’s seats in Muslim constituencies and one each in Sikh,
Indian Christian, and Anglo-Indian constituencies.
This division of the electoral pie brought out tensions along both gender and
religious lines. The major women’s organizations, such as the AIWC, including
some of its Muslim women members, protested that “the communal award will
divide us, Indian women” (quoted in Pearson 1989:210). Some male members of
“major” minorities, in turn, were disgruntled at having their quotas diluted by
women. Muslim leaders in Punjab, for instance, were “angry that of the few seats
for Muslims, one was reserved for a woman” (Forbes 1996:196). Muslim women
in particular fell between the cracks of the “major” and “minor” minorities
during this period prior to independence and the Partition of India and Pakistan;
in addition to facing such resistance from Muslim men, some Muslim women
were becoming estranged from the Hindu dominated women’s movement (Forbes
1996:196–203). For example, some Hindu women’s activists eventually gave up
on their preference for general rather than separate, communal electorates and
argued that if Muslims were to get separate electorates, then the general seats
reserved for women should be reserved for Hindu women only (Ali 2000:191).
When granted reservations in spite of themselves, women’s associations made
the most of the situation. The AIWC initially considered refusing to participate in
the new constitutional provisions, but eventually they resolved to take advantage
of them. Various women’s groups even lobbied for additional seats in their
areas. In the 1937 elections 80 women became legislators, giving India the third
highest number of female legislators in the world after the United States and the
Soviet Union (Visram 1992:39).
After independence in 1947, the new government retained legislative
reservations for the Scheduled Castes and Scheduled Tribes in the Lok Sabha
(the lower house of Parliament) and in the Vidhan Sabhas (the lower houses of
the state legislatures). Anglo-Indians also got a few nominated seats. The post-
independence Constitution did not reserve legislative seats for the Other
Backward Classes, religious minorities, or women. The Constitution also
outlawed separate electorates for national and state assemblies, so all voters in
the appointed districts elected candidates for reserved seats. Although women’s
reservations in the waning days of the colonial era were quite short-lived and
subdivided along religious lines, they gave women a foothold in Indian
legislative life and set a precedent which women could draw on decades later.
COMMUNITIES AND CATEGORIES: RENEWED

DEBATE OVER WOMEN'S RESERVATIONS
Independence brought a “lull” in feminist campaigns (Kumar 1995:60). The
Indian National Congress became the ruling Congress Party, and it incorporated
a number of feminists into the government. The 1950 Constitution declared
women to be equal and granted universal adult suffrage. But optimism about
freedom and development in the early years of Independence gradually waned as
the status of women seemed to stagnate or even decline. In spite of dramatic
exceptions, such as the election of Indira Gandhi as Prime Minister in 1966,
Indian women did not achieve the equality constitutionally granted to them and
remained grossly under-represented in legislatures. In the early 1970s, the
government of India carried out a comprehensive study of the status of Indian
women. The resulting 1974 report, Towards Equality, was the most
comprehensive government report on women in India. In the process of
documenting continuing inequalities and considering possible solutions, the issue
that most divided the committee was legislative reservations for women.
To place this report in context, the western feminist movement was at its peak
and the United Nations was starting to focus more on the status and development
of women in the Third World. Partly in response to this new UN agenda,
culminating in the 1979 Convention on the Elimination of All Forms of
Discrimination Against Women (CEDAW), the government of India in 1971
formed the Committee on the Status of Women in India in order to gather new
data as well as compile and analyze existing information (Bumiller 1990:26).
Towards Equality was released in time for the 1975 observance of the
International Women’s Year. In addition to the international push for a
reconsideration of women’s status, Indian women began to push for change
during the 1970s. Both the report and the renewed activism brought the
debate over the status of women back onto the national stage. The woman in
charge of the report, Political Science Professor Veena Mazumdar was
previously not a major player in the women’s movement, but she was so appalled
by her findings that she became a leading figure in the “new wave” of Indian
feminism (Bumiller 1990:125–7).
Towards Equality reported on women’s demographic, socio-cultural, legal,
economic, and educational status, evaluated current programs and policies, and
made several recommendations. Although it was a government report, it was
quite critical of the government. The report addressed the economic plight of
many women since Independence and the reluctance of legislators to put
Constitutional ideals into practice. “Large sections of women have suffered a
decline of economic status,” it concluded. “Every legal measure designed to
translate the Constitutional norm of equality or special protection into actual
practice has had to face tremendous resistance from the legislative and other
elites” (Committee on the Status of Women 1974:301). Such findings set the
stage for the debate in the committee over reserved seats for women in
legislative bodies. Those in charge of the report, like Mazumdar, came from
backgrounds far more privileged than those of most Indian women, yet they
attempted to reflect diverse viewpoints. The issue inspiring the most divergent
views, and necessitating the addition of several “notes of dissent” at the end of
the report, was legislative reservations for women.
Mazumdar and her committee, particularly those from the “pre-independence
generation,” initially had no intention of considering the issue of legislative
reservations for women. In the tradition of the nationalist women’s organizations,
they had “never been supporters of special representation or class representation
in any form” (Committee on the Status of Women 1974:355). They still
associated such reservations with colonial strategies and “in academic discussions
we had often criticized the system of reservations for Scheduled Castes and
Scheduled Tribes as a legacy of the historical period which institutionalized the
backwardness of certain sections of our population” (Committee on the Status of
Women 1974:355). The committee’s initial interviews and surveys did not
include any questions on the issue of reservations for women. “Only when the
problem kept being posed repeatedly before us by various groups of women in
the course of our discussions did we become aware that a problem like this was
real,” Mazumdar confessed (Committee on the Status of Women 1974:355). By
the time the report was released, Mazumdar had embraced the concept of
legislative reservations. She personally even espoused reservations in
Parliament, going beyond the committee’s recommendation to limit them to the
municipal level. Towards Equality signaled a shift towards more open
acceptance of reservations among some women, even those previously opposed
to such measures.7 At the same time, the committee was very guarded in its
limited endorsement and included many familiar arguments against
reservations. Although the report included summaries of arguments for

reservations up to the level of Parliament, the committee as a whole concluded
that they could not recommend such a step.
Major arguments against reservations for women continued to involve
concerns about other disadvantaged groups, namely religious minorities and
lower classes and castes. Two types of arguments along these lines emerged in
the debate over Towards Equality. One argument was that the inequalities faced
by these other “communities” outweighed those faced by the “category” of
women. This argument parallels the colonial distinction between major and minor
minorities and subsequent decisions to grant women’s reservations at a later date
than those for other groups and, even then, only with communal subdivisions. A
second type of argument in Towards Equality had precedents in the nationalist
movement’s anti-reservation stance. This argument raised the specter of national
disintegration in the wake of such official distinctions between groups, be they
religious, caste or gender based. Clearly, concerns about national unity continued
after independence.
The report’s conclusions drew on both types of arguments against reservations,
prioritizing “communities” over “categories” and expressing concern about
national integrity. In terms of communities, it considered fallacious “the entire
argument for separate representation for women. Women’s interests as such
cannot be isolated from the economic, social and political interests of groups,
strata and classes in the society” (Committee on the Status of Women 1974:304).
This conclusion depended on the idea that women are a “category,” which, in
contrast to other groups, does not constitute a “community.” The Committee
admitted, “Though they have some real problems of their own, they share with men
the problems of their groups, locality and community. Women are not
concentrated in certain areas confined to particular fields of activity. Under these
circumstances, there can be no rational basis for reservation for women”
(Committee on the Status of Women 1974:304). Although not using the word
“caste,” the Committee contrasted the plight of women with the spacial and
occupational segregation characterizing the caste system. The word
“community” in the Indian context had become a common euphemism for castes
and religious groups, reinforcing the Committee’s distinction between women as
a category and these other groups as communities.
The Committee also included the national unity argument against reservations:
“Such a system of special representation may precipitate similar demands from
various other interests and communities and threaten national integration”
(Committee on the Status of Women 1974:304). In spite of these concerns, it
made some limited recommendations for reservations for women at the local
level as a “transitional measure” (Committee on the Status of Women 1974:303).
This careful choice of phrase echoed the WIA’s initial support of reservations
for women in 1931 as a “transitional necessity,” prior to their decision to join

with the other two major women’s organizations in opposition to reservations.
Even the recommendation of limited and local level reservations sparked notes
of dissent, appended to the report. Phulrenu Guha seconded the Committee’s
discomfort with prioritizing the category of “women” over class divisions by
arguing that “there is a possibility that reservation of seats will only help women
of a particular class who are already privileged. It should be our aim to see that
the masses of women of all classes become equal partners with men in all senses
in our society” (Committee on the Status of Women 1974:354). Guha’s note of
dissent also embraced the national unity argument that “this type of reservation of
seats might lead other communities/classes to argue for reservation of seats.
This, to my mind, will encourage separatist tendencies and hamper national
integration” (Committee on the Status of Women 1974:354).
Not all notes of dissent, however, opposed women’s reservations. Some
women argued for more reservations. In their note of dissent, Vina Mazumdar
and Lotika Sarkar argued that the committee’s reported recommendations did
not go far enough. They advocated legislative reservations at higher levels. They
claimed that the number of women in Parliament was still less than 5 per cent,
only marginally higher than the proportion elected in the Central Legislature
after the 1935 Government of India Act. These dissenters discounted the
community versus category assumptions of many critics of women’s
reservations, arguing that helping women will help the women of various
communities as well. “Larger numbers” of women in Parliament, they proposed,
“will also help to break the somewhat exclusive class composition of this group”
(Committee on the Status of Women 1974:357). Mazumdar and Sarkar also
rebutted the argument that reservations would fracture the nation, or, as they
described it, “the argument that special representation might precipitate
fissiparous tendencies” (Committee on the Status of Women 1974:357). Here
they deployed the argument that because women constitute a widespread
category and not a community, they should have special reservations.
Contrasting women with other minority communities, they argued that
reservations for women could not create the “isolated pockets” feared by critics
of reservations (Committee on the Status of Women 1974:357).
Although the rhetoric shifted from a distinction between “major” and “minor”
minorities to one between “communities” and mere “categories,” the major
conclusions of the Committee on the Status of Women continued the tradition of
ambivalence towards women as a unitary category for the purposes of public
policies. From the beginning, the report noted that “the inequalities inherent in
our traditional social structure, based on caste, community and class have a very
significant influence on the status of women in different spheres” (Committee on
the Status of Women 1974:3).8 One legacy of colonial policies was the
continuing primacy of caste and religious communities, even in the eyes of many
members of the Committee on the Status of Women. This view of Indian society
undermined demands for a special reservation for “women” as a group. Since
independence, fears that the nationalist movement might disintegrate had
transformed into fears that the nation itself might disintegrate. This view of the
Indian nation and fears of “fissiparous tendencies” further undermined the
possibility of a new form of group-based reservation. Both types of arguments,
with their roots in the colonial era, limited the prospects for women’s reservations.
ELECTED BODIES: WOMEN'S RESERVATIONS IN

THE 1990s
It’s a step, but it’s not going to deal with all the problems that women face
because, then again, politics is not only elected bodies. Politics is what’s
happening around you and how you’re treated on the streets.
(Feminist leader Brinda Karat, on reservations for women, interview 10
December 1996)
Towards Equality's proposal for local level reservations for women eventually
became part of the 73rd Constitutional Amendment Act, which revitalized the
local government system known as panchayati raj. Introduced by Prime Minister
Rajiv Gandhi in 1989, the amendment passed in 1993. Consequently village
panchayats or councils must reserve one-third of their seats for women. They must
also reserve seats for Scheduled Castes and Tribes in proportion to their
populations in the area; one-third of SC and ST seats are reserved for SC and ST
women (Singh 1994, Bakshi 1996). Notably, the “category” receives the quite
arbitrary allocation of one-third of seats, while the “communities” receive
proportional representation.
Former Government Minister Margaret Alva nevertheless sees women’s
reservations, even limited to one-third of seats at the local level, as an important
step in a gradual process. Alva advocated reservations on all levels, including
Parliament, in the 1980s when she was Minister for Women.9 She viewed the 33
per cent reserved seats for women in the panchayats as a stepping stone to
reservations in Parliament. “Let us start with the panchayats,” she declared.
“Instead of taking on everybody, let’s start with the panchayats and see how it
runs, and then move upwards” (Alva interview 19 December 1996). Alva felt
such gradualism would quell the doubts of those who believed that “you won’t
find women to contest, you know; they are not educated; they are not trained.
How do they do it?” (Alva interview 19 December 1996). Alva could then point
to the success of the initiative:
In the course of the last three years, between ‘93 and ‘96, one million
women have been elected to local bodies. Now one million…is more than
the population of some countries. But one million women today are elected
representatives in local bodies all over the country. Now just imagine if
one million have been elected, at least five million have contested… Five
million women have gone through the political process.
(Alva interview 19 December 1996)
The adoption of this constitutional amendment, with the endorsement of major

members of the Congress Party, signaled the increasing acceptability of women’s
reservations; yet numerous bills attempting to extend such reservations to the
national level failed. The major arguments against these latest proposals repeated
those of the past.
The 1990s brought several movements pushing for more equal representation
into conflict with each other. The competing demands of women, Other
Backward Classes and Muslims for legislative reservations have led to political
gridlock. In 1990, the Indian government granted reservations in central
government jobs, although not in legislatures, to the Other Backward Classes. This
national recognition gave the previously more diffuse class-based and caste-
based critiques of women’s reservations a political grounding. Like the 1998
scene in the Indian Parliament described at the opening of this chapter, other
attempts in 1996 and 2000 to introduce a women’s reservation bill succumbed to
disputes over recognition of the Other Backward Classes within the category of
women.10 Minister of Parliament Madhukar Sarpotdar argued in the Lok Sabha
debates over the 1996 bill: “Would the skies have fallen on the nation if it [the
women’s reservation bill] had been kept pending or had been referred to a Select
or Standing Committee and then, once and for all, a comprehensive Bill in this
regard was brought forward?… It should have been brought after involving every
section and after proper deliberations…. [W]hat about the people from the Other
Backward Classes who have not been included in this bill?” (Lok Sabha debate of
13 September 1996 over the Constitution (Eighty-first Amendment) Bill, 1996).
Muslim demands for subquotas have also competed with women’s demands.
Competition between groups defined by gender, class, caste and religion have
stymied a purely gender-based reservation scheme at the national level.
The major critics of Parliamentary reservations for women, again, distinguish
between the claims of the “major minorities” or “communities” and those of
women. “National unity,” however, is less prominent than in previous debates.
Some critics do not necessarily want to thwart the bill but nevertheless point out
its limitations for Backward Classes and Muslims. Their concerns are reminiscent
of Towards Equality and its recognition of diversity within the category of
women. Other critics are incumbent politicians who do have an incentive to
thwart the bill because a 33 per cent reservation for women in a legislature that
peaked at 8 per cent women would put their seats at risk. Some incumbents have
rather abruptly become concerned about various disadvantaged communities and
squelched the bill on the floor, in spite of the decision of every major party to
support the bill in their 1996 party platforms. Such tactics resemble group-based
policies under the British, which strategically both appeased and divided various
groups. By endorsing the bill in party platforms and then failing to pass it out of
a sudden concern for backward citizens or Muslims, politicians court the
women’s vote, the backwards vote, and the Muslim vote and simultaneously
protect their own hopes of re-election.
Some women have raised concerns, even while supporting the bill, ranging
from prominent leftist activist Brinda Karat to Hindu nationalist Uma Bharati.
Karat remains skeptical about women’s reservations and has analyzed the
condition of the backward women under the new local level reservations in order
to discover the limits of such policies. Karat concluded that even lower caste
women have benefited a great deal from reservations in local councils, but she
noted that reservations are not always enough to grant them access. She
commented on both the progress and the predicaments of a subcategory of
women, the Scheduled Castes:
What we are finding is Scheduled Caste women who would never have
been given an opportunity to come into…politics, are now coming in.
Unfortunately, in many, many cases, they are, just as the Scheduled Caste
men have been all along, just a rubber stamp… They are not allowed to
participate…we had cases, where they hold the meeting deliberately in the
house of an upper caste person, so the Scheduled Caste women, because of
the social immobility, would censor herself…and so she will be sitting
outside and they would send her the register and she would put her
thumbprint on it. And so you see reservation on its own cannot be an
instrument to remove this.
(Karat interview 10 December 1996)11
Local level reservations sparked some concern even among those supportive of
such a measure, because these policies for “women” were blunt tools and not a
panacea for the problems particular to the women of the lowest castes. Likewise,
in spite of policies aimed at the socioeconomic uplift of the Scheduled Castes
and Tribes, the women within these groups remain “doubly disadvantaged.”
“The multiplicity of social categories in India often serves to obscure the status of
women in the most disadvantaged segments of the population” (Dunn 1993:66).
Vimla Farooqui of the National Forum for Indian Women recognized the mixed
success of the local level reservations but argued that even if only a few of the
elected women feel that they are an important part of the political process, that is
an advance (Farooqui interview 2 December 1996). Due to concerns about
backward women’s access, some local elections, in Goa and Mumbai for
example, have included reserved seats for OBC women within the women’s
quota (Prabhudesai 2000).
Many women’s organizations have failed to build bridges to other
disadvantaged groups. For example, many women’s organizations have not been
terribly supportive of lower castes or of the extension of central government job
reservations to Other Backward Classes. Inspired by minority women’s critiques
of the women’s movement in the West, Brinda Karat’s organization has
promoted the notion that “sisterhood means you have to come out and openly
support Dalit women” (Karat interview 10 December 1996). But when Other
Backward Classes were given central government job reservations, only a few
national women’s organizations defended this policy: There were “middle class
women in the streets of Delhi threatening to kill themselves and coming out with
the most obscene signs” as well as “a section who preferred to remain silent”
(Karat interview 10 December 1996). This political backdrop increases the
tensions between supporters of general women’s reservations and those of an
OBC women’s subquota in Parliament.
For example, Other Backward Class politician Sharad Yadav memorably said
the women’s reservation bill would only benefit “balkati auraten,” or short-
haired women, most likely a reference to upper class, urban feminists.12
Likewise, a newspaper editorial suggesting women’s organizations should
consider “reservations within reservations” critiqued “creamy layer feminism”
(The Hindu 4 September 1998). The leaders of various “communities” are
increasingly questioning feminists’ legitimacy and insinuating that they are not
“real” or “true” women (Kumar 1994: 283). The idea that identity can be
characterized as true or false is highly dubious; yet this notion is reminiscent of
the Supreme Court’s characterization of genuine and spurious identity claims,
discussed in Chapter 2, and can be a politically useful argument in the hands of
communal politicians (Kumar 1994:238).13
Women’s organizations have been ambivalent towards Muslim demands for
reservations as well. Karat claims that Muslims are under-represented in
Parliament, but a Muslim reservation would do little for Muslim women without
the women’s reservation as well. “This is the only way that Muslim women are
going to be able to come out into public life, because even if you have
community representation, they will never allow Muslim women to come in and
represent. No way” (Karat interview 10 December 1996). Muslims, on the other
hand, fear that a women’s reservation would essentially be a Hindu women’s
reservation. Such qualms parallel Muslim fears in the 1930s that extending even
the right to vote, let alone reserved seats, to women would increase the political
power of the Hindu majority, due to the larger number of educated Hindu
women. Thus the discomfort of some Muslims with political rights for women is
not only due to cultural conservatism but also to electoral calculations. In recent
years some Muslim groups have been demanding the reserved seats that they lost
at Independence (Wright 1997). This demand also came up in the context of
Parliamentary debates over women’s reservations in the form of proposals for a
subquota for Muslim women (Sonalkar 1999). An extreme example of the quota-
within-quota argument was “a demand that the quota for Muslim women be
reserved for ‘Backward Class’ Muslims” (The Hindu 12 October 2000).
Some women understand the logic behind the demands for reservations on a
class or religious basis, but feel that women’s reservations should come first.
Margaret Alva, a key proponent of the women’s reservation bill, argued for the
legal recognition of women as a legitimate group for reservations. “Whether one
is fighting for the Scheduled Castes, the Backward Classes, or the minorities—the
largest group that is affected is women,” she contended. “Women are the single
largest group of backward citizens in the country” (Nath 1996:11). Feminist and
Christian leader Jotsna Chatterjee admits that “we have no objection to the OBCs
getting reservations” but first, women should be given 33 per cent reservation,
and “automatically this will apply to every category.” That would mean “that
women will have to be given space in the Scheduled Caste, Scheduled Tribe
section, and then if it is extended to the OBCs, it will have to be also given in the
OBCs and also in the minorities” (Chatterjee interview 22 November 1996).
Chatterjee is a member of a religious minority community, Christians, but since
their educational and socio-economic status tends to be higher than that of
Muslims, Christians have not been inspired to lodge a similar demand for
subreservations within women’s reservations.
These activists remain supportive of women’s reservations, although they
recognize that this policy alone does not adequately address the doubly
disadvantaged backward or Muslim women. Politicians in Parliamentary
debates, however, may have raised concerns about Muslim and OBC women in
part as a way to defeat the women’s reservation bill. Margaret Alva has charged:
“When it was introduced…there was hullabaloo in the House… No man has the
courage to stand up in the House and say we don’t want it, so they had to
sabotage it. Now the only way they could sabotage it is to appeal to caste,
because caste cuts across women” (Alva interview 19 December 1996). Some
activists supportive of women’s reservations describe this as a strategy to divide
and rule women on the basis of caste, class and creed, “splitting hairs” to
continue to hold onto power in Parliament (Ganguli interview 3 December 1996,
Chatterjee interview 22 November 1996). Maneka Gandhi predicted that the
women’s reservation bill “will be diluted and further diluted till you have a law
that says you can have your one third reservation for women provided they have
pink hair, are totally backward, completely unheard of in any political arena”
(Gandhi 1996:18).
A possible example of what these women might call “division” or “dilution,”
Member of Parliament Uma Bharati demanded that Other Backward Classes be
included in the women’s reservation bill. She herself is difficult to categorize, a

backward class woman in the Hindu nationalist Bharatiya Janata Party (BJP),
which is dominated by upper caste men. Although her party’s platform supported
the bill, and she claimed to support it as well, she became a spokeswoman for the
demand to amend the bill to include the Other Backward Classes, an amendment
that led to its downfall. A few years earlier, the Bharatiya Janata Party had
vociferously decried the extension of reserved central government jobs to Other
Backward Classes, a campaign that led to several self-immolations by upper
caste students. At that point the Bharatiya Janata Party was quick to criticize the
reservations for Other Backward Classes as divisive and a threat to national
integrity. Although this sudden concern for OBCs in the context of the women’s
reservation bill seems suspicious, Uma Bharati’s credentials to make such a
proposal are hard to fault: “Since I am from the Backward Castes, I know from
experience that women from the oppressed classes are the weakest of the
weakest section of society” (quoted in Nath 1996:11).14
In short, the discourse of “communities” came up again in the debate over
women’s reservations in the 1990s, at times in ways reminiscent of British divide
and rule tactics and, at other times, in ways similar to the Committee on the
Status of Women’s measured support for reservations. National unity was a
minor theme in this latest round. Even the BJP remained unusually silent about
the threat such group-based policies might pose for Hindus or national integrity,
letting the women’s reservation bill fail largely due to squabbling over proposals
to include subquotas. The competing gender, class, caste and religious minority
movements for reservations have not yet reached a detente.
CONCLUSIONS
For women to get rights is not a very simple thing.

(Vimla Farooqui, National Forum for Indian Women, interview 2
December 1996)
The history of women’s reservations in India illustrates the many ways

administrators, politicians and activists have constructed women’s identities and
interests. Like the British administrator who categorized women as “minor
minorities” in comparison with caste and religious groups, the post-independence
Committee on the Status of Women in India contrasted women as a “category”
with “communities” based on caste or religion, which were still considered more
legitimate political groupings. The extension of reservations to women in local
level legislative bodies signaled some acceptance of the category of “women” as
a legitimate target of such public policies; yet the uproars in Parliament over the
subsequent bill to extend women’s reservations to the Parliament itself
demonstrate that women’s goals are still perceived as competing with other
groups rather than complementing them. Muslims fear Hindu women will
dominate the reserved seats; lower classes and castes argue that privileged
women will prevail. A few politicians see the parallels and linkages between
different disadvantaged groups. For example, former Prime Minister V.P.Singh,
who spearheaded the expansion of reservations for Other Backward Classes,
argued that India “can’t have social justice without justice to women” (Singh
interview 20 November 1996).
The history of debates over women’s reservations, particularly the competition
between various disadvantaged groups, is a rich example of the complexity of
overlapping identities and the tendency of political interests to bring different
identities into relief at different historical points. Attempting to simultaneously
appease and divide the “major minorities,” colonial officials subsumed vast
diversity under the categories of a Hindu “majority” and various religious and
caste “minorities.” The Government of India Act of 1935 superimposed
women’s reservations onto these primary categories.
This legacy lingered in the debate over women’s reservations for the 1974
report, Towards Equality. The Committee on the Status of Women agreed to
recommend limited, local level women’s reservations, in large part due to
overriding concerns about caste and religious communities. One major concern
expressed in the report was that such “communities” had more legitimate claims
to reservations, as opposed to the claims of women, who constitute a mere
“category.” Drawing on similar assumptions, another concern expressed was the
fear that any reservations could encourage national disintegration.
The more recent arguments against reservations for women at the national
level continue to echo these priorities. Throughout the historical and
contemporary debates over women’s reservations, other groupings (religious
minorities, lower classes or castes, or the nation itself) are given precedence.
What makes the persistence of these arguments remarkable is the very different
motivations and actors at play in the three time periods under consideration,
ranging from a colonial power trying to maintain control of an unruly colony, to
a government committee genuinely concerned about the plight of women, to
policy-makers hoping to retain their seats in Parliament.
Why is gender repeatedly singled out as a problematic category for group-
based policies? Is there a sound basis for past distinctions between the category
of gender and communities of religious minorities or lower castes? Do these
latter communities in India more neatly coincide with class, further legitimating
their claims to special policies for disadvantaged citizens? Every sort of group in
India has internal diversity. There are well off Muslims and relatively “forward”
backward classes; yet women have the most internal diversity, since they are a
substantial part of all class groups. If the only purpose of the policies is to help
redistribute power and resources to the poor, gender alone may not be an
appropriate category; yet ignoring gender and using other categories can be
equally problematic. Women who are also in other disadvantaged groups are
often the worst off, and even advantaged women face gender discrimination.
Moreover, legislative reservations in particular are not simply a redistributive
policy but also a means for group recognition and representation. For this
purpose, women may be as relevant a group as any, even if they are scattered
throughout the class hierarchy. As Benedict Anderson pointed out, even people
who may never meet can become “imagined communities” (Anderson 1991). On
the other hand, the internally diverse category of women is particularly prone to
politically motivated re-imagining. Thus politicians could appeal to half of the
Indian electorate when promising Parliamentary reservations for women, but in
the later debates they could argue, for any other constituency, that those seats
would be taken by the women of a rival group. Due, in part, to such political
manipulations and to the diffuse nature of the category itself, women’s
organizations in India have not achieved unity, let alone a broader solidarity of
disadvantaged citizens, encompassing women, Muslims and Backward Classes.
Delineating the minority or backward women, like efforts to target the
noncreamy Other Backward Classes, could avoid reifying a particular category
and reach some of the most disadvantaged members of society. Yet controversies
over defining categories and subcategories may prevent any policy innovations
at all. Policy-makers and activists must walk the thin line between exploring
more complex policies and courting political gridlock.
10
Conclusions
It is a very cruel divide when it comes to daily life. Consider a person of

lower caste in a village. From morning to evening the way he is called by
name, the way he is addressed, the way he is asked to sit—he goes through
101 humiliations in a day… As a human being he suffers. The only way to
really deal with the situation is not to put this reality under a carpet… It is
better to put it on the table: Yes it is there. Let us change the results.
(Former Prime Minister of India V.P.Singh interview 20 November 1996)
Reservation policies designed along group lines pose a dilemma for those who
hope to overcome group stratification. By identifying and classifying certain
groups in society—by putting these divisions on the table—such policies could
further entrench the very boundaries they are meant to diminish. This is a major
argument of reservation policy critics. Such critics include several groups
discussed in this book, ranging from the anti-Mandal student protesters and some
Hindu nationalists to organizations such as the Gandhi Caste Society. Others,
however, argue that alternatives to reservations, such as a caste-blind approach,
would sweep reality under the carpet. Defenders of group-based policies assert
that it is blindness to forms of discrimination like casteism and racism that
perpetuates these problems. Such advocates of reservations include the many
groups, also described in this book, fighting for their own reservations, including
some Muslim and Christian organizations, class or caste organizations, and
feminist activists. In the context of racial discrimination, one legal scholar
proposed a way out of this dilemma, arguing that an “alternative vision to color
blindness is a color-consciousness that seeks to destroy itself” (Wu 1996:185).
Can policies both reflect and destroy the categories associated with
disadvantage?
These arguments over reservation policies parallel some theoretical debates
about state identifications and constructions of identities. Critics of essentialist
notions of caste or race emphasize the fluidity and complexity of identities and
the potential for official classifications to reinforce these identities through static
and unyielding typologies. Other scholars recognize the utility of reconstructing

even negative categories of identity for purposes of political action and policy-
induced change. At the root of these debates over policy and theory are
assumptions about the relationship between identity and the state. People making
these varied arguments hold varying assumptions about whether state categories
can actually construct social divisions, whether state structures can lead to
positive changes in society, and whether societies are inert or active in response
to state identifications. It is vital to consider these common political and
theoretical arguments for and against state classifications empirically. As Henry
Lopez argues in his work on the “legal construction of whiteness,” “[t]here is no
indication that in and of itself the legal construction of race is normatively good
or bad.” This ultimately depends on “what role such construction plays in the
attainment or frustration of social justice” (Lopez 1996:115). My research on
reservations addresses these theoretical and political arguments by examining
official classifications in practice and by analyzing various political responses to
them.
Do reservations reinforce the categories they are meant to undermine? On the
basis of the first half of this book, on state simplifications, I conclude that the
processes of implementing reservations can solidify boundaries, but that these
practices could be improved without doing away with the policies. My research
on the adjudication of identities, the People of India project, the administration
of caste certificates and lists, and the national census demonstrates that
government classifications for the purpose of reservations at times solidify the
boundaries they are supposed to break down. Sometimes judges and bureaucrats
classify liminal people who do not fit neatly into existing categories by
privileging old or “original” identities, such as a convert’s original caste, a
migrant’s original status, or a tribe’s original culture. The implementation and
administration of caste certificates and lists involves labeling, scrutinizing and
ranking disadvantaged groups. Government data on disadvantaged groups,
whether ethnographic or numeric in nature, are a political tool too often used to
divide groups rather than empower them. In all of these arenas of reservation
policy implementation, the divisiveness associated with colonial categories
persists in contemporary India. Even as reservations weaken group distinctions
by opening opportunities to under-represented groups, the process of
implementing reservations can indeed give new emphasis and durability to these
very distinctions.
However, on the basis of the subsequent chapters, focused on political
complications, I conclude that, even given these rigid codifications, people are
not encapsulated by their classifications but rather defy them in a wide variety of
ways. My research on political challenges to reservation categories, on the basis
of religion, caste, class, and gender, demonstrates that the boundaries of official
categories spark debate rather than stop it. Debates over additional reservations
for religious minorities or women show that competition between identity-based

movements may result in political gridlock, impeding the demands of various
disadvantaged groups. Yet demands by religious minorities to be included in
existing categories and demands by women to have a category of their own throw
into question not only the official definitions of disadvantage but also the
definitions preferred by more politically powerful groups, such as the Hindu
nationalists. Mobilizations on the basis of various intersecting identities help to
prevent the reification of categories, so that even Hindu nationalists must juggle
the intertwined loyalties of citizens who vary by caste, class, and region. The
issue of including class and economic factors in the design of reservation
categories suggests that, at times, it is the state that changes the categories and
protest groups that resist, preferring the status quo. Based on these political
complications, I conclude that even strong systems of official identification do
not necessarily reinforce group identities along those lines but can, in fact, be a
catalyst for challenges to these boundaries.
STATE SIMPLIFICATIONS AND POLITICAL

COMPLICATIONS: A THEORETICAL SYNTHESIS
I apply two distinct theoretical approaches to the study of identity-based policies
and politics in my study of reservations in India. One theoretical approach,
introduced and applied in the chapters of Part I, highlights the problems of state
simplifications of complex social systems. Such insights, particularly criticisms
of colonial era reification of identities, shed light on continuities in the
contemporary judicial, anthropological, and administrative practices involved in
implementing reservations. As applied in the chapters of Part II, the other
theoretical approach focuses on the political utility of identity politics for
disadvantaged groups. Although it is often strategically useful to focus on a
single identity, identities are never singular. Thus the multiple, intersecting
movements organized along lines of religion, caste, class, or gender are the focus
of the second part of the book on political complications.
State simplifications for the purpose of policies such as reservations raise
several dilemmas. Forms of group-based discrimination such as casteism or
racism rely on oversimplified categories; therefore groupbased policies to counter
such discrimination often result in simplifications too. Liminal individuals or
groups are essential to undermine casteism or racism, yet group-based policies
necessitate the classification of people who seem to be on the margins of
categories or in multiple categories.
Striking examples of state simplifications are judges and bureaucrats treating
caste, tribe or community as a status determined by birth, with little room for
official recognition of the potential malleability of such a category due to
modernization, migration, marriage, adoption or religious conversion.
Bureaucrats issuing caste certificates collect information and verify identities in

order to ultimately slot each challenging case into an existing category.
Applications to be included on the lists of Other Backward Classes require groups
to rank themselves against other groups and to supply evidence of their place in
the social hierarchy. Census enumerators funnel infinitely varied responses into a
finite number of standard entries. Even state employed anthropologists carefully
organize their ethnographic accounts into the colonial administrative categories
of Scheduled Castes and Scheduled Tribes.
These examples are in keeping with theories emphasizing the state’s tendency
to oversimplify complex identities and the dangers of such state simplifications.
Those scholars who have drawn attention to the pernicious effects of colonial
state classifications can find the lingering effects of this legacy in postcolonial
reservation policy implementation. Those scholars who, in contrast, study the
strategic use of identity by or for disadvantaged groups would do well to
consider the ironic outcomes of implementing group-based reservation policies,
especially the judicial, administrative, and even “scientific” reinforcement of
traditionally disadvantaged categories.
Nevertheless, a group’s sense of identity is not determined by such official
identification processes. The second body of theory I draw on highlights the
political complications impeding such processes, in particular, the politicized
identity-based groups that challenge government classifications. State
simplifications do not negate the variety of identities embodied in each individual,
although they can influence which identities become salient at different
junctures. The editor of the feminist journal Manushi, Madhu Kishwar, writes of
“identity layers,” which at various times come into focus with the strategic use of
an identity: “For the most part, people take these identity layers for granted and
they find expression in their appropriate realms at different points of time,” but
group-based reservations can inspire people to assert the identities that could
increase their access to opportunities (Kishwar 1996a:6). In the preceding
chapters, various examples of political mobilizations to gain reservation benefits
demonstrate that people do tend to rally around the currently recognized
categories. However, in the process, many groups challenge the boundaries of
the existing categories, as in religious minority movements to include new
groups in the Scheduled Castes and Other Backward Classes. In other cases,
groups assert that different categories should be used, such as demands to add
gender-based reservations in national and state legislatures.
Members of subordinate groups reconstruct and utilize categories formerly
used to oppress them. Even caste identities that were previously reviled can be
reshaped into a tool of empowerment, as the testimony of people of Dalit
backgrounds illustrates. A professor of sociology, S.K. Thorat, vividly described
grappling with his stigmatized caste identity as a child and eventually embracing
the positive reconstruction of that identity as a “Dalit”:
An untouchable child, particularly in a village, is subjected to a stigmatized

identity from the time he can begin to walk and to touch things and
people… Later, as an adult, he may either accept or reject this identity, or
do both partially.
(Thorat 1993:67)
He discussed both accepting and rejecting his stigmatized identity, then forming
a new identity and eventually “outgrowing my caste’s boundaries” over the
course of his life (Thorat 1993:66, Thorat interview 29 April 1996). Bishop
Azariah argued that regaining a sense of their own identity is at the heart of the
Dalit movement. Untouchability, an imposed label, “simply takes away the
identity of these people,” and Dalits, a self-defined group, “want to get their
identity as human beings and as members of the human society” (Azariah
interview 10 July 1996). In addition to individuals such as Thorat and Azariah,
lively and diverse movements are challenging the boundaries and meanings of
identities, even in the process of using identity-based categories to fight
oppression.
Current movements, inspired by various forms of identity, complicate the
largely caste-based categories used by the state. Christian, Muslim and Hindu
groups, themselves divided along class, caste, and gender lines, propound
competing visions of the most appropriate reservation categories. Many of these
controversies involve differing views about the existence of castes within these
three religious communities in India. In addition to disputes within and between
these religious communities, some groups are arguing for economic or class-
based alternatives to the current parameters, while others resist such
considerations by denying the existence of any “creamy layer,” or economically
advanced individuals, among the backward classes. Women demanding reserved
Parliamentary seats face competing proposals from backward and Muslim
groups. Even debates over the future census categories and the sudden
population jump in certain Scheduled Tribes are throwing official “boxes” into
question.
All this churning of categories demonstrates that people are not simply
adopting the state’s identifications, however stringent, as their own identities.
Although none of these movements could ultimately ignore the hybrid nature of
identity, these examples are more in keeping with the theories that recognize the
utility of categories, even previously degraded or divided ones such as
untouchables or women, for mobilizing and gaining advantages for one’s group.
Even groups that challenged the boundaries of existing categories did not reject
the use of such categories altogether but, rather, mobilized along different lines
to offer alternatives to the state’s definitions of disadvantage. Most of the
activists I interviewed felt that the potential of benefitting from reservations
outweighed the problems of classification.
Those who note the dangers of official classifications of complex identities

might predict an entrenchment of existing reservation categories in society; yet
the complexity of identity politics, with movements along not just caste lines but
a variety of axes of identity, belies that assumption. At the same time, those who
emphasize the utility of simple categories for the political empowerment of the
disadvantaged would not have expected the degree of mobilization over the
finer, multilayered points of identity differentiation. For example, activists
creatively combine religion and caste, as in the Scheduled Caste Christian
movement, or gender and class, as in the demands for reserved seats for women
of the Other Backward Classes. Moreover, hybrid identities splinter not only
political movements of the disadvantaged but also dominant groups. Some of the
greatest political challenges facing the Hindu nationalists are from groups that cut
across the “Hindu” ranks. Resolving the theoretical debate between those
eulogizing the complexity of identities and those asserting the utility of categories
necessitates grounding these perspectives in contextual case studies. My study of
reservation policies and politics allows me to assess the ways these official
categories simultaneously reinforce and undermine group boundaries.
POLICY IMPLICATIONS
In addition to highlighting the need for a theoretical synthesis, case studies of
reservation implementation and political activism demonstrate that major policy
debates over reservations are not simply polarized between those who are for or
against reservations. Advocates and critics alike are pressing the government to
reformulate the reservation categories rather than embracing or rejecting them
outright. Therefore, I offer some reflections on this third alternative, revising
reservations, particularly focusing on the potential for policies that recognize
disadvantaged groups but better reflect the complexity of identity. Beyond
reservations, this balance is relevant for the design of other policies that involve
group recognition in diverse democracies, whether these policies are political,
economic, social or cultural.
Reservations or other policies dependant on social classifications could be
improved by making the categories more complex and the policies more fluid, as
well as by avoiding a group policy versus universal policy dichotomy.
Reservations or other forms of affirmative action are largely an attempt to
counter the oversimplifications associated with forms of discrimination such as
casteism or racism. Therefore, to a certain degree, these policies are bound to
involve oversimplified categories. Over time, “caste became rigid and water
tight, and it was no more based on ability and functional basis, but on birth. So
anybody who had the birth seal, he would be a Brahmin and [was] supposed to
be in a superior class, although he may be a bloody ruffian” (Prasad interview 18
September 1996). Justice Prasad’s description of the facile assumptions of

casteism has parallels in the following description of racism in the United States:
Race discrimination in this country historically has not allowed for racial
complexity. Government race classifications have been crude because, by
necessity, they reflect the problem they address. Government’s challenge
now is to keep fighting discrimination while recognizing some complexity
in racial identity.
(New York Times, “Multiracial Americans” 8 November 1997:A14)
How might governments face this challenge and recognize the complexity of
identities?
In India, various economic and social criteria feed into the Other Backward
Classes category. However, low caste or other similarly disadvantaged
communities, the original targets of these policies, are taken as the starting point.
This is important because complexity is being recognized without trammeling
those originally intended to benefit from reservations. Adopting “creamy layer”
rules allowed the government of India to recognize caste or community as an
organizing principle of the policy, while excluding individuals who have
arguably already managed to overcome caste hurdles. Notably these rules only
apply to Other Backward Classes, not to the more disadvantaged Scheduled
Castes and Tribes. Given continuing difficulties filling reserved quotas for
Scheduled Castes and Tribes, especially in upper level jobs and prestigious
academic programs, such restrictions are appropriate only for categories in which
it is possible to find “non-creamy” yet qualified applicants for positions at all
levels (Chanana 1993, Dushkin 1979, Planning Commission 1997–2002: Table 3.
9.6). Creamy layer rules enable more fluid classifications, reflecting changes in
status from generation to generation. Maintaining caste as a unit of analysis, to
which other considerations of disadvantage are added or from which a creamy
layer is subtracted, retains the still salient categories targeted by the policies but
adds other dimensions to better reflect change at both individual and societal
levels. With this added complexity, however, comes additional administrative
hurdles, as discussed in Chapters 4 and 8.
Tracing individual progress is not only a means of fine tuning the categories
of beneficiaries but also a way to foster success among those who do benefit.
Once reserved positions are filled, particularly in higher education, those in the
reserved categories can benefit from a more individualized approach to their
continued progress. Veena Das, an anthropologist at Delhi University, argues,
“[a]n affirmative action program means that there should be flexibility.” The
outcome of rather blunt categories is a pool of beneficiaries who may vary
widely in terms of their affluence and education: “Right now Delhi University
has quotas for SCs and STs. But there is no homogeneity in the various groups
for whom quotas have been set.” Thus, “[e]very group or student has to be
treated differently.” This flexibility could encompass the schedule of the degree
program, additional training or preparation, mentoring or fellowships (Das
quoted in “Class and Reservations,” Telegraph-Calcutta 25 June 1995). As
discussed in Chapter 4, employers are to regularly check on people hired in
government jobs in order to make sure their communities are still on the official
lists, and a government regulation demands that employees report any religious
conversion that might make them ineligible for their reserved position. Yet little
is done to track the successes or problems of people in their jobs or universities.
Treating people as a category must end when they are enrolled or employed, or
they too often fall by the wayside.
Additional categories may also better respond to liminal cases. Chapters 2 and
4 illustrate the problem of penalizing those in intercaste marriages in the sense that
the lower caste person may lose reservation eligibility. Such marriages, discussed
in the context of adjudicating and administering identities, are one of the most
positive steps toward breaking down the barriers of caste. Reservations for
people who marry outside of their caste, one proposal of a 1961 End Caste
Conference, continue to hold promise in the eyes of administrators and
politicians (Kumar 1992:298, n.20). To avoid penalizing marriages in which a
low caste women would take on the official status of her higher caste husband,
the Ministry of Welfare has stated that “Inter-caste marriage should be
encouraged. To encourage non-SC male youth marrying unemployed SC girls,
incentive of jobs outside the reserved quota may be considered” (Ministry of
Welfare 1990:75). Government Minister Ram Vilas Paswan advocated a
reservation category for those who have intercaste marriages:
In India you can change religion, you can change the party, you can become
rich, rich can become poor, but you can’t change your caste. So caste is just
like a rock. So the only process where the caste system can be weakened is
intercaste marriage… If the reservation is made on that ground, intercaste
marriage, then slowly, slowly caste system will be abolished. And if there
is no caste then there will be no reservation on the basis of caste.
(Ram Vilas Paswan interview 20 December 1996)
In this way additional categories cannot only expedite change but may eventually
make the old categories obsolete.
In addition to recognizing rather than penalizing those who transgress
boundaries, another way to incorporate social change in public policies is to
reconsider them at regularly scheduled times. The alternative, reconsidering the
policies when they are under attack, is less likely to result in incremental
changes, as advocates and critics entrench themselves in their positions. Syed
Shahabuddin, an activist for reservations for Muslims, proposed a survey every
ten years by a competent authority to determine who should benefit from

reservations. This would be a methodological challenge, although a survey is not
as daunting as the proposals to collect comprehensive data on all castes in the
decennial census, discussed in Chapter 5. A survey might provide data to help
adjust, increase, decrease, and eventually phase out reservations (Syed
Shahabuddin interview 11 September 1996). In the case of reservations for the
Other Backward Classes, a ten-year review of the OBC lists is supposed to be
carried out by 2003 to “take out those communities which may have ceased to be
backward” (D.L.Sheth interview 6 September 1996). Legislative reservations are
also to be reviewed for possible extension every ten years.
Such regular reviews can facilitate more fluid policy categories, as long as
they are used as an opportunity to comprehensively assess and fine tune the
policies rather than simply extend them (Bhatnagar 2001). Review becomes all
the more necessary when socio-economic changes affect not only the definition
of the targeted groups but also the relevance of the reserved opportunities, now
confined to the public sector. With economic liberalization, the administrative
service is no longer as dominant a gateway to power and status in India (Heeks
1996). As elites take key positions in the increasingly important private sector,
various groups and activists, including several I interviewed, have proposed
private sector reservations (Mehta 1997, Times of India, “Liberalization deprives
SC/STs of Rights” 9 December 2001). Private professional colleges, especially
medical colleges, also can reinforce traditional class and caste advantages
(Dushkin 1979:663). This question regarding types of reserved opportunities is
beyond the scope of this study of categorizing beneficiaries; yet changes
associated with liberalization are likely to throw both the definition of the
reservations and the definition of the beneficiaries into question. This changing
context reinforces the importance of regularly reconsidering reservations, rather
than over-hauling such policies only when they are under fire.
Polarized debates can degenerate into endless arguments pitting the merits of
group-based policies against the merits of universal policies. These types of
policies are not an either/or proposition:
Let everybody have the same educations, the same house, the clothing,
same food. There is no need of any reservation then. Nobody wants to lock
the society into compartments. Let the existing compartments break, these
compartments of thousands of years... But the existing reality has to be
reorganized.
(Prime Minister V.P.Singh, speech to Rajya Sabha,
27 August 1990)
Former Prime Minister V.P.Singh, a staunch supporter of reservations in India

who was responsible for extending national-level reservations to the Other
Backward Classes, also emphasized the key role of universal policies, going so
far as to say “I think a universal policy is the better way because then these
distinctions or classes don’t apply” (V.P.Singh interview 20 November 1996).
This is not a contradiction. Wider social and economic programs complement
reservations by reaching the many disadvantaged citizens that reservations
inevitably miss.
Policy-makers may be tempted to use reservations as a cheaper, symbolic
“fix” for wider social disparities. It is unfair to criticize reservations for not
helping each member of disadvantaged groups to gain a good education and job,
since reservations cannot solve the problems of inequality and mass poverty on
their own. Likewise, it is also a mistake to use the existence of reservations as an
excuse to avoid more fundamental, universal policies. In the words of sociologist
M.N.Srinivas, “It is unfortunate that reservation is widely regarded as a panacea
for ills such as poverty, and lack of access to education, government
employment and political power. Reservation has its uses but only up to a point”
(Srinivas 1997:4). In Chapter 9, discussion of continuing discrimination against
low castes, particularly women, included feminist activist Brinda Karat’s
argument that “reservation on its own cannot be an instrument to remove this”
(Karat interview 10 December 1996, emphasis added).
India, like many other countries, would avoid some of the extreme disparities
in life chances that such affirmative action programs can only partly alleviate by
providing and enforcing attendance in consistently good primary and secondary
educational facilities (V.P.Singh interview 20 November 1996). The sooner such
universal policies are effectively implemented, the sooner group-based policies
may become unnecessary. Yet “universal” policies to date have had a far from
uniform impact on the life chances of citizens. Despite a constitutional
commitment to universal primary education, about one half of the population as
a whole remains illiterate and almost two-thirds of Dalits remain illiterate
(Human Rights Watch 2001). According to reports between 1996 and 1998 by the
National Commission for Scheduled Castes and Scheduled Tribes, Dalit children
have a drop-out rate of close to 50 per cent at the primary level, a rate that climbs
to 78 per cent for secondary school (Human Rights Watch 2001). Since universal
policies seem unlikely to dramatically reshape Indian society any time soon, it is
important to recognize the continuing role of group-based policies and to see that
these two policy approaches work best in tandem.
Although the previous chapters inspired these policy suggestions for more
flexible categories, these chapters also highlight some of the political challenges
of changing reservation policies as well as the administrative and legal
challenges of implementing more malleable categories. Other group-based
policies in India and elsewhere have proven to be difficult to change (Jenkins
1998). “Once policy makers have redefined the disadvantaged to encompass the
majority, the decision becomes virtually irreversible” (Weiner and Katzenstein
1981). Political momentum behind current definitions of disadvantage may make

changes to the policies very difficult but not impossible. This is demonstrated in
Chapter 8 by the extension of Other Backward Class reservations and the
restrictions on the “creamy layer” in the face of political opposition to such
changes to the status quo. Another political challenge to policy adjustment is that
one proposed change sparks a plethora of other proposals, as seen in Chapters 6,
7 and 9, in which various movements prioritized categories based on religion,
caste, or gender.
In spite of the difficulty of implementing changes, both the Indian state and
Indian society, at least at times, recognize the complexity of socio-economic
disadvantage in India and the need for more flexible policies. In other words,
both government officials and political activists argue that policies should reflect
social complexity and change. For example, one rather striking government
memorandum goes so far as to describe India as a “multiple undulating society”
(Ministry of Personnel Office Memorandum No. 36012/31/90-Estt (SCT)).
Likewise, a Dalit political organization argues that “Programs for the Scheduled
Castes or Scheduled Tribes shall not be made like Procrustean beds” (National
Action Forum for Social Justice 1996:11, emphasis in the original). Procrustes,
according to Greek mythology, was a giant who stretched or squashed his
prisoners in order to fit them into standardized beds. Governments sometimes
stretch or squash people to fit them into official categories, but people do
continue to defy categorization.
CONCLUSION
Contemporary state practices of identification shape identity politics, yet
contemporary protest groups also demonstrate agency by challenging static
classifications. The official categorizations of identities used for reservation
policies not only are embedded in policies, but also are adjudicated by the courts,
given scientific clout by official anthropologists, listed and certified by
administrators, and tracked by the national census. Although judges,
anthropologists and administrators face the ambiguity and liminality of identities
in practice, the judicial, scientific and administrative processes associated with
implementing these categories reinforce neat and static definitions of
disadvantage.
Official categories imposed by the state shape identities but do not determine
them. Various protest groups in India are challenging the classifications of
citizens used to implement reservations. When abstract state classifications and
complex social identities clash, the resulting melange of competing demands
helps to prevent the reification of such classifications. Official categories do
create boundaries, yet disagreements over reservations in contemporary India are
often arguments about the boundaries themselves rather than disputes between
groups neatly divided by those boundaries. Since reservation policies are

ultimately supposed to make themselves obsolete, tensions and uncertainty over
the most appropriate categories to use can be a sign of progress rather than
trouble. Even ambiguous claims to belong in the ranks of the officially
disadvantaged challenge the rigid stratification and hierarchies these policies
were meant to undermine. State simplifications and political complications
together construct identities, as when individuals and groups in India claim to be
backward in order to move forward.
Appendices
182 APPENDICES
APPENDIX I:
GOVERNMENT DOCUMENTS
(a)
Application form for Scheduled Caste certificate
APPENDICES 183
184 APPENDICES
(b)
Form of certificate to be produced by a candidate
belonging to a Scheduled Caste or Scheduled Tribe in
support of his claim (Department of Personnel 1993:396)
APPENDICES 185
(c)
Application for a certificate of eligibility for reservation of
jobs for Other Backward Classes in civil posts and services
under Government of Delhi
186 APPENDICES
APPENDICES 187
188 APPENDICES
APPENDICES 189
190 APPENDICES
APPENDICES 191
(d)
Form of certificate to be produced by Other Backward
Classes applying for appointment to posts under the
Government of India (Department of Personnel 1993:487)
192 APPENDICES
(e)
Questionnaire for consideration of requests for inclusion
and complaints of under-inclusion in the central list of
Other Backward Classes
QUESTIONNAIRE
PART I
General Descriptive Da ta o f the State
1. Name of the State

2. Population of the State
3. Population of the State as percentage
of All India Population
4. Population of OBC in the State
5. Percentage of OBC population to the
total State population
6. Percentage of the State OBC
population to the All India OBC population
7. Scheduled Caste population in the State
8. Percentage of Scheduled Caste population
to the total State population
9. Percentage of Scheduled Caste
population to the All India Scheduled Caste population
10. Scheduled Tribe population in the State
11. Percentage of Scheduled Tribe population
12. Percentage of Scheduled Tribe population
to the All India Scheduled Tribe population
13. Total population of Backward Classes
(SCs+STs+OBC) in the State
14. Percentage of Backward Classes population
15. Percentage of total Backward Classes
population of the State to the All India
Backward Classes population
16. (a) When was reservation for OBC in the
services of the State started?
(b) What was the percentage of such
reservation for OBC then?
(c) Furnish changes if any, in respect

APPENDICES 193
of reservation for OBC and its percentage ,

made from time to time
(d) What is the present percentage of
reservation for OBC?
(e) Furnish statistical data of the quota of reservation
for OBC and its fulfilment year-wise for the
last ten years indicating shortfalls, if any
(f) Reasons for shortfalls and remedial measures
taken
17. Has the State government set up any machinery/cell for

monitoring the implementation of the reservation scheme
for the OBC in the State?
If so, furnish particulars
18. What steps has the State government taken and proposes
to take for helping and enabling the OBC to
successfully compete and secure appointments
in the Central Services (services under
Govt. of India, Public Sector Undertakings,
Financial Institutions etc.)
19. Literacy rate of the state:
20. Number of entries of castes/sub-castes etc. included in:
(a) State list:

194 APPENDICES
(b) Mandal list:

(c) Common list:
21. Particulars of Commissions/Committees appointed in the State up-to-date
(a) Name of the Commission/Committee

(b) Chairperson
(c) Number of Members including Chairperson
(d) Date of appointment
(e) Date of Report
(f) Whether the Report has been accepted by the Government or not
(g) Any other important information
22. Particulars of Commission/Committee appointed pursuant to the Mandal

Judgment of the Supreme Court
(a) Whether statutory or not

(b) Names of the Chairperson and other Members
(c) Whether any report has been presented by such Commission/Committee
and if so, what action has been taken on the same by the Government
23. (a) Total number of posts in services under the State Government
Year of Reference
i) Group A/Class I:
ii) Group B/Class H:
iii) Group C/Class III:
iv) Group D/Class IV:
(b) Out of the total number of posts number of posts held by the members of
all OBCs:
Year of Reference
i) Group A/Class I:
ii) Group B/Class II:
APPENDICES 195
(c) Out of the total number of posts, number of posts held by each caste/
community included in the list of OBCs of the State, separately in the
following format:
(d) Out of the total number of posts, number of posts held by SCs:
Year of Reference
i) Group A/Class I:
(e) Out of the total number of posts, number of posts held by STs:
Year of Reference
i) Group A/Class I:
(f) Number of posts held by the single OBC caste/community accounting

for the largest number/percentage of the posts held by all OBC castes/
communities
Year of Reference
i) Group A/Class I:
196 APPENDICES
(g) Number of castes/communities among the OBC not holding any post:
(specify names of such castes/ communities)
APPENDICES 197
PART II
General Data of the Caste/Community Under
Consideration
1. Name of the caste/sub-caste/community/social group/synonym in respect of

which the request for inclusion or complaint of under-inclusion has been
made
2. Name and address of the individual/organisation/representative making the
request or the complaint
3. Is it the main caste/community? If not, give the name of main caste/community
4. Give the names of other sub-castes/sub-communities, synonyms etc.
5. Is the caste/community/sub-caste/known by any other name?
6. Population in the State of the caste/sub-caste/community/ sub-community/
synonym etc. etc. in respect of which the request or complaint has been
made
7. a) Percentage of the population of the caste/sub-caste under consideration to
the total OBC population of the State
b) If the caste/community under consideration is a sub-caste/ sub-
community, then give the percentage of this sub-group to the total
population of the main caste/community in the State
8. Percentage of the population of the caste/sub-caste etc. to the total State

population
9. Whether the caste/community/sub-caste/social group is spread all over the
State or largely concentrated in some districts of the State? In case of the
latter position, give the names of the districts where there is large
concentration
10. Give district-wise population figures of the caste/sub-caste etc. under
consideration. Also furnish the total population of each such district
11. Whether the caste/sub-caste etc is included in:
(a) the State List (if included, give its serial number in the State list, with
date of its inclusion)
(b) the Mandal List for the State (if included, indicate the serial number in
the Mandal list)
12. Is the caste/sub-caste etc. listed as a backward class in any other State(s).
If yes, give the name(s) of the State(s) and the serial number(s) in the
State list(s)
13. Specify the religion/faith/sect if any which members of the caste/community
sub-caste etc. (on whose behalf request/complaint has been made) follow
14. Date of request/complaint
198 APPENDICES
Data Relating to Social and Other Factors
A. Social
1. Whether the caste/community under consideration is generally regarded

as socially backward or socially not backward
2. (a) What is the occupation on which the members of the caste/ community
mainly depend for livelihood? Specify the occupation
(b) Indicate whether the occupation is agricultural or non-agricultural

(c) If agricultural, what proportion(approximately) of the members of the
caste/community are land-holders with holdings of more than 50% of the
statutory ceiling limit for agricultural lands in the State
(d) (i) If agricultural: Do the activities of their occupation mainly involve

manual labour or do not mainly involve manual labour
(ii) If agricultural activities involve manual labour, whether it is manual

labour rendered for wage or manual labour on own land
(e) (i) If not agricultural: state which occupation are the members engaged
in
(ii) Do the activities of the above mentioned occupation mainly involve

manual labour or does not mainly involve manual labour
(iii) If the above-mentioned occupation involves manual labour,
whether it is manual labour rendered for wage or manual labour
rendered on own works
(iv) has the caste or community got or acquired significant resource base
in the form of
a) Infrastructure such as land, buildings, workshops, quarries,
etc.
b) Machines and equipment necessary for carrying out the
occupation. Are the machines and equipment traditional or are they
modern?
3. Whether there are any occupations other than the main occupation referred
to at 2(a), on which substantial numbers of members of the caste/
community depend for livelihood. If so, specify such occupations
4. a) Whether or not the women of the caste/community, as a general practice,
are, for their own or for their family’s livelihood, engaged in agricultural
labour for wage
APPENDICES 199
b) Whether or not the women of the caste/community, as a general

practice, are, for their own or for their family’s livelihood engaged in any
other manual, (i.e., non-agricultural) labour for wage
If so, specify the type/nature of manual labour:
5. a) Whether or not children of the caste/community, as a general practice,
are, for their family’s livelihood or for supplementing their family’s
income, engaged in agricultural labour
b) Whether or not the children of the caste/community, as a general
practice, are, for their family’s livelihood or for supplementing their
family’s low income, engaged in any other manual, i.e., non-agricultural,
labour
If so, specify the type/nature of the manual labour
6. What percentage of the population of the caste/community etc. (male,
female, children taken together) are landless manual (both agricultural and
non-agricultural) labourers?
Explanation: The term “landless” includes those who have not more
than 1 hectare of unirrigated land and have no irrigated land at all
7. Whether the caste/community is, in terms of the caste system, identified/
linked with any traditional craft
If so,
(a) Specify which traditional craft are they identified/linked with?
(b) The percentage(approximately) of the population of the adult
members (males and females taken together) of the caste/community,
actually engaged in that craft.
(c) Of the adult members of the caste/community who are actually not
engaged in that craft, what proportion (approximately) are engaged in:
i) agricultural labour for wage

ii) other manual labour for wage (specify which)
iii) cultivation of own land
iv) other occupations [like services (clerical/supervisory/ managerial/
academic) in Government, Public Sector, Universities, Colleges &
Schools, organised private sector, trade/commerce/contracts/
entrepreneurial manufacture, professions (lawyer, doctor, consultancy
etc.)]
Specify the occupation(s):
8. (a) Whether the caste/community is, in terms of the caste system,

identified/linked with any other traditional or hereditary occupation, (i.e.,
other than traditional crafts).
If so, specify which occupation(s)
200 APPENDICES
(b) Whether such traditional or hereditary occupation is, in terms of the

caste system, regarded to be lowly, undignified, unclean or stigmatised?
(c) The proportion (approximately) of the adult members (males and
females taken together) of the caste/community actually engaged in that
occupation
(d) Of the adult members of the caste/community who are actually not
engaged in that occupation, what proportion (approximately) are engaged
in:-
i) agricultural labour for wage

ii) other manual labour for wage (specify which kind of manual labour)
iii) cultivation of own land
iv) other occupations [like services (clerical/supervisory/managerial/
academic) in Government, Public Sector, Universities, Colleges &
Schools, organised private sector/ trade/commerce/contracts/
entrepreneurial manufacture, professions (lawyer, doctor, consultancy
etc:)]
Specify the occupation(s)
9. a) Is the caste/community categorised as:-
i) Nomadic caste/community/tribe?
If so,
name the Commission(s)/Committee(s)/Report(s) which has so
categorised it
ii) Semi-nomadic/caste/community/tribe?
If so,
name the Commission(s)/Committee(s)/Report(s) which has 50
categorised it
b) If the answer to (i) or (ii) is yes, What is the present occupation(s) of

the members of the caste/community?
10. a) Is the caste/community categorised as De-notified or Vimukta Jati caste/
community/tribe [in terms of Criminal Tribes (Repeal) Act, 1952, Act No.
XXIV of 1952]
b) If the answer is yes, what is the present occupation(s) of the members
of the caste/community?
11. Was the caste/community subject to bonded labour? If so, since when? Is
it still being so subjected?
12. a) Number of MLAs belonging to the caste/community on the date of
application and their proportion to the total strength of the Legislative
Assembly
APPENDICES 201
b) Furnish separately the number of MLAs belonging to the caste/

community during the twenty-five years preceding the date of application
13. a) State the number of members of the caste/community elected to the
elective bodies at the district level, i.e., panchayati raj institutions—Zilla
Parishad, Zilla Panchayat, District Council etc. Give the figures for the ten
years preceding the date of application
b) Give the total number of elected members in the State in such district
level elective bodies during the above period
202 APPENDICES
B. Educational
1. a) (i) Number of literates of the caste/community in the State
(ii) Literacy rate of the caste/community in the State

Specify year of reference
b) (i) Total number of literates in the State
(ii) Total Literacy rate of the State

Specify the year of reference
Where the caste/community is not spread over in the entire State but is largely
concentrated in one or a few districts, also furnish the following information.
c) (i) Number of literates of the caste/community in the district
(ii) Literacy rate of the caste/community in the district

d) (i) Total number of literates in the district
(ii) Total literacy rate of the district

2. Out of the total number of literates of the caste/community in the State, please
furnish the total number of female literates of the caste/community
3. a) Number of Matriculates (or equivalent High School Examination) among
the members of the caste/community in the State
b) Proportion of matriculates of the caste/community to the total population
of the caste/community in the State
c) Total matriculates in the State:
d) Proportion of total matriculates in the State to the population of the State
Where the caste/community is not spread over in the entire State but is largely
concentrated in one or a few districts, also furnish the following information
e) No. of matriculates among the members of the caste/community in each of

the concerned districts
APPENDICES 203
Specify the year of reference:

f) Proportion of matriculates of the caste/community in the district to the
total population of the caste/community in the district:
g) Total number of matriculates in the district
Specify the year of reference:
h) Proportion of total matriculates in the district to the total population in
the district
4. Out of the total .number of matriculates of the caste/community in the State,
please furnish the total number of female matriculates among the members of
the caste and community in the State.
5. a) Total number of all graduates (in arts, commerce, law, management,
science, applied, technological, technical, professional etc. fields) among the
members of the caste/community in the State
b) Proportion of total number of graduate of the caste/community to the
total population of the caste/community in the State
c) Total number of graduates in the State
d) Proportion of total number of graduates in the State to the total population
of the State
Where the caste/community b not spread over the entire State but is largely
concentrated in one or a few districts, furnish also the following information
e) Number of graduates among the members of the caste/community in

the district
f) Proportion of graduates of the caste/community in the district to the
total population of the caste/community in the district
g) Total number of graduates in the district
h) Proportion of total number of graduates in the district to the total
population in the district
6. Out of the total number of graduates of the caste/community in the State,
specify how many among them are female graduates (arts, science, all
other graduates taken together)
204 APPENDICES
C. Economic
1. What percentage (approximately) of the families of the caste/community in

the State live in the type of houses indicated below:
Kachha houses
(including huts and sheds)
Pucca Houses
(including Chawls)
Where the caste/community is not spread over the entire State but is largely
concentrated in one or a few districts, then the above information may also be
separately furnished districtwise in respect of the districts where the population is
concentrated
2. a) Total number of cases of surrender of agricultural land in the State under

the Land Ceiling Act of the State
b) Out of this, the number of cases pertaining to the caste/ community
c) The total area (in hectares or acres) involved in the cases at (a)
d) Out of this, the area pertaining to the members of the caste/ community
Where the caste/community is not spread over the entire State but is largely
concentrated in one or a few districts, the following information may be further
furnished:-
e) Number of cases of surrender of agricultural land in each district under the

Land Ceiling Act of the State
f) Out of this, the no. of cases pertaining to the caste/community
g) The total area (in hectares or acres) involved in the cases at (e)
h) Out of this the area pertaining to the members of the caste/community
3. Number of posts in the services of the State Government held by the

members of the caste/community under consideration
4. Besides State Government Services, state how many persons of the caste/
community under consideration are engaged in the following areas of
employment and professions
APPENDICES 205
a) State level Public Undertakings, autonomous/semi-autonomous

establishments
b) (i) Teachers in colleges and universities
(ii) Administrative Personnel in Colleges and Universities
c) Doctors
d) Lawyers
e) Engineers and Architects
f) Chartered Accountants
g) Income Tax, financial and managemet consultants
h) Media professionals
i) Defence services
(Major in the Army and above, equivalent ranks of Navy and Air-
force)
j) Any other important fields of employment or profession (Specify the
fields)
5. State what percentage of the members of the caste/community is income-

tax assessees on account of trade/business
206 APPENDICES
D. Representation in the Services of Central Government
1. Total number of posts under the Central Government:
2.
(a) Number of posts held by all OBC

in the present Central List (Common List)
(b) Number of posts held by each caste/community in the

Central List (Common List) separately
(Serial Number as given in the Central List may be
indicated)
3. Explanation: As and when any caste/community is added to the list, the data
against Q. 2(a) and 2(b) may be updated
Number of posts held by SCs:
APPENDICES 207
4. Number of posts held by STs:
5. Number of posts held by the caste/community under consideration:

208 APPENDICES
E. Miscellaneous
1. (a) What are the main reasons on account of which the caste/ community
consider itself to be backward
(b) What are the main reasons on account of which the caste/ community
is considered backward or not backward by the State Govt.
(c) Has there been any improvement in the condition of the caste/
community during the last twenty years? If so, in what respects?
(d) Has there been any deterioration in the condition of the caste/
community during the last twenty years?
If so, in what respects?
Note: Support your reasoning with authentic evidence as far as
possible
2. (a) Furnish the names of two castes/communities (whether from among

the backward or forward castes/communities) at a level immediately
higher than the caste/community under consideration
Give reasons
(b) Furnish the names of two designated backward castes/communities in

the State, along with serial number in the State List, which are more or less
at the same level as the caste/community under consideration.
Give reasons.
3. Any other points besides those covered by the questionnaire above which
need to be mentioned in respect of the request or complaint
APPENDICES 209
APPENDIX II: INTERVIEWEES

Ahmad, Imtiaz. Professor at Jawaharlal Nehru University, Delhi, editor of Caste
and Social Stratification among the Muslims in India (Delhi: Manohar, 1978) and
author of numerous other works on this topic, 14 September 1996, Delhi.
Ahmad, M.S. Ministry of Welfare administrator focusing on OBC and Minority
issues, 17 December 1996, Delhi.
Alva, Margaret. Congress M.P., formerly a Cabinet Minister under Indira
Gandhi, 19 December 1996, Delhi.
Azariah, M. Bishop. Co-ordinator, All India United Christians Movement for
Equal Rights and National coordination Committee for SC Christians, 10 July
1996, Madras.
Beteille, Andre. Sociologist, Delhi School of Economics, 9 September 1996,
Delhi.
Bhansoor, Krishna Devi. Assistant Director, Research Wing, National
Commission for Scheduled Castes and Scheduled Tribes, 3 September 1996, Lok
Nayak Bhavan, Delhi.
Bidwai, Praful. Journalist and Fellow at the Nehru Memorial Library, 3
December 1996, Delhi.
Bose, Ashish. Demographer and author of numerous works on the Indian census
including Population of India (Delhi: B.R.Publishing Corporation, 1991), 17 July
1996 and 4 November 1996, follow up conversation 17 September 2001, Delhi.
Burman, B.K. Roy. Anthropologist, Research Officer for first Backward Classes
Commission (Kalelkar Commission), Former Head of the Social Studies Division
of the Census, former Head of Cultural Research Institute, Tribal Welfare
Department, Government of West Bengal and involved in exploratory GOI study
of alternatives to caste-based identification of OBCs, Author of Beyond Mandal
and After: Backward Classes in Perspective (New Delhi: Mittal Publications,
1992), 31 October 1996, follow up conversation 15 September 2001, Delhi.
Chakravorty, C. Deputy director of the Census, Man Singh Road Office,
previously assigned to Social Studies Division of Census, Seva Bhavan,
responsible for responding to Ministry of Welfare queries for data to aid in
administration of SC lists, 16 December 1996, Delhi.
Chatterjee, Jotsna. Women’s movement leader and YWCA activist, 22
November 1996, follow up conversation 15 September 2001, Delhi.
Chatterjee, Saral K. President, All India Christian People’s Forum, 22
November 1996.
Choudhary, A.K. Joint Secretary to the Government of India, Ministry of
Welfare, 17 December 1996, Delhi.
Daniel, Jose. President, All India United Christians Movement for Equal Rights
and the National Coordination Committee for SC Christians, 30 January 1996,
Delhi.
Das, Bhagwan. Dalit Buddhist Lawyer, 30 April 1996, Delhi.
210 APPENDICES
David, Shakuntala. Counselor, Free Church (Church of North India) Sansad

Marg, 17 November 1996, Delhi.
Devasahayan, V. Dalit Theology Department, Chennai, 10 July 1996.
Farooqui, Vimla. National Forum for Indian Women (NFIW), 2 December
1996, Delhi.
Gandhi, Rajmohan. M.K.Gandhi’s grandson and biographer, former Member
of Parliament, Research Professor with the Centre for Policy Research, New Delhi,
21 September 1996, Delhi.
Gangurde, R.P. University Grants Commission, 19 June 1996, Delhi.
Hamid, Naviad. Secretary of Association for Promoting Education and
Employment for Muslims, Former Secretary of Markazi Janait Ulama-e-Hind, 31
January 1996, Delhi.
Hamid, Syed. President of Association for Promoting Education and
Employment for Muslims, Former Vice-Chancellor Aligarh Muslim University,
Former Chair of the Staff Selection Board, (largest civil service recruiting agency
for junior posts), career Indian Administrative Service (IAS) officer, 2 September
1996, Hamdard Educational Society Office, Delhi.
Jatiya, Satyanarayan. All India President of BJP SC and ST Morcha, 12
September 1996, Delhi.
Khan, I.A. Office Secretary BJP Minority Morcha, Muslim BJP member, 12
September 1996, BJP Central Office, Delhi.
Kananaikal, Father Jose. Former Director of the Programme for Scheduled
Castes, Indian Social Institute, Director of Bihar Dalit Vikas Samiti, PhD
dissertation on SCs, 11 January 1996, Indian Social Institute, Delhi.
Karat, Brinda. Women’s movement leader (AIDWA), CPM (Communist Party-
Marxist) political activist, 10 December 1996, Delhi.
Khan, Aziz. Ministry of Welfare, 17 December 1996, Delhi.
Koppad, K.B. Social Anthropologist, Office of the Registrar General, 30 July
1996, Delhi.
Khusro, Ali Mohammad. Former Vice Chancellor Aligarh Muslim University,
spoke at Hyderabad Third Conference of the Association for Promoting Education
and Employment for Muslims in Hyderabad, 25 September 1996, Delhi.
Kumar, Neeraj. Participant in Anti-Mandal Commission Forum during 1990
student demonstrations, 22 November 1996, Delhi.
Lourduswamy, Father S. Vice President/Organizing Secretary of the All India
United Christians Movement for Equal Rights and National Coordination
Committee for SC Christians, Executive Secretary for SC/STs of the Catholic
Bishop’s Conference (CBC), 19 December 1995, Delhi.
Mahanty, Neeti. Director Jigyansu Tribal Research Center, 27 March 1996,
Delhi.
APPENDICES 211
Mahesh, K. SDO (Sub-Divisional Officer) Executive Magistrate, in charge of

issuing caste certificates in part of Delhi, 29 November 1996 and 3 December 1996,
Patalia House, Delhi.
Massey, James. Author of numerous works on Dalit Christians, and co-editor
of Dalit Solidarity, Delhi: ISPCK, 1995, 15 October 1996, Delhi.
Mathur, M.L. Research Division of the National Commission for Backward
Classes, 18 July 1996, National Commission for Backward Classes, Delhi.
Nimesh, R.D. Vice President of All India Federation of Scheduled Castes/
Tribes, Backwards and Minorities Employees Welfare Associations, 26 March
1996, Delhi.
Paswan, Ram Vilas. Railway Minister, M.P. and Secretary General of Janata
Dal (Party), Head of Dalit Sena, 20 December 1996, Railway Bhavan, Delhi.
Prasad, Justice R.N. Former Head of National Commission for Backward
Classes, Head of “Creamy Layer” Committee, 18 September 1996, Delhi.
Raj, Ebenezer Sundar. 11 July 1996, Madras.
Rao, R. Sanghettha. National President of All India Federation of Scheduled
Castes/Tribes, Backwards and Minorities Welfare Associations, 26 March 1996,
Delhi.
Rizvi, Ajaz. All India President of BJP Minority Morcha, 12 September 1996,
BJP Central Offices, Delhi.
Sajwan, Leela. Delhi State President of the Gandhi Caste Society, 1 August
1996, Delhi.
Salmani, Islam Qamar. Secretary General of the All India Muslim OBC
Organization, 5 December 1996, Delhi.
Santram, Bishop Pritram. Bishop of Church of North India, 10 December 1996,
Delhi.
Shah, A.M. Professor of Sociology at the Delhi School of Economics, Delhi
University.
Shahabuddin, Syed. M.P., Janata Dal Party. One of the Founders of the
Association for Promoting Education and Employment for Muslims and the
Convention on Reservations for Muslims in Delhi in 1994. Editor of Muslim
India, 11 September 1996, Delhi.
Sharif, Salim. Pastor at Free Church (Church of North India), Sansad Marg, 17
November 1996, Delhi.
Sheth, D.L. Professor of Sociology at the Centre for the Study of Developing
Societies and Member of National Backward Classes Commission, 6 September
1996, follow up conversation 17 September 2001, Delhi.
Singh, A.K. Author of the introduction to concepts in the 1991 Census of India.
Office of the Registrar General. Numerous conversations in 1996, follow up
conversation 13 September 2001, Delhi.
Singh, Hare Ram. Headed the Anti-Mandal Commission Forum action
committee during 1990 student demonstrations, 22 November 1996, Delhi.
212 APPENDICES
Singh, K.S. Head of POI Project, Former Director of Anthropological Survey

of India, 11 April 1996, Delhi.
Singh, O.P. Headed the Anti-Mandal Commission Forum legal committee
during the 1990 student demonstrations, 22 November 1996, Delhi.
Singh, Salig. Office Superintendent, National Federation of Scheduled Castes/
Tribes, Backwards and Minorities Welfare Association, 26 March 1996, Delhi.
Singh, V.P. Former Prime Minister of India, 20 November 1996, Delhi.
Singh, Hoshian. Research Division of the National Commission for Backward
Classes, 18 July 1996, National Commission for Backward Classes, Delhi.
Srinivas, M.N. Sociologist, 8 November, St. Louis.
Sonavane, Vilas. On Advisory Board of the All India Muslim OBC
Organization, Maharashtrian non-party political activist, 2 December 1996, Delhi.
Thorat, S.K. Professor of Sociology at Jawaharlal Nehru University, 29 April
1996, Delhi.
Vijayanunni, M. Registrar General and Census Commissioner of India, 16
December 1996, Delhi.
Notes
1
Identity and identification
1 “Backward” is not my term but rather official jargon associated with reservations.
The term is also used by many political activists. The term will appear in the rest of
the book without the repeated use of quotation marks, which does not imply that I
agree with the negative connotation it suggests. When writing about official
categories and the protest groups contesting them, I use either the official language
in question or the terms groups choose for themselves.
2 Defining institutions for the purposes of this study as “formal arrangements for
aggregating individuals and regulating their behavior through the use of explicit
rules and decision processes,” I focus on the bureaucratic and legal institutions
most directly involved in implementing reservations (Levi 1990: 404–5). The “new
institutionalists” in political science, economics and sociology assert that
institutions “constrain and refract politics” (Thelen and Steinmo 1992:3), and a few
institutionalist scholars have noted the influence of contemporary state institutions
on the politics of identity (Gunther and Mughan 1993).
3 Tajfel offers classic social psychological work on the effects of external
categorization on individual identities (Tajfel 1981). Such research would be
another approach to these issues of categories and groups. When relevant and
available, secondary sources which include quantitative data will be cited or
summarized.
4 Reservations have been a valuable avenue of social mobility for some members of
oppressed groups. Although the procedures involved in implementation are at times
problematic due to the need to categorize people, these problems should not
overshadow the net positive impact of these policies for the categorized groups.
Quantitative data, some of which will be discussed later in this and future chapters,
demonstrate the progress facilitated by reservations, notably the increasing number
of Scheduled Castes and Scheduled Tribes employed in government jobs and their
guaranteed representation in legislatures (Planning Commission 1997–2002).
5 Nicholas Greenwood Onuf offers an overview of some “constructivist” approaches
in philosophy and social theory and applies them to international relations theory
(1989) (see especially pp. 35–65). An excellent collection of works applying
constructivism in comparative politics is Daniel Green’s edited volume (2002).
214 NOTES
Benedict Anderson’s influential work on “imagined communities” introduced the

idea of the historical construction of nationalism to a wide audience (1991).
Another classic, Edward Said’s Orientalism, examines constructions of the
“Orient” by the “West” (1978). Of the literature on race and caste, two notable
examples of constructivist approaches are Michael Omi and Howard Winant’s
work on “racial formation” (1986) and Ronald Inden’s discussions of colonial
codifications of caste in Imagining India (1990). These are just a few representative
examples from this varied literature on constructivism and identity.
6 Scott, also known for his work on resistance, is an inspiration for my focus on the
agency of liminal groups as well (Scott 1990).
7 Much of the literature on the instrumental uses of ethnic and other identities
contrasts this approach with a “newer” constructivist approach; however, many of
the classic examples of instrumentalism, such as “ethnic entrepreneurs” mobilizing
movements by selectively constructing a sense of shared traditions, could also be
seen as a strain of the constructivist approach (Young 1993:21–5, Tilley 1997). An
instrumental constructivist approach highlights how identity constructions are often
strategic, used to gain a competitive edge or advantage. Many such analyses are
attempts to explain ethnic violence and thus focus on the negative effects of
manipulating identities for political or economic gain (Horowitz 1985, Basu et al.
1993). The subset of the literature I will consider here underscores instead the
potentially positive impact of oppressed groups using- and, in the process,
reconstructing—certain identities.
8 This poem, received at a Dalit Women Writer’s Conference held in New Delhi in
1996, was written by Sushila Takbhaure and translated from Hindi by Laura
Dudley Jenkins and S.Thapar (personal copy).
9 In the United States the continuing use of race-based affirmative action has been
called a form of “liberal racism” that “does not expose racism; it recapitulates and,
sometimes, reinvents it” (Sleeper 1997). Another critic wonders “whether it is
possible to divorce any system of racial classification from the practice of racial
discrimination” (Cose 1997:26).
10 I learned about the approach of the Gandhi Caste Society or Gandhi Margam by
attending one of their meetings in Delhi on 1 August 1996, speaking with
members, and interviewing Leela Sajwan, Delhi State President of this
organization, on 17 September 1996.
11 Race and caste are contingent and ambiguous social constructions, whereas racism
and casteism are more persistent phenomena which, by definition, ignore the
fluidity of the former. Although “those who resist legal remedies for the history of
racism might use the nonexistence of races to argue in the United States, for example,
against affirmative action, that strategy is, as a matter of logic, easily opposed…the
existence of racism does not require the existence of races” (Appiah 1995:105).
12 The English word “caste” is a variation on the Portuguese casta (meaning species or
type). The related words jati and varna will be discussed later in this chapter in the
section on categories, but the term caste is also still used in contemporary India,
especially in official language and by English speakers.
13 See the numerous volumes entitled Subaltern Studies published by Oxford
University Press, the first volumes edited by Ranajit Guha. See also Antonio
Gramsci on subalterns (1980).
NOTES 215
14 Declan Quigley characterizes “caste” as “an extremely unhappy translation of two

quite different indigenous concepts” (Quigley 1993:4), but the term has since been
adopted by many Indians and has taken on important meanings politically, as in the
official lists of the Scheduled Castes. Clearly, even if it was an imported concept,
we cannot reject the term “caste” today “if we are to know how Indians (especially
those educated in English) see their own society” (Galanter 1984:7). The term
caste, usually referring to jati, is commonly used, especially in official language. I
use the official terms when writing about government categories and the terms
protest groups choose when writing about activism.
15 The varied meanings of jati are one example of the messiness of these concepts:
“Its meaning in most Indian languages has at least as large a number of
connotations as does the term caste in colloquial English. In Kannada, for example,
`jati' most simply means ‘kind.’ A man might say, for instance, that he prefers one
jati of cigarettes over another jati. If an individual is asked, ‘what is your jati?’ he
may respond with the name of the unit within which he confines his marriage
relations, with the name of a collection of these units that are similar (caste
category or caste complex in anthropological jargon), with his varna affiliation, or
even with the name of his religion, as for example, ‘Christian jati’” (Harper 1968:
51). Beth Roy lucidly discusses the idea that castes are both “fixed and mobile”
(Roy 1994:18).
16 The term “untouchables” was probably first used in the early 1900s by the
Maharaja of Baroda (Rao 2000). Mohandas Gandhi euphemistically, strategically—
and, many argue, patronizingly—dubbed them “Harijans,” or children of god,
during the Independence movement. Followers of the untouchable leader who
contributed greatly to the Indian Constitution, B.R.Ambedkar, preferred and
popularized the use of Dalit.
17 Mitra and Singh base this conclusion on the postelection poll data of 1996. Their
“deprivation index” includes both objective class differences (occupation, assets,
type of house and income) as well as a subjective satisfaction variable (Mitra and
Singh 1999:183, 188–9, 193–5).
18 The constitution leaves the contents of the schedules of tribes and castes up to the
President, in consultation with the governors; the Parliament can by law include or
exclude groups from the list (Indian Constitution, Articles 341 and 342).
19 The term Backward Classes “was used as far back as 1880 to describe a list of
groups, also called illiterate or indigent classes, entitled to allowances for study in
elementary schools” (Galanter 1984:154, n. 1).
20 1992 Supp (3) S.C.C. 217:1992 S.C.C. (L&S) Supp. 1: (1992) 22 A.T.C. 385.
21 Other policies targeting Scheduled Castes and Scheduled Tribes are supposed to
redistribute resources such as land, housing, grants, loans and health or legal aid, or
offer protection from “social disabilities” or caste-based discrimination and
persecution (Galanter 1986:131). See also the SC/ST (Prevention of Atrocities) Act
1989.
22 According to the 1991 census, Scheduled Castes are about 16.5 per cent of the
national population, and the Scheduled Tribes are about 8.1 per cent. The 15 per
cent and 7.5 per cent quota figures, based on the 1961 census, have been retained
since 1970, because the 1971 census figures did not necessitate changes, the 1981
census was not carried out in Assam and the 1991 census was not carried out in
Jammu and Kashmir; thus “all India” percentages of the scheduled groups could not
216 NOTES
be calculated. These quotas are used for people appointed to government jobs
through direct recruitment on a national basis. For some lower level jobs (classified
as group C and D on a scale from A through D), “normally attracting candidates
from a locality or region,” the quota is calculated on the basis of the percentage of
Scheduled Castes or Tribes in that state or union territory (Nabhi 2001:3–4).
23 The 50 per cent cap on reservations has not always been obeyed. The southern state
of Tamil Nadu, for example, has had 69 per cent reservations. The Tamil Nadu
Reservations Act is an attempt to protect these state-level reservations.
2
Adjudicating identities
1 (1994) 5 S.L.R. (S.C.) 206: (1994) 6 S.C.C. 241:1994 S.C.C. (L&S) 1349: (1994)
28 A.T.C. 259. The paranthetical page numbers in this section of the text refer to
quotations from this decision in the Services Law Reporter (S.L.R.). Citation to
Indian legal authority herein is based upon standard Indian legal form. For a
discussion of Indian legal citation, see Galanter’s “Note on Citation and
References” in Competing Equalities (Galanter 1984:xxix).
2 Srinivas defines sanskritization as “the process by which a ‘low’ Hindu caste, or
tribal or other group, changes its customs, ritual, ideology, and way of life in the
direction of a high…caste. Generally such changes are followed by a claim to a
higher position in the caste hierarchy than that traditionally conceded to the
claimant caste by the local community. The claim is usually made over a period of
time, in fact, a generation or two, before the ‘arrival’ is conceded. Occasionally a
caste claims a position which its neighbors are not willing to concede” (Srinivas
1966:6).
3 See also a government letter dated 19 September 2000, about whether the Kolis of
Maharashtra, a Backward Class, would be declared to be Mahadeo Koli, a
Scheduled Tribe [www.tn.gov.in/gorders/adtw81-t.htm] downloaded 10 March
2002. A news report addressing this ongoing demand entitled “Grand Birthday Fete
of Sharana Planned” appeared in The Hindu 13 January 2002.
4 For example, a court in the state of Kerala, in a case involving several people who
were alleged to be members of Other Backward Classes but were asserting a
Scheduled Caste identity, held that these questions should be brought before a
scrutiny committee, following the procedure outlined in the Kumari Madhuri Patil
case. Kerala Pattikajathi Samrakshana Samithy v. State of Kerala, I.L.R. 1995(3)
Kerala 1.
5 See Director of Tribunal Welfare, Government of A.P. v. Laveti Giri, Supreme
Court of India, Civil Appeal No. 4545 of 1995, decided 18 April 1995.
6 C. Sunil Krishnan v. State of Kerala, A.I.R. 1997 Kerala 63.
7 Gayatrilaxmi Bapurao Nagpure v. State of Maharashtra, Supreme Court of India
Civil Appeal No. 4377 of 1996, decided 15 March 1996.
8 Government of Andhra Pradesh v. R.K. Ragala, A.P. High Court (1994).
9 (1996) 3 S.C.C. 545: A.I.R. 1996 S.C. 1011. As in the preceding section, page
numbers in this section of the text refer to quotations from the Valsamma Paul
decision in the Supreme Court Cases (S.C.C.) reporter.
NOTES 217
10 Saroja notes the difficulty of assessing definitively the rates of intermarriages due
to the lack of enforcement of marriage registration (Saroja 1999:183).
11 Citing, Indra Sawhney v. Union of India, 1992 Supp (3) S.C.C. 217:1992 S.C.C.
(L&S) Supp 1: (1992) 22 A.T.C. 385.
12 Atul Chandra Adhikari v. State of Orissa, Orissa (1995). Interestingly, the US
Supreme Court refused to disturb a similar tribal rule under which the children of a
woman who married outside the tribe could not enjoy the rights of tribal
membership (i.e. voting rights, the right to hold office and the right to inherit
property), whereas children born to a man who married outside the tribe were
considered members of the tribe entitled to such benefits. See Santa Clara Pueblo
v. Martinez, 436 US 49 (1978).
13 (1996) 3 S.C.C. 100. As in the preceding sections, the parenthetical page numbers
in this section refer to quotations from the Swvigaradoss decision in the Supreme
Court Cases (S.C.C.) reporter.
14 Notably, Buddhism and Sikhism also have been used as escape routes from the
caste system. See Chapters 6 and 7 for a discussion of the politics behind this
official distinction between Christians and Muslims, who are ineligible for
Scheduled Caste status, and Hindus, Sikhs and neo-Buddhists, who may qualify for
this category.
15 Discussed in Soosai v. Union of India, A.I.R. 1986 S.C. 733.
16 Valsamma Paul v. Cochin University (1996) 3 S.C.C. 545, 567.
17 G.M.Arumugam v. S.Rajagopal (1976) 3 S.C.R. 82.
18 Id. at 94.
19 Id. at 94–5, citing Nathu v. Keshwaji, I.LR. 26 Bom. 174.
20 Guntur Medical College v. Mohan Rao, A.I.R. 1976 S.C. 1904, 1908.
21 Id. citing Dugaprasada Rao v. Sudarsanaswami, A.I.R. 1940 Mad. 513.
22 Citing Ganpat v. Presiding Officer, A.I.R. 1975 S.C. 420, 424.
3
Official anthropology
1 The Scheduled Caste volume of the People of India project was first published in
1993, and it is the second, revised, hardback edition of 1995 to which I refer. The
photograph of the woman and her children is on the protective paper cover of this
edition, a point I mention because the earlier edition features a collection of
different photographs.
2 Official ethnography elsewhere in Asia likewise has had an impact on
constructions of identity. See, for instance, the fascinating work on China by Dru
Gladney (1996) and Melissa Brown (2001) and the volume on anthropology and
colonialism edited by Bremen and Shimizu (1999). I presented an earlier version of
this chapter at the 2000 Association for Asian Studies meeting in a panel on Asian
anthropology, including cases from India, Japan and China, which further
convinced me of the fruitfulness of future comparative research.
3 Bates is summarizing the Report of the Ethnological Committee on Papers laid
before them and upon the examination of specimens of Aboriginal tribes brought to
the Jubbulpore Exhibition of 1866±67 (Nagpur 1868). This sort of disrespectful
display was not unusual for its era; consider the 1890 Chicago World’s Fair.
218 NOTES
4 Nicholas Dirks, although arguing that caste was “appropriated, and reconstructed,
by the British,” holds that “[n]either British administrators nor orientalists were
able to go to India and invent caste through sheer acts of will and rhetorical fancy,
however useful caste was as a social mechanism to assist in the management of an
immensely complex society” (Dirks 1992:61). Likewise, Prakash emphasizes “how
the categories of colonial discourse were revised in the process of their historical
articulation” and avoids “portraying British India as a place scorched by the power/
knowledge axis, leaving nothing of its history” (Prakash 1992:172).
5 Some valuable studies that have discussed the contemporary implications of past
practices include Susan Bayly (1999), especially Chapter 7, and Sumit Guha
(1998). Some specifically mention the current People of India Project, albeit briefly
(Bayly 1999, Bates 1995:219, Searle-Chatterjee 1996).
6 Raheja elucidates the types of changes which occurred after 1857, clarifying the
tight relationship between anthropological studies of caste and the administrative
concerns of the state: “Such fissures had certainly begun to appear long before
1857, but the rebellion so impressed itself upon the colonial imagination that
dramatic shifts in administrative policy occurred soon thereafter. Colonial
administrators began carefully recording caste identities on the decennial census,
commissioning the publication of region-by-region caste compendia, relying more
heavily on caste identities in formulating land revenue policy, and disciplining
certain groups as ‘criminal’ castes and tribes or as castes prone to rebellion”
(Raheja 1996:495).
7 This approach was rather standard for the day. See George W.Stocking (1982),
particularly Chapter 8 on “The Critique of Racial Formalism.” Not all colonial
anthropologists were as guilty of oversimplified typologies. In fact, as Susan Bayly
points out, some accounts of caste, such as Denzil Ibbetson’s 1881 introduction to
the Census of the Punjab, were quite nuanced (Bayly 1995). Yet Crispin Bates notes
that more contextual accounts such as Ibbetson’s were not always well received as
proper “science” in their day: “however popular his ideas may have become in
certain academic circles in more recent times, they sat awkwardly in the period in
which they were first formulated” (Bates 1995:231). Risley himself quotes
Ibbetson’s work, which he describes as “[a]n admirable picture of open-air work; it
has been drawn on the spot; it is full of local colour; and it breathes throughout
with the quaint humour of the peasantry of the Punjab, the manliest and most
attractive of all the Indian races. From this wealth of material it is not altogether
easy to disentangle the outlines of a cut-and-dried theory” (Risley 1915:263).
Risley’s own desire for “scientific” typologies rather than anecdotal nuance come
through in this commentary.
8 For more on the use of proverbs in colonial reference works on caste, including
Herbert Risley’s work, see Raheja (1996).
9 The criteria for being an “informed” informant are left unspecified, although the
relative numbers of men and women interviewed suggest a gender bias among
investigators and/or potential informants, perhaps based on a perception that
women are less informed or that what they know is not relevant to the study.
10 Keeping more detailed information about certain minority groups is a social
science/administrative tradition that is far from unique to India. Consider the
degree of scrutiny historically given to nonwhite minorities in the US census (Lee
1993:82).
NOTES 219
11 For more information on the portrait building system and the Anthropological
Survey of India’s involvement in it, see the webpage of the NCRB, particularly the
section on the Facial Analysis and Criminal Identification System at [http://
www.ncrbindia.org/bound.htm] downloaded 27 September 2000.
12 This work was recommended to me by Syed Hamid, who is a leader in the
Hamdard Education Society in Delhi (Ahmad 1995, Hamid interview 2 September
1996).
13 This tendency to emphasize the distinctiveness of groups while simultaneously
pointing to a unified or “composite culture” is not unique to the Anthropological
Survey. The introduction to an exhibit of tribal art, organized by the Government
of India, attempted to “highlight the cohesiveness of our rich and diverse heritage,”
noting, about the art forms represented, that “as diverse as they are, there are
linkages, a homogeneity which is perhaps the result of the ancient living composite
cultural crucible we exist in” (Program from “Usha Abhil Asha: An exhibition of
folk, tribal and traditional paintings and sculpture,” 24 March to 7 April 1996, Lalit
Kala Academi Gallery, New Delhi. Organized by the Department of Culture,
Ministry of Human Resource Development, and the Zonal Cultural Centers). From
the Anthropological Survey of India to the Department of Culture, the official study
and display of “culture” in India is carefully presented to emphasize national unity.
See also Srirupa Roy (1999) on official nationalism in the form of parades and
publications.
14 The full title of Volume 7 is Identity, Ecology, Social Organization, Economy,
Development Process and Linkages: A Quantitative Profile (1996). The other
volumes Singh described as the “soul of the project” are Volume 1, the
Introduction (1992), and Volume 8, Communities, Segments, Synonyms, Surnames
and Titles (1996). For a critique of “trait distributions” as an anthropological tool,
see Moerman (1965).
15 Ironically, in spite of “sensitivity tests” applied to naming practices in the new
People of India project, the governmental category of “backward” citizens remains
curiously untouched, a remnant of an evolutionary model of countering inequality
and social divisions (Searle-Chatterjee 1996).
16 The status ranking of Scheduled Tribes could not rely on “varna” since “only 11.8
per cent of them recognize their place in it” (Singh 1994:7), so the regional
hierarchy was measured using the categories of high, medium and low status.
Interestingly, both self-perception and perception by others was noted: “When it
comes to self-perception of a tribal community in the regional hierarchy we find
that 171 tribes, i.e. 26.9 per cent see themselves as being of high status, while 298
tribes (46.9 per cent) perceive themselves as being in the middle position. About
25.3 per cent, i.e. 161 tribes see themselves as being of low status. Others see over
11.2 per cent of the tribes as high, 39.2 per cent as medium and 49.4 per cent as of
low status” (Singh 1994:7). It is this last figure which makes the quoted conclusion
about the Scheduled Tribes’ lack of social stigma so troubling.
17 Anthropologists and sociologists in India have repeatedly reflected on the
dilemmas of postcolonial anthropology and differ in their conclusions about how to
move beyond colonial practices. As quoted earlier, Andre Beteille is skeptical about
the value of the People of India project, whereas K.S.Singh is confident that the
project can supersede the legacies of Risley. Such perspectives reflect in part
institutional affiliations, a view from academia on the one hand and a view from
220 NOTES
government on the other; however, many universities were involved in the project,
and even eminent sociologist M.N.Srinivas was positive about it as long as it was
considered a starting point for more in-depth fieldwork (Srinivas interview 8
November 1997). Sociologists and anthropologists in India from different
institutional backgrounds have joined periodically to debate the future of their
disciplines, and the issues raised shed light not only on the broader politics of these
disciplines in India but also on the tensions ignited by the People of India project;
in one recent workshop, both Beteille and Singh were participants (Uberoi 2000;
Sundar et al. 2000). One recurring debate is over “relevance” and the proper
relationship between anthropology and practical concerns or policy prescriptions,
in short, the relationship between the field and the government (Srivastava 1999:
545–52, Debnath 1999).
18 [http://www.ad2000.org/uters2.htm] 18 December 2001.
4
Caste certificates and lists
1 (1994) 5 S.L.R. (S.C.) 206: (1994) 6 S.C.C. 241:1994 S.C.C. (L&S) 1349: (1994)
28 A.T.C. 259.
2 The survey of “Harijan Elites” was carried out in Azamgarh in eastern Uttar
Pradesh. The sample included political leaders, caste organization/caste leaders,
bureaucrats, doctors, engineers, lawyers, teachers, and businessmen (Roy and
Singh 1987:23, 29).
3 Government of Andhra Pradesh v. R.K.Ragala, A.P. High Court (1994).
4 An additional example of the repeated identification and labeling of reservation
beneficiaries comes from another bureaucracy devoted to these policies, the
National Backward Classes Finance and Development Corporation (NBCFDC).
This is a nonprofit corporation, wholly owned by the Government of India, which
extends credit to the Backward Classes to assist in skill development, employment
schemes and other economic projects. Although this type of affirmative action is
not my primary focus, the relatively new procedure of “pre-identification” initiated
by the Corporation demonstrates the increasing scrutiny of beneficiaries’ identities
by the government. As of 1995, “Pre-identification of beneficiaries has been made
compulsory… The factor of unknown beneficiary has also been eliminated”
(National Backward Classes Finance and Development Corporation Objective,
Organization, Operating Procedure 1996:12). This preidentification is to help in
“establishing the authenticity of the loanee,” who is issued a “beneficiary card”
(NBCFDC Guidelines for Implementation of NBCFDC Schemes 1996:3).
5 The Caste/Tribe certificate will only be accepted as “sufficient proof in support of a
candidate’s claim as belonging to the Scheduled Caste or Scheduled Tribe” if it is
issued by one of the following authorities: District Magistrate, Additional District
Magistrate, Collector, Deputy Commissioner, Additional Deputy Commissioner,
Deputy Collector, 1st Class Stipendary Magistrate, Sub Divisional Magistrate,
Taluka Magistrate, Executive Magistrate, Extra Assistant Commissioner; Chief
Presidency Magistrate, Additional Chief Presidency Magistrate, Presidency
Magistrate; Revenue Officer not below the rank of Tehsildar; or Sub-Divisional
NOTES 221
Officer of the area where the candidate and/or his family normally resides
(Ministry of Personnel 1993:248).
6 (1994) 5 Supreme Court Cases 244.
7 The application form for a Scheduled Caste certificate as well as several other forms
are included in the appendix. These are from a caste certificate issuing office in
Delhi unless otherwise indicated. These documents include:
a. Application form for Scheduled Caste Certificate.

b. Form of Certificate to be Produced by a Candidate Belonging to a
Scheduled Caste or Scheduled Tribe in Support of His Claim
(Department of Personnel 1993:396).
c. Application Form for a Certificate of Eligibility for Reservation of
Jobs for Other Backward Classes in Civil Posts and Services
Under Government of Delhi.
d. Form of Certificate to be Produced by Other Backward Classes
Applying for Appointment to Posts Under the Government of India
(Department of Personnel 1993:487).
e. Questionnaire for Consideration of Requests for Inclusion and
Complaints of Under-Inclusion in the Central List of Other
Backward Classes (from the Office of the National Commission
for Backward Classes).
8 Often these memoranda use the term inter-caste in reference to both Scheduled
Castes and Tribes. I will use the term intercommunity marriages to more clearly
encompass both. By using terms like mixed marriage or even intercommunity
marriage I do not mean to imply that I endorse the idea that these castes or tribes
are innately distinct.
9 See the government order replacing the older rule with the newer one at [http://
www.tn.gov.in/gorders/adtw17-e.htm] 8 March 2002.
10 The Ministry of Social Justice and Empowerment was formerly the Ministry of
Welfare.
11 [http://socialjustice.nic.in/schedule/faq.htm] 15 March 2002.
12 (1996) 3 S.C.C. 100.
13 In the case of petitions to be included on the lists of Scheduled Castes or Tribes, the
group would submit data to the “List Cell” within the SC/ST Division of the
Ministry of Social Justice and Empowerment (known as the Ministry of Welfare
prior to 1998). In contrast to the National Commission for Backward Classes,
which is responsible for decisions about which groups are to be added to or
removed from the official national lists of OBCs, the National Commission for
Scheduled Castes and Scheduled Tribes is less involved in determining which
groups are on the lists and more involved in acting as a watchdog, monitoring the
implementation of SC/ST reservations. Unlike the OBCs, the SCs and STs can only
be added to or deleted from lists through an Act of Parliament. According to Mrs
Krishna Devi Bhansoor, Assistant Director of the Research Wing of the SC/ST
Commission, it is not the SC/ST Commission that supplies data to Parliament to
help them make these determinations, but rather the “List Cell” within the (then)
Ministry of Welfare’s SC/ST Division (Bhansoor interview 3 September 1996).
222 NOTES
The Ministry of Welfare’s SC/ST Division deals with questions not only over
which groups belong on the lists but also over which individuals belong in which
groups. According to the official instructions on caste certificates, “Cases in which
a doubt arises whether a person is a Scheduled Caste/Tribe or not may be referred
to the Ministry of Welfare (Scheduled Castes and Scheduled Tribes Division),
Shastri Bhavan, A-Wing, New Delhi” (Ministry of Personnel 1993:247).
14 This study’s focus on bureaucratic processes raises interesting questions for future
research. What is the impact of the processes on the perceptions of those who
experience them? Interviews with people who have gone through these processes
might shed light on the impact of the policies on applicants. A related question is
how do experiences with petitions for certificates or listings influence applicants’
sense of their own agency in their dealings with the state? A contemporary
newspaper editorial paints a dire picture of the government’s treatment of Dalits
through procedures such as caste certificates: “At the state level, the attempts are
made by the state agencies to dissolve particularly the Dalits into social
insignificance where they have to prove their identity not through their authentic
self but through certain identity cards or a piece of paper such as a caste certificate
as specified by the state” (The Hindu, “Editorial: Hindutva’s Passive Revolution”
21 September 2000).
15 See this document in the appendix.
16 OBC reservations are used not only for central government jobs, but also in several
states, as well as the government of the capital, Delhi, and can vary at these
different levels. The following discussion draws on regulations and forms from the
central government and Delhi government.
17 See these documents in the appendix.
18 The confidentiality of the application seems not to be a major concern. Something
that came up in my interview with Mahesh raised questions about the
confidentiality, or lack thereof, associated with the caste certificate process.
Mahesh noted that another role of his office is to aid the government when a
criminal remains unidentified, although, when asked more specifically, he said that
the information he gathers in the caste certificate process is not used to help
identify criminals (Mahesh interview 29 November 1996).
19 See [http://socialjustice.nic.in/obcs/welcome/htm] 15 March 2002.
20 The following description refers to the process of determining who is an OBC prior
the reconstitution of the membership of the National Backward Classes
Commission in 1998. Although their task is unchanged, the increase in the number
of cases processed under the new commission suggests that the rather in-depth
procedures described here may have been subsequently abbreviated. A new
member of the commission, Akshaybhai Sahu, has boasted, for example, that “The
commission, under the chairmanship of justice P.K.Shyam, cleared 471 cases
within months of taking over, while the remaining cases are expected to be cleared
by January-end, Sahu said, claiming that the last commission could clear only 320
cases in its tenure—40 cases every year” (as paraphrased in Indian Express,
“NCBC Member Raps Government” 7 December 1998). See “126 More Castes on
OBC List” (The Hindu 20 November 1999) on the government’s implementation
of some of the Commission’s recommendations.
NOTES 223
21 This research may be done by “independent professional bodies, academic bodies,

research institutes, university departments, individual scholars who work in that
area” (Sheth interview 6 September 1996).
23 Dugger describes a new computer in a village in Madhya Pradesh, where “For 25–
35 cents, villagers buy printouts of documents they might have spent days trying to
get from local bureaucrats: land records, caste certificates and proof of income,
among others” (Dugger 2000). The Orissa Government launched its official
website and announced that “Many more innovative features, including the
payment of Government dues, taxes, utility bills, issue of driving licenses, birth,
death and caste certificate etc. would be enabled over the Internet in future” (“State
Government Launches Website” The Hindu 5 September 2000). See the Delhi
government’s Application Form for a Certificate for Eligibility for Reservation of
Jobs for Other Backward Classes in Civil Posts and Services Under Government of
Delhi [http://delhigovt.nic.in/ dept/district/anx8.pdf] 25 January 2002. As
Government Minister Ram Vilas Paswan points out, however, “The large majority,
especially in remote, rural and semi-urban areas, does not yet have access to even
basic telephones, let alone IT services” (Paswan 2000:45).
25 Planning Commission 1997–2002, “Participation of SCs and STs in Administration
and Decision Making,” Section 3.9.21–2.
5
Categorizing and counting on the census
1 In the Household Schedule for the 2001 Indian census, question 8 reads: “If
Scheduled Caste, write name of the Scheduled Caste from the list supplied” and
question nine reads: “If Scheduled Tribe, write name of the Scheduled Tribe from
the list supplied” [http://www.censusindia.net/census2001/qpopenu. html] 5
October 2000.
2 See Chapter 1, Note 22 supra.
3 The postapartheid South African census and its utility for affirmative action, in
comparison with the US and India, was one subject of discussion at an
international conference I attended on “Rethinking Equality in the Global Society”
at Washington University in St. Louis, 8–10 November 1997.
4 The trend toward census taking and the date of the first Indian census are discussed
in the following passage from the Government of India Home Department (Public)
Programs from 26 November 1870: “That a census of the whole people is most
desirable, or rather we may say is absolutely necessary, as a sound basis of almost
every economical reform, has long since been admitted as a simple truism, but it is
only recently that a definite project has been formed to accomplish it. Local
attempts had indeed been previously made, but it was only in 1856, that the late
Court of Directors urged upon the Indian Government the importance of a measure
which is decennially carried out in Great Britain, the United States of America, and
many of the countries of Europe, and it was proposed that the general census of ‘our
Indian territories’ should be taken in 1861, to correspond with the census of Great
Britain and thenceforward to be repeated every ten years. But the mutiny stopped
224 NOTES
this project, and a census was only attempted partially in some Provinces and in a
very rough and imperfect manner. The Government had more pressing cares to
attend to, and it was not until 1865 that the general measure was revived, and the
1st of January 1871 was then fixed for the undertaking” (Natarajan 1972:3).
5 For a discussion of the assumptions associated with modernization theories, as well
as the demise of the assumption that caste would fade away with modernity, see
Randall and Theobald 1998, particularly their discussion of Rudolph and
Rudolph’s classic, The Modernity of Tradition (1967).
6 1992 Supp (3) S.C.C. 217:1992 S.C.C. (L&S) Supp 1:(1992) 22 A.T.C. 385.
7 The 1941 census was abbreviated due to World War Two; the 1951 count of OBCs
was “entirely provisional” (Gopalswami 1953:2); and subsequent censuses
recorded caste or tribe only for the scheduled communities.
8 One example of political party involvement can be found in the article
“Memorandum seeks caste-wise census” in The Hindu 30 March 2000. Other
articles discussed the debates over a castewise census (Indian Express, “Caste in
Census” 1 May 1998, Shah 1998). On the initial announcement that caste would be
counted see, India Today, “But the die is caste” 11 May 1998. On the decision not
to count caste after all, see Narayan 1999.
9 Shah’s suggestion to deal with the paucity of OBC data is a survey (Shah 1998).
10 After the first post-independence lists in 1950, the schedules or lists have been
modified or supplemented as in 1956 when the states were reorganized,
necessitating reorganization of the lists, which are done on a state-by-state basis.
Examples of other changes include a 1976 amendment removing some area
restrictions and a 1990 amendment allowing neo-Buddhists to be SCs. Other
changes were made to certain state lists, such as the STs in Uttar Pradesh Order of
1967, Meghalaya STs in 1987, and Kashmir STs in 1989 (Nanda 1994).
11 Also note the strategy to extract the “correct” name from the citizen, namely
“persuading” them by suggesting that reservation benefits may be at stake. This
practice reinforces a longstanding association in the public imagination between
census enumeration and reservation eligibility, even though census slips are
confidential and cannot be used as evidence in a court of law (The Census Act
1948). This perception helps to explain the rising number of people claiming to be
in Scheduled Tribes in recent censuses, to be discussed later in this chapter.
12 Not all states had lists of Other Backward Classes, so each state was to compile
either a provisional list of Backward Classes or Non-Backward Classes for the
enumerators. The census report on “Special Groups” did not include OBC data;
rather it emphasized that this information was to “be treated as entirely
provisional” and was “given for use by the Backward Classes Commission”
(Gopalaswami 1953:2).
13 This tendency to keep more detailed census data on certain minority groups is
widespread. See Lee 1993:82, for similar findings in the US context. For example,
the goal of the multiracial categories of 1890 (Mulatto, Quadroon, Octaroon) was
not to represent multiculturalism as in the demand for multi-racial recognition on
the US census in the 1990s, but rather was an attempt to monitor racial mixing.
14 Anand Patwardhan’s documentary film Father, Son and Holy War (New York:
First Run Icarus Films, 1994) includes some striking examples of such rhetoric in
political speeches.
NOTES 225
15 Studies have shown that educating women correlates to successful family planning
and population control. See Sen (1994), especially Chapter 8 on “Women’s Agency
and Social Change.” Demographer Ashish Bose criticized the nonsensical yet
popular argument that Muslim polygamy leads to higher population growth (Bose
interviews 19 July 1996 and 4 November 1996).
16 The delay in processing the caste data collected in the past is another argument
used against the idea of an OBC caste count. P.Radhakrishnan, Professor at Madras
Institute of Development Studies, argues: “if the department has not completed
processing the data collected on the SCs and STs (accounting for only about one-
fourth of the total population) even nearly a decade after its collection, how will it
be able to process, not to speak of making available to the data-users in the
foreseeable future, caste data on 75 per cent of the total population?”
(Radhakrishnan 1999).
17 The census did gather information on types of economic activity, land ownership
and education, however, which may prove more useful than income data, given the
perhaps more extreme challenges of accuracy with the latter type of data.
18 Another factor in the jump in numbers was a switch in the official definition of
Scheduled Tribe status. From 1950, when the president promulgated lists of
Scheduled Tribes and Scheduled Areas, the special benefits for Scheduled Tribes
were only available if they lived in the specified areas. The removal of intrastate
area restrictions in the Scheduled Castes and Scheduled Tribes Orders
(Amendment) Act, 1976, removed the geographical element from the identification
of Scheduled Tribe members within states. See Chaube (1999). Yet, this too cannot
account for the entire population jump.
6
ªBackwardº Muslims and ªScheduled Casteº Ch ristians
1 This use of categories to challenge categories is not unusual. Class consciousness

has been used in attempts to destroy class itself; working class solidarity is the
basis of most efforts to undermine the class system. Likewise, some groups trying
to undermine racism and casteism are themselves based on low caste or racial
minority solidarity. Gayatri Chakravorty Spivak’s discussion of strategic
essentialisms touches on these ironies. Spivak discusses the use of strategic
essentialisms by those contributing to subaltern studies: “I would read it, then, as a
strategic use of positivist essentialism in a scrupulously viable political interest”
(Spivak 1987:205, emphasis in the original). Strategic essentialisms are also used
by subalterns themselves: “Class-consciousness on the descriptive level is itself a
strategic and artificial rallying awareness which, on the transformative level, seeks
to destroy the mechanics which come to construct the outlines of the very class of
which a collective conscious has been situationally developed” (Spivak 1987:205,
emphasis in the original).
2 The many lines along which reservation categories could be drawn has resulted in
so many claims that some are beginning to wonder what will be the next category
under contention. See for example Swapan Dasgupta’s article entitled, “Madness
Unreserved: Caste, Gender, Religion: Where Will it Stop?” (India Today 9 June
1997:27). Although many people challenging the official categories may be
226 NOTES
motivated by a sense that these categories do not reflect their unique senses of
identity, some are also very cognizant of the material benefits and opportunities that
are at stake if a reclassification occurs. Identities and interests are intertwined and
impossible to neatly untangle. Such untangling is not attempted here, but their
interaction is evident in the following case studies.
3 For a balanced perspective on the ways these group-based policies both overcome
group disparities and cause group tensions, see Mitra 1987, 1994.
4 This assumption pervaded many media and academic analyses of the decision to
extend central government reservations to the OBCs in 1990. See also Note 12 in
Chapter 8 for more sources of similar arguments against reservations. For example,
see Singh and Sharma’s argument in their book, tellingly entitled, Reservation
Politics in India: Mandalization of the Society (Singh and Sharma 1995).
5 Government officials and citizens are now struggling with the policy ramifications
of a debate which has raged among academics, classic examples being Louis
Dumont and F.G.Bailey: Is caste a religious institution rooted in Hinduism or a
sociological phenomenon characteristic of a variety of societies? (Dumont 1980,
Bailey 1957, 1960). I thank an anonymous reviewer of one of my articles (Jenkins
200 1a) for encouraging me to place this political debate in the context of this
academic debate.
6 Other groups have made similar demands for Muslim reservations, such as the
Islamic Council of India and the All-India Milli Council. See The Hindu 13 April
1996 and Indian Express 22 July 1996. An even larger group of Muslim
organizations has included the demand for reservations for Muslims as a religious
minority group in their Agenda 1999 for Muslims. See “Backward, Dalit Muslims
campaign to secure rights sparks controversy,” Times of India 12 September 1999.
7 Vilas Sonavane is a member of Advisory Board of the All India Muslim OBC
Organisation. See also V.Date, “Reservation Demand for ‘Muslim OBCs’ Gains
Momentum,” Times of India 10 December 1999.
8 Memorandum from Shabbir Ansari, President of the All India Muslim OBC
Organisation, to Welfare Minister B.S.Ramoowalia, 16 September 1996. Other
organizations have made similar demands, including the All India Muslim
Congress. See “Reservation for Dalit Muslims Sought,” Times of India 16
December 1999.
9 In addition to the broader associations discussed here, the extension of Scheduled
Caste reservations to Sikhs and Buddhists has inspired “similarly situated groups
among Muslims,” such as Halal Khors, to also seek official SC status (Times of
India 18 January 1996:10).
10 Notably, there is no corresponding religious bar to Scheduled Tribe or Other
Backward Class membership, perhaps because non-Hindu tribals or lower classes are
less of a challenge than non-Hindu untouchables to this ideological association
between caste and Hinduism.
11 See an excellent overview of the Dalit Christian movement and its ideology (Wyatt
1998). For a list of some of the rallies, strikes, bills and meetings with public
officials over this issue, see the National Coordination Committee for Dalit
Christians’ pamphlet (for distribution to Members of Parliament), Demand for
Restoration of Reservation for Christian Dalits (1996:17–18). The United Front
Government’s 1996 Common Minimum Program included the extension of
reservation benefits to Dalit Christians. Deccan Herald-Bangalore 16 July 1996.
NOTES 227
12 The National Coordination Committee for SC Christians is sometimes called the

National Coordination Committee for Dalit Christians. Other organizations
involved in these issues include the All India Christian People’s Forum and the All
India United Christians Movement for Equal Rights.
13 The author attended some meetings of this organization in Delhi in 1995–6.
Examples of speeches and information about the Dalit Solidarity Programme can
be found in a book written by two of the Programme’s founders, Bhagwan Das, a
Buddhist, and James Massey, a Christian (Das and Massey 1995).
7
Hindu nationalism and selective inclusion
1 For an excellent brief introduction to the ideology and politics of the Hindu
nationalists, see Tapan Basu et al. (1993). For further analysis see Christoffe
Jaffrelot (1996) and Thomas Blom Hansen (1999).
2 Eleanor Zelliot notes that these conversions started in 1956 but also continued
afterward. Exact numbers of converts are difficult to estimate but census
figures document a large leap in the number of Buddhists in India: 180,823 in 1951,
before Ambedkar’s conversion movement began, and 3,250,227 in 1961 (Zelliot
1996:223).
3 This speech was delivered at the last meeting of the Minorities Committee, 13
November 1931. See also Gandhi’s correspondence on this subject (Pyarelal 1984).
B.R. Ambedkar’s critique of Gandhi’s approach can be found in Ambedkar’s What
Congress and Gandhi Have Done to the Untouchables (1945).
4 Limiting Scheduled Caste status to Hindus, Sikhs and Buddhists, argues Dieter
Conrad, can “be understood only as a sanction against apostasy from Hinduism,”
part of a trend toward what he calls legal Hindutva (Conrad 1995).
5 The strict association of the SC category with Hinduism is not just an assumption
and argument of Hindu extremists. As discussed above, Gandhi opposed
untouchability but always felt that “Harijans” were a part of Hindu society.
Preserving unity in this way was very important to him. Gandhi’s grandson
Rajmohan Gandhi has drawn a parallel to Abraham Lincoln, who wanted to end
slavery in the United States but also, if possible, preserve the union (Gandhi
interview 21 September 1996, Gandhi 1995:242).
6 See the following news coverage of the expanding rolls of OBCs on the national list:
“51 castes included in OBC Central List” (Hindustan Times 5 September 2000).
“90 Castes Added to OBC List for Jobs” (Indian Express 18 March 2000). “123
More Castes in OBC List” (The Hindu 20 November 1999). “Sheila puts Delhi Jats
on OBC List,” a response to the Vajpayee government putting Rajastani Jats on the
central OBC list (Indian Express 23 October 1999). “96 More Castes May Be
Included in OBC List” (Hindustan Times 15 June 1999).
8
Class, classification and creamy layers
1 The phrase “truly backward,” used by the Supreme Court of India in Indra
Sawhney v. Union of India (1992) 1992 Supp (3) S.C.C. 217:1992 S.C.C. (L&S)
228 NOTES
Supp 1:(1992) 22 A.T.C. 385, echoes that of American sociologist William Julius
Wilson, who in a different context, argued that a race-based analysis might
overlook the “truly disadvantaged” by ignoring class distinctions (Wilson 1987).
Advocates of a class approach continue to add to the debate over affirmative action
in the United States (Kahlenberg 1996).
2 Indra Sawney v. Union of India, supra note 1.
3 See the Reports of the Backward Classes Commissions of 1955 and 1980. These
are known as the Kalelkar Commission and Mandal Commission, respectively,
named after their Chairmen. There are also numerous state level commissions and
reports.
4 “Communities” in this context can be considered both a euphemism for castes
within Hindu society and a term for the caste-like subdivisions within other
religions. Notably the OBC category, unlike the SC category, is open to all
religions, so communities within each religious community can be declared OBCs
by state or national level commissions. Whether disadvantaged subcommunities
within these other religions are actually “castes” and not just communities is not only
a linguistic and sociological question but also a political question, due to the
current demands for Scheduled Caste status for Muslims and Christians, as
described in Chapters 6 and 7.
5 Triloki Nath v. State of Jammu and Kashmir (1969) I S.C.R. 103 A 1960 S.C.I.
6 Based on the postelection poll survey of 1996, Mitra and Singh charted the
correlation between “caste” and “class.” The official SC, ST, OBC and other
categories were used a proxies for caste. Data used to define class included
occupation, assets, type of house and income. Mitra and Singh found a
close correlation between caste and class: The Scheduled Castes and Tribes were
most highly represented in the “very poor class” category, while the Other
Backward Classes dominated the “poor class” category, and the other (upper)
castes were represented at more than twice their percentage in the population in the
“upper class” category. According to their data, Scheduled Castes (19 per cent of
the population) are 30 per cent of the very poor, and Scheduled Tribes (10 per cent
of the population) are 15 per cent of the very poor. The OBCs (38 per cent of the
population) are 40 per cent of the very poor, 42 per cent of the poor, 23 per cent of
the middle class and 23 per cent of the upper class. The upper castes, in contrast,
who are only 34 per cent of the population, make up almost 70 per cent of the
upper class category (Mitra and Singh 1999:189 Table 6.1).
7 Chiranjit Lal v. Union of India (1950) S.C.R. 869.
8 State of Madras v. Champakam Dorairajan (1951) 1 S.C.R. 226, and
Venkataramana v. State of Madras (1951) 1 S.C.R. 229.
9 This language was added in the Constitution (First Amendment) Act, 1951, Sec. 2.
See Bakshi (1996:23).
10 See the Mandal Commission Report on Backward Classes (1980). The current
National Backward Classes Commission uses similar criteria on its questionnaire
for those groups claiming backward status. Forty-six questions on a variety of
social, educational, economic, and other factors range from the number of
community members holding government jobs to the number of members in
primitive housing. See National Commission for Backward Classes (1994).
11 More specifically, “[t]he reservation is applicable to all civil posts and services under
the Govt. of India, Public Sector Undertakings, Financial Institutions including
NOTES 229
Public Sector Banks, autonomous bodies, statutory and semi govt. bodies and
voluntary agencies receiving grants from the Govt.” (Nabhi 2001: 374). The Other
Backward Classes for purposes of this new policy were defined as the castes and
communities on both the Mandal Commission’s list and lists prepared by state
governments.
12 These were not the first anti-reservation agitations. See Mitra (1987). On the
relationship between the anti-Mandal protests and the dispute between Hindus and
Muslims over the Babri Masjid in Ayodhya, see Parikh “Religion, Reservations and
Riots” (1998). The India Today coverage of the protests in 1990 is notable for its
anti-Mandal slant; the Seminar issue of November 1990, entitled “Reserved
Futures,” is devoted to reservations, including Ghanshyam Shah’s piece on the
“Agitations in Gujarat,” which also discusses the broader anti-Mandal protest.
Other coverage of the opposition to Mandal includes: Balagopal, K. (1990) “This
Anti-Mandal Mania,” Economic and Political Weekly 25:40, 6 October, pp. 2231–
4; Chopra Pran (1990) “Reservations: the economic criteria,” Hindustan Times 1
October; Karlekar, Hiranmay (1990) “The Wrath of Youth,” Indian Express 4
October; Mammen, Matthew (1990) “Rising Militancy in Bihar: Mandal Protests
spark caste war,” Hindustan Times 16 October.
13 Indra Sawhney v. Union of India (1992) 1992 Supp (3) S.C.C. 217:1992 S.C.C.
(L&S) Supp 1:(1992) 22 A.T.C. 385.
14 Justice R.N.Prasad, Chair of “creamy layer” committee and author of creamy layer
ruling, interview, New Delhi, 18 September 1996. Whether one’s wife holds such a
position is not accounted for on the “Creamy layer” forms in Delhi. Pressure to
raise the one lakh rupee income ceiling has subsequently occurred (The Hindu
“India ceiling for OBC creamy layer to be raised” 24 July 1998).
15 From Creamy Layer Schedule, Ministry of Personnel Office Memorandum No.
36012/22/93-Estt. (SCT). Subject: Reservation for Other Backward Classes in
Civil Posts and Services under the Government of India.
16 See also Chapter 4 in this volume, on caste certificates. A telling example of such
problems with identity documents and corruption is the growing number of people
losing their official identities entirely when relatives manage to get them declared
dead in order to inherit their property. This has sparked an Association of Dead
People fighting to be declared alive. Barry Bearak, “Back to Life in India, Without
Reincarnation,” New York Times 24 October 2000.
17 Many of these concerns over the use of economic data have also been raised by
political scientist Sunita Parikh at a conference on affirmative action in India, the
United States and South Africa in a discussion of the “creamy layer” rule in India
(8 November 1997). The conference was entitled “Rethinking Equality in the
Global Society” and was held at Washington University.
18 Indra Sawhney v. Union of India (1992) 1992 Supp (3) S.C.C. 217:1992 S.C.C.
(L&S) Supp 1:(1992) 22 A.T.C. 385.
230 NOTES
9
Women's reservations and representation
1 Times of India (1998) “RJD-SP Stall Introduction of Women’s Quota Bill,” 14 July.
Similar scenes erupted in 1996 and 2000 as the bill has been repeatedly introduced
only to die again.
2 The Women’s Reservation bill would reserve one-third of seats in the lower houses
of the national parliament (Lok Sabha) and state assemblies (Vidham Sabhas). The
reserved seats would be rotated randomly so the same districts would not be
permanently reserved for female candidates only.
3 Kishwar herself is critical of the Women’s Reservation Bill (Kishwar 1996b) and
offers an alternative (Narayan et al. 2002).
4 An emphasis on “difference” (such as recognition of distinct class or racial
experiences of women) is a trend in western feminist scholarship, offsetting a
history of inattention to such differences; yet the study of women’s movements
elsewhere reveals that the “politics of difference” itself plays out in different ways
(Tripp 2000). In the Indian context, the government pays much attention to religion
and caste, so the importance of “difference” is hardly a new revelation and could be
viewed instead as a longstanding political challenge for feminist activists.
5 On the relationship between the nationalist and women’s movements in India, see
also Nair 1996 and Chatterjee 1989.
6 After elections to the central legislature under this plan, 3.4 per cent of the legislators
were women (Committee on the Status of Women 1974:356).
7 Examples in this report of the arguments for legislative reservations include the
need for political empowerment of women to precede socio-economic
empowerment in the Indian context, the need to compel political parties to shift
strategies in candidate selection, and the value of “a body of spokesmen of the
women’s cause” in legislatures (Committee on the Status of Women 1974:302–3).
8 Some low caste organizations in the 1970s drew parallels between their oppression
and women’s oppression, undeterred by continuing distinctions between
communities and categories. Yet the more frequent pattern of activism by women
and other disadvantaged groups was continuing division between urban elite
organizations and rural, grassroots movements (Kumar 1995:63).
9 Alva was appointed to the upper house of Parliament by Indira Gandhi in 1974.
She later was appointed by Rajiv Gandhi as head of the Women’s Department
within the Ministry of Human Resource Development. She has also served as
Minister of State for Youth Affairs, Sports and Women.
10 The debate in 2000 is described in an article entitled “Sparks Fly Over Delay in
Tabling Women’s Bill.” The Hindu (26 August 2000).
11 Such reports about the poor treatment of some female council members help
explain survey data suggesting that women tend to trust local government less than
men, even after the implementation of women’s reservations at the local level
(Mitra and Singh 1999:237)
12 “More Anti-Women than the Common Man?: Demanding Sub-Quotas,” The
Statesman, 6 January 2000.
13 Arguments about whether “real” women benefit from reservations have also been
raised in cases in which eunuchs have been elected to local offices in seats reserved
for women. For example, four eunuchs were elected in the state of Madya Pradesh,
including one mayor. Characterized in the media as a sign of voters’ discontent
with politics and politicians, this also suggests ambivalent public views of
women’s reserved seats (Daily Excelsior (2000) “Eunuch’s Election Shows Voters
NOTES 231
Disgust With Politics,” 2 January). A judge ruled that Mayor Kamla Jaan is male,
based on voter registration records, a photo identity card and other official
documents, and declared Jaan’s election null and void (Statesman (2002) “Court
verdict triggers gender debate on Eunuchs,” 1 September).
14 BJP MP Uma Bharati, quoted in Meenakshi Nath (1996) “Cutting Across Party
Lines: Women Members of Parliament Explain Their Stand on Reservation
Quotas,” Manushi, no. 96, September-October, p. 11.
Bibliography
Agrawal, Puru Shottam (1994) “‘Kan Kan Mein Vyape Hein Ram’: The slogan as
metaphor of cultural interrogation,” The Oxford Literary Review, 16.
Ahmad, Imtiaz (1966) “Ashraf-Ajlaf dichotomy in Muslim social structure in India,”
Indian Economic and Social History Review, 3, pp. 344–50.
—(ed.) (1978) Caste and Social Stratification Among Muslims in India, Delhi: Manohar.
—(1980) “Job reservation for Muslims?,” The Hindustan Times, 25 January.
—(1981) “Muslim educational backwardness: An inferential analysis,” Economic and
Political Weekly, 16, 36, pp. 1457–65, 5 September.
Ahmad, Tufail (1997) “Redefining minority,” The Observer, 1 September.
Aijazuddin, Ahmad (1995) Muslims in India: Their Educational, Demographic and Socio-
Economic Status with Comparative Indicators for Hindus, Sikhs, Christians and Other
Communities, Based on a Singular and Systematic Field Survey 1990±1993. Volume 3.
National Capital Territory of Delhi, New Delhi: Inter-India Publications.
Alborn, Timothy L. (1999) “Age and empire in the Indian census, 1871–1931,” Journal
of Interdisciplinary History, 30:1, Summer, pp. 61–89.
Ali, Azra Asghar (2000) The Emergence of Feminism Among Indian Muslim Women
1920±1947, Karachi: Oxford University Press.
Ali, M.M. (1998) “India’s new BJP government postpones its hard-line agenda,” The
Washington Report on Middle East Affairs, May/June, pp. 59–91.
Alonso, William and Paul Starr (eds) (1987) The Politics of Numbers, New York: Russell
Sage.
Ambedkar, B.R. (1945) What Congress and Gandhi Have Done to the Untouchables,
Bombay: Thacker and Co. Ltd.
Anderson, Benedict (1991) Imagined Communities, London: Verso.
—(1998) The Spectre of Comparisons: Nationalism, Southeast Asia and the World,
London: Verso.
Anderson, J.D. (1982) The Peoples of India, Delhi: Bimla Publishing House (first
published in 1913).
Ansari, Ghaus (1960) Muslim Caste in Uttar Pradesh, Lucknow: Ethnographic and Folk
Culture Society.
Ansari, Shabbir (1996) Memorandum from President of the All India Muslim OBC
Organisation to Welfare Minister B.S.Ramoowalia, 16 September.
Anthropological Survey of India (1996) People of India National Series, vol. VIII.
Communities, Segments, Synonyms, Surnames and Titles, New Delhi: Oxford
University Press.
Appadurai, Arjun (1993) “Number in the colonial imagination,” In: Breckenridge, Carol
A. and Peter Van der Veer (eds) Orientalism and the Post-Colonial Predicament,
Philadelphia: University of Pennsylvania Press.
Appiah, Kwame Anthony (1995) “African identities,” In: Nicholson, Linda and Steven
Seidman (eds) Social Postmodernism: Beyond Identity Politics, Cambridge: Cambridge
University Press, pp. 103–15.
BIBLIOGRAPHY 233
Association for Promoting Education and Employment of Muslims (1996) Resolution on

Promotion of Employment Reservation in Public Employment and Management of
Wakf Properties, distributed at Third National Conference, Hyderabad, India, 18
August.
Backward Classes Commission (1953) Report (The Kalelkar Commission), New Delhi:
Government of India.
Backward Classes Commission (1980) Report (The Mandal Commission), New Delhi:
Bailey, F.G. (1957) Caste and the Economic Frontier, Manchester University Press.
—(1960) Tribe, Caste and Nation, Manchester University Press.
Bakshi, P.M. (1996) The Constitution of India with Comments and Subject Index, Delhi:
Universal Law Publishing Co. Pvt. Ltd.
Ballhatchet, Kenneth (1998) Caste, Class and Catholicism in India 1789±1914, London:
Curzon Press.
Bandyopadhyay, Sekhar (1997) Caste, Protest and Identity in Colonial India: The
Namasudras of Bengal, 1872±1947, London: Curzon Press.
Banerjee, Kakoli (1999) “Gender stratification and the contemporary marriage market in
India,” Journal of Family Issues, 20:5, pp. 648–76.
Banton, Michael (1997) Ethnic and Racial Consciousness, London: Longman.
Barrier, N.Gerald (1981) The Census in British India: New Perspectives, New Delhi:
Manohar.
Basu, Mahadeb Prasad (1990) “A short note on the portrait building system,” Journal of
the Anthropological Survey of India, 39, pp. 113–16.
Basu, T., P.Datta, S.Sarkar, T.Sarkar and S.Sen (1993) Khaki Shorts and Saffron Flags: A
Critique of the Hindu Right, New Delhi: Orient Longman.
Bates, Crispin (1995) “Race, caste and tribe in Central India: The early origins of Indian
anthropometry,” In: Robb, Peter (ed.) The Concept of Race in South Asia, Delhi:
Oxford University Press.
Bayly, Susan (1995) “Caste and ‘Race’ in the colonial ethnography of India,” In: Robb,
Peter (ed.) The Concept of Race in South Asia, Delhi: Oxford University Press.
—(1999) Caste, Society and Politics in India from the Eighteenth Century to the Modern
Age, Cambridge: Cambridge University Press.
Berreman, Gerald D. (1960) “Caste in India and the United States,” American Journal of
Sociology, 66:4, pp. 120–7.
Bhabha, Homi K. (1994) The Location of Culture, London and New York: Routledge.
Bhagavan, Manu Belur (1994) “Creating identities: Educating the backward classes at
Mysore University,” M.A.Thesis, University of Texas at Austin.
Bharatiya Janata Party (1998) Manifesto [http://bjp.org/manifesto/chap2.htm].
Bhatnagar, Rakesh (2001) “Plea in SC Challenges Reservations,” Times of India Online,
28 May.
Bidwai, Praful (1996) “Age of Empowerment: Muslim OBCs Discover Mandal,” The Times
of India, 12 September, p. 12.
Bose, Ashish (1991) Population of India, Delhi: B.R.Publishing Corporation.
Brass, Paul (1985) Ethnic Groups and the State, Totowa, New Jersey: Barnes and Noble
Books.
—(1994) The Politics of India Since Independence, Cambridge: Cambridge University
Press.
234 BIBLIOGRAPHY
Bremen, Jan van and Akitoshi Shimizu (1999) Anthropology and Colonialism in Asia and
Oceania, London: Curzon Press.
Bright, Kristy (1994) “Classified ethnographies: The production and collection of racial
‘Evidence’ in India, 1908–1915,” typescript presented at “Revisioning Culture”
University of California Santa Cruz.
Brown, Melissa (2001) “Ethnic classification and culture: The case of the Tujia in Hubei,
China,” Asian Ethnicity, 2:1, March, pp. 55–72.
—(2002) “Local government agency: Manipulating Tujia identity,” Modern China
(forthcoming July).
Budman, Pamela (1997) “The long goodbye: The University of California bids farewell to
affirmative action,” Lingua Franca, June/July, p. 29.
Bumiller, Elisabeth (1990) May You Be the Mother of a Hundred Sons: A Journey Among
the Women of India, New York: Random House.
Bureau of National Affairs (1997) “Race, ethnic targeted tests show little use of
multiracial choice, Census Bureau finds,” Daily Labor Report, 95, 16 May, p. A-5.
Bureau of National Affairs, Inc. (1999) “Changes to EEO, affirmative action reporting
pending,” Fair Employment Practices, 35:875, 10 June, p. 72.
Burman, B.K.Roy (1977) “Census of India—scheduled castes and scheduled tribes,” In:
Bose, Ashish et al. (eds) Population Statistics in India, New Delhi: Vikas Publishing
House.
—(1992) Beyond Mandal and After: Backward Classes in Perspective, New Delhi: Mittal
Publications.
Caroll, Lucy (1978) “Colonial perceptions of Indian society and the emergence of caste(s)
associations,” Journal of Asian Studies, 37:2, pp. 233–50.
Chandra, Kanchan (2000) “The transformation of ethnic politics in India: The decline of
Congress and the rise of the Bahujan Samaj Party in Hoshiarpur,” Journal of Asian
Studies, 59:1, February, pp. 26–61.
Chanana, Karuna (1993) “Accessing higher education—the dilemma of schooling
women, minorities, scheduled castes and scheduled tribes in contemporary India,” In:
Altbach, Philip and Suma Chitnis (eds) Higher Education Reform in India, New Delhi:
Sage, pp. 115–54.
Charsley, Simon (1996) “‘Untouchable’: What is in a Name?” Journal of the Royal
Anthropological Institute, 2:1, March, pp. 1–23.
Chatterjee, Partha (1989) “The nationalist resolution of the women’s question,” In
Kumkum Sangari and Sudesh Vaid (eds) Recasting Women: Essays in Indian Colonial
History, New Brunswick, NJ, : Rutgers University Press, pp. 233–53.
Chaube, S.K. (1999) “The scheduled tribes and Christianity in India,” Economic and
Political Weekly, 27 February, p. 524.
Cohen, Anthony P. (1994) Self Consciousness: An Alternative Anthropology of Identity,
London and New York: Routledge.
Cohn, Bernard (1987) An Anthropologist Among the Historians and Other Essays, New
Delhi: Oxford University Press.
Cohn, Bernard (1996) Colonialism and its Forms of Knowledge: The British in India,
Princeton: Princeton University Press.
Committee on the Status of Women (1974) Towards Equality: Report of the Committee
on the Status of Women, New Delhi: Government of India, Ministry of Education and
Social Welfare, Department of Social Welfare.
BIBLIOGRAPHY 235
Conlon, Frank F. (1981) “The census of India as a source for the historical study of
religion and caste,” In: Barrier, N.Gerald (ed.) The Census in British India: New
Perspectives, New Delhi: Manohar, pp. 103–18.
Conrad, Dieter (1995) “The personal law question and Hindu nationalism,” In: Dalmia,
V. and H.Von Stieten (eds) Representing Hinduism: The Construction of Religious
Tradition and National Identity, New Delhi: Sage Publications.
Constable, Pamela and Rama Laxmi (2000) “Counting on complications in tally of India’s
1 billion,” The Washington Post, 26 February, p. A12.
Corbridge, Stuart (2000) “Competing equalities: The scheduled tribes and reservations
system in India’s Jharkhand,” Journal of Asian Studies, 59:1, February, pp. 62–85.
Cose, Ellis (1997) Color-Blind: Seeing Beyond Race in a Race-Obsessed World, New
York: Harper Perennial.
Coutin, S.B. (1998) “From refugees to immigrants: The legalization strategies of
Salvadoran immigrants and activists,” International Immigration Review, 32:4, Winter,
pp. 901–25.
Cunningham, Clark D. (1991) “Why American lawyers should go to India: Retracing
Galanter’s intellectual odyssey,” Law and Social Inquiry, 16:4, Fall, pp. 777–808.
Dale, Angela and Clare Holdsworth (1995) “The construction of ethnicity in the 1991
British census: Evidence from Microdata,” In: Bureau of the Census. 1995 Annual
Research Conference: Proceedings, Washington DC: US Department of Commerce.
Daniel, E.Valentine (1984) Fluid Signs: Being a Person the Tamil Way, Berkeley:
University of California Press.
Daniel, Jose (date uncertain, received 1996) Demand to Extend Statutory and Non-Statutory
Benefits to the Scheduled Castes Converted to Christianity at Par with the SCs in
Hinduism, Sikhism and Buddhism by Amending Para 3 of the President's Constitution
(Scheduled Castes) Order 1950, Document of the All India United Christians
Movement for Equal Rights and National Coordination Committee for SC Christians.
Das, Bhagvan (2000) “Moments in a history of reservations,” Economic and Political
Weekly, 28 October, pp. 3831–4.
Das, Bhagwan and James Massey (1995) Dalit Solidarity, Delhi: ISPCK.
Das, Veena (1996) Critical Events: An Anthropological Perspective on Contemporary
India, Delhi: Oxford University Press.
Dasgupta, Swapan (1997) “Madness unreserved: Caste gender, religion: Where will it
stop?” India Today, 9 June, p. 27.
Davidoff, Judith (1997) “Rethinking race,” Isthmus, 13 June, pp. 35–7.
Davis, Adrienne D. (1996) “Identity notes part one: Playing in the light,” The American
University Law Review, 45:3, February, pp. 695–720.
Debnath, Biswanth (1999) “Crisis of Indian anthropology,” Economic and Political
Weekly, 30 October.
Des Chene, Mary (1999) “Military ethnology in British India,” South Asia Research, 19:
2, Autumn, pp. 121–36.
Deshmukh, V.S. (1988) “Scrutiny of scheduled certificates: Why and how?” Tribal
Research Bulletin, 10:2, pp. 23–6.
Deshpande, Satish and Nandini Sundar (1998) “Caste and the census: Implications for
society and the social sciences,” Economic and Political Weekly, 33:32, 8 August,
pp. 2157–9.
Deshpande, V.S. (1990) “What does SC mean?” Hindustan Times, 28 May.
236 BIBLIOGRAPHY
de Zwart, Frank (2000) “The logic of affirmative action: Caste, class and quotas in India,”
Acta Sociologica, 43, pp. 235–49.
Dirks, Nicholas (1992a) “Castes of mind,” Representations, 37, pp. 56–78.
—(1992b) Colonialism and Culture, Ann Arbor: University of Michigan Press.
—(1997) “The policing of tradition: Colonialism and anthropology in southern India,”
Comparative Studies in Society and History, 39:1, January, pp. 182–212.
—(2001) “Discriminating difference: The postcolonial politics of caste in India,” In:
Burguiere, Andre and Raymond Grew (eds) The Construction of Minorities, Ann
Arbor: University of Michigan Press.
Drass, K.A., P.R. Gregware and M. Musheno (1997) “Social, cultural, and temporal
dynamics of the AIDS case congregation: Early years of the epidemic,” Law & Society
Review, 31:2, pp. 267–99.
Dreze, Jean and Amartya Sen (1995) India: Economic Development and Social
Opportunity, Delhi: Oxford University Press.
—(eds) (1997) Indian Development: Selected Regional Perspectives, New Delhi: Oxford
University Press.
Dudley, William (ed.) (1997) African Americans: Opposing Viewpoints, San Diego:
Greenhaven Press.
Dugger, Celia W. (2000) “Connecting rural India to the world,” The New York Times, 28
May.
Dumont, Louis (1970) Homo Hierarcicus: The Caste System and Its Implications,
Chicago: University of Chicago Press.
Dunn, Dana (1993) “Gender inequality in education and employment in the scheduled
castes and tribes of India,” Population Research and Policy Review, 12, pp. 53–70.
Durkheim, Emile (1957) Professional Ethics and Civic Morals, Translated by Cornelia
Brookfield, London and NewYork: Routledge.
Dushkin, Lela (1979) “Backward class benefits and social class in India, 1920–1970,”
Economic and Political Weekly, 14:7, April, pp. 661–7.
Engineer, Asghar Ali (1996) “Elections and the Muslims,” The Hindu, 13 April, p. 10.
Engineer, Asghar Ali (ed.) (1991) Mandal Commission Controversy, Delhi: Ajanta
Publications.
Escobar, Arturo and Sonia E.Alvarez (eds) (1992) The Making of Social Movements in Latin
America: Identity, Strategy and Democracy, Boulder: Westview.
Essed, Philomena (2001) “Multi-identifications and transformations: Reaching beyond
racial and ethnic reductionisms,” Social Identities, 7:4, pp. 493–509.
Fanon, Frantz (1967) Black Skin, White Masks, New York: Grove Press, Inc.
Faundez, J. (1994) Affirmative Action: International Perspectives, Geneva: International
Labour Office.
Fernandez-Armesto, Felipe (1995) Millennium: A History of Our Last Thousand Years,
London: Black Swan.
Fletcher, Pamela R. (1994) “A mythology with a mission: African-American intellectuals
debate race,” Utne Reader, November/December, pp. 89–90.
Fogel, Aaron (1996) “Bruegel’s The Census at Bethlehem and the visual anticensus,”
Representations, 54, Spring, pp. 1–27.
Forbes, Geraldine (1996) Women in Modern India, Cambridge: Cambridge University
Press.
Ford, Christopher A. (1994) “Administering identity: The determination of ‘Race’ in race-
conscious law,” California Law Review, 82:4, July, pp. 1231–85.
BIBLIOGRAPHY 237
Forrester, Duncan B. (1980) Caste and Christianity: Attitudes and Policies of Anglo-
Saxon Protestant Missionaries in India, London and Dublin: Curzon Press; Atlantic
Highlands, New Jersey: Humanities Press.
Foucault, Michel (1979) Discipline and Punish: The Birth of the Prison, Translation by
Alan Sheridan, New York: Vintage Books.
Frankel, Francine R. (1988) “Middle classes and castes in India’s politics: Prospects for
political accommodation,” In: Atul Kohli (ed.) India's Democracy: An Analysis of
Changing State-Society Relations, Princeton: Princeton University Press, pp. 225–61.
Fredrickson, George (1981) White Supremacy: A Comparative Study in American and
South African History, Oxford: Oxford University Press.
Friedman, Susan Stanford (1995) “Beyond white and other: Relationality and narratives
of race in feminist discourse,” Signs, 21:1, pp. 1–49.
Frost, Robert (1995) Collected Poems, Prose and Plays, New York: Library of America.
Frykenberg, Robert Eric (1985) “Caste, morality and western religion under the Raj,”
Modern Asian Studies, 19:2, pp. 321–52.
—(1987) “The concept of ‘Majority’ as a devilish force in the politics of modern India,”
Journal of Commonwealth and Comparative Politics, 25:3, November, pp. 267–74.
—(1993) “Constructions of Hinduism at the nexus of history and religion,” Journal of
Interdisciplinary History, 23:3, Winter, pp. 523–50.
Gaikwad, J.S. (1986) “A demographic profile of tribals in Maharashtra state,” Tribal
Research Bulletin, 9:1, pp. 20–1.
Galanter, Marc (1974) “Why the ‘Haves’ come out ahead: Speculations on the limits of
legal change,” Law and Society Review, 9:1, Fall, pp. 95–160.
—(1984) Competing Equalities: Law and the Backward Classes in India, Delhi: Oxford
University Press.
—(1986) “Pursuing equality: An assessment of India’s policy of compensatory
discrimination for disadvantaged groups,” In Dilip K.Basu and Richard Sisson (eds)
Social and Economic Development in India: A Reassessment, University of California:
Sage Publications.
—(1992) Law and Society in Modern India, Delhi: Oxford University Press.
Gandhi, Maneka (1996) “And the one who differs… Maneka Gandhi on the Women’s
Reservation Bill,” Manushi, 96, September-October, p. 18.
Gandhi, Rajmohan (1995) The Good Boatman: A Portrait of Gandhi, New Delhi: Viking.
Geertz, Clifford (1973) The Interpretation of Cultures, New York: Basic Books.
George, Nirmala (1999) “Government drops move on collating caste data,” Indian
Express, 5 March.
Ghosh, S.K. (1996) Reservation in Services: A Manual for Government and Public Sector,
Delhi: Akshar Publications.
Giddens, Anthony (1977) Studies in Social and Political Theory, London: Hutchinson.
Gladney, Dru (1996) Muslim Chinese: Ethnic Nationalism in the People's Republic,
Cambridge: Harvard University Press.
Gopalswami, R.A., Registrar General (1953a) Census of India, 1951 Volume 1 India, Part
1-A- Report, New Delhi: Government of India.
—(1953b) “Special groups—1951 census,” In: Census of India Paper No. 4, New Delhi:
Gottschalk, Peter (2000) Beyond Hindu and Muslim: Multiple Identity in Narratives from
Village India, Oxford: Oxford University Press.
Government of India Act (1935) [25 & 26 Geo. 5. Ch. 42], Great Britain India Office.
238 BIBLIOGRAPHY
Government of India, Ministry of Welfare (1996) Annual Report 1995±1996, New Delhi:
Government of India, Ministry of Education and Social Welfare, Department of Social
Welfare (1975) Government Order Ms. No. 477, 27 June.
Green, Daniel M. (2002) Constructivism and Comparative Politics, Armonk, NY:
M.E.Sharpe.
Guinier, Lani (1994) The Tyrany of the Majority: Fundamental Fairness in
Representative Democracy, New York: The Free Press.
Guha, Ranajit (1996) “The small voice of history,” In: Amin, Shahid and Dipesh
Chakrabarty (eds) Subaltern Studies IX, Delhi: Oxford Univerisity Press.
Guha, Sumit (1998) “Lower strata, older races, and aboriginal peoples: Racial
anthropology and mythical history past and present,” Journal of Asian Studies, 57:2,
May, pp. 423–41.
—(1999) Environment and Ethnicity in India 1200±1991, Cambridge: Cambridge
University Press.
—(2001) “The politics of identity and enumeration in India c. 1600–1981 Mughal,
Maratha, British and Indian,” typescript.
—(2001) Health and Population in South Asia from the Earliest Times to the Present,
London: Hurst.
Gunther, R. and A.Mughan (1993) “Political institutions and cleavage management,” In:
Weaver, K. and B.Rockman (eds) Do Institutions Matter?, Washington, DC: Brookings
Institution, pp. 272–301.
Hacking, Ian (1990) The Taming of Chance, Cambridge: Cambridge University Press.
Hamid, Navaid (1996) “Association for promoting education and employment of
Muslims: A brief note,” Distributed to press at conference in Hyderabad, 18 August.
Handler, Joel F. (1992) “Postmodernism, protest, and the new social movements,” Law
and Society Review, 26, p. 697.
Hansen, Thomas Blom (1999) The Saffron Wave, Princeton: Princeton University Press.
Harper, Edward B. (1968) “A comparative analysis of caste: The United States and
India,” In: Singer, Milton and Bernard S.Cahn (eds) Structure and Change in Indian
Society, Chicago: Aldine Publishing Company, pp. 51–77.
Harper, Susan Billington (1992) “The politics of conversion: The Azariah-Gandhi
controversy over Christian mission to the depressed classes in the 1930s,” Indo-British
Review: A Journal of History, pp. 147–75.
Hasan, Zoya (2000) “Representation and redistribution: the new lower caste politics of
North India,” In: Frankel, Francine R. et al. (eds) Transforming India: Social and
Political Dynamics of Democracy, Delhi: Oxford University Press, pp. 146–75.
Heath, Shirley Brice (1995) “Race, ethnicity, and the defiance of categories,” In: Hawley,
Willis D. and Anthony W.Jackson (eds) Toward a Common Destiny, San Francisco:
Jossey-Bass Publishers.
Heeks, Richard (1996) India's Software Industry, Thousand Oaks: Sage Publications.
Horowitz, Donald L. (1985) Ethnic Groups in Conflict, Berkeley: University of California
Press.
—(1994) “Democracy in divided societies,” In: Diamond, Larry and Marc F. Plattner
(eds) Nationalism, Ethnic Conflict and Democracy, Baltimore: Johns Hopkins
University Press.
Hoskins, Catherine and Shirin M.Rai (1998) “Gender, class and representation: India and
the European Union,” European Journal of Women's Studies, V, pp. 345–65.
BIBLIOGRAPHY 239
Hudson, R.B. (1999) “Conflict in today’s aging politics: New population encounters old
ideology,” Social Service Review, 73:3, September, pp. 358–79.
Hunter, C. and Nixon, J. (1999) “The discourse of housing debt—the social construction
of landlords, lenders, borrowers and tenants,” Housing Theory and Society, 16:4,
pp. 165–78.
Hutton, J.H. (1946) Caste in India, its nature, function and origins, Cambridge: The
University Press.
Human Rights Watch (2001) Caste Discrimination: A Global Concern, Report for the
United Nations World Conference Against Racism, Racial Discrimination, Xenophobia
and Related Intolerance, Durban, South Africa, September, [http://www.hrw.org/
reports/2001/globalcaste/].
Ilaiah, Kancha (1996) Why I am Not a Hindu: A Sudra Critique of Hindutva Philosophy,
Culture and Political Economy, Calcutta: Samya.
Inden, Ronald (1990) Imagining India, Oxford: Basil Blackwell.
India Missions Association (1995) Scheduled Caste People Group Profiles, Madras: India
Missions Association.
Irschick, Eugene (1969) Politics and Social Conflict in South India: The Non-Brahman
Movement and Tamil Separatism, 1916±1929, Berkely and Los Angeles: University of
California Press.
—(1994) Dialogue and History: Constructing South India, 1795–1895, Berkeley:
University of California Press.
Jaffrelot, Christophe (1996) The Hindu Nationalist Movement and Indian Politics,
London: Hurst and Co.
—(2000) “The rise of other backward classes in the Hindi belt,” Journal of Asian Studies,
59:1, February, pp. 86–108.
Jalali, Rita (1993) “Preferential policies and the movement of the disadvantaged: The case
of the scheduled castes in India,” Ethnic and Racial Studies, 16:1, January, pp. 95–120.
Japhet, S. (1998) “Caste oppression in the Catholic church,” In: Prabhakar, M.E. (ed.)
Toward a Dalit Theology, Delhi: ISPCK, pp. 176–80.
Jaquette, Jane S. (1997) “Women in power: From tokenism to critical mass,” Foreign
Policy, 108, Fall, pp. 23–7.
Jayaram, N. (1996) “Caste and Hinduism: Changing protean relationship,” In: Srinivas,
M.N. (ed.) Caste: Its Twentieth Century Avatar, New Delhi: Viking, pp. 69–86.
Jenkins, Craig J. (1985) The Politics of Insurgency: The Farm Worker Movement in the
1960s, New York: Columbia University Press.
Jenkins, Laura Dudley (1998) “Preferential policies for disadvantaged ethnic groups:
Employment and education,” In: Young, M.Crawford (ed.) Ethnic Diversity and Public
Policy: A Comparative Inquiry, London: Macmillan Press.
—(1999) “Competing inequalities: The struggle over reserved legislative seats for women
in India,” International Review of Social History, 44, Supplement, pp. 53–75.
—(2000) “Caste, class and Islam: Debating the boundaries of backwardness in India,”
Eastern Anthropologist, 53:3–4, July-December, pp. 327–39.
—(2001a) “Becoming backward: Preferential policies and religious minorities in India,”
Commonwealth and Comparative Politics, 39:2, July, pp. 32–50.
—(2001b) “Personal law and reservations: Volition and religion in contemporary India,”
In: Larson, Gerald J. (ed.) Religion and Personal Law in Secular India: A Call to
Judgment, Bloomington: Indiana University Press.
240 BIBLIOGRAPHY
Jenkins, Laura Dudley and Christian A.Jenkins (1998) “Affirmative action law in India
and the United States: The politics of defining the ‘Haves’ and the ‘Have Nots,’” Paper
presented at the conference “Do the ‘Haves’ Still Come Out Ahead?” Institute for
Legal Studies, University of Wisconsin-Madison, 1–2 May.
Jones, James E. Jr. (1993) “The rise and fall of affirmative action,” In: Hill, Herbert and
James E.Jones, Jr. (eds) Race in America: The Struggle for Equality, Madison:
University of Wisconsin Press.
Jones, James E. Jr., William P.Murphy and Robert Belton (1993) Cases and Materials on
Discrimination in Employment (5th Edition), St. Paul: West Publishing Co.
Kahlenberg (1996) The Remedy: Class, Race, And Affirmative Action, New York: Basic
Books.
Kalelkar Commission (1955) Report of the Backward Classes Commission, Simla:
Government of India Press.
Kananaikil, Jose (1983) Christians of Scheduled Caste Origin, New Delhi: India Social
Institute Monograph Series.
—(1990) “Discrimination against Dalit converts: With specific reference to Dalit converts
to Christianity,” Religion and Society, 37:4, December, pp. 60–4.
—(1990) Scheduled Caste Converts and Social Disabilities: A Survey of Tamil Nadu, New
Delhi: Indian Social Institute.
Karanth, G.K. (1996) “Caste in contemporary rural India,” In: Srinivas, M.N. (ed.) Caste:
Its Twentieth Century Avatar, New Delhi: Viking, pp. 87–109.
Karlsson, B.G. (2001) “Indigenous politics: Community formation and indigenous
peoples’ struggle for self-determination in Northeast India,” Identities, 8:1, pp. 7–45.
Kennard, Edward A. and Gordon Macgregor (1953) “Applied anthropology in government:
United States,” In: Kroebler, A.L. (ed.) Anthropology Today: An Encyclopedic Inquiry,
Chicago: University of Chicago Press.
Kertzer, David I. and Doinique Arel (eds) (2002) Census and Identity: The Politics of
Race, Ethnicity, and Language in National Censuses, New York: Cambridge University
Press.
Khan, Masood Ali (1996) “Masood Ali Khan on reservation of Muslims,” Muslim India,
161, May, pp. 225–6.
Khan, M.Yusuf (1996) “Muslims must aim higher than quota,” Times of India, 27
December.
Khilnani, Sunil (1997) The Idea of India, New York: Farrar, Straus and Girouz.
Kishwar, Madhu (1996a) “Who am I? Living identities vs. acquired ones,” Manushi, 94,
May-June, pp. 6–17.
Kishwar, Madhu (1996b) “Women and politics: Beyond quotas,” Economic and Political
Weekly, 26 October, pp. 2867–74.
Klandermans, Bert and Sidney Tarrow (1988) “Mobilization into social movements:
synthesizing European and American approaches,” In: Klandermans, Bert, Hanspeter
Kriesi and Sidney Tarrow (eds) From Structure to Action: Comparing Social
Movements Across Cultures, Greenwich: JAI Press.
Kohli, Atul, Peter Evans, Peter J.Katzenstein, Adam Przeworski, Susanne Hoeber
Rudolph, James C.Scott and Theda Skocpol (1995) “The role of theory in comparative
politics: A symposium,” World Politics, 48, October, pp. 1–49.
Kondo, Dorinne (1990) Crafting Selves: Power, Gender, and Discourses of Identity in a
Japanese Workplace, Chicago: University of Chicago Press.
BIBLIOGRAPHY 241
Kooiman, Dick (1989) Conversion and Social Equality in India: The London Missionary
Society in South Travancore in the 19th Century, Amsterdam: Free University Press.
Koshy, Ninan (1968) Caste in the Kerala Churches, Bangalore: The Christian Institute for
the Study of Religion and Society.
Kothari, Rajni (1968) “The tasks within,” Seminar 112, pp. 14–19.
—(1994) “Rise of the Dalits and the renewed debate on caste,” Economic and Political
Weekly, 29:26, 25 June, pp. 1589–94.
Kulkarni, Sharad (1991) “Distortion of census data on scheduled tribes,” Economic and
Political Weekly, 26:5, 2 February, pp. 205–8.
Kumar, Dharma (1992) “The affirmative action debate in India,” Asian Survey, XXXII:3,
March, pp. 290–302.
—(1994) “The backward classes: Affirmative action and ethnicity,” Paper prepared for
the Seminar on Ethnicity and Nation Building organized by the US Educational
Foundation in India and the Centre for South and Southeast Asian Studies, University of
Madras, 21–23 March.
Kumar, Radha (1993) The History of Doing: An Illustrated Account of Movements for
Women's Rights and Feminism in India 1800±1990, London: Verso.
—(1994) “Identity politics and the contemporary Indian feminist movement,” In:
Moghadam, Valentine M. (ed.) Identity Politics and Women, Boulder: Westview Press.
—(1995) “From Chipko to Sati: The contemporary Indian women’s movement,” In:
Basu, Amrita (ed.) The Challenges of Local Feminisms: Women's Movements in
Global Perspective, Boulder: Westview Press.
Kundu, Shampa and Kazi Sherif (1982) “The post-graduate science students of Calcutta
University on intercaste marriage,” Eastern Anthropologist, 35:4, pp. 323–6.
Larana, Enrique, Hank Johnston and Joseph R. Gusfield (1994) New Social Movements:
From Ideology to Identity, Philadelphia: Temple University Press.
Laxminarayana (1992) “The problem of pseudo-tribals in Maharashtra,” Tribal Research
Bulletin, 14:2, p. 12.
Lee, Sharon M. (1993) “Racial classification in the U.S. census: 1890–1990,” Ethnic and
Racial Studies, 16:1, pp. 74–94.
Leonard, Jonathan S. (1987) “The effectiveness of equal employment law and affirmative
action regulation,” In: Ehrenberg, R.G. (ed.) Research in Labor Economics, Vol. 8,
Part B, Greenwich, CT: JAI Press, pp. 319–50.
Leopold, Joan (1974) “British applications of the Aryan theory of race to India, 1850–
1870,” English History Review, July, pp. 578–603.
Levi, M. (1990) “A logic of institutional change,” In: Cook, K.S. and M.Levi (eds) The
Limits of Rationality, Chicago: University of Chicago Press.
Lieberman, Robert C. (1995) “Comment” and Anne Schneider and Helen Ingram
“Response” in “Social Construction (Continued),” American Political Science Review
89:2.
Lopez, Ian Fidencio Haney (1996) White by Law: The Legal Construction of Whiteness,
New York: New York University Press.
Luce, Edward (2001) “Hyderabad to use computers to bypass bureaucracy,” Financial
Times, 27 August.
Ludden, David (1985) Peasant History in South India, Princeton: Princeton University
Press.
Maclean, Kama Kellie (1999) “Embracing the untouchables: The BJP and scheduled caste
votes,” Asian Studies Review, 23:4, December, pp. 488–509.
242 BIBLIOGRAPHY
Major, Andrew (1999) “State and criminal tribes in colonial Punjab: Surveillance,
control, and reclamation of the ‘Dangerous Classes,’” Modern Asian Studies, 33:3,
pp. 657–88.
Malik, Kenan (1996) “Universalism and difference: Race and the postmodernists,” Race
and Class, 37:3.
Mandal Commission (1980) Report of the Backward Classes Commission, Delhi: Akalank
Publications (republished 1990).
Mann, E.A. (1992) Boundaries and Identities: Muslims, Work and Status in Aligarh, New
Delhi: Sage.
Marquez, Benjamin (2001) “Choosing issues, choosing sides: Constructing identities in
Mexican-American social movement organizations,” Ethnic and Racial Studies, 24:2,
pp. 218–35.
Marx, Anthony (1998) Making Race and Nation, Cambridge: Cambridge University
Press.
Massey, James (1991) “Scheduled castes: A special reference to Christians of scheduled
caste origin,” Religion and Society, 38:1, March, pp. 26–36.
—(1996) “Emerging trends and challenges to the Dalit movement in India today,”
Typescript.
McAdam, Doug (1982) Political Process and the Development of Black Insurgency,
1930±1970, Chicago: University of Chicago Press.
Mehta, Pratap B. (1997) “India: fragmentation and consensus,” Journal of Democracy, 8:
1, January, pp. 56–69.
Mendelsohn, Oliver and Marika Vicziany (1994) “The untouchables,” In: Mendelsohn,
Oliver and Upendra Baxi (eds) The Rights of Subordinated Peoples, Delhi: Oxford
Merelman, Richard (1995) Representing Black Culture: Racial Conflict and Cultural
Politics in the United States, New York: Routledge.
Metcalf, Barbara (1995) “Presidential address: Too little and too much: Reflections on
Muslims in the history of India.,” Journal of Asian Studies, 54:4, November,
pp. 951–67.
Metcalf, Thomas, R. (1995) Ideologies of the Raj, New Delhi: Cambridge University
Press.
Minault, Gail (ed.) (1981) The Extended Family: Women and Political Participation in
India and Pakistan, Delhi: South Asia Books.
Ministry of Home Affairs (1975) “Circular Letter No. 35/1/72-RU,” 2 May.
—(1977) “Memorandum to State Governments No. B.C. 12025/2/76-SCT.I,” 22 March.
—(1977) “Memorandum to States No. 39/37/73-SCT.I,” 21 May.
Ministry of Personnel, Public Grievances and Pensions, Department of Personnel and
Training (1993) Brochure on Reservations for Scheduled Castes and Scheduled Tribes
in Services, New Delhi: Government of India.
Ministry of Welfare (1990) Report of the Working Group on Development and Welfare of
Scheduled Castes During Eighth Five Year Plan, 1990±1995, New Delhi: Government
of India.
—(1994) “Letter No. 12017/1/91 SCD-R Cell,” 17 January (also issued 8 April 1991).
—(1996) Annual Report 1995±1996, New Delhi: Government of India.
Minor, David (1982) Anthropology as Administrative Tool: The Use of Anthropology by
the War Relocation Authority, M.A. Thesis, North Texas State University.
BIBLIOGRAPHY 243
Minwalla, Shabnam (2002) “Muslim academics look inward for answers to questions,”
Times of India, 31 March.
Mitra, A., Registrar General (1961) “Special tables for scheduled tribes,” Census of India,
vol. 1, India Part V-A(ii), New Delhi: Government of India.
Mitra, Subrata (1987) “The perils of promoting equality: The latent significance of the
anti-reservation movement in India,” Commonwealth and Comparative Politics, 25:3,
pp. 292–312.
—(1990) Politics of Positive Discrimination: A Cross National Perspective, Bombay:
Sangam Press.
—(1994) “Caste, democracy and the politics of community formation in India,” In: Searle-
Chatterjee, M. and U.Sharma (eds) Contextualizing Caste: Post-Dumontian
Approaches, Oxford: Oxford University Press, pp. 49–71.
Mitra, Subrata and V.B.Singh (1999) Democracy and Social Change in India: A Cross-
Sectional Analysis of the National Electorate, New Delhi: Sage Publications.
Moerman, Michael (1965) “Ethnic identification in a complex civilization: Who are the
Lue?” American Anthropologist, 67, pp. 1215–31.
Mohanty, S.P. and A.R.Momim (eds) (1996) Census as Social Document, Jaipur and New
Delhi: Rawat Publications.
Mondal, Sekh Rahim (1996) “Tradition versus modernity in Muslim society (An analysis
of social movements among Muslims in India),” Islam and the Modern Age, 27:2,
May, pp. 105–26.
Moon, Vasant (2001) Growing Up Untouchable in India: A Dalit Biography, Lanham,
Maryland: Rowman and Littlefield Publishers, Inc.
Mujahid, Abdul Malik (1989) Conversion to Islam: Untouchables' Strategy for Protest in
India, Chambersburg, PA: Anima Publications.
Mukherjee, R. (1999) “Caste in itself, Caste and Class, or Caste in Class,” Economic and
Political Weekly, 34:27, July, pp. 1759–61.
Murugkar, Lata (1991) Dalit Panther Movement in Maharashtra: A Sociological
Appraisal, Bombay: Popular Prakashan.
Muthuswamy and Brinda (1996) Swamy's Compilation on Reservations and Concessions
for SCs/STs, Ex-Servicemen, Sportsmen, Compassionate Appointments, Physically
Handicapped and SEBC (Mandal Commission), [Incorporating Orders received up to
February, 1996], Madras and Delhi: Swamy Publishers (P) Ltd.
Nabhi, Kumar Jain (2001) Reservation and Concession for SC/ST, Other Backward
Classes, Ex-Servicemen, Compassionate Appointments and Physically Handicapped,
New Delhi: Nabhi Publications.
Nader, Laura (1997) “The phantom factor: The impact of the cold war on anthropology,”
In: Noam Chomsky et al. The Cold War and the University, New York: The New
Press, pp. 107–46.
Nair, Janaki (1996) Women and Law in Colonial India, Kali for Women in Association
with National Law School of India University: New Delhi.
Nagel, Joane (1995) “American Indian Ethnic Renewal: Politics and the resurgence of
identity,” American Sociological Review, 60:6, December, pp. 947–65.
Nanada, A.R. (1994) Primary Census Abstract: General Population, Series I, India Part II
B (I), Delhi: Government of India.
Narayan, Hamendra (1999) “2001 census will not record caste,” The Statesman, 15
February.
244 BIBLIOGRAPHY
Narayan, J., D.Sheth, Y.Yadav and M.Kishwar (2002) Enhancing Women's

Representation in Legislatures: An Alternative to the Government Bill for Women's
Reservation, internet publication [http://free.freespeech.org/manushi/1116/
alterbill.html] 16/alterbill.html] 13 March.
Natarajan, D. (1972) Indian Census Through A Hundred Years, Census Centenary
Monograph no. 2, part 1. New Delhi: Office of the Registrar General.
Nath, Meenakshi (1996) “Cutting across party lines: Women Members of Parliament
explain their stand on reservation quotas,” Manushi, 96, September-October, p. 11.
Nath, Ram (1995) Dalit-Muslim Unity: Why? And How? Bangalore: Dalit Sahitya
Akademy.
National Action Forum for Social Justice (1996) Dalit Manifesto (Incorporating the
Rights and Entitlements of Scheduled Castes, Scheduled Tribes and Backward Classes),
Delhi.
National Backward Classes Finance and Development Corporation (NBCFDC) (1996)
Objective, Organization, Operating Procedure, New Delhi: Government of India.
—(1996) Guidelines for Implementation of NBCFDC Scheme, Pamphlet from NBCFDC
offices, New Delhi: Government of India.
National Commission for Backward Classes (1994) Questionnaire for Consideration of
Requests for Inclusion and Complaints of Under-Inclusion in the Central List of Other
Backward Classes, New Delhi: Government of India.
National Commission for Scheduled Castes and Scheduled Tribes (1993) Historical
Background, Functions, General Information, New Delhi: Government of India.
National Coordination Committee for Dalit Christians (1996) Demand for Restoration of
Reservation for Christian Dalits, Pamphlet published by the National Coordination
Committee for distribution to Members of Parliament, August.
National Coordination Committee for SC Christians (NCCSCC) (1995a) Cobbler
Constitution Confusion, Pamphlet published by the NCCSCC, November.
—(1995b) End Apartheid Against SC Christians, pamphlet published by the NCCSCC, 1
January.
Nesiah, Devanesan (1997) Discrimination with Reason?: The Policy of Reservations in
the United States, India and Malaysia, Delhi: Oxford University Press.
Nicholson, Linda and Steven Seidman (eds) (1995) Social Postmodernism: Beyond
Identity Politics, Cambridge: Cambridge University Press.
Nobles, Melissa (2000) Shades of Citizenship: Race and the Census in Modern Politics,
Stanford: Stanford University Press.
Oddie, Geoffrey A. (1996) “Christianity and social mobility in South India 1840–1920: A
continuing debate,” South Asia, 19, pp. 143–59.
—(1997) Religious Conversion Movements in South Asia: Continuities and Change 1800±
1900, Surrey: Curzon Press.
O’Hanlon, Rosalind and David Washbrook (1992) “After orientalism: Culture, criticism,
and politics in the third world,” Comparative Studies in Society and History, 34:1,
January, pp. 141–67.
Omi, Michael and Howard Winant (1986) Racial Formation in the United States from the
1960s to the 1980s, New York: Routledge.
Omvedt, Gail (1993) Reinventing Revolution: New Social Movements and the Socialist
Tradition in India, Armonk and London: M.E.Sharpe.
—(1994) Dalits and the Democratic Revolution, New Delhi: Sage.
BIBLIOGRAPHY 245
Onuf, Nicholas Greenwood (1989) World of Our Making: Rules and Rule in Social
Theory and International Relations, Columbia: University of South Carolina Press.
Osella, Filippo and Caroline Osella (2000) Social Mobility in Kerala: Modernity and
Identity in Conflict, London: Pluto.
Ouwerkerk, Louise (1994) No Elephants for the Maharaja, Social and Political Change
in the Princely State of Travancore (1921±1947), New Delhi: Manohar.
Padmanabha, P. (1978) Indian Census and Anthropological Investigations, New Delhi:
The Registrar General and Census Commissioner of India (for the Tenth International
Congress of Anthropological and Ethnological Sciences).
Pandey, Gyanendra (1992) The Construction of Communalism in Colonial North India,
Delhi: Oxford University Press.
—(1995) “Voices from the edge: The struggle to write subaltern histories,” Ethnos, 60:
3–4, pp. 225–42.
Pandian, M.S.S. (2000) “Myth of creamy layer,” Economic and Political Weekly, 22
April, p. 1417.
Pant, Rashmi (1987) “The cognitive status of caste in colonial ethnography: A review of
some literature on the north west provinces and Oudh,” The Indian Economic and
Social History Review, 24:2, pp. 145–62.
Parikh, Sunita (1990) Preferences for Equality: A Comparative Analysis of Higher
Education Policy in the United States and India, Doctoral Dissertation, University of
Chicago.
—(1997) The Politics of Preference: Democratic Institutions and Affirmative Action in
the United States and India, Ann Arbor: The University of Michigan Press.
—(1998) “Religion, reservations, and riots: The politics of ethnic violence in India,” In:
Basu, Amrita and Atul Kohli (eds) Community Conflict and the State in India, Delhi:
Oxford University Press.
Paswan, Ram Vilas (2000) “International focus: Moving India into the information age,”
Government Technology, 13:3, February, p. 45.
Pearson, Gail (1989) “Reserved seats—women and the vote in Bombay,” In:
Krishnamurty, J. (ed.) Women in Colonial India: Essays on Survival, Work and the
State, Delhi.
Peterson, William (1997) Ethnicity Counts, New Brunswick, NJ: Transaction Publishers.
Phelan, Shane (1995) “The space of justice,” In: Nicholson, Linda and Steven Seidman
(eds) Social Postmodernism: Beyond Identity Politics, New York: Cambridge
Pickett, J.Waskom (1933) Christian Mass Movements in India: A Study With
Recommendations, Lucknow: Lucknow Publishing House.
Pinney, Christopher (1990) “Colonial anthropology in the ‘Laboratory of Mankind,’” In:
Bayly, C.A. (ed.) The Raj: India and the British 1600±1947, pp. 252–63, London:
National Portrait Gallery Publications.
—(1994) “Castes of thousands,” Times Higher Education Supplement, 2 September,
p. 22.
Planning Commission, Government of India (1997) Ninth Five Year Plan 1997±2002
[http://planningcommission.nic.r/welcome.html].
Pontusson, Jonas (1995) “From comparative public policy to political economy: Putting
political institutions in their place and taking interests seriously,” Comparative
Political Studies, 28:1, April, pp. 117–47.
246 BIBLIOGRAPHY
Potter, David C. (1996) India's Political Administrators: From ICS to IAS, Delhi: Oxford
University Press.
Prabhudesai, Sandesh (2000) “ZP polls raise controversy over OBC women reservation,”
Goanews.com, 5 January [www.goanews.com/5jan00.htm].
Prakash, Gyan (1992) “Science ‘Gone Native’ in colonial India,” Representations, 40,
Fall, pp. 153–78.
—(1995) After Colonialism: Imperial Histories and Postcolonial Displacements,
Princeton: Princeton University Press.
Punekar, V.B. (1986) “Ethnic relations in India: Reinforcing and re-grouping of ethnic
identities,” Man in India, 66:2, June, pp. 109–20.
Pyarelal (1984) The Epic Fast, New Delhi: Akhil Bharat Anusuchit Jati Parishad.
Quigley, Declan (1993) The Interpretation of Caste, Oxford: Clarendon Press.
Radhakrishnan, P. (1990) “OBCs and central commissions,” Seminar, 375, November,
pp. 22–6.
—(1999) “Cast off caste in census 2000,” Indian Express, 9 December.
Raheja, G.G. (1996) “Caste, colonialism, and the speech of the colonized:
Entextualization and disciplinary control in India,” American Ethnologist, 23:3,
August, pp. 494–513.
Rahman, Maseeh (2000) “Conversions: ‘We Need to Bring India’s Muslims Into the
BJP’s Fold,’ Interview with Bangaru Laxman, president of India’s ruling Bharatiya
Janata Party,” Time Asia, 26 September.
Ramusack, Barbara (1990) “Cultural missionaries, maternal imperialists, feminist allies:
British women activists in India, 1865–1945,” Women's Studies International Forum,
XIII, p. 316.
Randall, Vicky and Robin Theobald (1998) Political Change and Underdevelopment,
Durham: Duke University Press.
Rao, Anupama (2000) “Note on nomenclature,” Typescript.
Rao, Narasimha (1993) “The Prime Minister’s remarks on the occasion of the release of
the book, People of India: An Introduction, published by Anthropological Survey of
India at 9:30 a.m. on 22 September, 1992 at Prime Minister’s residence, New Delhi,”
Journal of the Anthropological Survey of India, 42, pp. 1–7.
Rattansi, Ali (1995) “Just framing: Ethnicities and racisms in a postmodern framework,”
In: Nicholson, Linda and Steven Seidman (eds) Social Post-modernism: Beyond
Identity Politics, Cambridge: Cambridge University Press, pp. 250–86.
Reade, Steven G. and Rosemary Maxwell (1996) “Federal affirmative action programs,
following Adarand, are being revamped to promote diversity through measures that
don’t use racial or gender preferences,” National Law Journal, 19 February, p. B6.
Report of the Criminal Tribes Enquiry Committee (1947) United Provinces.
Riley, Denise (1988) Am I that Name? Feminism and the Category of `Woman' in History,
London: MacMillan.
Risley, Herbert (1915) The People of India, Calcutta, Simla, London: Thacker, Spink & Co.
Risley, Herbert (1969) The People of India (Second ed.) Crooke, W. (ed.) Delhi: Oriental
Books Reprint Corporation.
Robison, Robert K., Ross L.Fink and Billie Morgan Allen (1988) “Affirmative action in
the public sector: The increasing burden of ‘Strict Scrutiny,’” Labor Law Journal,
pp. 801–9.
Roy, Arundhati (1997) The God of Small Things, New York: Random House.
Roy, Beth (1994) Some Trouble with Cows, Berkeley: University of California Press.
BIBLIOGRAPHY 247
Roy, Ramashray and V.B.Singh (1987) Between Two Worlds: A Study of Harijan Elites,
Delhi: Discovery Publishing House.
Roy, Srirupa (1999) “Instituting diversity: Official nationalism in post-independence
India,” South Asia, xxii:1, June, pp. 79–100.
Rudolph, Lloyd I. and Susanne Hoeber Rudolph (1967) The Modernity of Tradition:
Political Development in India, Chicago: University of Chicago Press.
—(1996) “Occidentalism and orientalism: Perspectives on legal pluralism,” In: Sally
Humphreys (ed.) Cultures of Scholarship, Ann Arbor, Michigan: University of
Michigan Press.
Rushdie, Salman (1995) The Moor's Last Sigh, London: Jonathan Cape.
Sabatier, P. (2000) “Theories of the policy process,” Journal of European Public Policy,
7:1, pp. 135–40.
Sachchidananda (1977) The Harijan Elite, New Delhi: Thomson Press (India) Limited.
Sahay, K.B. (1995) “Impact on politics,” Hindustan Times, 7 August.
Said, Edward (1978) Orientalism, New York: Pantheon.
Sarkar, R.M. (1992) “Book review” (of K.S. Singh’s People of India: An Introduction), Man
in India, 72, September, pp. 365–70.
—(1999) “Conversions and politics of Hindu right,” Economic and Political Weekly, 34:
26, pp. 1691–1700.
Saroja, K. (1999) “Intercaste marriage and social dynamics in India: A critique,” The
Indian Journal of Social Work, 60:2, pp. 183–92.
Schneider, Anne and Helen Ingram (1993) “Social construction of target populations:
Implications for politics and policy,” American Political Science Review, 87:2, June,
pp. 334–47.
—(1995) “Social construction (Continued),” American Political Science Review, 89:2,
June, pp. 437–46.
Scott, James C. (1990) Domination and the Arts of Resistance, New Haven: Yale.
—(with Atul Kohli, Peter Evans, Peter J.Katzenstein, Adam Przeworski, Susanne Hoeber
Rudolph, and Theda Skocpol) (1995) “The role of theory in comparative politics: A
symposium,” World Politics, 48, October, pp. 1–49.
—(1998) Seeing Like a State: How Certain Schemes to Improve the Human Condition Have
Failed, New Haven: Yale University Press.
Searle-Chatterjee, Mary (1994) “Urban ‘Untouchables’ and Hindu nationalism,”
Immigrants and Minorities, 13:1, pp. 12–25.
—(1996) review of K.S.Singh’s “People of India: National Series Volume II The
Scheduled Castes,” In: Ethnic and Racial Studies, 19:3, pp. 733–4.
Searle-Chatterjee, Mary and U.Sharma (eds) (1994) Contextualizing Caste: Post-
Dumontian Approaches, Oxford: Blackwell.
Sen, Amartya (1994) Development as Freedom, New York: Knopf.
Shah, A.M. (1989) “Caste and the intelligensia,” Hindustan Times, 24 March, p. 13.
—(1996) “The judicial and sociological view of the other backward classes,” In: Srinivas,
M.N. (ed.) Caste: Its Twentieth Century Avatar, New Delhi: Viking, pp. 174–94.
Shah, Ghanshyam Shah. (1998) “Caste-based census will compound past blunders,” The
Times of India, 22 May.
Shah, Hemant (1994) “News and the ‘Self-Production’ of society: Times of India
coverage of caste conflict and job reservations in India,” Journalism Monographs, 144,
Columbia, SC: Association for Education in Journalism and Mass Communication.
Shahabuddin, Syed (1996) Muslim India, February, pp. 77–8.
248 BIBLIOGRAPHY
Sharma, K. (2000) “Power and representation: Reservation for women in India,” Asian
Journal of Women's Studies, 6:1, pp. 47–87.
Sharma, S.L. (1981) “Conversions,” Seminar (Issue on “Reservations: A Symposium on
the Ramifications of a Policy”) 268, December, pp. 27–34.
Shatrugna, M. (1994) “All Kapus as BCs: Reducing reservation to a farce,” Economic and
Political Weekly, 29:37, 10 September, pp. 2397–400.
Sheth, D.L. (1999) “Secularization of Caste and Making of New Middle Class,”
Economic and Political Weekly, 34:34–5, September, pp. 2502–10.
—(2000) “Changing terms of elite discourse: The case of reservations for ‘Other
Backward Classes,’” In: Hasan, Zoya (ed.) Politics and the State in India, New Delhi:
Sage Publications.
Shourie, Arun (1999) “Body-counts have become big, compelling business,” [http://
www.pragma.org/ISS02403.html] 19 December 2001.
Simmel, Georg (1955) Conflict and The Web of Group Affiliations, Glencoe, Illinois: The
Free Press.
Singh, Gopal and Hari Lal Sharma (1995) Reservation Politics in India: Mandalization of
the Society, New Delhi: Deep & Deep Publications.
Singh, Hoshiar (1994) “Constitutional base for panchayati raj in India: The 73rd
Amendment Act,” Asian Survey, XXXIV, pp. 824–5.
Singh, K.S. (1985) Tribal Society in India, New Delhi: Manohar.
—(1987) “The people of India a national project: Its perspectives and prospects,” Man in
India, 63:3, September, pp. 232–49.
—(1990) “A report on the release of quantitative data on people of India by honourable
Minister of Human Resource Development Shri Arjun Singh on 24, December 1991,”
Journal of the Anthropological Survey of India, 39, pp. 100–8.
—(1992a) People of India: An Introduction, Calcutta: Anthropological Survey of India.
—(1992b) Visual Anthropology and India: Proceedings of a Seminar, Calcutta:
Anthropological Survey of India.
—(1993) “A report on the presentation on the ‘People of India’ project made at the World
Bank, Washington from 26 to 31 July, 1992,” Journal of the Anthropological Survey of
India, 42, pp. 59–64.
—(1995) The Scheduled Castes (People of India, National Series Volume 2), Delhi:
Oxford University Press and Anthropological Survey of India.
—(1996a) Communities, Segments, Synonyms, Surnames and Titles (People of India,
National Series Volume 8), Delhi: Oxford University Press and Anthropological Survey
of India.
—(1996b) Identity, Ecology, Social Organization, Economy, Development Process and
Linkages: A Quantitative Profile (People of India, National Series Volume 7), Delhi:
Oxford University Press and Anthropological Survey of India.
Singh, K.S. (1998) People of India: India's Communities (Volumes 4–6), Delhi:
Anthropological Survey of India and Oxford University Press.
—(2000) “A perspective on the ASI,” Seminar, 495, pp. 40–4.
Singh, K.S., V.Bhalla and V.Kaul (1994) The Biological Variations in Indian
Populations, (People of India, National Series Volume 10), Anthropological Survey of
India, Delhi: Oxford University Press.
Singh, S.N. (1996) Reservation Policy for Backward Classes, Jaipur and New Delhi:
Rawat Publications.
BIBLIOGRAPHY 249
Skerry, Peter (2000) Counting on the Census? Race, Group Identity, and the Evasion of
Politics, Washington, DC: Brookings Institution Press.
Skocpol, Theda (1985) “Bringing the state back in: Strategies of analysis in current
research,” In: Evans, Peter B., Dietrich Rueschemeyer and Theda Skocpol (eds)
Bringing the State Back In, Cambridge: Cambridge University Press.
Sleeper, Jim (1997) Liberal Racism, New York: Viking.
Sonalkar, Wandana (1999) “An agenda for gender politics,” Economic and Political
Weekly, 2–9 January [http://www.epw.org.in/34–0102 /persp.htm].
Spaeth, Anthony (1996) “Society,” From Special Report on India, Time (International
Edition), 25 March, pp. 42–44.
Spitz, Douglas (1993) “Cultural pluralism, revivalism, and modernity in South Asia: The
Rashtriya Swayamsevak Sangh,” In: Young, M.Crawford (ed.) The Rising Tide of
Cultural Pluralism: The Nation-State at Bay? Madison: The University of Wisconsin
Press.
Spivak, Gayatri Chakravorty (ed.) (1987) “Subaltern studies: Deconstructing
historiography,” In: In Other Worlds: Essays in Cultural Politics, New York and
London: Methuen.
Srinivas, M.N. (1966) Social Change in Modern India, Berkeley: University of California
Press.
—(1989) The Cohesive Role of Sanskritization and Other Essays, New Delhi: Oxford
University Press.
—(1992) “On the people of India project,” In: Singh, K.S. (ed.) People of India: An
Introduction (National Series Volume 1) Calcutta: Anthropological Survey of India,
pp. 1–12.
—(1997) “The pangs of change,” Frontline, 14:16, 9–22 August.
Srivastava, Vinay Kumar (1999) “The future of anthropology,” Economic and Political
Weekly, 34:9, pp. 545–52.
Stepan, Alfred (ed.) (1989) Democratizing Brazil: Problems of Transition and
Consolidation, New York: Oxford University Press.
Stocking, George W., Jr. (1982) Race, Culture and Evolution: Essays in the History of
Anthropology, Chicago: University of Chicago Press.
Stoler, Ann Laura and Frederick Cooper (1996) Between Metropole and Colony, In:
Stoler, Ann Laura and Frederick Cooper (eds) Tensions of Empire: Colonial Cultures in
a Bourgeois World, Berkeley: University of California Press.
Strachey, John (1885) India, London: K.Paul.
Sudarsen, V. (1994) “Scheduled castes and scheduled tribes of India: Constitutional
safeguards and ground realities,” Paper prepared for the Seminar on Ethnicity and
Nation Building organized by the US Educational Foundation in India and the Centre
for South and Southeast Asian Studies, University of Madras, 21–23 March.
Sundar, Nandini (1999) “The Indian census, identity and inequality,” In: Guha,
Ramachandra and Jonathan P.Parry (eds) Institutions and Inequalities: Essays in
Honour of Andre Beteille, New Delhi, New York: Oxford University Press.
Sunder, Nandini, Satishe Deshpande and Patricia Oberoi (2000) “Indian anthropology and
sociology: Towards a history,” Economic and Political Weekly, 10–16 June.
Tajfel, Henry (1981) Human Groups and Social Categories: Studies in Social
Psychology, Cambridge: Cambridge University Press.
Tarrow, Sidney (1994) Power in Movement: Social Movements, Collective Action and
Politics, New York: Cambridge University Press.
250 BIBLIOGRAPHY
Taylor, Charles (1992) “Multiculturalism and the Politics of Recognition,” In: Amy
Gutmann (ed.) Multiculturalism and the Politics of Recognition: An Essay, Princeton:
Princeton University Press.
—(1994) Multiculturalism and the Politics of Recognition (Second edn), Princeton:
Princeton University Press.
Telles, Edward E. (1995) “Who are the Morenas?” Social Forces, 73:4, June,
pp. 1609–11.
Tharoor, Shashi (1997) India: From Midnight to the Millenium, New York: Harper
Perennial.
Thelen, K. and S.Steinmo (1992) “Historical institutionalism in comparative politics,” In:
Steinmo, S., K. Thelen and F.Longstreth (eds) Structuring Politics: Historical
Institutionalism in Comparative Analysis, New York: Cambridge University Press,
pp. 1–32.
Thorat, S.K. (1993) “Passage to adulthood: Perceptions from below,” In: Sudhir Kakar
(ed.) Identity and Adulthood, Bombay: Oxford University Press (originally published in
1979).
Tilak, Jandhyala B.G. (1981) “A new criterion,” Seminar (Issue on “Reservations: A
Symposium on Ramifications of a Policy”), 268, December, pp. 35–6.
Tilley, Virginia (1997) “The terms of the debate: Untangling language about ethnicity and
ethnic movements,” Ethnic and Racial Studies, 20:3, July, pp. 497–522.
Tripp, Aili (2000) “Rethinking difference: Comparative perspectives from Africa,” Signs,
25:3, pp. 649–75.
Uberoi, Patricia (2000) “Deja Vu?” Seminar, 495, pp. 14–19.
Upreti, H.C. and N.Upreti (1982) “Parental authority and attitude of women students
toward marriage and dowry,” Indian Journal of Social Work, 43:3, pp. 247–52.
Van den Berghe, Pierre (1995) “Does race matter?” Nations and Nationalism, 1:3,
pp. 357–68.
Van der Veer, Peter (1997) Imperial Encounters, Princeton: Princeton University Press.
Varshney, Ashutosh (1998) “Why democracy survives,” Journal of Democracy, 9: 3, July,
pp. 36–50.
—(2000) “Is India becoming more democratic?” Journal of Asian Studies, 59: 1,
February, pp. 3–25.
Venugopal, P. (2000) “Kerala minorities’ stand puts Narendran panel in a spot,” Indian
Express, 3 February.
Vijayanunni, M., Reg. General and Census Commissioner (1991a) Union Primary Census
Abstract for Scheduled Castes and Scheduled Tribes, Census of India 1991, series-1.
Vijayanunni, M. (1991b) Religion, Census of India 1991, series-1.
—(1999) “Caste can’t be cast off,” Indian Express 19 November.
Visram, Rozina (1992) Women in India and Pakistan: the struggle for independence from
British rule, Cambridge: Cambridge University Press.
Viswanathan, Gauri (1998) Outside the Fold: Conversion, Modernity and Belief, New
Jersey: Princeton University Press.
Wakin, Eric (1992) Anthropology Goes to War: Professional Ethics and
Counterinsurgency in Thailand, Madison: Center for Southeast Asian Studies
(Monograph Series), University of Wisconsin-Madison.
Wald, Kenneth D., James W.Button and Barbara A.Rienzo (1996) “The politics of gay
rights in American communities: Explaining antidiscrimination ordinances and
policies,” American Journal of Political Science, XL , p. 1169.
BIBLIOGRAPHY 251
Walzer, Michael (1997) On Toleration, New Haven: Yale University Press.

Watson, John Forbes and John William Kaye (1868–75) The People of India: A Series of
Photographic Illustrations, London: India Office Library.
Webster, John C.B. (1994) The Dalit Christians: A History, Delhi: ISPCK.
Weiner, Myron (1983) “The political consequences of preferential policies: A
comparative perspective,” Comparative Politics, 16:1, October, pp. 58–61.
Weiner, Myron and Mary Fiansod Katzenstien (1981) India's Preferential Policies:
Migrants, the Middle Classes, and Ethnic Equality, Chicago: University of Chicago
Press.
White, Michael J. and Sharon Sassler (1995) “Ethnic definition, social mobility and
residential segregation in the United States,” In: Calvin Goldscheider (ed.) Population,
Ethnicity and Nation-Building, Brown University Studies in Population and
Development, Boulder: Westview Press.
White, P. (1999) “Ethnicity in the 1991 census, vol. 3, social, geography and ethnicity by
P.Ratcliff,” Ethnic and Racial Studies, 22:1, pp. 182–3.
Wilson, William Julius (1987) The Truly Disadvantaged: The Inner City, the Underclass,
and Public Policy, Chicago: University of Chicago Press.
Wolfe, A.B. (1932) “Population censuses before 1790,” Journal of the American
Statistical Association, 27:180, December, pp. 357–70.
Wright, Lawrence (1994) “One drop of blood,” The New Yorker, July, pp. 46–55.
Wright, Theodore P. (1997) “A new demand for Muslim reservations in India,” Asian
Survey, 37:9, September, pp. 852–58.
—(1983) “The ethnic numbers game in India: Hindu-Muslim conflicts over conversion,
family planning, migration, and the census,” In: McCready, William C. (ed.) Culture,
Ethnicity and Identity: Current Issues in Research, New York: Academic Press.
Wu, Frank H. (1996) “From black to white and back again,” Asian Law Journal, 3:1, May,
pp. 185–215.
Wyatt, Andrew (1998) “Dalit Christians and identity politics in India,” Bulletin of
Concerned Asian Scholars, 30:4, pp. 16–23.
Wyzan, Michael L. (ed.) (1990) The Political Economy of Ethnic Discrimination and
Affirmative Action: A Comparative Perspective, New York: Praeger.
Yin, Tung (1997) “A carbolic smoke ball for the nineties: Class-based affirmative action,”
31 Loy. L.A.L. Rev. 213, November.
Young, M. Crawford (1976) The Politics of Cultural Pluralism, Madison: University of
Wisconsin Press.
—(1985) “Ethnicity and the colonial and post-colonial state in Africa,” In: Brass, Paul (ed.)
Ethnic Groups and the State, London: Croom Helm.
—(1993) The Rising Tide of Cultural Pluralism, Madison: The University of Wisconsin
Press.
—(1994a) “In search of civil society,” In: Harbeson, John et al. (eds) Civil Society and the
State in Africa, Boulder: Lynne Reinner.
—(1994b) The African Colonial State in Comparative Perspective, New Haven: Yale
University Press.
Young, Robert J.C. (1995) Colonial Desire: Hybridity in Theory, Culture and Race,
London and New York: Routledge.
Zelliot, Eleanor (1996) From Untouchable to Dalit, New Delhi: Manohar.
Index
academic colonialism 64 All India Backward Muslim Morcha

achhut 98 (AIBMM) 117, 138
AdDharm 70 All India Christian People’s Forum
Adhivakta Parishad 138; (AICPF) 122, 136
and SC Christian demands 138–9 All India Muslim OBC Organisation 115–
Adi-Andhra 70 16;
Adi-Dravida 35, 37, 70 And caste/class 116, 118;
Adi-Karnataka 70 and Dalits 117;
adivasi see Scheduled Tribes surveys 116
Adivasi Koli Mahasangh (AKM) 27 All India Women’s Conference (AIWC)
administration and identity 10–12, 24, 47, 160, 161, 162
51–3, 67–9 Alva, Margaret 167–8, 171
see also administrators; Ambedkar, B.R. 35, 55, 70, 130, 133, 148;
caste certificates; and conversion movement 35, 120, 130,
census; 141;
Other Backward Classes: and Indian Constitution 131;
and identifying and labeling 69–74; and reservations 128, 130–1
and reservation policies 86–8; see also Mohandas Gandhi
and special concessions 70 Anderson, Benedict 94, 107, 174
administrators (colonial and post colonial): Anderson, J.D. 41
as classifiers of society 43–4, 47, 51–3, Andhara Pradesh 117–18, 135
89, 92–4 Ansari, Shabbir Ahmed 115, 117
adoption 33, 78, 80 Anthropological Survey of India 17–18,
Advani, Lal Krishna 135, 140 41;
affirmative action policies 1–2, 8–9, 10, criminal identification in 52;
180–5 and People of India survey 49–57;
The African Colonial State in Comparative “Portrait Building System” 51
Perspective 43 anthropologist/administrators 42, 47, 51–2
AIBMM see All India Backward Muslim see also Risley
Morcha anthropology:
AICPF see All India Christian People’s colonial and post compared 43–5, 64,
Forum 100
Ain-i-Akbari 92 see also People of India;
AIWC see All India Women’s Conference in contemporary India 42, 44, 64–5;
Ali, Aijaz 117, 138 in courts 28–30
252
INDEX 253
anthropometric data collected 44, 45, 50, BJP Scheduled Caste and Scheduled Tribe
51 Morcha 134
see also bio-anthropological parameters Blackmun, J. 7
anti-creamy layer see Kerala Bose, Ashish 105
Anti-Mandal Commission Forum 147 British Raj see colonial policy
anti-Mandal protests 54, 96, 147 Buddhists:
see also Mandal Commission Report and conversion 35, 79;
Appadurai, Arjun 6, 44, 94, 118 and Dalit Solidarity Programme 123;
Aprishya Sudras 70 and Hinduism 132–3;
Association for Promoting Education and and Scheduled Castes 120, 133
Employment of Muslims 115, 118, 123; Bureau of Police Research and
convention on reservations 117 Development and Portrait Building
Atul Chandra Adhikari v. State of Orissa System 51
76 bureaucracy
avarna 70 see also administration and identity:
Azariah, M. 121, 123, 179 and liberalization 183;
and representation 88;
Backward Classes Commission see and reservations 15
National Backward Classes Commission Burman, B.K. Roy 55, 104
Backward Muslim Morcha see All India
Backward Muslim Morcha Calcutta University:
backwardness/backward classes defined/ and attitudes to intercaste marriage 32
discussed 2, 14–15, 71, 73, 81–6, 145–6 caste
Banton, Michael 2, 70 see also Scheduled Castes:
Bates, Crispin 12, 44 in Christian society 1–2, 31, 35, 121,
Beda Jangam 55, 65 179;
Bengal census 98 defined 13;
Beteille, Andre 54–5, 62 in Muslim society 116, 179;
Bharati, Uma 169, 172 and nationalism 45, 57–63, 93, 128–30,
Bharatiya Janata Party (BJP) 134
see also Hindu nationalism; see also Hindu nationalism
Hindu nationalists: caste associations:
and categories 9; certificate of membership 30;
and conversion 131; and Other Backward Class numbers
and Dalit Christians 121; 101;
and moderation 139; as response to census 48, 93, 101
and Muslims 133–5; caste census 89–90, 95–7, 101
and OBCs 172; see also census;
and People of India project 65; colonial census
platform 127; caste certificates 18, 67–70, 72–3, 81, 87;
and political savvy 138; and conversion 79–81;
and reservations 150 and creamy layer 82;
Bihar 155 false certificates 26, 33, 72–3;
bio-anthropological parameters 60 and intercommunity marriage and
BJP see Bharatiya Janata Party adoption 76–9
BJP Minority Morcha: Muslims/Christians/ see also inter-caste marriages;
Zoroastrians 133 and migrants 74–6
254 INDEX
caste endogamy 32 colonial/postcolonial compared 89–90,

caste identity: 92–5, 96–8;
and birth 33, 36, 37; counting caste and religion 95, 100–4;
and Christians 2; first colonial 92;
as club-like 37; first post-Independence 93, 95, 99;
and employment files 71, 87; and identity politics 90, 93;
fluidity 13–14, 98; modernization of 93–4;
as Hindu 2, 134; monographs 100
paradox of 8; see also Risley;
reconstructing 7–8, 130, 179 People of India;
see also constructivism; numbers or social indicators 99, 102–4;
repeated verification of 29, 69, 72–3, of Bengal 98;
79 and Other Backward Classes 89–90, 95–
Caste in India 93 6, 101–2;
casteism 10 and policy uses 106;
categories and religion 89, 93, 102–4, 106;
see also colonial categories; and reservations 18, 90–1, 92, 95–100,
community; 102, 105–6;
gender; and Scheduled Castes/Tribes 89, 90,
constructivism: 98–9, 102–6
boundary keeping 105–6; see also retribalization;
changing 105; and simplifications 94, 97–8, 104–5;
and class 19–20; and “Special Groups” 89–90, 98, 99
dangers of 5–6, 9, 85, 177–8; Charsley, Simon 70, 71
government use of 6, 8–10, 10–13 Chatterjee, Jotsna 171
see also administration and identity; Chatterjee, S.K. 122
periodic re-evaluation needed 10, 183; Christian-Muslim cooperation/discord
political or “scientific” 29–30, 54–5, 116–17, 137–8
97; Christians:
racial 2, 6–7, 11, 91, 111–12, 175–6 and caste identity 1, 2, 19, 31
see also race; see also Dalit Christians;
reconstruction of 105, 107, 111, 112; Swigaradoss;
for reservations 13–15 and caste recognition 113, 119, 121;
see also Scheduled Castes; and census 89;
Scheduled Tribes; and Dalit Solidarity Programme 123;
Other Backward Classes; demands 119;
utility of 5, 6–8, 178–80 and Hindu nationalists 19, 119, 120,
Catholic Bishops Conference of India 135– 121, 127–9, 131, 133–9;
6 Mother Theresa 120;
CEDAW see Convention on the and People of India project 55–6, 65;
Elimination of All Forms of protesters 1, 4;
Discrimination Against Women and reservations 119–20, 123, 124,
census 134, 137–9;
see also colonial census: and Scheduled Castes 35–6, 65, 69, 89,
administrators and commissioners 89, 134, 137
91–3, 100, 101, 104; Church of North India (Protestant) 123
of Bengal 98; class
classifications 89–92, 97–100; see also creamy layer;
INDEX 255
economic conditions; and competition of the disadvantaged

Indra Sawhney v. Union of India: 173;
Mandal Commission; and legislative reservations 164–5;
Other Backward Classes; and Other Backward Classes 15, 84,
Supreme Court decisions; 145;
defined 14–15; samudaya 59;
legality of 146 versus category 165, 166, 172, 173, 174
Cochin see Kerala community studies 55–6
Cohn, Bernard 47 Congress Party:
Cokhamela 62 and People of India survey 65;
colonial administrators see administrators and reservations 150;
(colonial and postcolonial) 73rd constitutional amendment 167–8
colonial anthropology Constitution:
see also anthropology; amendment for special provisions 146;
Risley: and defining Hindu 133;
critics of 43–4 and discrimination 146;
colonial categories/lists (schedules) 11, 30– and reservations 15, 145–6, 163;
1, 50–1 and separate electorates 163;
see also categories: 73rd constitutional amendment 167–8
and caste association lobbying 48, 93 constructivism 5–8, 111–12, 177–80, 186
colonial census Convention on the Elimination of All
see also census: Forms of Discrimination Against
academic scrutiny of 94–5; Women (CEDAW) 34, 163
compared to post colonial 89–90, 92–5, conversion
96–8; see also Ambedkar;
critics of 92, 93 Buddhists;
colonial ethnography see anthropology; Christians:
Risley administrative view of 79–80;
colonial legal system 24, 37 and caste status 25, 35–6, 79–80, 121–
colonial policy 2;
see also Government of India Acts: Gandhi’s view of 129, 131;
backward class reservations 143, 151; Hindu nationalist view of 127, 135–6;
Christian reservations 119; legal view of 24, 35–40;
divide and rule 24, 47, 57–8, 160, 173; as political protest 35;
legislative rights 160; reasons for 35–6, 38;
Muslim reservations 114, 160–2; and reconversion 35, 79–80
Scheduled Caste/Tribe reservations 14, cooperation among minority religions 116–
70–1, 160, 164; 7
women’s reservations 20, 159–63 see also Dalit Solidarity Programme:
colonial states and ethnic identities 10–11 and Hindu nationalism 137–8
Committee on the Status of Women 20, courts see colonial legal system;
163–7 Supreme Court of India
Common Minimum Programme and Dalit creamy layer 19, 125, 179, 181, 185
Christians 120 see also Narendran Commission;
community: for anti-creamy layer see Kerala:
acceptance by 76, 78, 80; administration 82;
as caste 41; criteria and economic factors 149, 154;
defined 149–50;
256 INDEX
and resistance 150–4; diversity and categories 8–10, 113, 124–6,

and reservations 148 127–8, 157–9, 175–80
Creamy Layer Committee 149–50; see also categories
and transparency 150 Dreze, Jean 158
“criminal tribes/categories” 47, 51, 100 Dushkin, Leila 16, 69–70
Dalit Christians 1, 19, 56, 119–24, 134–9 economic conditions/criteria

see also Christians; see also class;
Dalits creamy layer;
Dalit conversion movement see Buddhists; Indra Sawhey v. Union of India;
conversion Other Backward Classes:
Dalit converts to Christianity 121, 122 and census 104;
Dalit Jesuits 123 and creamy layer ruling 142, 148–50;
Dalit liberation theology 123 and Mandal Commission 145–8;
Dalit Muslims 115, 118, 125 and policy changes 144, 149–53;
Dalit Solidarity Programme: and reservation categories 10, 14–15,
and fear of Hindu nationalism 137; 16, 82, 84, 142–4, 153–4, 181
and interfaith groupings 123, 125 education 15–16, 103, 184
Dalit Theology Department (Chennai) 123 see also higher education
The Dalit Voice 61–2 effectiveness of reservations 15–17, 63,
Dalits 65, 70, 98 184
see also Ambedkar; End Caste Conference 182
Scheduled Castes: Essed, Philomena 9–10
defined 13; ethnic identities/tensions 10–11, 124, 125
and identity politics 7–8, 61–2, 70–1, see also categories;
130, 179, 185; diversity;
and illiteracy 184; identity
unity of 71, 98, 121–2 ethnographic mapping of India and
“Danger Signal for Hindus” 103 nationalism 58
Das, Bhagwan 79 see also anthropology;
Das, Veena 181–2 People of India
David, Shakuntala 123, 133 Ethnological Survey of India 47
Davis, Adrienne D. 4, 112 see also anthropology;
Delhi: People of India
and caste certificates 75; “experts” and status verification 29–30
and reservations 143 exterior castes 70
Delhi University and quotas 182 ex-untouchables 70
“depressed classes” 70
see also Scheduled Castes false certificates 26, 33, 72–3
desanskritization 25, 27–8 see also caste certificates
Devasahayan, V. 123, 130; family planning see fertility rates
and meaning of Dalit 130 Farooqui, Vimla 170
Dirks, Nicholas 43, 45, 58 female legislators 162
discrimination see Caste; feminist legitimacy 170
constitution; see also gender;
gender; women;
race women’s associations/organizations
INDEX 257
fertility rates and minority scrutiny 102–4 and Muslim seats 114, 160
see also census Government of India Act (1935): 14;
Fifth Caste 70 and Other Backward Classes
Ford, Christopher A. 23–4 and Scheduled Castes 14, 70–1, 119;
Foucault, Michel 111 and Scheduled Tribe category 14
Franchise Committee 162 Government of India Acts (1919; 1935):
Frykenberg, Robert Eric 12–13, 38, 102, nominated seats 160;
121, 130 reserved seats 162;
separate electorates 160;
Galanter, Marc 120, 145 suffrage 162
Gandhi Caste Society/Gandhi Sangam 9, Government of India (Scheduled Castes)
19, 143, 148 Order:
Gandhi, Indira 49, 163 and Christians 119–20
Gandhi, Maneka 171–2 group and individual processes 81–7
Gandhi, Mohandas 70, 98, 130, 161; see also identity
and caste discrimination/reform 19, group certification see schedules
129; Guha, Phulrenu 166
and conversion 131; Guha, Sumit 92, 105
“Gandhi caste” proposed 148;
group-based policies 9, 130–1; Halba/Halbi tribe:
and Harijans 129, 130, 131; population/category change 105
and inclusive Hinduism 128–31; Hamid, Syed 115, 138
and reservations 130–1; Harijans (untouchables) 117, 70, 98
see also Ambedkar; see also caste;
Dalit; Dalit;
Harijan Scheduled Castes:
Gandhi, Rajiv 51, 167 and Hindu absorption 130;
Gandhi, Rajmohan 129–30 and Mohandas Gandhi 129
gazetteers 92 higher education:
Geertz, Clifford 5 and reservations 15, 16, 147, 181–2,
gender: 183–4;
as category 7, 157–9, 173–4; and Supreme Court decision 26
discrimination 20, 34, 49, 77, 158; Hindu Marriage Act 133
and inter-caste marriages 32, 77 Hindu nationalism
see also inter-caste marriages and see also Bharatiya Janata Party (BJP);
reserved seats 173–4 Hinduism;
Goa: Hindutva;
Christian community and status 55; Vishwa Hindu Parishad (VHP);
and reserved seats 170 Rashtriya Swayamsevak Sangh (RSS):
God of Small Things 23 and B.R. Ambedkar 131, 141;
Government of India and census 102–4;
see also administration; and conversion/reconversion 79, 125,
census; 135–6, 141
courts; see also conversion;
People of India: and cross-cutting identities 127–8, 139,
Planning Commission 180;
Government of India Act (1909): defined 128–9;
258 INDEX
exclusiveness 9, 132–6, 140; identity politics;

and Mohandas Gandhi 9, 129–31; self identification:
and Muslim/Christian demands 125, group/unofficial 1–2, 3–4, 18–20, 111–
127–8, 130–6; 12;
organizations 128–9, 132; layers 8–10, 19, 112–13, 127–8, 178;
and People of India project 62, 65 reconstructing 178–80
see also People of India; see also constructivism;
political compromise 127, 139–40; desanskritization;
version of history 134–6 retribalization;
Hinduisation 62 theories of 5–8, 9–10, 11–13, 24–5, 43–
Hinduism 4, 67–8, 94, 111–12, 127–8, 157–9,
see also Hindu nationalism: 177–80
and Christianity and Islam 113, 133, identity construction see identification
134–6; identity fraud see Supreme Court decisions
and Dalit Solidarity Programme 123; identity politics/political activism 3, 11,
and Harijans 130; 18–20, 175–80
philosophy/religion 38, 133, 140; see also caste, Christians, class,
and Scheduled Castes 89, 120, 113, gender, Hindu nationalists, identity,
128–32; Muslims, race:
subsuming Buddhism/Jainisn/Sikhism and colonial/postcolonial practices 10–
120, 132–3 13
Hindus: Ilaiah, Kancha 141
defining 38, 129; illiteracy see literacy
majoritarianism 125 Imagined Communities 94
see also Hindu nationalism; Inden, Ronald 42, 48
minorities India: diverse democracy 2;
Hindustan Times 113; an “ethnographic universe” 58;
and scheduled castes 133 secular state 114
Hindutva 129 Indian Christians see Christians
see also Hindu nationalism Indian constitution see constitution
Horowitz, Donald 125 Indian Councils Act 160
“Hum Dalit” 7–8 Indian ethnography and caste see
Hutton, J.H. 70, 93 anthropology
Indian Express-New Delhi 54
identification Indian government see Government of
see also administration and identity; India
census; Indian Muslims and reservations see
People of India; Muslims
Supreme Court of India: Indian National Congress 161;
identifying beneficiaries 69–70, 72–3 and feminists 163;
see also caste certificates; and 1st Round Table Conference 161;
outsider/insider 2, 12; and special electoral rights 160
by state 1, 2–3, 4, 6, 8–9, 11–12, 17–18, Indian Planning Commission 15–16, 49, 51
23, 157, 175–7 Indian Supreme Court see Supreme Court
identity individual and group processes 81–7
see also caste identity; Indra Sawhney v. Union of India 15;
gender; and reservations 148, 149, 154
Ingram, Helen 68
INDEX 259
inter-caste/intercommunity marriages 31– Kishwar, Madhu 158, 178

4; Koli of Maharashtra see Adivasi Koli
attitudes to 32; Mahasangh
and caste status 76–9, 80–1, 34–5, 37; Koli status 26, 27
offspring of 76–8; Kondo, Dorinne 8
and reservations 32, 182; Koshti subtribe 105
and social mobility 33; Kumar, Neeraj 147
and Supreme Court 31–4; Kumar, Ramesh 81
and women 32–3, 77 Kumaran, K.V. 137
Islam Kumari Madhuri Patil v. Additional
see also caste; Commissioner Tribal Development 25–
Muslims: 31, 36, 69
and caste 116;
and Dalit Solidarity Programme 123 Laxman, Bangaru 139, 140
Iyer, Ganapathi 37 legal system/disputes and identity 23–5
Izhava community 152 see also colonial legal system;
Supreme Court of India
Janata Dal Party 147; legislative assemblies:
and reservations 150 reservations in 15, 16
jati 11 legislative change 40
see also caste: legislative franchise 160
and community 59; legislative reservations for Scheduled
defined 13 Castes and Scheduled Tribes:
Jatiya, Satyanarayan 134 pre and post independence 15, 162–3
Jats 85 legislative reservations/seats for women
Jesuits 123 and religious minorities
Joseph Commission 152 see also Lok Sabha:
see also Kerala after independence 156–7, 163, 164–72;
before independence 159–162;
Kalelkar Commission 97, 145 competing demands for 168–9;
see also National Commission for suffrage 160;
Backward Classes; 73rd Constitutional Amendment Act
Other Backward Classes and panchayat reservations for women
Kananaikil, Jose 122, 137 167;
Karat, Brinda 169, 184; and subquotas for OBC-Muslims 157;
and Muslim and Dalit women 170 suffrage 160
Kaye, J.W. 41 legislators see female legislators
Kerala 89, 153; Let My People Go: Scheduled Caste
and commissions 151; People Group Profiles 56
and creamy layer 143–4, 151–2, 154; liberalization 183
and economic data 151, 152; “Linkages: a Quantitative Profile” 59–60
and Muslim reservations 114, 152; lists see schedules
and OBC reservations 143; literacy 103, 184
and Supreme Court 151–2 litigation about categories 23–4, 69–70
Kerala High Court, 95 see also Supreme Court of India:
Khan, M. Yusuf 117 conversion 35–8
Khurani, Madan Lal 135 inter-caste marriage 31–4;
260 INDEX
“spurious” tribe 25–31 on Risley 44

Lok Sabha (House of the People): migration 74–6, 80–1;
reservations in 15, 16; and caste/tribe certificates 75–6
women in 156, 166 Minister of Women 167
Lopez, Henry 176 Ministry of Home Affairs:
Lourduswamy, S. 121 and category verification 69, 75, 97;
and population growth 102
Madurai 123 Ministry of Social Justice and
Mahadeo Koli status 26, 27, 28, 30 Empowerment 95
Maharashtra: see also Ministry of Welfare:
and census 105; and category verification 77
infiltration of Scheduled Tribes 105; Ministry of Welfare
and Muslim surveys 116 see also Ministry of Social Justice and
Mahars 55 Empowerment:
Mahesh, K. 81–2 and Anthropological Survey of India
major and minor minorities 159–60 data 52;
major minorities (Muslims) and reserved and category verification 76–7, 79, 80;
seats 160 and inter-caste marriage 182
majority see Hindus; minorities:
minorities and census 102–5;
Manavadharmasastra 13 and Hindu nationalism 128–32, 139–41
Mandal Commission 9, 55 see also Hindu nationalism;
Mandal Commission Report 14, 114, 144– major vs. minor 159
7 modernization theory 93
see also Anti-Mandal Commission Mother Theresa 120, 136
Forum; Mughals 92
anti-Mandal protests; Mujahid, Abdul Malik 103
National Commission for Backward multicultural societies
Classes: see also diversity and categories;
backwardness/class defined 145; identity:
and Muslims 114; and toleration 9–10
OBCs and economic status 146–7; multidisciplinary theories 5;
OBC list 83; of identity 6–7
OBC reservations 53–54; see also identity
survey 146 multifaceted reservation categories 142–3,
Mandalization 9, 148 154, 180–5;
Manushi 178 caste or economic criteria 147–9
Marathas 92 see also class
Marquez, Benjamin 13 “multiple undulating society” 185
marriage and caste status 34, 36, 37 Mumbai reserved seats 170
marriage see inter-caste marriage Muslim League and reserved seats 160–1
Mathur, M.L. 83 Muslim OBC organization 125
Mazumdar, Veena 164, 166 Muslims
men and inter-caste marriages 32 see also Dalit Muslims;
see inter-caste marraige OBC Muslims;
Metcalf, Barbara 134 upper class Muslims:
Metcalf, Thomas R. 24, 94; and caste/class recognition 113, 115–
18;
INDEX 261
and Christians 123, 137–9; and ethnographic mapping 58;

census/population of 102–3 and Risley 45, 57–8;
see also census; and Singh (K. S.) 45, 58–9
demands for reservations 114–19; Nawaz, Begum Jahanara Shah:
electoral seats 162; and women’s reservations 160–1
as outsiders/invaders 134–5 NCCSCC see National Coordination
see also Hindu nationalism; Committee for SC Christians
and Partition 114, 135; NCWI see National Council of Women in
and proportional quotas 118; India
and reservations 19, 56–7, 114, 116, Nehru, Jawaharlal 93;
117; and secular state 114
and Scheduled Castes 89; neo-Buddhists see Buddhists
and women’s reservations 170–1 Non Resident Indian (NRI) 152
Mysore and OBC reservations 143 NRI see Non Resident Indian
numbers:
Naaga Holeya 70 census and reservations 100;
Naidu, Sarojini 161 minority counts 102;
Narayanan, K.R. 89 and OBCs 101;
Narendran Commission 152 power of 104;
National Backward Classes list: admittance reinforcing divisions/categories 100–1,
to 18, 83–8; 105
and People of India data 53
National Commission for Backward OBC see Other Backward Classes
Classes 14, 115 OBC Muslims 115–16, 125
see also Kalelkar Commission; official backwardness see backwardness/
Mandal Commission: backward classes
national list 18, 83–88, 149; official classifications
permanent commission established 149 see also identifications:
National Commission for Scheduled Castes debates over 176–7;
and Scheduled Tribes 81; in practice 177
anthropological data 84; Order of the President:
and census 84, 95; and Christians 120
certification process by 81–3; Other Backward Classes (OBCs) 2;
data/list compilation 83, 97; economic criteria 19–20, 142–4;
Research Division 83; definitions of 10, 14–15, 143, 145;
reservations 63; group and individual criteria 149;
social data 184 and intermarriage 31;
National Coordination Committee for SC lists and group petitions 81–6, 149;
(Dalit) Christians 138; numbers 101–2;
and Dalit identity 122; parliamentary quotas 159;
protests 122 and People of India project 53–4;
National Council of Women in India and political power 71;
(NCWI) 160, 161 and reservations 14, 15, 16, 53, 64, 94,
National Forum for India Women 170 114;
nationalism and Supreme Court cases 25–31, 31–4;
see also Hindu nationalism; see also creamy layer;
Mohandas Gandhi: and caste 57–9;
262 INDEX
National Backward Classes list legacies of colonial anthropology 42;

National Commission for Backward nationalism 42, 44, 57–8;
Classes; political (imperial) uses 41, 44, 46, 51–
Socially and Economically Backward 7, 85;
Classes (SEBCs) scientific claims 44
overlapping identities see identity People of India (Risley) 17–18;
administrative uses of 47, 48;
Panchama 70 caste as India 45;
Pancham Bandum 70 caste preeminence 43, 45–6;
pan-Dalit identity see caste identity; caste system and race 46;
Dalit criminal tribes 47;
Pandey, Gyanendra 24 ideology 41–2;
pan-religious identity/movement 117, 123 and Indian nationalism 57–8;
pan-Scheduled Caste identity 71 physical and social types 46;
see also Scheduled Castes and scientific claims 43, 44
Pant, Govind 93 Phelan, Shane 158
pariah 70 Planning Commission 15–16, 49, 51
parliamentary reservations for women see police and category verification 29
legislative reservations for women political activism see identity politics/
Paswan, Ram Vilas 182 political activism
Patel, Sardar Vallabhbhai 93 politics of recognition/politics of
Patil sisters (Suchita and Madhuri): redistribution 158;
father 30; and multilayered identities 158
status verification of 25–30, 39 population growth 103;
Paul, Valsamma 31–4 see also census:
People of India (Anthropological Survey of and family planning
India) 17–18, 41; interest in 102;
administrative uses of 51–2; projections of 103
and caste 42; Portrait Building System: and criminals 51;
caste and nations 45; “ethnic groups” classified 51–2
continuity with colonial survey 48, 50, Prakash, Gyan 43–4
51, 55, 65; Prasad Committee 149
and nationalism 58–63; see also creamy layer
petitions resulting 55; Prasad, R.N. 115, 149–50, 181
political uses 51–7; privatization/private sector reservations
Portrait Building System 51–2; 183
process 49–50; Procrustes 185
scientific methodology 61; professional schools and reservations 16,
self-identification 53, 65, 105 25–31, 31–4, 183
People of India projects compared/
contrasted quotas 15 15–17
see also anthropology: see also reservations:
administrative uses 47, 48, 51–2; filled and unfilled
agendas 62, 65–6;
criminal tribes/portrait building 47, 51– race/racial categories 5, 7, 9, 10, 11, 111–
2; 12, 175–6
influence of first on second 49, 50–1; see also categories;
INDEX 263
identifications see also anti-Mandal protests;

race/racial identities 111–12; and the constitution 15
see also identities see also constitution;
among Mexican Americans 13; critics of 9, 16–17
postmodern explosion of 6 see also Anti-Mandal Commission
racial theory in India 12; Forum;
and caste 46 anti-Mandal protests;
racism 10 Gandhi Caste Society;
Raheja, G.G. 45–6 Hindu nationalists;
Rajastan 85 and Dalit Christians 119–24
Rajastan Univedrsity: see also Christians;
attitudes toward marriage 32 defined 1, 15–17;
Rajputs 85 effectiveness of 15–17, 63, 184;
Rao, Narasimha 62 and group-based policies elsewhere 2,
Rashtriya Swayamsevak Sangh (RSS) 128– 11, 91, 175–6, 180
9; see also affirmative action;
and caste 132 in higher education see higher
Rattansi, Ali 11 education;
Reddi, Muthulakshmi 161 and identity see identity;
Regents of the University of California v. identification;
Bakke 7 and legislative seats see legislative
religion see Buddhists; reservations;
Christians; and litigation 23–4
Hinduism; see also litigation about categories;
Hindus; and locale 74–6, 150–1;
Muslims; political repercussions 3–4, 112–13,
Sikhs 124–6, 127–8, 140–1, 143–4, 156–7,
religious conversion/reconversion, 172–4, 177–80;
converts see conversion Supreme Court on 15
religious identity and birth 38 see also Supreme Court of India;
reservation policies/reservations 85; revisions of 143, 180–5
administration of 67, 153 restrictions of social fluidity see Supreme
see also administration and identity; Court decisions
advocates for 7, 175; retribalization:
and categories 2–3, 5–8, 13±15 and self identity 105
see also categories; rights:
creamy layer; and recognition 1, 12;
Other Backward Classes; and representation 174
Scheduled Castes; Risley, Herbert 17, 41, 45, 47, 49, 50
Scheduled Tribes; see also anthropology;
and census 18 People of India:
see also census; compared to Kumar Suresh Singh 42,
class/economic criteria see economic 50;
conditions/criteria; and criminal tribes 50;
colonial 10–15, 70–1, 114, 119, 143, government uses of work 47;
151, 159–64 167 hierarchical caste list 93;
see also colonial policy; and nationalism 57–8;
conflict 2, 125 People of India projects 41, 64;
264 INDEX
and politicization 47–8 and varna system 50

Rizvi, Aizaz 133 Scheduled Tribes (STs; adivasis) 2, 14;
Rudolph, Lloyd 111 and census see census;
Rudolph, Susanne 111 retribalization;
and certificates 71–4;
Said, Edward 111 defined/described 14;
Sajwan, Leela 148 litigation over membership 25–31;
samudaya 59 and People of India project 52
see also community see also criminal tribes;
Sangh Parivar reservations for 15, 16;
see also Hindu nationalists: and varna system 50;
blood family 129; verification 28–31, 72–3
and political savvy 138, 140 schedules:
sanskritization/desanskritization 25, 27–8 application/certification 81, 87;
Sarkar, Lotika 166 data gathering process 84–5
Sarkar, Shivaji 102 Schneider, Anne 68
Sarpotdar, Madhukar 168 Scott, James C. 6;
Sassler, Sharon 11 and official categories 24, 80, 87
Scheduled Castes and Scheduled Tribes 2nd Round Table conference 161
Joint Action Council: self identification
and minority religion inclusion 137 see also People of India
Scheduled Caste Order of 1950 14 (Anthropological Survey of India);
The Scheduled Castes (People of India identity:
project) 41, 53; category change 105;
revision 55 census 91;
Scheduled Caste (SC) Christians see corrections to 90, 98
Christians; Seminar 144
Dalit Christians Sen, Amartya 158
Scheduled Castes (SCs; Dalits; Harijans) 2 separate electorates 131, 160
see also Christians; 73rd Constitutional Amendment Act:
Christian Dalits: and panchayat reservations for women
defined 14; 167
and Buddhists 120; Shah, Ghanshyam, 96
and caste certificates 71–4 Shahabuddin, Syed, 115, 183;
see also caste certificates; and Christian Dalit movement 138;
and census 89, 90, 98–9; and Muslim quotas 118
and conversion/reconversion 79–81 Sharif, Salim 123
see also conversion; Sharma, P.S. 103
and Muslim reservations 117, 124, 137 Sharma, Rohit 151
see also Muslims; Sheth, D.L. 53, 83, 84, 85, 95–6;
number of names/labels 70; and economic criteria 153
and People of India project 41, 50–6, Sikhism:
59–60, 62–3, 65–6; and Dalit Solidarity Programme 123;
and political power 71; and definitions of Hinduism 133
and religions of 56, 120, 132–3; Sikhs:
reservations for 15, 16; and conversion/reconversion 79;
scheduled caste associations 98; and Scheduled Castes 120, 133
Simmel, Georg 127–8
INDEX 265
Simon Commission 119 and Franchise Committee 162;

Singh, Arjun 58 and reserved seats 161–2
Singh, Hare Ram 147 Supreme Court of India
Singh, Hosain 83 see also litigation about categories:
Singh, K.S.: and classifications 146;
anthropologist/administrator 50; and conversions 38;
and caste fluidity 60–1; and creamy layers/truly backward
ethnographer 58; (OBC) 142, 143, 148, 149;
and lower caste progress 60–1; defines class 145;
and nationalism 58–9; and government distinctions 155;
and OBC definitions 53; and group boundaries 39;
and People of India survey 49, 52, 53, and identity categories 24–5;
63, 64; and migrants 75;
unity of Indian peoples 59–60 and reconversion 37;
Singh, O.P. 147 and reservation policy 95, 154;
Singh, V.P. 7, 96–7; suppressed classes 70
and Hindu nationalist opposition 132; Supreme Court of India cases/decisions:
and Mandal report findings 144; Action Committee on Issue of Caste
and OBC reservations 53–4, 64, 95, Certificate to Scheduled Castes and
114, 147; Scheduled Tribes in the State of
and reservation policy 53, 95, 147, 184; Maharashtra and Another v. Union of
and women 173 India 75;
Singh, Yogendra 64 Indra Sawhney v. Union of India 95,
Singhal, Ashok 136 148, 149, 154;
Skocpol, Theda 68 Kumari Madhuri Patil v. Additional
social fluidity: Commissioner Tribal Development 25–
and the Supreme Court 25 8, 39;
social identities see identity S. Swvigaradoss v. Zonal Manager 35–
Socially and Economically Backward 8, 39, 80;
Classes (SEBCs) 17 Valsamma Paul v Cochin University
see also Other Backward Classes 31–4, 39, 76
Sonavane, Vilas 116 Swaraj, Sushma 136
special concessions 70 S. Swvigaradoss v. Zonal Manager 35–8,
spurious tribe 25–6 39, 80
Srinivas, M.N. 27, 28, 184 Swvigaradoss, S. 35
state identification 1;
see also identification: Tamil Nadu 103, 155
groups see also Tirunelveli District:
official and theoretical boundaries 4; Christians of 123;
and social identity 12 Muslim conversions in 135
state legislative assemblies: target populations 68, 73–4
reservations in 15, 16 Taylor, Charles 12
state simplifications 17, 67–8, 177–8 Tharoor, Shashi 133
see also identification; theories see identity
and political complications 18–20 Thorat, S.K. 179
see also identity Tirunelveli District in Tamil Nadu 37
The Statesman 54 Towards Equality 20, 163, 164, 165, 169;
Subbarayan, Radhabai 161;
266 INDEX
and debates over reservations 164–6, and activism 163–4;

173 and arranged marriages 32;
traits: associations, before independence 160;
correlations of 59–60 as disadvantaged 169–70, 174;
Travancore see also Kerala: identity 7–8, 172–3;
Public Service Commission 151 and inter-caste marriages 31–2, 39;
medical students 26;
UN and women 163 and overlapping and competing
United Nations Convention on the categories 159;
Elimination of Discrimination Against religious minorities and legislative
Women (CEDAW) 34, 163 seats, 156–7, 159, 160;
see also Dalits; reservations 16, 20;
untouchables 70 reserved seats and equality 160–2, 179;
Harijans; as a rightful group 157;
Scheduled Castes in Third World 163;
upper class Muslims 117; and UN 163;
and backward status 117–18 as wives 31
Uttar Pradesh 155 women’s associations/organizations:
and Muslim reservations 170;
Valsamma Paul v. Cochin University 31–4, and politics/lobbying 160–1
36 Women’s India Association (WIA) 160,
Van den Berghe, Pierre 111–12 161, 165–6
varna 11, 48, 50 The Women’s Reservation Bill and
see also caste: legislative seats 156–7
defined 13;
and Scheduled Castes Yadav, Sharad 170
and Scheduled Tribes 50 Young, M. Crawford 43
Varshney, Ashutosh 125
Vedas 13
VHP see Vishwas Hindu Parishad
Vijayanunni, M. 96, 104
Vishwas Hindu Parishad (VHP) 135;
see also Hindu nationalism:
and Hindu missionaries
and Mother Theresa 136
Walzer, Michael 9–10

Watson, J. Forbes 41
Weber, Max 5
“What Does SC Mean?” 133–4
White, Michael 11
Why I am not a Hindu 141
Wright, Theodore P. 114, 115
women
see also gender;
reserved legislative seats;
2nd Round Table conference:

Laura Dudley Jenkins - Identity and Identification in India - Defining The Disadvantaged (2002, Routledge)

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Laura Dudley Jenkins - Identity and Identification in India - Defining The Disadvantaged (2002, Routledge)

Загружено:

Авторское право:

Доступные форматы

Identity and Identification in India

LONDON AND NEW YORK

ISBN 0-203-40193-X Master e-book ISBN

ISBN 0-203-40837-3 (Adobe eReader Format)

1 Identity and identification 1

PART I State simplifications 21

PART II Political complications 106

In 1990, I was headed to Delhi when my train screeched to a halt. The

Where am I to be classified? Or, if you prefer, tucked away?

I had two surprises in store for me on 27 November 1996, as I stopped in central

identity politics. It is important to distinguish between categories, as externally

the imperatives of group-based mobilizations and policies, regardless of

reservations, cross-cutting groups are demanding changes to governmental

CATEGORIES AND IDENTITIES

The problem arises…when all “identities,” of whatever form, are treated as

Such a reduction downplays the unique social or political problems associated

Although feminists contend strongly among themselves as to whether the

You will have to compensate

for the cost of every thing…

religion, gender or caste. Electoral policies drawing on such categories include

COLONIAL AND POSTCOLONIAL STATE-SOCIETY

In addition to the importance of recognizing elite members of colonial and

activists to identity politics, as in the “new social movements” literature or

RESERVATION POLICIES IN INDIA

challenges of untangling identities in applications involving migration, intercaste

overlapping identities result in competing perspectives on reservation categories

women as a “category” as opposed to a “community” in their consideration of

They used to make pickles, squashes, jams, curry powders and

challenges of adjudicating and administering identity is applying these

The judges to be discussed here thus participate, perhaps unwittingly, in a

court’s decisions mandate strict state surveillance of identity claims. In addition,

A ªSPURIOUSº TRI BE?

“traditional moorings and customary beliefs and practices” as its defining

membership is involuntary… Endogamy is its main characteristic” (554).11 None

by birth. At the same time, it reinforced a contradictory ideology about a

find themselves in the difficult position of trying to assess converts’ motivations.

In a bookstore in Delhi, a salesperson, apprised of my interest in lower caste

politically useful focus for an ethnography sponsored by the central government

ANTHROPOLOGY, COLONIALISM AND THE STATE

In the process, colonial ethnographers themselves became one such political

ethnological congress. It was to be a sort of living exhibition, in which people,

RISLEY'S PEOPLE OF INDIA

censuses, particularly Risley’s, tried to record additional information about the

curiously close correspondence between the gradations of racial type

THE NEW PEOPLE OF INDIA PROJECT

I trust this series on the People of India, which is based on a

The current People of India project is a national ethnographic survey of “all

The attributes of physical diversity are discussed in this volume by pooling

In order to have diversity of features, 2234 normal adult persons were

government categories also organize it. B.K.Roy Burman, an anthropologist and

describes disadvantages of untouchable groups, including Christians, countering

Dr Vidhyarti, the famous anthropologist studying the religion of

Another publication likewise argues, on “anthropological” grounds, that the

Reply the Christian Dalit anthropologists “every word of it is wrong. First

Groups in society are objecting to their categorization by the government and,

science of backwardness,” which, it is claimed, does not do enough to compare

NATIONALISM AND THE PEOPLE OF INDIA

nationalist awakening, even as it worked to reify categories of social

In Arthashastra and Mahabharata, they mention communities which are

The word community or samudaya as it is called in some states like Kerala

In contrast, the “old” ethnographers “seemed to take a peculiar delight in

If you look at traditional ethnography, some people are called degraded. Or